Flask docs
2 likes2,102 views
This document is the Flask documentation, which provides information on how to use the Flask web framework. It covers topics like installation, basic usage, routing, templates, testing, debugging errors, configuration, signals, pluggable views, application and request contexts, modular applications with blueprints, extensions, and patterns for building larger Flask applications. The documentation contains tutorials, API references, and best practices for developing Flask applications.
![2. URL building will handle escaping of special characters and Unicode data trans-
parently for you, so you don’t have to deal with them.
3. If your application is placed outside the URL root (say, in /myapplication instead
of /), url_for() will handle that properly for you.
4.3.3 HTTP Methods
HTTP (the protocol web applications are speaking) knows different methods for ac-
cessing URLs. By default, a route only answers to GET requests, but that can be
changed by providing the methods argument to the route() decorator. Here are some
examples:
@app.route(’/login’, methods=[’GET’, ’POST’])
def login():
if request.method == ’POST’:
do_the_login()
else:
show_the_login_form()
If GET is present, HEAD will be added automatically for you. You don’t have to deal
with that. It will also make sure that HEAD requests are handled as the HTTP RFC (the
document describing the HTTP protocol) demands, so you can completely ignore that
part of the HTTP specification. Likewise, as of Flask 0.6, OPTIONS is implemented for
you automatically as well.
You have no idea what an HTTP method is? Worry not, here is a quick introduction to
HTTP methods and why they matter:
The HTTP method (also often called “the verb”) tells the server what the clients wants
to do with the requested page. The following methods are very common:
GET The browser tells the server to just get the information stored on that page and
send it. This is probably the most common method.
HEAD The browser tells the server to get the information, but it is only interested in
the headers, not the content of the page. An application is supposed to handle
that as if a GET request was received but to not deliver the actual content. In
Flask you don’t have to deal with that at all, the underlying Werkzeug library
handles that for you.
POST The browser tells the server that it wants to post some new information to that
URL and that the server must ensure the data is stored and only stored once.
This is how HTML forms usually transmit data to the server.
PUT Similar to POST but the server might trigger the store procedure multiple times
by overwriting the old values more than once. Now you might be asking why
this is useful, but there are some good reasons to do it this way. Consider that
the connection is lost during transmission: in this situation a system between the
browser and the server might receive the request safely a second time without
breaking things. With POST that would not be possible because it must only be
triggered once.
16](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-26-320.jpg)
![4.6.2 The Request Object
The request object is documented in the API section and we will not cover it here in
detail (see request). Here is a broad overview of some of the most common operations.
First of all you have to import it from the flask module:
from flask import request
The current request method is available by using the method attribute. To access form
data (data transmitted in a POST or PUT request) you can use the form attribute. Here
is a full example of the two attributes mentioned above:
@app.route(’/login’, methods=[’POST’, ’GET’])
def login():
error = None
if request.method == ’POST’:
if valid_login(request.form[’username’],
request.form[’password’]):
return log_the_user_in(request.form[’username’])
else:
error = ’Invalid username/password’
# the code below is executed if the request method
# was GET or the credentials were invalid
return render_template(’login.html’, error=error)
What happens if the key does not exist in the form attribute? In that case a special
KeyError is raised. You can catch it like a standard KeyError but if you don’t do that, a
HTTP 400 Bad Request error page is shown instead. So for many situations you don’t
have to deal with that problem.
To access parameters submitted in the URL (?key=value) you can use the args at-
tribute:
searchword = request.args.get(’key’, ’’)
We recommend accessing URL parameters with get or by catching the KeyError be-
cause users might change the URL and presenting them a 400 bad request page in that
case is not user friendly.
For a full list of methods and attributes of the request object, head over to the request
documentation.
4.6.3 File Uploads
You can handle uploaded files with Flask easily. Just make sure not to forget to
set the enctype="multipart/form-data" attribute on your HTML form, otherwise the
browser will not transmit your files at all.
Uploaded files are stored in memory or at a temporary location on the filesystem. You
can access those files by looking at the files attribute on the request object. Each
uploaded file is stored in that dictionary. It behaves just like a standard Python file
20](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-30-320.jpg)
![object, but it also has a save() method that allows you to store that file on the filesys-
tem of the server. Here is a simple example showing how that works:
from flask import request
@app.route(’/upload’, methods=[’GET’, ’POST’])
def upload_file():
if request.method == ’POST’:
f = request.files[’the_file’]
f.save(’/var/www/uploads/uploaded_file.txt’)
...
If you want to know how the file was named on the client before it was uploaded to
your application, you can access the filename attribute. However please keep in mind
that this value can be forged so never ever trust that value. If you want to use the file-
name of the client to store the file on the server, pass it through the secure_filename()
function that Werkzeug provides for you:
from flask import request
from werkzeug import secure_filename
@app.route(’/upload’, methods=[’GET’, ’POST’])
def upload_file():
if request.method == ’POST’:
f = request.files[’the_file’]
f.save(’/var/www/uploads/’ + secure_filename(f.filename))
...
For some better examples, checkout the Uploading Files pattern.
4.6.4 Cookies
To access cookies you can use the cookies attribute. To set cookies you can use the
set_cookie method of response objects. The cookies attribute of request objects is a
dictionary with all the cookies the client transmits. If you want to use sessions, do not
use the cookies directly but instead use the Sessions in Flask that add some security on
top of cookies for you.
Reading cookies:
from flask import request
@app.route(’/’)
def index():
username = request.cookies.get(’username’)
# use cookies.get(key) instead of cookies[key] to not get a
# KeyError if the cookie is missing.
Storing cookies:
21](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-31-320.jpg)
![4.8 About Responses
The return value from a view function is automatically converted into a response ob-
ject for you. If the return value is a string it’s converted into a response object with the
string as response body, a 200 OK status code and a text/html mimetype. The logic
that Flask applies to converting return values into response objects is as follows:
1. If a response object of the correct type is returned it’s directly returned from the
view.
2. If it’s a string, a response object is created with that data and the default param-
eters.
3. If a tuple is returned the items in the tuple can provide extra information.
Such tuples have to be in the form (response, status, headers) or (response,
headers) where at least one item has to be in the tuple. The status value will over-
ride the status code and headers can be a list or dictionary of additional header
values.
4. If none of that works, Flask will assume the return value is a valid WSGI appli-
cation and convert that into a response object.
If you want to get hold of the resulting response object inside the view you can use the
make_response() function.
Imagine you have a view like this:
@app.errorhandler(404)
def not_found(error):
return render_template(’error.html’), 404
You just need to wrap the return expression with make_response() and get the re-
sponse object to modify it, then return it:
@app.errorhandler(404)
def not_found(error):
resp = make_response(render_template(’error.html’), 404)
resp.headers[’X-Something’] = ’A value’
return resp
4.9 Sessions
In addition to the request object there is also a second object called session which
allows you to store information specific to a user from one request to the next. This is
implemented on top of cookies for you and signs the cookies cryptographically. What
this means is that the user could look at the contents of your cookie but not modify it,
unless they know the secret key used for signing.
In order to use sessions you have to set a secret key. Here is how sessions work:
23](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-33-320.jpg)
![from flask import Flask, session, redirect, url_for, escape, request
app = Flask(__name__)
@app.route(’/’)
def index():
if ’username’ in session:
return ’Logged in as %s’ % escape(session[’username’])
return ’You are not logged in’
@app.route(’/login’, methods=[’GET’, ’POST’])
def login():
if request.method == ’POST’:
session[’username’] = request.form[’username’]
return redirect(url_for(’index’))
return ’’’
<form action="" method="post">
<p><input type=text name=username>
<p><input type=submit value=Login>
</form>
’’’
@app.route(’/logout’)
def logout():
# remove the username from the session if it’s there
session.pop(’username’, None)
return redirect(url_for(’index’))
# set the secret key. keep this really secret:
app.secret_key = ’A0Zr98j/3yX R~XHH!jmN]LWX/,?RT’
The escape() mentioned here does escaping for you if you are not using the template
engine (as in this example).
How to generate good secret keys
The problem with random is that it’s hard to judge what is truly random. And a secret
key should be as random as possible. Your operating system has ways to generate
pretty random stuff based on a cryptographic random generator which can be used to
get such a key:
>>> import os
>>> os.urandom(24)
’xfd{Hxe5<x95xf9xe3x96.5xd1x01O<!xd5xa2xa0x9fR"xa1xa8’
Just take that thing and copy/paste it into your code and you’re done.
A note on cookie-based sessions: Flask will take the values you put into the session
object and serialize them into a cookie. If you are finding some values do not per-
sist across requests, cookies are indeed enabled, and you are not getting a clear error
message, check the size of the cookie in your page responses compared to the size
24](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-34-320.jpg)
![PASSWORD=’default’
))
app.config.from_envvar(’FLASKR_SETTINGS’, silent=True)
The Config object works similar to a dictionary so we can update it with new values.
Database Path
Operating systems know the concept of a current working directory for each process.
Unfortunately you cannot depend on this in web applications because you might have
more than one application in the same process.
For this reason the app.root_path attribute can be used to get the path to the applica-
tion. Together with the os.path module files can then easily be found. In this example
we place the database right next to it.
For a real-work application it’s recommended to use Instance Folders instead.
Usually, it is a good idea to load a separate, environment specific configuration file.
Flask allows you to import multiple configurations and it will use the setting defined
in the last import. This enables robust configuration setups. from_envvar() can help
achieve this.
app.config.from_envvar(‘FLASKR_SETTINGS’, silent=True)
Simply define the environment variable FLASKR_SETTINGS that points to a config file to
be loaded. The silent switch just tells Flask to not complain if no such environment
key is set.
In addition to that you can use the from_object() method on the config object and
provide it with an import name of a module. Flask will the initialize the variable
from that module. Note that in all cases only variable names that are uppercase are
considered.
The SECRET_KEY is needed to keep the client-side sessions secure. Choose that key
wisely and as hard to guess and complex as possible. The debug flag enables or dis-
ables the interactive debugger. Never leave debug mode activated in a production system,
because it will allow users to execute code on the server!
We will also add a method that allows for easily connecting to the specified database.
This can be used to open a connection on request and also from the interactive Python
shell or a script. This will come in handy later. We create a simple database connection
through SQLite and then tell it to use the sqlite3.Row object to represent rows. This
allows us to treat the rows as if they were dictionaries instead of tuples.
def connect_db():
"""Connects to the specific database."""
rv = sqlite3.connect(app.config[’DATABASE’])
rv.row_factory = sqlite3.Row
return rv
Finally we just add a line to the bottom of the file that fires up the server if we want to
run that file as a standalone application:
30](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-40-320.jpg)
![5.7 Step 5: The View Functions
Now that the database connections are working we can start writing the view func-
tions. We will need four of them:
5.7.1 Show Entries
This view shows all the entries stored in the database. It listens on the root of the
application and will select title and text from the database. The one with the highest
id (the newest entry) will be on top. The rows returned from the cursor look a bit like
tuples because we are using the sqlite3.Row row factory.
The view function will pass the entries as dicts to the show_entries.html template and
return the rendered one:
@app.route(’/’)
def show_entries():
db = get_db()
cur = db.execute(’select title, text from entries order by id desc’)
entries = cur.fetchall()
return render_template(’show_entries.html’, entries=entries)
5.7.2 Add New Entry
This view lets the user add new entries if they are logged in. This only responds
to POST requests, the actual form is shown on the show_entries page. If everything
worked out well we will flash() an information message to the next request and redi-
rect back to the show_entries page:
@app.route(’/add’, methods=[’POST’])
def add_entry():
if not session.get(’logged_in’):
abort(401)
db = get_db()
db.execute(’insert into entries (title, text) values (?, ?)’,
[request.form[’title’], request.form[’text’]])
db.commit()
flash(’New entry was successfully posted’)
return redirect(url_for(’show_entries’))
Note that we check that the user is logged in here (the logged_in key is present in the
session and True).
Security Note
Be sure to use question marks when building SQL statements, as done in the example
above. Otherwise, your app will be vulnerable to SQL injection when you use string
formatting to build SQL statements. See Using SQLite 3 with Flask for more.
34](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-44-320.jpg)
![5.7.3 Login and Logout
These functions are used to sign the user in and out. Login checks the username and
password against the ones from the configuration and sets the logged_in key in the ses-
sion. If the user logged in successfully, that key is set to True, and the user is redirected
back to the show_entries page. In addition, a message is flashed that informs the user
that he or she was logged in successfully. If an error occurred, the template is notified
about that, and the user is asked again:
@app.route(’/login’, methods=[’GET’, ’POST’])
def login():
error = None
if request.method == ’POST’:
if request.form[’username’] != app.config[’USERNAME’]:
error = ’Invalid username’
elif request.form[’password’] != app.config[’PASSWORD’]:
error = ’Invalid password’
else:
session[’logged_in’] = True
flash(’You were logged in’)
return redirect(url_for(’show_entries’))
return render_template(’login.html’, error=error)
The logout function, on the other hand, removes that key from the session again. We
use a neat trick here: if you use the pop() method of the dict and pass a second param-
eter to it (the default), the method will delete the key from the dictionary if present or
do nothing when that key is not in there. This is helpful because now we don’t have
to check if the user was logged in.
@app.route(’/logout’)
def logout():
session.pop(’logged_in’, None)
flash(’You were logged out’)
return redirect(url_for(’show_entries’))
Continue with Step 6: The Templates.
5.8 Step 6: The Templates
Now we should start working on the templates. If we request the URLs now we would
only get an exception that Flask cannot find the templates. The templates are using
Jinja2 syntax and have autoescaping enabled by default. This means that unless you
mark a value in the code with Markup or with the |safe filter in the template, Jinja2 will
ensure that special characters such as < or > are escaped with their XML equivalents.
We are also using template inheritance which makes it possible to reuse the layout of
the website in all pages.
Put the following templates into the templates folder:
35](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-45-320.jpg)
![6.4 Controlling Autoescaping
Autoescaping is the concept of automatically escaping special characters of you. Spe-
cial characters in the sense of HTML (or XML, and thus XHTML) are &, >, <, " as well
as ’. Because these characters carry specific meanings in documents on their own you
have to replace them by so called “entities” if you want to use them for text. Not doing
so would not only cause user frustration by the inability to use these characters in text,
but can also lead to security problems. (see Cross-Site Scripting (XSS))
Sometimes however you will need to disable autoescaping in templates. This can be
the case if you want to explicitly inject HTML into pages, for example if they come
from a system that generate secure HTML like a markdown to HTML converter.
There are three ways to accomplish that:
• In the Python code, wrap the HTML string in a Markup object before passing it to
the template. This is in general the recommended way.
• Inside the template, use the |safe filter to explicitly mark a string as safe HTML
({{ myvariable|safe }})
• Temporarily disable the autoescape system altogether.
To disable the autoescape system in templates, you can use the {% autoescape %}
block:
{% autoescape false %}
<p>autoescaping is disabled here
<p>{{ will_not_be_escaped }}
{% endautoescape %}
Whenever you do this, please be very cautious about the variables you are using in
this block.
6.5 Registering Filters
If you want to register your own filters in Jinja2 you have two ways to do that.
You can either put them by hand into the jinja_env of the application or use the
template_filter() decorator.
The two following examples work the same and both reverse an object:
@app.template_filter(’reverse’)
def reverse_filter(s):
return s[::-1]
def reverse_filter(s):
return s[::-1]
app.jinja_env.filters[’reverse’] = reverse_filter
41](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-51-320.jpg)
![CHAPTER
SEVEN
TESTING FLASK APPLICATIONS
Something that is untested is broken.
The origin of this quote is unknown and while it is not entirely correct, it is also not
far from the truth. Untested applications make it hard to improve existing code and
developers of untested applications tend to become pretty paranoid. If an application
has automated tests, you can safely make changes and instantly know if anything
breaks.
Flask provides a way to test your application by exposing the Werkzeug test Client
and handling the context locals for you. You can then use that with your favourite
testing solution. In this documentation we will use the unittest package that comes
pre-installed with Python.
7.1 The Application
First, we need an application to test; we will use the application from the Tutorial. If
you don’t have that application yet, get the sources from the examples.
7.2 The Testing Skeleton
In order to test the application, we add a second module (flaskr_tests.py) and create a
unittest skeleton there:
import os
import flaskr
import unittest
import tempfile
class FlaskrTestCase(unittest.TestCase):
def setUp(self):
self.db_fd, flaskr.app.config[’DATABASE’] = tempfile.mkstemp()
flaskr.app.config[’TESTING’] = True
self.app = flaskr.app.test_client()
43](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-53-320.jpg)
![flaskr.init_db()
def tearDown(self):
os.close(self.db_fd)
os.unlink(flaskr.app.config[’DATABASE’])
if __name__ == ’__main__’:
unittest.main()
The code in the setUp() method creates a new test client and initializes a new database.
This function is called before each individual test function is run. To delete the
database after the test, we close the file and remove it from the filesystem in the
tearDown() method. Additionally during setup the TESTING config flag is activated.
What it does is disabling the error catching during request handling so that you get
better error reports when performing test requests against the application.
This test client will give us a simple interface to the application. We can trigger test
requests to the application, and the client will also keep track of cookies for us.
Because SQLite3 is filesystem-based we can easily use the tempfile module to create a
temporary database and initialize it. The mkstemp() function does two things for us: it
returns a low-level file handle and a random file name, the latter we use as database
name. We just have to keep the db_fd around so that we can use the os.close() func-
tion to close the file.
If we now run the test suite, we should see the following output:
$ python flaskr_tests.py
----------------------------------------------------------------------
Ran 0 tests in 0.000s
OK
Even though it did not run any actual tests, we already know that our flaskr applica-
tion is syntactically valid, otherwise the import would have died with an exception.
7.3 The First Test
Now it’s time to start testing the functionality of the application. Let’s check that the
application shows “No entries here so far” if we access the root of the application (/).
To do this, we add a new test method to our class, like this:
class FlaskrTestCase(unittest.TestCase):
def setUp(self):
self.db_fd, flaskr.app.config[’DATABASE’] = tempfile.mkstemp()
self.app = flaskr.app.test_client()
flaskr.init_db()
44](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-54-320.jpg)
![def tearDown(self):
os.close(self.db_fd)
os.unlink(flaskr.app.config[’DATABASE’])
def test_empty_db(self):
rv = self.app.get(’/’)
assert ’No entries here so far’ in rv.data
Notice that our test functions begin with the word test; this allows unittest to auto-
matically identify the method as a test to run.
By using self.app.get we can send an HTTP GET request to the application with the
given path. The return value will be a response_class object. We can now use the
data attribute to inspect the return value (as string) from the application. In this case,
we ensure that ’No entries here so far’ is part of the output.
Run it again and you should see one passing test:
$ python flaskr_tests.py
.
----------------------------------------------------------------------
Ran 1 test in 0.034s
OK
7.4 Logging In and Out
The majority of the functionality of our application is only available for the adminis-
trative user, so we need a way to log our test client in and out of the application. To do
this, we fire some requests to the login and logout pages with the required form data
(username and password). And because the login and logout pages redirect, we tell
the client to follow_redirects.
Add the following two methods to your FlaskrTestCase class:
def login(self, username, password):
return self.app.post(’/login’, data=dict(
username=username,
password=password
), follow_redirects=True)
def logout(self):
return self.app.get(’/logout’, follow_redirects=True)
Now we can easily test that logging in and out works and that it fails with invalid
credentials. Add this new test to the class:
def test_login_logout(self):
rv = self.login(’admin’, ’default’)
assert ’You were logged in’ in rv.data
45](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-55-320.jpg)
![rv = self.logout()
assert ’You were logged out’ in rv.data
rv = self.login(’adminx’, ’default’)
assert ’Invalid username’ in rv.data
rv = self.login(’admin’, ’defaultx’)
assert ’Invalid password’ in rv.data
7.5 Test Adding Messages
We should also test that adding messages works. Add a new test method like this:
def test_messages(self):
self.login(’admin’, ’default’)
rv = self.app.post(’/add’, data=dict(
title=’<Hello>’,
text=’<strong>HTML</strong> allowed here’
), follow_redirects=True)
assert ’No entries here so far’ not in rv.data
assert ’<Hello>’ in rv.data
assert ’<strong>HTML</strong> allowed here’ in rv.data
Here we check that HTML is allowed in the text but not in the title, which is the in-
tended behavior.
Running that should now give us three passing tests:
$ python flaskr_tests.py
...
----------------------------------------------------------------------
Ran 3 tests in 0.332s
OK
For more complex tests with headers and status codes, check out the MiniTwit Exam-
ple from the sources which contains a larger test suite.
7.6 Other Testing Tricks
Besides using the test client as shown above, there is also the test_request_context()
method that can be used in combination with the with statement to activate a request
context temporarily. With this you can access the request, g and session objects like
in view functions. Here is a full example that demonstrates this approach:
app = flask.Flask(__name__)
with app.test_request_context(’/?name=Peter’):
assert flask.request.path == ’/’
assert flask.request.args[’name’] == ’Peter’
46](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-56-320.jpg)
![For a test it would be nice to override this user from the outside without hav-
ing to change some code. This can trivially be accomplished with hooking the
flask.appcontext_pushed signal:
from contextlib import contextmanager
from flask import appcontext_pushed
@contextmanager
def user_set(app, user):
def handler(sender, **kwargs):
g.user = user
with appcontext_pushed.connected_to(handler, app):
yield
And then to use it:
from flask import json, jsonify
@app.route(’/users/me’)
def users_me():
return jsonify(username=g.user.username)
with user_set(app, my_user):
with app.test_client() as c:
resp = c.get(’/users/me’)
data = json.loads(resp.data)
self.assert_equal(data[’username’], my_user.username)
7.8 Keeping the Context Around
New in version 0.4.
Sometimes it is helpful to trigger a regular request but still keep the context around
for a little longer so that additional introspection can happen. With Flask 0.4 this is
possible by using the test_client() with a with block:
app = flask.Flask(__name__)
with app.test_client() as c:
rv = c.get(’/?tequila=42’)
assert request.args[’tequila’] == ’42’
If you were to use just the test_client() without the with block, the assert would fail
with an error because request is no longer available (because you are trying to use it
outside of the actual request).
48](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-58-320.jpg)
![7.9 Accessing and Modifying Sessions
New in version 0.8.
Sometimes it can be very helpful to access or modify the sessions from the test client.
Generally there are two ways for this. If you just want to ensure that a session has
certain keys set to certain values you can just keep the context around and access
flask.session:
with app.test_client() as c:
rv = c.get(’/’)
assert flask.session[’foo’] == 42
This however does not make it possible to also modify the session or to access the ses-
sion before a request was fired. Starting with Flask 0.8 we provide a so called “session
transaction” which simulates the appropriate calls to open a session in the context of
the test client and to modify it. At the end of the transaction the session is stored. This
works independently of the session backend used:
with app.test_client() as c:
with c.session_transaction() as sess:
sess[’a_key’] = ’a value’
# once this is reached the session was stored
Note that in this case you have to use the sess object instead of the flask.session
proxy. The object however itself will provide the same interface.
49](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-59-320.jpg)
![What you want instead is a mail the second the exception happened. Then you get an
alert and you can do something about it.
Flask uses the Python builtin logging system, and it can actually send you mails for
errors which is probably what you want. Here is how you can configure the Flask
logger to send you mails for exceptions:
ADMINS = [’yourname@example.com’]
if not app.debug:
import logging
from logging.handlers import SMTPHandler
mail_handler = SMTPHandler(’127.0.0.1’,
’server-error@example.com’,
ADMINS, ’YourApplication Failed’)
mail_handler.setLevel(logging.ERROR)
app.logger.addHandler(mail_handler)
So what just happened? We created a new SMTPHandler that will send mails with
the mail server listening on 127.0.0.1 to all the ADMINS from the address server-
error@example.com with the subject “YourApplication Failed”. If your mail server re-
quires credentials, these can also be provided. For that check out the documentation
for the SMTPHandler.
We also tell the handler to only send errors and more critical messages. Because we
certainly don’t want to get a mail for warnings or other useless logs that might happen
during request handling.
Before you run that in production, please also look at Controlling the Log Format to put
more information into that error mail. That will save you from a lot of frustration.
8.2 Logging to a File
Even if you get mails, you probably also want to log warnings. It’s a good idea to keep
as much information around that might be required to debug a problem. Please note
that Flask itself will not issue any warnings in the core system, so it’s your responsi-
bility to warn in the code if something seems odd.
There are a couple of handlers provided by the logging system out of the box but not
all of them are useful for basic error logging. The most interesting are probably the
following:
• FileHandler - logs messages to a file on the filesystem.
• RotatingFileHandler - logs messages to a file on the filesystem and will rotate
after a certain number of messages.
• NTEventLogHandler - will log to the system event log of a Windows system. If
you are deploying on a Windows box, this is what you want to use.
• SysLogHandler - sends logs to a UNIX syslog.
52](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-62-320.jpg)
![Once you picked your log handler, do like you did with the SMTP handler above, just
make sure to use a lower setting (I would recommend WARNING):
if not app.debug:
import logging
from themodule import TheHandlerYouWant
file_handler = TheHandlerYouWant(...)
file_handler.setLevel(logging.WARNING)
app.logger.addHandler(file_handler)
8.3 Controlling the Log Format
By default a handler will only write the message string into a file or send you that
message as mail. A log record stores more information, and it makes a lot of sense to
configure your logger to also contain that information so that you have a better idea
of why that error happened, and more importantly, where it did.
A formatter can be instantiated with a format string. Note that tracebacks are ap-
pended to the log entry automatically. You don’t have to do that in the log formatter
format string.
Here some example setups:
8.3.1 Email
from logging import Formatter
mail_handler.setFormatter(Formatter(’’’
Message type: %(levelname)s
Location: %(pathname)s:%(lineno)d
Module: %(module)s
Function: %(funcName)s
Time: %(asctime)s
Message:
%(message)s
’’’))
8.3.2 File logging
from logging import Formatter
file_handler.setFormatter(Formatter(
’%(asctime)s %(levelname)s: %(message)s ’
’[in %(pathname)s:%(lineno)d]’
))
53](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-63-320.jpg)
![getLogger(’otherlibrary’)]
for logger in loggers:
logger.addHandler(mail_handler)
logger.addHandler(file_handler)
55](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-65-320.jpg)
![CHAPTER
TEN
CONFIGURATION HANDLING
New in version 0.3.
Applications need some kind of configuration. There are different settings you might
want to change depending on the application environment like toggling the debug
mode, setting the secret key, and other such environment-specific things.
The way Flask is designed usually requires the configuration to be available when the
application starts up. You can hardcode the configuration in the code, which for many
small applications is not actually that bad, but there are better ways.
Independent of how you load your config, there is a config object available which
holds the loaded configuration values: The config attribute of the Flask object. This
is the place where Flask itself puts certain configuration values and also where exten-
sions can put their configuration values. But this is also where you can have your own
configuration.
10.1 Configuration Basics
The config is actually a subclass of a dictionary and can be modified just like any
dictionary:
app = Flask(__name__)
app.config[’DEBUG’] = True
Certain configuration values are also forwarded to the Flask object so you can read
and write them from there:
app.debug = True
To update multiple keys at once you can use the dict.update() method:
app.config.update(
DEBUG=True,
SECRET_KEY=’...’
)
59](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-69-320.jpg)
![Here for example a helper context manager that can be used to figure out in a unittest
which templates were rendered and what variables were passed to the template:
from flask import template_rendered
from contextlib import contextmanager
@contextmanager
def captured_templates(app):
recorded = []
def record(sender, template, context, **extra):
recorded.append((template, context))
template_rendered.connect(record, app)
try:
yield recorded
finally:
template_rendered.disconnect(record, app)
This can now easily be paired with a test client:
with captured_templates(app) as templates:
rv = app.test_client().get(’/’)
assert rv.status_code == 200
assert len(templates) == 1
template, context = templates[0]
assert template.name == ’index.html’
assert len(context[’items’]) == 10
Make sure to subscribe with an extra **extra argument so that your calls don’t fail if
Flask introduces new arguments to the signals.
All the template rendering in the code issued by the application app in the body of
the with block will now be recorded in the templates variable. Whenever a template is
rendered, the template object as well as context are appended to it.
Additionally there is a convenient helper method (connected_to()). that allows you
to temporarily subscribe a function to a signal with a context manager on its own.
Because the return value of the context manager cannot be specified that way one has
to pass the list in as argument:
from flask import template_rendered
def captured_templates(app, recorded, **extra):
def record(sender, template, context):
recorded.append((template, context))
return template_rendered.connected_to(record, app)
The example above would then look like this:
templates = []
with captured_templates(app, templates, **extra):
...
template, context = templates[0]
68](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-78-320.jpg)
![def close_db_connection(sender, **extra):
session.close()
from flask import appcontext_tearing_down
appcontext_tearing_down.connect(close_db_connection, app)
This will also be passed an exc keyword argument that has a reference to the
exception that caused the teardown if there was one.
flask.appcontext_pushed
This signal is sent when an application context is pushed. The sender is the
application. This is usually useful for unittests in order to temporarily hook in
information. For instance it can be used to set a resource early onto the g object.
Example usage:
from contextlib import contextmanager
from flask import appcontext_pushed
@contextmanager
def user_set(app, user):
def handler(sender, **kwargs):
g.user = user
with appcontext_pushed.connected_to(handler, app):
yield
And in the testcode:
def test_user_me(self):
with user_set(app, ’john’):
c = app.test_client()
resp = c.get(’/users/me’)
assert resp.data == ’username=john’
New in version 0.10.
flask.appcontext_popped
This signal is sent when an application context is popped. The sender is the
application. This usually falls in line with the appcontext_tearing_down signal.
New in version 0.10.
flask.message_flashed
This signal is sent when the application is flashing a message. The messages is
sent as message keyword argument and the category as category.
Example subscriber:
recorded = []
def record(sender, message, category, **extra):
recorded.append((message, category))
from flask import message_flashed
message_flashed.connect(record, app)
72](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-82-320.jpg)
![class MyView(View):
methods = [’GET’, ’POST’]
def dispatch_request(self):
if request.method == ’POST’:
...
...
app.add_url_rule(’/myview’, view_func=MyView.as_view(’myview’))
12.3 Method Based Dispatching
For RESTful APIs it’s especially helpful to execute a different function for each HTTP
method. With the flask.views.MethodView you can easily do that. Each HTTP method
maps to a function with the same name (just in lowercase):
from flask.views import MethodView
class UserAPI(MethodView):
def get(self):
users = User.query.all()
...
def post(self):
user = User.from_form_data(request.form)
...
app.add_url_rule(’/users/’, view_func=UserAPI.as_view(’users’))
That way you also don’t have to provide the methods attribute. It’s automatically set
based on the methods defined in the class.
12.4 Decorating Views
Since the view class itself is not the view function that is added to the routing system
it does not make much sense to decorate the class itself. Instead you either have to
decorate the return value of as_view() by hand:
def user_required(f):
"""Checks whether user is logged in or raises error 401."""
def decorator(*args, **kwargs):
if not g.user:
abort(401)
return f(*args, **kwargs)
return decorator
77](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-87-320.jpg)
![view = user_required(UserAPI.as_view(’users’))
app.add_url_rule(’/users/’, view_func=view)
Starting with Flask 0.8 there is also an alternative way where you can specify a list of
decorators to apply in the class declaration:
class UserAPI(MethodView):
decorators = [user_required]
Due to the implicit self from the caller’s perspective you cannot use regular view dec-
orators on the individual methods of the view however, keep this in mind.
12.5 Method Views for APIs
Web APIs are often working very closely with HTTP verbs so it makes a lot of sense
to implement such an API based on the MethodView. That said, you will notice that the
API will require different URL rules that go to the same method view most of the time.
For instance consider that you are exposing a user object on the web:
URL Method Description
/users/ GET Gives a list of all users
/users/ POST Creates a new user
/users/<id> GET Shows a single user
/users/<id> PUT Updates a single user
/users/<id> DELETE Deletes a single user
So how would you go about doing that with the MethodView? The trick is to take
advantage of the fact that you can provide multiple rules to the same view.
Let’s assume for the moment the view would look like this:
class UserAPI(MethodView):
def get(self, user_id):
if user_id is None:
# return a list of users
pass
else:
# expose a single user
pass
def post(self):
# create a new user
pass
def delete(self, user_id):
# delete a single user
pass
def put(self, user_id):
78](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-88-320.jpg)
![# update a single user
pass
So how do we hook this up with the routing system? By adding two rules and explic-
itly mentioning the methods for each:
user_view = UserAPI.as_view(’user_api’)
app.add_url_rule(’/users/’, defaults={’user_id’: None},
view_func=user_view, methods=[’GET’,])
app.add_url_rule(’/users/’, view_func=user_view, methods=[’POST’,])
app.add_url_rule(’/users/<int:user_id>’, view_func=user_view,
methods=[’GET’, ’PUT’, ’DELETE’])
If you have a lot of APIs that look similar you can refactor that registration code:
def register_api(view, endpoint, url, pk=’id’, pk_type=’int’):
view_func = view.as_view(endpoint)
app.add_url_rule(url, defaults={pk: None},
view_func=view_func, methods=[’GET’,])
app.add_url_rule(url, view_func=view_func, methods=[’POST’,])
app.add_url_rule(’%s<%s:%s>’ % (url, pk_type, pk), view_func=view_func,
methods=[’GET’, ’PUT’, ’DELETE’])
register_api(UserAPI, ’user_api’, ’/users/’, pk=’user_id’)
79](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-89-320.jpg)
![If you check the rules registered on the application, you will find these:
[<Rule ’/static/<filename>’ (HEAD, OPTIONS, GET) -> static>,
<Rule ’/<page>’ (HEAD, OPTIONS, GET) -> simple_page.show>,
<Rule ’/’ (HEAD, OPTIONS, GET) -> simple_page.show>]
The first one is obviously from the application ifself for the static files. The other two
are for the show function of the simple_page blueprint. As you can see, they are also
prefixed with the name of the blueprint and separated by a dot (.).
Blueprints however can also be mounted at different locations:
app.register_blueprint(simple_page, url_prefix=’/pages’)
And sure enough, these are the generated rules:
[<Rule ’/static/<filename>’ (HEAD, OPTIONS, GET) -> static>,
<Rule ’/pages/<page>’ (HEAD, OPTIONS, GET) -> simple_page.show>,
<Rule ’/pages/’ (HEAD, OPTIONS, GET) -> simple_page.show>]
On top of that you can register blueprints multiple times though not every blueprint
might respond properly to that. In fact it depends on how the blueprint is imple-
mented if it can be mounted more than once.
15.5 Blueprint Resources
Blueprints can provide resources as well. Sometimes you might want to introduce a
blueprint only for the resources it provides.
15.5.1 Blueprint Resource Folder
Like for regular applications, blueprints are considered to be contained in a folder.
While multiple blueprints can originate from the same folder, it does not have to be
the case and it’s usually not recommended.
The folder is inferred from the second argument to Blueprint which is usually
__name__. This argument specifies what logical Python module or package corre-
sponds to the blueprint. If it points to an actual Python package that package (which is
a folder on the filesystem) is the resource folder. If it’s a module, the package the mod-
ule is contained in will be the resource folder. You can access the Blueprint.root_path
property to see what the resource folder is:
>>> simple_page.root_path
’/Users/username/TestProject/yourapplication’
To quickly open sources from this folder you can use the open_resource() function:
with simple_page.open_resource(’static/style.css’) as f:
code = f.read()
93](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-103-320.jpg)
![17.2 Firing Before/After Request
By just creating a request context, you still don’t have run the code that is normally
run before a request. This might result in your database being unavailable if you are
connecting to the database in a before-request callback or the current user not being
stored on the g object etc.
This however can easily be done yourself. Just call preprocess_request():
>>> ctx = app.test_request_context()
>>> ctx.push()
>>> app.preprocess_request()
Keep in mind that the preprocess_request() function might return a response object,
in that case just ignore it.
To shutdown a request, you need to trick a bit before the after request functions (trig-
gered by process_response()) operate on a response object:
>>> app.process_response(app.response_class())
<Response 0 bytes [200 OK]>
>>> ctx.pop()
The functions registered as teardown_request() are automatically called when the
context is popped. So this is the perfect place to automatically tear down resources
that were needed by the request context (such as database connections).
17.3 Further Improving the Shell Experience
If you like the idea of experimenting in a shell, create yourself a module with stuff you
want to star import into your interactive session. There you could also define some
more helper methods for common things such as initializing the database, dropping
tables etc.
Just put them into a module (like shelltools and import from there):
>>> from shelltools import *
100](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-110-320.jpg)
![def create_app(config_filename):
app = Flask(__name__)
app.config.from_pyfile(config_filename)
from yourapplication.model import db
db.init_app(app)
from yourapplication.views.admin import admin
from yourapplication.views.frontend import frontend
app.register_blueprint(admin)
app.register_blueprint(frontend)
return app
The downside is that you cannot use the application object in the blueprints at import
time. You can however use it from within a request. How do you get access to the
application with the config? Use current_app:
from flask import current_app, Blueprint, render_template
admin = Blueprint(’admin’, __name__, url_prefix=’/admin’)
@admin.route(’/’)
def index():
return render_template(current_app.config[’INDEX_TEMPLATE’])
Here we look up the name of a template in the config.
Extension objects are not initially bound to an application. Using db.init_app, the
app gets configured for the extension. No application-specific state is stored on the
extension object, so one extension object can be used for multiple apps. For more
information about the design of extensions refer to Flask Extension Development.
Your model.py might look like this when using Flask-SQLAlchemy:
from flask.ext.sqlalchemy import SQLAlchemy
# no app object passed! Instead we use use db.init_app in the factory.
db = SQLAlchemy()
# create some models
18.2.2 Using Applications
So to use such an application you then have to create the application first. Here an
example run.py file that runs such an application:
from yourapplication import create_app
app = create_app(’/path/to/config.cfg’)
app.run()
104](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-114-320.jpg)
![class SubdomainDispatcher(object):
def __init__(self, domain, create_app):
self.domain = domain
self.create_app = create_app
self.lock = Lock()
self.instances = {}
def get_application(self, host):
host = host.split(’:’)[0]
assert host.endswith(self.domain), ’Configuration error’
subdomain = host[:-len(self.domain)].rstrip(’.’)
with self.lock:
app = self.instances.get(subdomain)
if app is None:
app = self.create_app(subdomain)
self.instances[subdomain] = app
return app
def __call__(self, environ, start_response):
app = self.get_application(environ[’HTTP_HOST’])
return app(environ, start_response)
This dispatcher can then be used like this:
from myapplication import create_app, get_user_for_subdomain
from werkzeug.exceptions import NotFound
def make_app(subdomain):
user = get_user_for_subdomain(subdomain)
if user is None:
# if there is no user for that subdomain we still have
# to return a WSGI application that handles that request.
# We can then just return the NotFound() exception as
# application which will render a default 404 page.
# You might also redirect the user to the main page then
return NotFound()
# otherwise create the application for the specific user
return create_app(user)
application = SubdomainDispatcher(’example.com’, make_app)
18.3.4 Dispatch by Path
Dispatching by a path on the URL is very similar. Instead of looking at the Host header
to figure out the subdomain one simply looks at the request path up to the first slash:
107](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-117-320.jpg)
![from threading import Lock
from werkzeug.wsgi import pop_path_info, peek_path_info
class PathDispatcher(object):
def __init__(self, default_app, create_app):
self.default_app = default_app
self.create_app = create_app
self.lock = Lock()
self.instances = {}
def get_application(self, prefix):
with self.lock:
app = self.instances.get(prefix)
if app is None:
app = self.create_app(prefix)
if app is not None:
self.instances[prefix] = app
return app
def __call__(self, environ, start_response):
app = self.get_application(peek_path_info(environ))
if app is not None:
pop_path_info(environ)
else:
app = self.default_app
return app(environ, start_response)
The big difference between this and the subdomain one is that this one falls back to
another application if the creator function returns None:
from myapplication import create_app, default_app, get_user_for_prefix
def make_app(prefix):
user = get_user_for_prefix(prefix)
if user is not None:
return create_app(user)
application = PathDispatcher(default_app, make_app)
18.4 Implementing API Exceptions
It’s very common to implement RESTful APIs on top of Flask. One of the first thing
that developers run into is the realization that the builtin exceptions are not expressive
enough for APIs and that the content type of text/html they are emitting is not very
useful for API consumers.
The better solution than using abort to signal errors for invalid API usage is to im-
plement your own exception type and install an error handler for it that produces the
108](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-118-320.jpg)
![errors in the format the user is expecting.
18.4.1 Simple Exception Class
The basic idea is to introduce a new exception that can take a proper human readable
message, a status code for the error and some optional payload to give more context
for the error.
This is a simple example:
from flask import jsonify
class InvalidUsage(Exception):
status_code = 400
def __init__(self, message, status_code=None, payload=None):
Exception.__init__(self)
self.message = message
if status_code is not None:
self.status_code = status_code
self.payload = payload
def to_dict(self):
rv = dict(self.payload or ())
rv[’message’] = self.message
return rv
A view can now raise that exception with an error message. Additionally some extra
payload can be provided as a dictionary through the payload parameter.
18.4.2 Registering an Error Handler
At that point views can raise that error, but it would immediately result in an internal
server error. The reason for this is that there is no handler registered for this error
class. That however is easy to add:
@app.errorhandler(InvalidUsage)
def handle_invalid_usage(error):
response = jsonify(error.to_dict())
response.status_code = error.status_code
return response
18.4.3 Usage in Views
Here is how a view can use that functionality:
109](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-119-320.jpg)
![if app.url_map.is_endpoint_expecting(endpoint, ’lang_code’):
values[’lang_code’] = g.lang_code
The method is_endpoint_expecting() of the URL map can be used to figure out if it
would make sense to provide a language code for the given endpoint.
The reverse of that function are url_value_preprocessor()s. They are executed right
after the request was matched and can execute code based on the URL values. The
idea is that they pull information out of the values dictionary and put it somewhere
else:
@app.url_value_preprocessor
def pull_lang_code(endpoint, values):
g.lang_code = values.pop(’lang_code’, None)
That way you no longer have to do the lang_code assignment to g in every function.
You can further improve that by writing your own decorator that prefixes URLs with
the language code, but the more beautiful solution is using a blueprint. Once the
’lang_code’ is popped from the values dictionary and it will no longer be forwarded
to the view function reducing the code to this:
from flask import Flask, g
app = Flask(__name__)
@app.url_defaults
def add_language_code(endpoint, values):
if ’lang_code’ in values or not g.lang_code:
return
if app.url_map.is_endpoint_expecting(endpoint, ’lang_code’):
values[’lang_code’] = g.lang_code
@app.url_value_preprocessor
def pull_lang_code(endpoint, values):
g.lang_code = values.pop(’lang_code’, None)
@app.route(’/<lang_code>/’)
def index():
...
@app.route(’/<lang_code>/about’)
def about():
...
18.5.2 Internationalized Blueprint URLs
Because blueprints can automatically prefix all URLs with a common string it’s easy
to automatically do that for every function. Furthermore blueprints can have per-
blueprint URL processors which removes a whole lot of logic from the url_defaults()
111](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-121-320.jpg)
![your application into a package, head over to the Larger Applications pattern to see how
this can be done.
A working deployment with distribute is the first step into more complex and more
automated deployment scenarios. If you want to fully automate the process, also read
the Deploying with Fabric chapter.
18.6.1 Basic Setup Script
Because you have Flask running, you either have setuptools or distribute available on
your system anyways. If you do not, fear not, there is a script to install it for you:
distribute_setup.py. Just download and run with your Python interpreter.
Standard disclaimer applies: you better use a virtualenv.
Your setup code always goes into a file named setup.py next to your application. The
name of the file is only convention, but because everybody will look for a file with that
name, you better not change it.
Yes, even if you are using distribute, you are importing from a package called setuptools.
distribute is fully backwards compatible with setuptools, so it also uses the same import
name.
A basic setup.py file for a Flask application looks like this:
from setuptools import setup
setup(
name=’Your Application’,
version=’1.0’,
long_description=__doc__,
packages=[’yourapplication’],
include_package_data=True,
zip_safe=False,
install_requires=[’Flask’]
)
Please keep in mind that you have to list subpackages explicitly. If you want distribute
to lookup the packages for you automatically, you can use the find_packages function:
from setuptools import setup, find_packages
setup(
...
packages=find_packages()
)
Most parameters to the setup function should be self explanatory, include_package_data
and zip_safe might not be. include_package_data tells distribute to look for a MANI-
FEST.in file and install all the entries that match as package data. We will use this to
distribute the static files and templates along with the Python module (see Distribut-
ing Resources). The zip_safe flag can be used to force or prevent zip Archive creation.
113](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-123-320.jpg)
![In general you probably don’t want your packages to be installed as zip files because
some tools do not support them and they make debugging a lot harder.
18.6.2 Distributing Resources
If you try to install the package you just created, you will notice that folders like static
or templates are not installed for you. The reason for this is that distribute does not
know which files to add for you. What you should do, is to create a MANIFEST.in file
next to your setup.py file. This file lists all the files that should be added to your tarball:
recursive-include yourapplication/templates *
recursive-include yourapplication/static *
Don’t forget that even if you enlist them in your MANIFEST.in file, they won’t be
installed for you unless you set the include_package_data parameter of the setup function
to True!
18.6.3 Declaring Dependencies
Dependencies are declared in the install_requires parameter as list. Each item in that list
is the name of a package that should be pulled from PyPI on installation. By default
it will always use the most recent version, but you can also provide minimum and
maximum version requirements. Here some examples:
install_requires=[
’Flask>=0.2’,
’SQLAlchemy>=0.6’,
’BrokenPackage>=0.7,<=1.0’
]
I mentioned earlier that dependencies are pulled from PyPI. What if you want to de-
pend on a package that cannot be found on PyPI and won’t be because it is an internal
package you don’t want to share with anyone? Just still do as if there was a PyPI en-
try for it and provide a list of alternative locations where distribute should look for
tarballs:
dependency_links=[’http://example.com/yourfiles’]
Make sure that page has a directory listing and the links on the page are pointing to
the actual tarballs with their correct filenames as this is how distribute will find the
files. If you have an internal company server that contains the packages, provide the
URL to that server there.
18.6.4 Installing / Developing
To install your application (ideally into a virtualenv) just run the setup.py script with
the install parameter. It will install your application into the virtualenv’s site-packages
folder and also download and install all dependencies:
114](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-124-320.jpg)
![$ python setup.py install
If you are developing on the package and also want the requirements to be installed,
you can use the develop command instead:
$ python setup.py develop
This has the advantage of just installing a link to the site-packages folder instead of
copying the data over. You can then continue to work on the code without having to
run install again after each change.
18.7 Deploying with Fabric
Fabric is a tool for Python similar to Makefiles but with the ability to execute com-
mands on a remote server. In combination with a properly set up Python package
(Larger Applications) and a good concept for configurations (Configuration Handling) it
is very easy to deploy Flask applications to external servers.
Before we get started, here a quick checklist of things we have to ensure upfront:
• Fabric 1.0 has to be installed locally. This tutorial assumes the latest version of
Fabric.
• The application already has to be a package and requires a working setup.py file
(Deploying with Distribute).
• In the following example we are using mod_wsgi for the remote servers. You
can of course use your own favourite server there, but for this example we chose
Apache + mod_wsgi because it’s very easy to setup and has a simple way to reload
applications without root access.
18.7.1 Creating the first Fabfile
A fabfile is what controls what Fabric executes. It is named fabfile.py and executed by
the fab command. All the functions defined in that file will show up as fab subcom-
mands. They are executed on one or more hosts. These hosts can be defined either in
the fabfile or on the command line. In this case we will add them to the fabfile.
This is a basic first example that has the ability to upload the current sourcecode to the
server and install it into a pre-existing virtual environment:
from fabric.api import *
# the user to use for the remote commands
env.user = ’appuser’
# the servers where the commands are executed
env.hosts = [’server1.example.com’, ’server2.example.com’]
def pack():
115](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-125-320.jpg)
![$ fab -H newserver.example.com bootstrap
To setup a new server you would roughly do these steps:
1. Create the directory structure in /var/www:
$ mkdir /var/www/yourapplication
$ cd /var/www/yourapplication
$ virtualenv --distribute env
2. Upload a new application.wsgi file to the server and the configuration file for the
application (eg: application.cfg)
3. Create a new Apache config for yourapplication and activate it. Make sure to
activate watching for changes of the .wsgi file so that we can automatically reload
the application by touching it. (See mod_wsgi (Apache) for more information)
So now the question is, where do the application.wsgi and application.cfg files come
from?
18.7.3 The WSGI File
The WSGI file has to import the application and also to set an environment variable so
that the application knows where to look for the config. This is a short example that
does exactly that:
import os
os.environ[’YOURAPPLICATION_CONFIG’] = ’/var/www/yourapplication/application.cfg’
from yourapplication import app
The application itself then has to initialize itself like this to look for the config at that
environment variable:
app = Flask(__name__)
app.config.from_object(’yourapplication.default_config’)
app.config.from_envvar(’YOURAPPLICATION_CONFIG’)
This approach is explained in detail in the Configuration Handling section of the docu-
mentation.
18.7.4 The Configuration File
Now as mentioned above, the application will find the correct configuration file by
looking up the YOURAPPLICATION_CONFIG environment variable. So we have to
put the configuration in a place where the application will able to find it. Configura-
tion files have the unfriendly quality of being different on all computers, so you do not
version them usually.
A popular approach is to store configuration files for different servers in a sep-
arate version control repository and check them out on all servers. Then sym-
117](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-127-320.jpg)
![18.8.2 Easy Querying
Now in each request handling function you can access g.db to get the current open
database connection. To simplify working with SQLite, a row factory function is use-
ful. It is executed for every result returned from the database to convert the result.
For instance in order to get dictionaries instead of tuples this could be inserted into
get_db:
def make_dicts(cursor, row):
return dict((cursor.description[idx][0], value)
for idx, value in enumerate(row))
db.row_factory = make_dicts
Or even simpler:
db.row_factory = sqlite3.Row
Additionally it is a good idea to provide a query function that combines getting the
cursor, executing and fetching the results:
def query_db(query, args=(), one=False):
cur = get_db().execute(query, args)
rv = cur.fetchall()
cur.close()
return (rv[0] if rv else None) if one else rv
This handy little function in combination with a row factory makes working with the
database much more pleasant than it is by just using the raw cursor and connection
objects.
Here is how you can use it:
for user in query_db(’select * from users’):
print user[’username’], ’has the id’, user[’user_id’]
Or if you just want a single result:
user = query_db(’select * from users where username = ?’,
[the_username], one=True)
if user is None:
print ’No such user’
else:
print the_username, ’has the id’, user[’user_id’]
To pass variable parts to the SQL statement, use a question mark in the statement and
pass in the arguments as a list. Never directly add them to the SQL statement with
string formatting because this makes it possible to attack the application using SQL
Injections.
120](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-130-320.jpg)
![>>> from yourapplication.database import init_db
>>> init_db()
You can insert entries into the database like this:
>>> from yourapplication.database import db_session
>>> from yourapplication.models import User
>>> u = User(’admin’, ’admin@localhost’)
>>> db_session.add(u)
>>> db_session.commit()
Querying is simple as well:
>>> User.query.all()
[<User u’admin’>]
>>> User.query.filter(User.name == ’admin’).first()
<User u’admin’>
18.9.3 Manual Object Relational Mapping
Manual object relational mapping has a few upsides and a few downsides versus the
declarative approach from above. The main difference is that you define tables and
classes separately and map them together. It’s more flexible but a little more to type.
In general it works like the declarative approach, so make sure to also split up your
application into multiple modules in a package.
Here is an example database.py module for your application:
from sqlalchemy import create_engine, MetaData
from sqlalchemy.orm import scoped_session, sessionmaker
engine = create_engine(’sqlite:////tmp/test.db’, convert_unicode=True)
metadata = MetaData()
db_session = scoped_session(sessionmaker(autocommit=False,
autoflush=False,
bind=engine))
def init_db():
metadata.create_all(bind=engine)
As for the declarative approach you need to close the session after each request or
application context shutdown. Put this into your application module:
from yourapplication.database import db_session
@app.teardown_appcontext
def shutdown_session(exception=None):
db_session.remove()
Here is an example table and model (put this into models.py):
123](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-133-320.jpg)
![These results are also dict-like tuples:
>>> r = users.select(users.c.id == 1).execute().first()
>>> r[’name’]
u’admin’
You can also pass strings of SQL statements to the execute() method:
>>> engine.execute(’select * from users where id = :1’, [1]).first()
(1, u’admin’, u’admin@localhost’)
For more information about SQLAlchemy, head over to the website.
18.10 Uploading Files
Ah yes, the good old problem of file uploads. The basic idea of file uploads is actually
quite simple. It basically works like this:
1. A <form> tag is marked with enctype=multipart/form-data and an <input
type=file> is placed in that form.
2. The application accesses the file from the files dictionary on the request object.
3. use the save() method of the file to save the file permanently somewhere on the
filesystem.
18.10.1 A Gentle Introduction
Let’s start with a very basic application that uploads a file to a specific upload folder
and displays a file to the user. Let’s look at the bootstrapping code for our application:
import os
from flask import Flask, request, redirect, url_for
from werkzeug.utils import secure_filename
UPLOAD_FOLDER = ’/path/to/the/uploads’
ALLOWED_EXTENSIONS = set([’txt’, ’pdf’, ’png’, ’jpg’, ’jpeg’, ’gif’])
app = Flask(__name__)
app.config[’UPLOAD_FOLDER’] = UPLOAD_FOLDER
So first we need a couple of imports. Most should be straightforward, the
werkzeug.secure_filename() is explained a little bit later. The UPLOAD_FOLDER
is where we will store the uploaded files and the ALLOWED_EXTENSIONS is the set
of allowed file extensions. Then we add a URL rule by hand to the application. Now
usually we’re not doing that, so why here? The reasons is that we want the webserver
(or our development server) to serve these files for us and so we only need a rule to
generate the URL to these files.
125](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-135-320.jpg)
![Why do we limit the extensions that are allowed? You probably don’t want your users
to be able to upload everything there if the server is directly sending out the data to the
client. That way you can make sure that users are not able to upload HTML files that
would cause XSS problems (see Cross-Site Scripting (XSS)). Also make sure to disallow
.php files if the server executes them, but who has PHP installed on his server, right? :)
Next the functions that check if an extension is valid and that uploads the file and
redirects the user to the URL for the uploaded file:
def allowed_file(filename):
return ’.’ in filename and
filename.rsplit(’.’, 1)[1] in ALLOWED_EXTENSIONS
@app.route(’/’, methods=[’GET’, ’POST’])
def upload_file():
if request.method == ’POST’:
file = request.files[’file’]
if file and allowed_file(file.filename):
filename = secure_filename(file.filename)
file.save(os.path.join(app.config[’UPLOAD_FOLDER’], filename))
return redirect(url_for(’uploaded_file’,
filename=filename))
return ’’’
<!doctype html>
<title>Upload new File</title>
<h1>Upload new File</h1>
<form action="" method=post enctype=multipart/form-data>
<p><input type=file name=file>
<input type=submit value=Upload>
</form>
’’’
So what does that secure_filename() function actually do? Now the problem is that
there is that principle called “never trust user input”. This is also true for the filename
of an uploaded file. All submitted form data can be forged, and filenames can be dan-
gerous. For the moment just remember: always use that function to secure a filename
before storing it directly on the filesystem.
Information for the Pros
So you’re interested in what that secure_filename() function does and what the prob-
lem is if you’re not using it? So just imagine someone would send the following infor-
mation as filename to your application:
filename = "../../../../home/username/.bashrc"
Assuming the number of ../ is correct and you would join this with the UP-
LOAD_FOLDER the user might have the ability to modify a file on the server’s filesys-
tem he or she should not modify. This does require some knowledge about how the
application looks like, but trust me, hackers are patient :)
Now let’s look how that function works:
126](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-136-320.jpg)
![>>> secure_filename(’../../../../home/username/.bashrc’)
’home_username_.bashrc’
Now one last thing is missing: the serving of the uploaded files. As of Flask 0.5 we
can use a function that does that for us:
from flask import send_from_directory
@app.route(’/uploads/<filename>’)
def uploaded_file(filename):
return send_from_directory(app.config[’UPLOAD_FOLDER’],
filename)
Alternatively you can register uploaded_file as build_only rule and use the
SharedDataMiddleware. This also works with older versions of Flask:
from werkzeug import SharedDataMiddleware
app.add_url_rule(’/uploads/<filename>’, ’uploaded_file’,
build_only=True)
app.wsgi_app = SharedDataMiddleware(app.wsgi_app, {
’/uploads’: app.config[’UPLOAD_FOLDER’]
})
If you now run the application everything should work as expected.
18.10.2 Improving Uploads
New in version 0.6.
So how exactly does Flask handle uploads? Well it will store them in the webserver’s
memory if the files are reasonable small otherwise in a temporary location (as returned
by tempfile.gettempdir()). But how do you specify the maximum file size after
which an upload is aborted? By default Flask will happily accept file uploads to an un-
limited amount of memory, but you can limit that by setting the MAX_CONTENT_LENGTH
config key:
from flask import Flask, Request
app = Flask(__name__)
app.config[’MAX_CONTENT_LENGTH’] = 16 * 1024 * 1024
The code above will limited the maximum allowed payload to 16 megabytes. If a
larger file is transmitted, Flask will raise an RequestEntityTooLarge exception.
This feature was added in Flask 0.6 but can be achieved in older versions as well by
subclassing the request object. For more information on that consult the Werkzeug
documentation on file handling.
127](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-137-320.jpg)
![If you want to use memcached, make sure to have one of the memcache modules
supported (you get them from PyPI) and a memcached server running somewhere.
This is how you connect to such an memcached server then:
from werkzeug.contrib.cache import MemcachedCache
cache = MemcachedCache([’127.0.0.1:11211’])
If you are using App Engine, you can connect to the App Engine memcache server
easily:
from werkzeug.contrib.cache import GAEMemcachedCache
cache = GAEMemcachedCache()
18.11.2 Using a Cache
Now how can one use such a cache? There are two very important operations: get()
and set(). This is how to use them:
To get an item from the cache call get() with a string as key name. If something is in
the cache, it is returned. Otherwise that function will return None:
rv = cache.get(’my-item’)
To add items to the cache, use the set() method instead. The first argument is the
key and the second the value that should be set. Also a timeout can be provided after
which the cache will automatically remove item.
Here a full example how this looks like normally:
def get_my_item():
rv = cache.get(’my-item’)
if rv is None:
rv = calculate_value()
cache.set(’my-item’, rv, timeout=5 * 60)
return rv
18.12 View Decorators
Python has a really interesting feature called function decorators. This allow some
really neat things for web applications. Because each view in Flask is a function dec-
orators can be used to inject additional functionality to one or more functions. The
route() decorator is the one you probably used already. But there are use cases for
implementing your own decorator. For instance, imagine you have a view that should
only be used by people that are logged in to. If a user goes to the site and is not logged
in, they should be redirected to the login page. This is a good example of a use case
where a decorator is an excellent solution.
129](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-139-320.jpg)
![When you are working with WTForms you have to define your forms as classes first.
I recommend breaking up the application into multiple modules (Larger Applications)
for that and adding a separate module for the forms.
Getting most of WTForms with an Extension
The Flask-WTF extension expands on this pattern and adds a few handful little helpers
that make working with forms and Flask more fun. You can get it from PyPI.
18.13.1 The Forms
This is an example form for a typical registration page:
from wtforms import Form, BooleanField, TextField, PasswordField, validators
class RegistrationForm(Form):
username = TextField(’Username’, [validators.Length(min=4, max=25)])
email = TextField(’Email Address’, [validators.Length(min=6, max=35)])
password = PasswordField(’New Password’, [
validators.Required(),
validators.EqualTo(’confirm’, message=’Passwords must match’)
])
confirm = PasswordField(’Repeat Password’)
accept_tos = BooleanField(’I accept the TOS’, [validators.Required()])
18.13.2 In the View
In the view function, the usage of this form looks like this:
@app.route(’/register’, methods=[’GET’, ’POST’])
def register():
form = RegistrationForm(request.form)
if request.method == ’POST’ and form.validate():
user = User(form.username.data, form.email.data,
form.password.data)
db_session.add(user)
flash(’Thanks for registering’)
return redirect(url_for(’login’))
return render_template(’register.html’, form=form)
Notice that we are implying that the view is using SQLAlchemy here (SQLAlchemy in
Flask) but this is no requirement of course. Adapt the code as necessary.
Things to remember:
1. create the form from the request form value if the data is submitted via the HTTP
POST method and args if the data is submitted as GET.
2. to validate the data, call the validate() method which will return True if the data
validates, False otherwise.
133](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-143-320.jpg)
![{% endblock %}
{% block content %}
<h1>Index</h1>
<p class="important">
Welcome on my awesome homepage.
{% endblock %}
The {% extends %} tag is the key here. It tells the template engine that this template
“extends” another template. When the template system evaluates this template, first
it locates the parent. The extends tag must be the first tag in the template. To render
the contents of a block defined in the parent template, use {{ super() }}.
18.15 Message Flashing
Good applications and user interfaces are all about feedback. If the user does not get
enough feedback they will probably end up hating the application. Flask provides a
really simple way to give feedback to a user with the flashing system. The flashing
system basically makes it possible to record a message at the end of a request and
access it next request and only next request. This is usually combined with a layout
template that does this.
18.15.1 Simple Flashing
So here is a full example:
from flask import Flask, flash, redirect, render_template,
request, url_for
app = Flask(__name__)
app.secret_key = ’some_secret’
@app.route(’/’)
def index():
return render_template(’index.html’)
@app.route(’/login’, methods=[’GET’, ’POST’])
def login():
error = None
if request.method == ’POST’:
if request.form[’username’] != ’admin’ or
request.form[’password’] != ’secret’:
error = ’Invalid credentials’
else:
flash(’You were successfully logged in’)
return redirect(url_for(’index’))
return render_template(’login.html’, error=error)
136](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-146-320.jpg)
![It is also possible to provide categories when flashing a message. The default cate-
gory if nothing is provided is ’message’. Alternative categories can be used to give
the user better feedback. For example error messages could be displayed with a red
background.
To flash a message with a different category, just use the second argument to the
flash() function:
flash(u’Invalid password provided’, ’error’)
Inside the template you then have to tell the get_flashed_messages() function to also
return the categories. The loop looks slightly different in that situation then:
{% with messages = get_flashed_messages(with_categories=true) %}
{% if messages %}
<ul class=flashes>
{% for category, message in messages %}
<li class="{{ category }}">{{ message }}</li>
{% endfor %}
</ul>
{% endif %}
{% endwith %}
This is just one example of how to render these flashed messages. One might also use
the category to add a prefix such as <strong>Error:</strong> to the message.
18.15.3 Filtering Flash Messages
New in version 0.9.
Optionally you can pass a list of categories which filters the results of
get_flashed_messages(). This is useful if you wish to render each category in a sepa-
rate block.
{% with errors = get_flashed_messages(category_filter=["error"]) %}
{% if errors %}
<div class="alert-message block-message error">
<a class="close" href="#">×</a>
<ul>
{%- for msg in errors %}
<li>{{ msg }}</li>
{% endfor -%}
</ul>
</div>
{% endif %}
{% endwith %}
138](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-148-320.jpg)
![18.16.4 The HTML
Your index.html template either has to extend a layout.html template with jQuery
loaded and the $SCRIPT_ROOT variable set, or do that on the top. Here’s the HTML
code needed for our little application (index.html). Notice that we also drop the script
directly into the HTML here. It is usually a better idea to have that in a separate script
file:
<script type=text/javascript>
$(function() {
$(’a#calculate’).bind(’click’, function() {
$.getJSON($SCRIPT_ROOT + ’/_add_numbers’, {
a: $(’input[name="a"]’).val(),
b: $(’input[name="b"]’).val()
}, function(data) {
$("#result").text(data.result);
});
return false;
});
});
</script>
<h1>jQuery Example</h1>
<p><input type=text size=5 name=a> +
<input type=text size=5 name=b> =
<span id=result>?</span>
<p><a href=# id=calculate>calculate server side</a>
I won’t go into detail here about how jQuery works, just a very quick explanation of
the little bit of code above:
1. $(function() { ... }) specifies code that should run once the browser is done
loading the basic parts of the page.
2. $(’selector’) selects an element and lets you operate on it.
3. element.bind(’event’, func) specifies a function that should run when the user
clicked on the element. If that function returns false, the default behavior will not
kick in (in this case, navigate to the # URL).
4. $.getJSON(url, data, func) sends a GET request to url and will send the con-
tents of the data object as query parameters. Once the data arrived, it will call
the given function with the return value as argument. Note that we can use the
$SCRIPT_ROOT variable here that we set earlier.
If you don’t get the whole picture, download the sourcecode for this example from
github.
18.17 Custom Error Pages
Flask comes with a handy abort() function that aborts a request with an HTTP error
code early. It will also provide a plain black and white error page for you with a basic
141](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-151-320.jpg)
![url(’/’, ’views.index’)
url(’/user/<username>’, ’views.user’)
One thing to keep in mind is that before and after request handlers have to be in a file
that is imported upfront to work properly on the first request. The same goes for any
kind of remaining decorator.
18.19 MongoKit in Flask
Using a document database rather than a full DBMS gets more common these days.
This pattern shows how to use MongoKit, a document mapper library, to integrate
with MongoDB.
This pattern requires a running MongoDB server and the MongoKit library installed.
There are two very common ways to use MongoKit. I will outline each of them here:
18.19.1 Declarative
The default behavior of MongoKit is the declarative one that is based on common
ideas from Django or the SQLAlchemy declarative extension.
Here an example app.py module for your application:
from flask import Flask
from mongokit import Connection, Document
# configuration
MONGODB_HOST = ’localhost’
MONGODB_PORT = 27017
# create the little application object
app = Flask(__name__)
app.config.from_object(__name__)
# connect to the database
connection = Connection(app.config[’MONGODB_HOST’],
app.config[’MONGODB_PORT’])
To define your models, just subclass the Document class that is imported from Mon-
goKit. If you’ve seen the SQLAlchemy pattern you may wonder why we do not have
a session and even do not define a init_db function here. On the one hand, MongoKit
does not have something like a session. This sometimes makes it more to type but
also makes it blazingly fast. On the other hand, MongoDB is schemaless. This means
you can modify the data structure from one insert query to the next without any prob-
lem. MongoKit is just schemaless too, but implements some validation to ensure data
integrity.
Here is an example document (put this also into app.py, e.g.):
145](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-155-320.jpg)
![def max_length(length):
def validate(value):
if len(value) <= length:
return True
raise Exception(’%s must be at most %s characters long’ % length)
return validate
class User(Document):
structure = {
’name’: unicode,
’email’: unicode,
}
validators = {
’name’: max_length(50),
’email’: max_length(120)
}
use_dot_notation = True
def __repr__(self):
return ’<User %r>’ % (self.name)
# register the User document with our current connection
connection.register([User])
This example shows you how to define your schema (named structure), a valida-
tor for the maximum character length and uses a special MongoKit feature called
use_dot_notation. Per default MongoKit behaves like a python dictionary but with
use_dot_notation set to True you can use your documents like you use models in nearly
any other ORM by using dots to separate between attributes.
You can insert entries into the database like this:
>>> from yourapplication.database import connection
>>> from yourapplication.models import User
>>> collection = connection[’test’].users
>>> user = collection.User()
>>> user[’name’] = u’admin’
>>> user[’email’] = u’admin@localhost’
>>> user.save()
Note that MongoKit is kinda strict with used column types, you must not use a com-
mon str type for either name or email but unicode.
Querying is simple as well:
>>> list(collection.User.find())
[<User u’admin’>]
>>> collection.User.find_one({’name’: u’admin’})
<User u’admin’>
146](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-156-320.jpg)
![18.19.2 PyMongo Compatibility Layer
If you just want to use PyMongo, you can do that with MongoKit as well. You may
use this process if you need the best performance to get. Note that this example does
not show how to couple it with Flask, see the above MongoKit code for examples:
from MongoKit import Connection
connection = Connection()
To insert data you can use the insert method. We have to get a collection first, this is
somewhat the same as a table in the SQL world.
>>> collection = connection[’test’].users
>>> user = {’name’: u’admin’, ’email’: u’admin@localhost’}
>>> collection.insert(user)
MongoKit will automatically commit for us.
To query your database, you use the collection directly:
>>> list(collection.find())
[{u’_id’: ObjectId(’4c271729e13823182f000000’), u’name’: u’admin’, u’email’: u’admin@localhost’}
>>> collection.find_one({’name’: u’admin’})
{u’_id’: ObjectId(’4c271729e13823182f000000’), u’name’: u’admin’, u’email’: u’admin@localhost’}
These results are also dict-like objects:
>>> r = collection.find_one({’name’: u’admin’})
>>> r[’email’]
u’admin@localhost’
For more information about MongoKit, head over to the website.
18.20 Adding a favicon
A “favicon” is an icon used by browsers for tabs and bookmarks. This helps to distin-
guish your website and to give it a unique brand.
A common question is how to add a favicon to a flask application. First, of course,
you need an icon. It should be 16 × 16 pixels and in the ICO file format. This is not a
requirement but a de-facto standard supported by all relevant browsers. Put the icon
in your static directory as favicon.ico.
Now, to get browsers to find your icon, the correct way is to add a link tag in your
HTML. So, for example:
<link rel="shortcut icon" href="{{ url_for(’static’, filename=’favicon.ico’) }}">
That’s all you need for most browsers, however some really old ones do not support
this standard. The old de-facto standard is to serve this file, with this name, at the
website root. If your application is not mounted at the root path of the domain you
147](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-157-320.jpg)
![def generate():
yield ’Hello ’
yield request.args[’name’]
yield ’!’
return Response(stream_with_context(generate()))
Without the stream_with_context() function you would get a RuntimeError at that
point.
18.22 Deferred Request Callbacks
One of the design principles of Flask is that response objects are created and passed
down a chain of potential callbacks that can modify them or replace them. When the
request handling starts, there is no response object yet. It is created as necessary either
by a view function or by some other component in the system.
But what happens if you want to modify the response at a point where the response
does not exist yet? A common example for that would be a before-request function
that wants to set a cookie on the response object.
One way is to avoid the situation. Very often that is possible. For instance you can try
to move that logic into an after-request callback instead. Sometimes however moving
that code there is just not a very pleasant experience or makes code look very awk-
ward.
As an alternative possibility you can attach a bunch of callback functions to the g object
and call them at the end of the request. This way you can defer code execution from
anywhere in the application.
18.22.1 The Decorator
The following decorator is the key. It registers a function on a list on the g object:
from flask import g
def after_this_request(f):
if not hasattr(g, ’after_request_callbacks’):
g.after_request_callbacks = []
g.after_request_callbacks.append(f)
return f
18.22.2 Calling the Deferred
Now you can use the after_this_request decorator to mark a function to be called at the
end of the request. But we still need to call them. For this the following function needs
to be registered as after_request() callback:
150](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-160-320.jpg)
![@app.after_request
def call_after_request_callbacks(response):
for callback in getattr(g, ’after_request_callbacks’, ()):
callback(response)
return response
18.22.3 A Practical Example
Now we can easily at any point in time register a function to be called at the end of this
particular request. For example you can remember the current language of the user in
a cookie in the before-request function:
from flask import request
@app.before_request
def detect_user_language():
language = request.cookies.get(’user_lang’)
if language is None:
language = guess_language_from_request()
@after_this_request
def remember_language(response):
response.set_cookie(’user_lang’, language)
g.language = language
18.23 Adding HTTP Method Overrides
Some HTTP proxies do not support arbitrary HTTP methods or newer HTTP methods
(such as PATCH). In that case it’s possible to “proxy” HTTP methods through another
HTTP method in total violation of the protocol.
The way this works is by letting the client do an HTTP POST request and set the
X-HTTP-Method-Override header and set the value to the intended HTTP method (such
as PATCH).
This can easily be accomplished with an HTTP middleware:
class HTTPMethodOverrideMiddleware(object):
allowed_methods = frozenset([
’GET’,
’HEAD’,
’POST’,
’DELETE’,
’PUT’,
’PATCH’,
’OPTIONS’
])
bodyless_methods = frozenset([’GET’, ’HEAD’, ’OPTIONS’, ’DELETE’])
151](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-161-320.jpg)
![def __init__(self, app):
self.app = app
def __call__(self, environ, start_response):
method = environ.get(’HTTP_X_HTTP_METHOD_OVERRIDE’, ’’).upper()
if method in self.allowed_methods:
method = method.encode(’ascii’, ’replace’)
environ[’REQUEST_METHOD’] = method
if method in self.bodyless_methods:
environ[’CONTENT_LENGTH’] = ’0’
return self.app(environ, start_response)
To use this with Flask this is all that is necessary:
from flask import Flask
app = Flask(__name__)
app.wsgi_app = HTTPMethodOverrideMiddleware(app.wsgi_app)
18.24 Request Content Checksums
Various pieces of code can consume the request data and preprocess it. For instance
JSON data ends up on the request object already read and processed, form data ends
up there as well but goes through a different code path. This seems inconvenient when
you want to calculate the checksum of the incoming request data. This is necessary
sometimes for some APIs.
Fortunately this is however very simple to change by wrapping the input stream.
The following example calculates the SHA1 checksum of the incoming data as it gets
read and stores it in the WSGI environment:
import hashlib
class ChecksumCalcStream(object):
def __init__(self, stream):
self._stream = stream
self._hash = hashlib.sha1()
def read(self, bytes):
rv = self._stream.read(bytes)
self._hash.update(rv)
return rv
def readline(self, size_hint):
rv = self._stream.readline(size_hint)
self._hash.update(rv)
return rv
152](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-162-320.jpg)
![def generate_checksum(request):
env = request.environ
stream = ChecksumCalcStream(env[’wsgi.input’])
env[’wsgi.input’] = stream
return stream._hash
To use this, all you need to do is to hook the calculating stream in before the request
starts consuming data. (Eg: be careful accessing request.form or anything of that
nature. before_request_handlers for instance should be careful not to access it).
Example usage:
@app.route(’/special-api’, methods=[’POST’])
def special_api():
hash = generate_checksum(request)
# Accessing this parses the input stream
files = request.files
# At this point the hash is fully constructed.
checksum = hash.hexdigest()
return ’Hash was: %s’ % checksum
18.25 Celery Based Background Tasks
Celery is a task queue for Python with batteries included. It used to have a Flask inte-
gration but it became unnecessary after some restructuring of the internals of Celery
with Version 3. This guide fills in the blanks in how to properly use Celery with Flask
but assumes that you generally already read the First Steps with Celery guide in the
official Celery documentation.
18.25.1 Installing Celery
Celery is on the Python Package Index (PyPI), so it can be installed with standard
Python tools like pip or easy_install:
$ pip install celery
18.25.2 Configuring Celery
The first thing you need is a Celery instance, this is called the celery application. It
serves the same purpose as the Flask object in Flask, just for Celery. Since this instance
is used as the entry-point for everything you want to do in Celery, like creating tasks
and managing workers, it must be possible for other modules to import it.
For instance you can place this in a tasks module. While you can use Celery without
any reconfiguration with Flask, it becomes a bit nicer by subclassing tasks and adding
153](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-163-320.jpg)
![support for Flask’s application contexts and hooking it up with the Flask configura-
tion.
This is all that is necessary to properly integrate Celery with Flask:
from celery import Celery
def make_celery(app):
celery = Celery(app.import_name, broker=app.config[’CELERY_BROKER_URL’])
celery.conf.update(app.config)
TaskBase = celery.Task
class ContextTask(TaskBase):
abstract = True
def __call__(self, *args, **kwargs):
with app.app_context():
return TaskBase.__call__(self, *args, **kwargs)
celery.Task = ContextTask
return celery
The function creates a new Celery object, configures it with the broker from the ap-
plication config, updates the rest of the Celery config from the Flask config and then
creates a subclass of the task that wraps the task execution in an application context.
18.25.3 Minimal Example
With what we have above this is the minimal example of using Celery with Flask:
from flask import Flask
flask_app = Flask(__name__)
flask_app.config.update(
CELERY_BROKER_URL=’redis://localhost:6379’,
CELERY_RESULT_BACKEND=’redis://localhost:6379’
)
celery = make_celery(flask_app)
@celery.task()
def add_together(a, b):
return a + b
This task can now be called in the background:
>>> result = add_together.delay(23, 42)
>>> result.wait()
65
154](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-164-320.jpg)
![two problematic values in the WSGI environment usually are REMOTE_ADDR and
HTTP_HOST. You can configure your httpd to pass these headers, or you can fix them
in middleware. Werkzeug ships a fixer that will solve some common setups, but you
might want to write your own WSGI middleware for specific setups.
Here’s a simple nginx configuration which proxies to an application served on local-
host at port 8000, setting appropriate headers:
server {
listen 80;
server_name _;
access_log /var/log/nginx/access.log;
error_log /var/log/nginx/error.log;
location / {
proxy_pass http://127.0.0.1:8000/;
proxy_redirect off;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
}
}
If your httpd is not providing these headers, the most common setup invokes the host
being set from X-Forwarded-Host and the remote address from X-Forwarded-For:
from werkzeug.contrib.fixers import ProxyFix
app.wsgi_app = ProxyFix(app.wsgi_app)
Trusting Headers
Please keep in mind that it is a security issue to use such a middleware in a non-proxy
setup because it will blindly trust the incoming headers which might be forged by
malicious clients.
If you want to rewrite the headers from another header, you might want to use a fixer
like this:
class CustomProxyFix(object):
def __init__(self, app):
self.app = app
def __call__(self, environ, start_response):
host = environ.get(’HTTP_X_FHOST’, ’’)
if host:
environ[’HTTP_HOST’] = host
return self.app(environ, start_response)
162](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-172-320.jpg)
![RewriteCond %{REQUEST_FILENAME} !-f
RewriteRule ^(.*)$ yourapplication.fcgi/$1 [QSA,L]
</IfModule>
Set yourapplication.fcgi:
#!/usr/bin/python
#: optional path to your local python site-packages folder
import sys
sys.path.insert(0, ’<your_local_path>/lib/python2.6/site-packages’)
from flup.server.fcgi import WSGIServer
from yourapplication import app
class ScriptNameStripper(object):
def __init__(self, app):
self.app = app
def __call__(self, environ, start_response):
environ[’SCRIPT_NAME’] = ’’
return self.app(environ, start_response)
app = ScriptNameStripper(app)
if __name__ == ’__main__’:
WSGIServer(app).run()
19.4.3 Configuring lighttpd
A basic FastCGI configuration for lighttpd looks like that:
fastcgi.server = ("/yourapplication.fcgi" =>
((
"socket" => "/tmp/yourapplication-fcgi.sock",
"bin-path" => "/var/www/yourapplication/yourapplication.fcgi",
"check-local" => "disable",
"max-procs" => 1
))
)
alias.url = (
"/static/" => "/path/to/your/static"
)
url.rewrite-once = (
"^(/static($|/.*))$" => "$1",
"^(/.*)$" => "/yourapplication.fcgi$1"
)
Remember to enable the FastCGI, alias and rewrite modules. This configuration binds
166](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-176-320.jpg)
![#!/usr/bin/python
from wsgiref.handlers import CGIHandler
from yourapplication import app
CGIHandler().run(app)
19.5.2 Server Setup
Usually there are two ways to configure the server. Either just copy the .cgi into a
cgi-bin (and use mod_rewrite or something similar to rewrite the URL) or let the server
point to the file directly.
In Apache for example you can put something like this into the config:
ScriptAlias /app /path/to/the/application.cgi
On shared webhosting, though, you might not have access to your Apache config. In
this case, a file called .htaccess, sitting in the public directory you want your app to be
available, works too but the ScriptAlias directive won’t work in that case:
RewriteEngine On
RewriteCond %{REQUEST_FILENAME} !-f # Don’t interfere with static files
RewriteRule ^(.*)$ /path/to/the/application.cgi/$1 [L]
For more information consult the documentation of your webserver.
169](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-179-320.jpg)
![For example if your application is defined in yourapplication/app.py you should
create it with one of the two versions below:
app = Flask(’yourapplication’)
app = Flask(__name__.split(’.’)[0])
Why is that? The application will work even with __name__, thanks to how re-
sources are looked up. However it will make debugging more painful. Certain
extensions can make assumptions based on the import name of your application.
For example the Flask-SQLAlchemy extension will look for the code in your ap-
plication that triggered an SQL query in debug mode. If the import name is not
properly set up, that debugging information is lost. (For example it would only
pick up SQL queries in yourapplication.app and not yourapplication.views.frontend)
New in version 0.7: The static_url_path, static_folder, and template_folder parame-
ters were added.
New in version 0.8: The instance_path and instance_relative_config parameters
were added.
Parameters
• import_name – the name of the application package
• static_url_path – can be used to specify a different path for the
static files on the web. Defaults to the name of the static_folder
folder.
• static_folder – the folder with static files that should be served
at static_url_path. Defaults to the ’static’ folder in the root
path of the application.
• template_folder – the folder that contains the templates that
should be used by the application. Defaults to ’templates’
folder in the root path of the application.
• instance_path – An alternative instance path for the applica-
tion. By default the folder ’instance’ next to the package or
module is assumed to be the instance path.
• instance_relative_config – if set to True relative filenames for
loading the config are assumed to be relative to the instance
path instead of the application root.
add_template_filter(*args, **kwargs)
Register a custom template filter. Works exactly like the template_filter()
decorator.
Parameters name – the optional name of the filter, otherwise the
function name will be used.
add_template_global(*args, **kwargs)
Register a custom template global function. Works exactly like the
template_global() decorator.
178](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-188-320.jpg)
![New in version 0.10.
Parameters name – the optional name of the global function, oth-
erwise the function name will be used.
add_template_test(*args, **kwargs)
Register a custom template test. Works exactly like the template_test()
decorator.
New in version 0.10.
Parameters name – the optional name of the test, otherwise the
function name will be used.
add_url_rule(*args, **kwargs)
Connects a URL rule. Works exactly like the route() decorator. If a
view_func is provided it will be registered with the endpoint.
Basically this example:
@app.route(’/’)
def index():
pass
Is equivalent to the following:
def index():
pass
app.add_url_rule(’/’, ’index’, index)
If the view_func is not provided you will need to connect the endpoint to a
view function like so:
app.view_functions[’index’] = index
Internally route() invokes add_url_rule() so if you want to customize the
behavior via subclassing you only need to change this method.
For more information refer to URL Route Registrations.
Changed in version 0.2: view_func parameter added.
Changed in version 0.6: OPTIONS is added automatically as method.
Parameters
• rule – the URL rule as string
• endpoint – the endpoint for the registered URL rule. Flask
itself assumes the name of the view function as endpoint
• view_func – the function to call when serving a request to the
provided endpoint
• options – the options to be forwarded to the underlying Rule
object. A change to Werkzeug is handling of method options.
methods is a list of methods this rule should be limited to
179](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-189-320.jpg)
![Changed in version 0.9: Added the exc argument. Previously this was al-
ways using the current exception information.
enable_modules = True
Enable the deprecated module support? This is active by default in 0.7 but
will be changed to False in 0.8. With Flask 1.0 modules will be removed in
favor of Blueprints
endpoint(*args, **kwargs)
A decorator to register a function as an endpoint. Example:
@app.endpoint(’example.endpoint’)
def example():
return "example"
Parameters endpoint – the name of the endpoint
error_handler_spec = None
A dictionary of all registered error handlers. The key is None for error
handlers active on the application, otherwise the key is the name of the
blueprint. Each key points to another dictionary where the key is the status
code of the http exception. The special key None points to a list of tuples
where the first item is the class for the instance check and the second the
error handler function.
To register a error handler, use the errorhandler() decorator.
errorhandler(*args, **kwargs)
A decorator that is used to register a function give a given error code. Ex-
ample:
@app.errorhandler(404)
def page_not_found(error):
return ’This page does not exist’, 404
You can also register handlers for arbitrary exceptions:
@app.errorhandler(DatabaseError)
def special_exception_handler(error):
return ’Database connection failed’, 500
You can also register a function as error handler without using the
errorhandler() decorator. The following example is equivalent to the one
above:
def page_not_found(error):
return ’This page does not exist’, 404
app.error_handler_spec[None][404] = page_not_found
Setting error handlers via assignments to error_handler_spec however is
discouraged as it requires fiddling with nested dictionaries and the special
case for arbitrary exception types.
183](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-193-320.jpg)
![The first None refers to the active blueprint. If the error handler should be
application wide None shall be used.
New in version 0.7: One can now additionally also register custom excep-
tion types that do not necessarily have to be a subclass of the HTTPException
class.
Parameters code – the code as integer for the handler
extensions = None
a place where extensions can store application specific state. For exam-
ple this is where an extension could store database engines and similar
things. For backwards compatibility extensions should register themselves
like this:
if not hasattr(app, ’extensions’):
app.extensions = {}
app.extensions[’extensionname’] = SomeObject()
The key must match the name of the extension module. For example in case
of a “Flask-Foo” extension in flask_foo, the key would be ’foo’.
New in version 0.7.
full_dispatch_request()
Dispatches the request and on top of that performs request pre and post-
processing as well as HTTP exception catching and error handling.
New in version 0.7.
get_send_file_max_age(filename)
Provides default cache_timeout for the send_file() functions.
By default, this function returns SEND_FILE_MAX_AGE_DEFAULT from the con-
figuration of current_app.
Static file functions such as send_from_directory() use this function,
and send_file() calls this function on current_app when the given
cache_timeout is None. If a cache_timeout is given in send_file(), that
timeout is used; otherwise, this method is called.
This allows subclasses to change the behavior when sending files based on
the filename. For example, to set the cache timeout for .js files to 60 seconds:
class MyFlask(flask.Flask):
def get_send_file_max_age(self, name):
if name.lower().endswith(’.js’):
return 60
return flask.Flask.get_send_file_max_age(self, name)
New in version 0.9.
got_first_request
This attribute is set to True if the application started handling the first re-
quest.
184](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-194-320.jpg)
![jinja_loader
The Jinja loader for this package bound object.
New in version 0.5.
jinja_options = ImmutableDict({‘extensions’: [’jinja2.ext.autoescape’, ‘jinja2.ext.with_’]})
Options that are passed directly to the Jinja2 environment.
json_decoder
The JSON decoder class to use. Defaults to JSONDecoder.
New in version 0.10.
alias of JSONDecoder
json_encoder
The JSON encoder class to use. Defaults to JSONEncoder.
New in version 0.10.
alias of JSONEncoder
log_exception(exc_info)
Logs an exception. This is called by handle_exception() if debugging is
disabled and right before the handler is called. The default implementation
logs the exception as error on the logger.
New in version 0.8.
logger
A logging.Logger object for this application. The default configuration is to
log to stderr if the application is in debug mode. This logger can be used to
(surprise) log messages. Here some examples:
app.logger.debug(’A value for debugging’)
app.logger.warning(’A warning occurred (%d apples)’, 42)
app.logger.error(’An error occurred’)
New in version 0.3.
logger_name
The name of the logger to use. By default the logger name is the package
name passed to the constructor.
New in version 0.4.
make_config(instance_relative=False)
Used to create the config attribute by the Flask constructor. The in-
stance_relative parameter is passed in from the constructor of Flask (there
named instance_relative_config) and indicates if the config should be relative
to the instance path or the root path of the application.
New in version 0.8.
make_default_options_response()
This method is called to create the default OPTIONS response. This can be
186](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-196-320.jpg)
![@app.template_filter()
def reverse(s):
return s[::-1]
Parameters name – the optional name of the filter, otherwise the
function name will be used.
template_global(*args, **kwargs)
A decorator that is used to register a custom template global function. You
can specify a name for the global function, otherwise the function name will
be used. Example:
@app.template_global()
def double(n):
return 2 * n
New in version 0.10.
Parameters name – the optional name of the global function, oth-
erwise the function name will be used.
template_test(*args, **kwargs)
A decorator that is used to register custom template test. You can specify a
name for the test, otherwise the function name will be used. Example:
@app.template_test()
def is_prime(n):
if n == 2:
return True
for i in range(2, int(math.ceil(math.sqrt(n))) + 1):
if n % i == 0:
return False
return True
New in version 0.10.
Parameters name – the optional name of the test, otherwise the
function name will be used.
test_client(use_cookies=True)
Creates a test client for this application. For information about unit testing
head over to Testing Flask Applications.
Note that if you are testing for assertions or exceptions in your application
code, you must set app.testing = True in order for the exceptions to prop-
agate to the test client. Otherwise, the exception will be handled by the
application (not visible to the test client) and the only indication of an As-
sertionError or other exception will be a 500 status code response to the test
client. See the testing attribute. For example:
app.testing = True
client = app.test_client()
194](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-204-320.jpg)
![The test client can be used in a with block to defer the closing down of the
context until the end of the with block. This is useful if you want to access
the context locals for testing:
with app.test_client() as c:
rv = c.get(’/?vodka=42’)
assert request.args[’vodka’] == ’42’
See FlaskClient for more information.
Changed in version 0.4: added support for with block usage for the client.
New in version 0.7: The use_cookies parameter was added as well as the
ability to override the client to be used by setting the test_client_class
attribute.
test_client_class = None
the test client that is used with when test_client is used.
New in version 0.7.
test_request_context(*args, **kwargs)
Creates a WSGI environment from the given values (see
werkzeug.test.EnvironBuilder for more information, this function ac-
cepts the same arguments).
testing
The testing flag. Set this to True to enable the test mode of Flask extensions
(and in the future probably also Flask itself). For example this might activate
unittest helpers that have an additional runtime cost which should not be
enabled by default.
If this is enabled and PROPAGATE_EXCEPTIONS is not changed from the
default it’s implicitly enabled.
This attribute can also be configured from the config with the TESTING con-
figuration key. Defaults to False.
trap_http_exception(e)
Checks if an HTTP exception should be trapped or not. By default
this will return False for all exceptions except for a bad request key er-
ror if TRAP_BAD_REQUEST_ERRORS is set to True. It also returns True if
TRAP_HTTP_EXCEPTIONS is set to True.
This is called for all HTTP exceptions raised by a view function. If it returns
True for any exception the error handler for this exception is not called and
it shows up as regular exception in the traceback. This is helpful for debug-
ging implicitly raised HTTP exceptions.
New in version 0.8.
update_template_context(context)
Update the template context with some commonly used variables. This in-
jects request, session, config and g into the template context as well as every-
thing template context processors want to inject. Note that the as of Flask
195](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-205-320.jpg)
![0.6, the original values in the context will not be overridden if a context
processor decides to return a value with the same key.
Parameters context – the context as a dictionary that is updated in
place to add extra variables.
url_build_error_handlers = None
A list of functions that are called when url_for() raises a BuildError. Each
function registered here is called with error, endpoint and values. If a function
returns None or raises a BuildError the next function is tried.
New in version 0.9.
url_default_functions = None
A dictionary with lists of functions that can be used as URL value prepro-
cessors. The key None here is used for application wide callbacks, otherwise
the key is the name of the blueprint. Each of these functions has the chance
to modify the dictionary of URL values before they are used as the key-
word arguments of the view function. For each function registered this one
should also provide a url_defaults() function that adds the parameters
automatically again that were removed that way.
New in version 0.7.
url_defaults(*args, **kwargs)
Callback function for URL defaults for all view functions of the applica-
tion. It’s called with the endpoint and values and should update the values
passed in place.
url_map = None
The Map for this instance. You can use this to change the routing converters
after the class was created but before any routes are connected. Example:
from werkzeug.routing import BaseConverter
class ListConverter(BaseConverter):
def to_python(self, value):
return value.split(’,’)
def to_url(self, values):
return ’,’.join(BaseConverter.to_url(value)
for value in values)
app = Flask(__name__)
app.url_map.converters[’list’] = ListConverter
url_rule_class
The rule object to use for URL rules created. This is used by add_url_rule().
Defaults to werkzeug.routing.Rule.
New in version 0.7.
alias of Rule
url_value_preprocessor(*args, **kwargs)
196](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-206-320.jpg)
![• value – the value of the cookie.
• max_age – should be a number of seconds, or None (default)
if the cookie should last only as long as the client’s browser
session.
• expires – should be a datetime object or UNIX timestamp.
• domain – if you want to set a cross-domain cookie. For exam-
ple, domain=".example.com" will set a cookie that is readable
by the domain www.example.com, foo.example.com etc. Oth-
erwise, a cookie will only be readable by the domain that set
it.
• path – limits the cookie to a given path, per default it will
span the whole domain.
21.5 Sessions
If you have the Flask.secret_key set you can use sessions in Flask applications. A
session basically makes it possible to remember information from one request to an-
other. The way Flask does this is by using a signed cookie. So the user can look at the
session contents, but not modify it unless they know the secret key, so make sure to
set that to something complex and unguessable.
To access the current session you can use the session object:
class flask.session
The session object works pretty much like an ordinary dict, with the difference
that it keeps track on modifications.
This is a proxy. See Notes On Proxies for more information.
The following attributes are interesting:
new
True if the session is new, False otherwise.
modified
True if the session object detected a modification. Be advised that modifica-
tions on mutable structures are not picked up automatically, in that situation
you have to explicitly set the attribute to True yourself. Here an example:
# this change is not picked up because a mutable object (here
# a list) is changed.
session[’objects’].append(42)
# so mark it as modified yourself
session.modified = True
permanent
If set to True the session lives for permanent_session_lifetime seconds. The
206](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-216-320.jpg)
![This object provides dumping and loading methods similar to simplejson but it
also tags certain builtin Python objects that commonly appear in sessions. Cur-
rently the following extended values are supported in the JSON it dumps:
•Markup objects
•UUID objects
•datetime objects
•tuples
Notice
The PERMANENT_SESSION_LIFETIME config key can also be an integer starting with Flask
0.8. Either catch this down yourself or use the permanent_session_lifetime attribute
on the app which converts the result to an integer automatically.
21.7 Test Client
class flask.testing.FlaskClient(application, response_wrapper=None,
use_cookies=True, al-
low_subdomain_redirects=False)
Works like a regular Werkzeug test client but has some knowledge about how
Flask works to defer the cleanup of the request context stack to the end of a with
body when used in a with statement. For general information about how to use
this class refer to werkzeug.test.Client.
Basic usage is outlined in the Testing Flask Applications chapter.
session_transaction(*args, **kwds)
When used in combination with a with statement this opens a session trans-
action. This can be used to modify the session that the test client uses. Once
the with block is left the session is stored back.
with client.session_transaction() as session: session[’value’] = 42
Internally this is implemented by going through a temporary test request
context and since session handling could depend on request variables this
function accepts the same arguments as test_request_context() which are
directly passed through.
21.8 Application Globals
To share data that is valid for one request only from one function to another, a global
variable is not good enough because it would break in threaded environments. Flask
provides you with a special object that ensures it is only valid for the active request
and that will return different values for each request. In a nutshell: it does the right
thing, like it does for request and session.
210](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-220-320.jpg)
![to the target location. Supported codes are 301, 302, 303, 305, and 307. 300 is not
supported because it’s not a real redirect and 304 because it’s the answer for a
request with a request with defined If-Modified-Since headers.
New in version 0.6: The location can now be a unicode string that is encoded
using the iri_to_uri() function.
Parameters
• location – the location the response should redirect to.
• code – the redirect status code. defaults to 302.
flask.make_response(*args)
Sometimes it is necessary to set additional headers in a view. Because views
do not have to return response objects but can return a value that is converted
into a response object by Flask itself, it becomes tricky to add headers to it. This
function can be called instead of using a return and you will get a response object
which you can use to attach headers.
If view looked like this and you want to add a new header:
def index():
return render_template(’index.html’, foo=42)
You can now do something like this:
def index():
response = make_response(render_template(’index.html’, foo=42))
response.headers[’X-Parachutes’] = ’parachutes are cool’
return response
This function accepts the very same arguments you can return from a view func-
tion. This for example creates a response with a 404 error code:
response = make_response(render_template(’not_found.html’), 404)
The other use case of this function is to force the return value of a view function
into a response which is helpful with view decorators:
response = make_response(view_function())
response.headers[’X-Parachutes’] = ’parachutes are cool’
Internally this function does the following things:
•if no arguments are passed, it creates a new response argument
•if one argument is passed, flask.Flask.make_response() is invoked with
it.
•if more than one argument is passed, the arguments are passed to the
flask.Flask.make_response() function as tuple.
New in version 0.6.
214](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-224-320.jpg)
![flask.after_this_request(f)
Executes a function after this request. This is useful to modify response objects.
The function is passed the response object and has to return the same or a new
one.
Example:
@app.route(’/’)
def index():
@after_this_request
def add_header(response):
response.headers[’X-Foo’] = ’Parachute’
return response
return ’Hello World!’
This is more useful if a function other than the view function wants to modify
a response. For instance think of a decorator that wants to add some headers
without converting the return value into a response object.
New in version 0.9.
flask.send_file(filename_or_fp, mimetype=None, as_attachment=False, attach-
ment_filename=None, add_etags=True, cache_timeout=None, con-
ditional=False)
Sends the contents of a file to the client. This will use the most efficient
method available and configured. By default it will try to use the WSGI
server’s file_wrapper support. Alternatively you can set the application’s
use_x_sendfile attribute to True to directly emit an X-Sendfile header. This how-
ever requires support of the underlying webserver for X-Sendfile.
By default it will try to guess the mimetype for you, but you can also explicitly
provide one. For extra security you probably want to send certain files as at-
tachment (HTML for instance). The mimetype guessing requires a filename or an
attachment_filename to be provided.
Please never pass filenames to this function from user sources without checking
them first. Something like this is usually sufficient to avoid security problems:
if ’..’ in filename or filename.startswith(’/’):
abort(404)
New in version 0.2.
New in version 0.5: The add_etags, cache_timeout and conditional parameters were
added. The default behavior is now to attach etags.
Changed in version 0.7: mimetype guessing and etag support for file objects was
deprecated because it was unreliable. Pass a filename if you are able to, otherwise
attach an etag yourself. This functionality will be removed in Flask 1.0
Changed in version 0.9: cache_timeout pulls its default from application config,
when None.
Parameters
215](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-225-320.jpg)
![• filename_or_fp – the filename of the file to send. This is rel-
ative to the root_path if a relative path is specified. Alterna-
tively a file object might be provided in which case X-Sendfile
might not work and fall back to the traditional method. Make
sure that the file pointer is positioned at the start of data to
send before calling send_file().
• mimetype – the mimetype of the file if provided, otherwise
auto detection happens.
• as_attachment – set to True if you want to send this file with a
Content-Disposition: attachment header.
• attachment_filename – the filename for the attachment if it dif-
fers from the file’s filename.
• add_etags – set to False to disable attaching of etags.
• conditional – set to True to enable conditional responses.
• cache_timeout – the timeout in seconds for the headers. When
None (default), this value is set by get_send_file_max_age() of
current_app.
flask.send_from_directory(directory, filename, **options)
Send a file from a given directory with send_file(). This is a secure way to
quickly expose static files from an upload folder or something similar.
Example usage:
@app.route(’/uploads/<path:filename>’)
def download_file(filename):
return send_from_directory(app.config[’UPLOAD_FOLDER’],
filename, as_attachment=True)
Sending files and Performance
It is strongly recommended to activate either X-Sendfile support in your web-
server or (if no authentication happens) to tell the webserver to serve files for
the given path on its own without calling into the web application for improved
performance.
New in version 0.5.
Parameters
• directory – the directory where all the files are stored.
• filename – the filename relative to that directory to download.
• options – optional keyword arguments that are directly for-
warded to send_file().
flask.safe_join(directory, filename)
Safely join directory and filename.
216](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-226-320.jpg)
![Example usage:
@app.route(’/wiki/<path:filename>’)
def wiki_page(filename):
filename = safe_join(app.config[’WIKI_FOLDER’], filename)
with open(filename, ’rb’) as fd:
content = fd.read() # Read and process the file content...
Parameters
• directory – the base directory.
• filename – the untrusted filename relative to that directory.
Raises NotFound if the resulting path would fall out of directory.
flask.escape(s)
Convert the characters &, <, >, ‘ and ” in string s to HTML-safe sequences. Use
this if you need to display text that might contain such characters in HTML.
Marks return value as markup string.
class flask.Markup
Marks a string as being safe for inclusion in HTML/XML output without need-
ing to be escaped. This implements the __html__ interface a couple of frame-
works and web applications use. Markup is a direct subclass of unicode and pro-
vides all the methods of unicode just that it escapes arguments passed and always
returns Markup.
The escape function returns markup objects so that double escaping can’t happen.
The constructor of the Markup class can be used for three different things: When
passed an unicode object it’s assumed to be safe, when passed an object with
an HTML representation (has an __html__ method) that representation is used,
otherwise the object passed is converted into a unicode string and then assumed
to be safe:
>>> Markup("Hello <em>World</em>!")
Markup(u’Hello <em>World</em>!’)
>>> class Foo(object):
... def __html__(self):
... return ’<a href="#">foo</a>’
...
>>> Markup(Foo())
Markup(u’<a href="#">foo</a>’)
If you want object passed being always treated as unsafe you can use the
escape() classmethod to create a Markup object:
>>> Markup.escape("Hello <em>World</em>!")
Markup(u’Hello <em>World</em>!’)
Operations on a markup string are markup aware which means that all argu-
ments are passed through the escape() function:
217](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-227-320.jpg)
![>>> em = Markup("<em>%s</em>")
>>> em % "foo & bar"
Markup(u’<em>foo & bar</em>’)
>>> strong = Markup("<strong>%(text)s</strong>")
>>> strong % {’text’: ’<blink>hacker here</blink>’}
Markup(u’<strong><blink>hacker here</blink></strong>’)
>>> Markup("<em>Hello</em> ") + "<foo>"
Markup(u’<em>Hello</em> <foo>’)
classmethod escape(s)
Escape the string. Works like escape() with the difference that for sub-
classes of Markup this function would return the correct subclass.
striptags()
Unescape markup into an unicode string and strip all tags. This also re-
solves known HTML4 and XHTML entities. Whitespace is normalized to
one:
>>> Markup("Main » <em>About</em>").striptags()
u’Main xbb About’
unescape()
Unescape markup again into an unicode string. This also resolves known
HTML4 and XHTML entities:
>>> Markup("Main » <em>About</em>").unescape()
u’Main xbb <em>About</em>’
21.10 Message Flashing
flask.flash(message, category=’message’)
Flashes a message to the next request. In order to remove the flashed mes-
sage from the session and to display it to the user, the template has to call
get_flashed_messages().
Changed in version 0.3: category parameter added.
Parameters
• message – the message to be flashed.
• category – the category for the message. The following values
are recommended: ’message’ for any kind of message, ’error’
for errors, ’info’ for information messages and ’warning’ for
warnings. However any kind of string can be used as category.
flask.get_flashed_messages(with_categories=False, category_filter=[])
Pulls all flashed messages from the session and returns them. Further calls in the
same request to the function will return the same messages. By default just the
messages are returned, but when with_categories is set to True, the return value
will be a list of tuples in the form (category, message) instead.
218](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-228-320.jpg)
![DEBUG = True
SECRET_KEY = ’development key’
app.config.from_object(__name__)
In both cases (loading from any Python file or loading from modules), only up-
percase keys are added to the config. This makes it possible to use lowercase
values in the config file for temporary values that are not added to the config or
to define the config keys in the same file that implements the application.
Probably the most interesting way to load configurations is from an environment
variable pointing to a file:
app.config.from_envvar(’YOURAPPLICATION_SETTINGS’)
In this case before launching the application you have to set this environment
variable to the file you want to use. On Linux and OS X use the export statement:
export YOURAPPLICATION_SETTINGS=’/path/to/config/file’
On windows use set instead.
Parameters
• root_path – path to which files are read relative from. When
the config object is created by the application, this is the appli-
cation’s root_path.
• defaults – an optional dictionary of default values
from_envvar(variable_name, silent=False)
Loads a configuration from an environment variable pointing to a configu-
ration file. This is basically just a shortcut with nicer error messages for this
line of code:
app.config.from_pyfile(os.environ[’YOURAPPLICATION_SETTINGS’])
Parameters
• variable_name – name of the environment variable
• silent – set to True if you want silent failure for missing files.
Returns bool. True if able to load config, False otherwise.
from_object(obj)
Updates the values from the given object. An object can be of one of the
following two types:
•a string: in this case the object with that name will be imported
•an actual object reference: that object is used directly
Objects are usually either modules or classes.
223](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-233-320.jpg)
![This function however can help you keep the context around for longer:
from flask import stream_with_context, request, Response
@app.route(’/stream’)
def streamed_response():
@stream_with_context
def generate():
yield ’Hello ’
yield request.args[’name’]
yield ’!’
return Response(generate())
Alternatively it can also be used around a specific generator:
from flask import stream_with_context, request, Response
@app.route(’/stream’)
def streamed_response():
def generate():
yield ’Hello ’
yield request.args[’name’]
yield ’!’
return Response(stream_with_context(generate()))
New in version 0.9.
21.16 Useful Internals
class flask.ctx.RequestContext(app, environ, request=None)
The request context contains all request relevant information. It is created at the
beginning of the request and pushed to the _request_ctx_stack and removed at
the end of it. It will create the URL adapter and request object for the WSGI
environment provided.
Do not attempt to use this class directly, instead use test_request_context()
and request_context() to create this object.
When the request context is popped, it will evaluate all the functions registered
on the application for teardown execution (teardown_request()).
The request context is automatically popped at the end of the request for you.
In debug mode the request context is kept around if exceptions happen so that
interactive debuggers have a chance to introspect the data. With 0.4 this can also
be forced for requests that did not fail and outside of DEBUG mode. By setting
’flask._preserve_context’ to True on the WSGI environment the context will
not pop itself at the end of the request. This is used by the test_client() for
example to implement the deferred cleanup functionality.
You might find this helpful for unittests where you need the information from
the context local around for a little longer. Make sure to properly pop() the stack
225](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-235-320.jpg)
![class flask.signals.Namespace
An alias for blinker.base.Namespace if blinker is available, otherwise a dummy
class that creates fake signals. This class is available for Flask extensions that
want to provide the same fallback system as Flask itself.
signal(name, doc=None)
Creates a new signal for this namespace if blinker is available, otherwise
returns a fake signal that has a send method that will do nothing but will
fail with a RuntimeError for all other operations, including connecting.
21.18 Class-Based Views
New in version 0.7.
class flask.views.View
Alternative way to use view functions. A subclass has to implement
dispatch_request() which is called with the view arguments from the URL rout-
ing system. If methods is provided the methods do not have to be passed to the
add_url_rule() method explicitly:
class MyView(View):
methods = [’GET’]
def dispatch_request(self, name):
return ’Hello %s!’ % name
app.add_url_rule(’/hello/<name>’, view_func=MyView.as_view(’myview’))
When you want to decorate a pluggable view you will have to either do that
when the view function is created (by wrapping the return value of as_view())
or you can use the decorators attribute:
class SecretView(View):
methods = [’GET’]
decorators = [superuser_required]
def dispatch_request(self):
...
The decorators stored in the decorators list are applied one after another when
the view function is created. Note that you can not use the class based decorators
since those would decorate the view class and not the generated view function!
classmethod as_view(name, *class_args, **class_kwargs)
Converts the class into an actual view function that can be used with the
routing system. Internally this generates a function on the fly which will in-
stantiate the View on each request and call the dispatch_request() method
on it.
The arguments passed to as_view() are forwarded to the constructor of the
class.
229](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-239-320.jpg)
![decorators = []
The canonical way to decorate class-based views is to decorate the return
value of as_view(). However since this moves parts of the logic from the
class declaration to the place where it’s hooked into the routing system.
You can place one or more decorators in this list and whenever the view
function is created the result is automatically decorated.
New in version 0.8.
dispatch_request()
Subclasses have to override this method to implement the actual view func-
tion code. This method is called with all the arguments from the URL rule.
methods = None
A for which methods this pluggable view can handle.
class flask.views.MethodView
Like a regular class-based view but that dispatches requests to particular meth-
ods. For instance if you implement a method called get() it means you will
response to ’GET’ requests and the dispatch_request() implementation will au-
tomatically forward your request to that. Also options is set for you automati-
cally:
class CounterAPI(MethodView):
def get(self):
return session.get(’counter’, 0)
def post(self):
session[’counter’] = session.get(’counter’, 0) + 1
return ’OK’
app.add_url_rule(’/counter’, view_func=CounterAPI.as_view(’counter’))
21.19 URL Route Registrations
Generally there are three ways to define rules for the routing system:
1. You can use the flask.Flask.route() decorator.
2. You can use the flask.Flask.add_url_rule() function.
3. You can directly access the underlying Werkzeug routing system which is ex-
posed as flask.Flask.url_map.
Variable parts in the route can be specified with angular brackets (/user/<username>).
By default a variable part in the URL accepts any string without a slash however a
different converter can be specified as well by using <converter:name>.
Variable parts are passed to the view function as keyword arguments.
The following converters are available:
230](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-240-320.jpg)
![return ’Hello World!’
index.provide_automatic_options = False
index.methods = [’GET’, ’OPTIONS’]
app.add_url_rule(’/’, index)
New in version 0.8: The provide_automatic_options functionality was added.
233](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-243-320.jpg)
![problems and the only thing that could possibly break are very large responses that
can cause some kind of denial of service at the receiver’s side.
However there is a catch. Due to how browsers work the CSRF issue comes up with
JSON unfortunately. Fortunately there is also a weird part of the JavaScript specifica-
tion that can be used to solve that problem easily and Flask is kinda doing that for you
by preventing you from doing dangerous stuff. Unfortunately that protection is only
there for jsonify() so you are still at risk when using other ways to generate JSON.
So what is the issue and how to avoid it? The problem are arrays at top-level in JSON.
Imagine you send the following data out in a JSON request. Say that’s exporting the
names and email addresses of all your friends for a part of the user interface that is
written in JavaScript. Not very uncommon:
[
{"username": "admin",
"email": "admin@localhost"}
]
And it is doing that of course only as long as you are logged in and only for you. And
it is doing that for all GET requests to a certain URL, say the URL for that request is
http://example.com/api/get_friends.json.
So now what happens if a clever hacker is embedding this to his website and social
engineers a victim to visiting his site:
<script type=text/javascript>
var captured = [];
var oldArray = Array;
function Array() {
var obj = this, id = 0, capture = function(value) {
obj.__defineSetter__(id++, capture);
if (value)
captured.push(value);
};
capture();
}
</script>
<script type=text/javascript
src=http://example.com/api/get_friends.json></script>
<script type=text/javascript>
Array = oldArray;
// now we have all the data in the captured array.
</script>
If you know a bit of JavaScript internals you might know that it’s possible to patch
constructors and register callbacks for setters. An attacker can use this (like above)
to get all the data you exported in your JSON file. The browser will totally ignore
the application/json mimetype if text/javascript is defined as content type in the
script tag and evaluate that as JavaScript. Because top-level array elements are allowed
(albeit useless) and we hooked in our own constructor, after that page loaded the data
from the JSON response is in the captured array.
247](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-257-320.jpg)
![Flask extensions must be licensed under a BSD, MIT or more liberal license to be able
to be enlisted in the Flask Extension Registry. Keep in mind that the Flask Extension
Registry is a moderated place and libraries will be reviewed upfront if they behave as
required.
26.2 “Hello Flaskext!”
So let’s get started with creating such a Flask extension. The extension we want to
create here will provide very basic support for SQLite3.
First we create the following folder structure:
flask-sqlite3/
flask_sqlite3.py
LICENSE
README
Here’s the contents of the most important files:
26.2.1 setup.py
The next file that is absolutely required is the setup.py file which is used to install your
Flask extension. The following contents are something you can work with:
"""
Flask-SQLite3
-------------
This is the description for that library
"""
from setuptools import setup
setup(
name=’Flask-SQLite3’,
version=’1.0’,
url=’http://example.com/flask-sqlite3/’,
license=’BSD’,
author=’Your Name’,
author_email=’your-email@example.com’,
description=’Very short description’,
long_description=__doc__,
py_modules=[’flask_sqlite3’],
# if you would be using a package instead use packages instead
# of py_modules:
# packages=[’flask_sqlite3’],
zip_safe=False,
include_package_data=True,
platforms=’any’,
254](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-264-320.jpg)
![install_requires=[
’Flask’
],
classifiers=[
’Environment :: Web Environment’,
’Intended Audience :: Developers’,
’License :: OSI Approved :: BSD License’,
’Operating System :: OS Independent’,
’Programming Language :: Python’,
’Topic :: Internet :: WWW/HTTP :: Dynamic Content’,
’Topic :: Software Development :: Libraries :: Python Modules’
]
)
That’s a lot of code but you can really just copy/paste that from existing extensions
and adapt.
26.2.2 flask_sqlite3.py
Now this is where your extension code goes. But how exactly should such an exten-
sion look like? What are the best practices? Continue reading for some insight.
26.3 Initializing Extensions
Many extensions will need some kind of initialization step. For example, consider an
application that’s currently connecting to SQLite like the documentation suggests (Us-
ing SQLite 3 with Flask). So how does the extension know the name of the application
object?
Quite simple: you pass it to it.
There are two recommended ways for an extension to initialize:
initialization functions:
If your extension is called helloworld you might have a function called
init_helloworld(app[, extra_args]) that initializes the extension for that
application. It could attach before / after handlers etc.
classes:
Classes work mostly like initialization functions but can later be used to
further change the behavior. For an example look at how the OAuth exten-
sion works: there is an OAuth object that provides some helper functions
like OAuth.remote_app to create a reference to a remote application that uses
OAuth.
What to use depends on what you have in mind. For the SQLite 3 extension we will
use the class-based approach because it will provide users with an object that handles
opening and closing database connections.
255](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-265-320.jpg)
![What’s important about classes is that they encourage to be shared around on mod-
ule level. In that case, the object itself must not under any circumstances store any
application specific state and must be shareable between different application.
26.4 The Extension Code
Here’s the contents of the flask_sqlite3.py for copy/paste:
import sqlite3
from flask import current_app
# Find the stack on which we want to store the database connection.
# Starting with Flask 0.9, the _app_ctx_stack is the correct one,
# before that we need to use the _request_ctx_stack.
try:
from flask import _app_ctx_stack as stack
except ImportError:
from flask import _request_ctx_stack as stack
class SQLite3(object):
def __init__(self, app=None):
self.app = app
if app is not None:
self.init_app(app)
def init_app(self, app):
app.config.setdefault(’SQLITE3_DATABASE’, ’:memory:’)
# Use the newstyle teardown_appcontext if it’s available,
# otherwise fall back to the request context
if hasattr(app, ’teardown_appcontext’):
app.teardown_appcontext(self.teardown)
else:
app.teardown_request(self.teardown)
def connect(self):
return sqlite3.connect(current_app.config[’SQLITE3_DATABASE’])
def teardown(self, exception):
ctx = stack.top
if hasattr(ctx, ’sqlite3_db’):
ctx.sqlite3_db.close()
@property
def connection(self):
ctx = stack.top
if ctx is not None:
if not hasattr(ctx, ’sqlite3_db’):
256](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-266-320.jpg)
![CHAPTER
TWENTYSEVEN
POCOO STYLEGUIDE
The Pocoo styleguide is the styleguide for all Pocoo Projects, including Flask. This
styleguide is a requirement for Patches to Flask and a recommendation for Flask ex-
tensions.
In general the Pocoo Styleguide closely follows PEP 8 with some small differences and
extensions.
27.1 General Layout
Indentation: 4 real spaces. No tabs, no exceptions.
Maximum line length: 79 characters with a soft limit for 84 if absolutely necessary.
Try to avoid too nested code by cleverly placing break, continue and return state-
ments.
Continuing long statements: To continue a statement you can use backslashes in
which case you should align the next line with the last dot or equal sign, or
indent four spaces:
this_is_a_very_long(function_call, ’with many parameters’)
.that_returns_an_object_with_an_attribute
MyModel.query.filter(MyModel.scalar > 120)
.order_by(MyModel.name.desc())
.limit(10)
If you break in a statement with parentheses or braces, align to the braces:
this_is_a_very_long(function_call, ’with many parameters’,
23, 42, ’and even more’)
For lists or tuples with many items, break immediately after the opening brace:
items = [
’this is the first’, ’set of items’, ’with more items’,
’to come in this line’, ’like this’
]
261](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-271-320.jpg)
![Blank lines: Top level functions and classes are separated by two lines, everything
else by one. Do not use too many blank lines to separate logical segments in
code. Example:
def hello(name):
print ’Hello %s!’ % name
def goodbye(name):
print ’See you %s.’ % name
class MyClass(object):
"""This is a simple docstring"""
def __init__(self, name):
self.name = name
def get_annoying_name(self):
return self.name.upper() + ’!!!!111’
27.2 Expressions and Statements
General whitespace rules:
• No whitespace for unary operators that are not words (e.g.: -, ~ etc.) as well
on the inner side of parentheses.
• Whitespace is placed between binary operators.
Good:
exp = -1.05
value = (item_value / item_count) * offset / exp
value = my_list[index]
value = my_dict[’key’]
Bad:
exp = - 1.05
value = ( item_value / item_count ) * offset / exp
value = (item_value/item_count)*offset/exp
value=( item_value/item_count ) * offset/exp
value = my_list[ index ]
value = my_dict [’key’]
Yoda statements are a no-go: Never compare constant with variable, always variable
with constant:
Good:
262](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-272-320.jpg)
![29.4.3 Manual Error Handler Attaching
While it is still possible to attach error handlers to Flask.error_handlers it’s discour-
aged to do so and in fact deprecated. In general we no longer recommend custom
error handler attaching via assignments to the underlying dictionary due to the more
complex internal handling to support arbitrary exception classes and blueprints. See
Flask.errorhandler() for more information.
The proper upgrade is to change this:
app.error_handlers[403] = handle_error
Into this:
app.register_error_handler(403, handle_error)
Alternatively you should just attach the function with a decorator:
@app.errorhandler(403)
def handle_error(e):
...
(Note that register_error_handler() is new in Flask 0.7)
29.4.4 Blueprint Support
Blueprints replace the previous concept of “Modules” in Flask. They provide better
semantics for various features and work better with large applications. The update
script provided should be able to upgrade your applications automatically, but there
might be some cases where it fails to upgrade. What changed?
• Blueprints need explicit names. Modules had an automatic name guesssing
scheme where the shortname for the module was taken from the last part of
the import module. The upgrade script tries to guess that name but it might fail
as this information could change at runtime.
• Blueprints have an inverse behavior for url_for(). Previously .foo told
url_for() that it should look for the endpoint foo on the application. Now it
means “relative to current module”. The script will inverse all calls to url_for()
automatically for you. It will do this in a very eager way so you might end up
with some unnecessary leading dots in your code if you’re not using modules.
• Blueprints do not automatically provide static folders. They will also no longer
automatically export templates from a folder called templates next to their loca-
tion however but it can be enabled from the constructor. Same with static files: if
you want to continue serving static files you need to tell the constructor explic-
itly the path to the static folder (which can be relative to the blueprint’s module
path).
• Rendering templates was simplified. Now the blueprints can provide template
folders which are added to a general template searchpath. This means that you
271](https://image.slidesharecdn.com/flask-docs-140615143817-phpapp01/85/Flask-docs-281-320.jpg)
Ad
More Related Content
What's hot (19)
Similar to Flask docs (20)
Ad
Recently uploaded (20)
![Metal dept project[462].ppt metal manufacturing assembly](https://cdn.slidesharecdn.com/ss_thumbnails/metaldeptproject462-250429085341-3dc6625e-thumbnail.jpg?width=560&fit=bounds)
Ad
Flask docs
- 1. Flask Documentation Release 0.11-dev February 08, 2014
- 3. CONTENTS I User’s Guide 1 1 Foreword 3 1.1 What does “micro” mean? . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.2 Configuration and Conventions . . . . . . . . . . . . . . . . . . . . . . . 3 1.3 Growing with Flask . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2 Foreword for Experienced Programmers 5 2.1 Thread-Locals in Flask . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.2 Develop for the Web with Caution . . . . . . . . . . . . . . . . . . . . . . 5 2.3 The Status of Python 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3 Installation 7 3.1 virtualenv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 3.2 System-Wide Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3.3 Living on the Edge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 3.4 pip and distribute on Windows . . . . . . . . . . . . . . . . . . . . . . . . 9 4 Quickstart 11 4.1 A Minimal Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 4.2 Debug Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 4.3 Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 4.4 Static Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 4.5 Rendering Templates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 4.6 Accessing Request Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 4.7 Redirects and Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 4.8 About Responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 4.9 Sessions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 4.10 Message Flashing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 4.11 Logging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 4.12 Hooking in WSGI Middlewares . . . . . . . . . . . . . . . . . . . . . . . 25 4.13 Deploying to a Web Server . . . . . . . . . . . . . . . . . . . . . . . . . . 26 5 Tutorial 27 5.1 Introducing Flaskr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 5.2 Step 0: Creating The Folders . . . . . . . . . . . . . . . . . . . . . . . . . 28 5.3 Step 1: Database Schema . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 i
- 4. 5.4 Step 2: Application Setup Code . . . . . . . . . . . . . . . . . . . . . . . . 29 5.5 Step 3: Database Connections . . . . . . . . . . . . . . . . . . . . . . . . . 31 5.6 Step 4: Creating The Database . . . . . . . . . . . . . . . . . . . . . . . . 32 5.7 Step 5: The View Functions . . . . . . . . . . . . . . . . . . . . . . . . . . 34 5.8 Step 6: The Templates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 5.9 Step 7: Adding Style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 5.10 Bonus: Testing the Application . . . . . . . . . . . . . . . . . . . . . . . . 38 6 Templates 39 6.1 Jinja Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 6.2 Standard Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 6.3 Standard Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 6.4 Controlling Autoescaping . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 6.5 Registering Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 6.6 Context Processors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 7 Testing Flask Applications 43 7.1 The Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 7.2 The Testing Skeleton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 7.3 The First Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 7.4 Logging In and Out . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 7.5 Test Adding Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 7.6 Other Testing Tricks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 7.7 Faking Resources and Context . . . . . . . . . . . . . . . . . . . . . . . . 47 7.8 Keeping the Context Around . . . . . . . . . . . . . . . . . . . . . . . . . 48 7.9 Accessing and Modifying Sessions . . . . . . . . . . . . . . . . . . . . . . 49 8 Logging Application Errors 51 8.1 Error Mails . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 8.2 Logging to a File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 8.3 Controlling the Log Format . . . . . . . . . . . . . . . . . . . . . . . . . . 53 8.4 Other Libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 9 Debugging Application Errors 57 9.1 When in Doubt, Run Manually . . . . . . . . . . . . . . . . . . . . . . . . 57 9.2 Working with Debuggers . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 10 Configuration Handling 59 10.1 Configuration Basics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 10.2 Builtin Configuration Values . . . . . . . . . . . . . . . . . . . . . . . . . 60 10.3 Configuring from Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 10.4 Configuration Best Practices . . . . . . . . . . . . . . . . . . . . . . . . . 63 10.5 Development / Production . . . . . . . . . . . . . . . . . . . . . . . . . . 63 10.6 Instance Folders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 11 Signals 67 11.1 Subscribing to Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 11.2 Creating Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 11.3 Sending Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 ii
- 5. 11.4 Signals and Flask’s Request Context . . . . . . . . . . . . . . . . . . . . . 70 11.5 Decorator Based Signal Subscriptions . . . . . . . . . . . . . . . . . . . . 70 11.6 Core Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 12 Pluggable Views 75 12.1 Basic Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 12.2 Method Hints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 12.3 Method Based Dispatching . . . . . . . . . . . . . . . . . . . . . . . . . . 77 12.4 Decorating Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 12.5 Method Views for APIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 13 The Application Context 81 13.1 Purpose of the Application Context . . . . . . . . . . . . . . . . . . . . . 81 13.2 Creating an Application Context . . . . . . . . . . . . . . . . . . . . . . . 82 13.3 Locality of the Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 13.4 Context Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 14 The Request Context 85 14.1 Diving into Context Locals . . . . . . . . . . . . . . . . . . . . . . . . . . 85 14.2 How the Context Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 14.3 Callbacks and Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 14.4 Teardown Callbacks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 14.5 Notes On Proxies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 14.6 Context Preservation on Error . . . . . . . . . . . . . . . . . . . . . . . . 88 15 Modular Applications with Blueprints 91 15.1 Why Blueprints? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 15.2 The Concept of Blueprints . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 15.3 My First Blueprint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 15.4 Registering Blueprints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 15.5 Blueprint Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 15.6 Building URLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 16 Flask Extensions 97 16.1 Finding Extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 16.2 Using Extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 16.3 Flask Before 0.8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 17 Working with the Shell 99 17.1 Creating a Request Context . . . . . . . . . . . . . . . . . . . . . . . . . . 99 17.2 Firing Before/After Request . . . . . . . . . . . . . . . . . . . . . . . . . 100 17.3 Further Improving the Shell Experience . . . . . . . . . . . . . . . . . . . 100 18 Patterns for Flask 101 18.1 Larger Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 18.2 Application Factories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 18.3 Application Dispatching . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 18.4 Implementing API Exceptions . . . . . . . . . . . . . . . . . . . . . . . . 108 18.5 Using URL Processors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 iii
- 6. 18.6 Deploying with Distribute . . . . . . . . . . . . . . . . . . . . . . . . . . 112 18.7 Deploying with Fabric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 18.8 Using SQLite 3 with Flask . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 18.9 SQLAlchemy in Flask . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 18.10 Uploading Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 18.11 Caching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 18.12 View Decorators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 18.13 Form Validation with WTForms . . . . . . . . . . . . . . . . . . . . . . . 132 18.14 Template Inheritance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 18.15 Message Flashing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 18.16 AJAX with jQuery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 18.17 Custom Error Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 18.18 Lazily Loading Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 18.19 MongoKit in Flask . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 18.20 Adding a favicon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 18.21 Streaming Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 18.22 Deferred Request Callbacks . . . . . . . . . . . . . . . . . . . . . . . . . . 150 18.23 Adding HTTP Method Overrides . . . . . . . . . . . . . . . . . . . . . . 151 18.24 Request Content Checksums . . . . . . . . . . . . . . . . . . . . . . . . . 152 18.25 Celery Based Background Tasks . . . . . . . . . . . . . . . . . . . . . . . 153 19 Deployment Options 157 19.1 mod_wsgi (Apache) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 19.2 Standalone WSGI Containers . . . . . . . . . . . . . . . . . . . . . . . . . 160 19.3 uWSGI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 19.4 FastCGI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164 19.5 CGI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 20 Becoming Big 171 20.1 Read the Source. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 20.2 Hook. Extend. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 20.3 Subclass. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 20.4 Wrap with middleware. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 20.5 Fork. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 20.6 Scale like a pro. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 20.7 Discuss with the community. . . . . . . . . . . . . . . . . . . . . . . . . . 173 II API Reference 175 21 API 177 21.1 Application Object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 21.2 Blueprint Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 21.3 Incoming Request Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 21.4 Response Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 21.5 Sessions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 21.6 Session Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 21.7 Test Client . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 21.8 Application Globals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 iv
- 7. 21.9 Useful Functions and Classes . . . . . . . . . . . . . . . . . . . . . . . . . 211 21.10 Message Flashing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 21.11 JSON Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 21.12 Template Rendering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 21.13 Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 21.14 Extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 21.15 Stream Helpers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 21.16 Useful Internals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 21.17 Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 21.18 Class-Based Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 21.19 URL Route Registrations . . . . . . . . . . . . . . . . . . . . . . . . . . . 230 21.20 View Function Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 III Additional Notes 235 22 Design Decisions in Flask 237 22.1 The Explicit Application Object . . . . . . . . . . . . . . . . . . . . . . . . 237 22.2 The Routing System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 22.3 One Template Engine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 22.4 Micro with Dependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . 239 22.5 Thread Locals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 22.6 What Flask is, What Flask is Not . . . . . . . . . . . . . . . . . . . . . . . 240 23 HTML/XHTML FAQ 241 23.1 History of XHTML . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 23.2 History of HTML5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 23.3 HTML versus XHTML . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 23.4 What does “strict” mean? . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 23.5 New technologies in HTML5 . . . . . . . . . . . . . . . . . . . . . . . . . 244 23.6 What should be used? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 24 Security Considerations 245 24.1 Cross-Site Scripting (XSS) . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 24.2 Cross-Site Request Forgery (CSRF) . . . . . . . . . . . . . . . . . . . . . . 246 24.3 JSON Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 25 Unicode in Flask 249 25.1 Automatic Conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 25.2 The Golden Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250 25.3 Encoding and Decoding Yourself . . . . . . . . . . . . . . . . . . . . . . . 250 25.4 Configuring Editors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250 26 Flask Extension Development 253 26.1 Anatomy of an Extension . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 26.2 “Hello Flaskext!” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 26.3 Initializing Extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 26.4 The Extension Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256 26.5 Using _app_ctx_stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 v
- 8. 26.6 Teardown Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 26.7 Learn from Others . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 26.8 Approved Extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 26.9 Extension Import Transition . . . . . . . . . . . . . . . . . . . . . . . . . 260 27 Pocoo Styleguide 261 27.1 General Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 27.2 Expressions and Statements . . . . . . . . . . . . . . . . . . . . . . . . . . 262 27.3 Naming Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 27.4 Docstrings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 27.5 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 28 Python 3 Support 265 28.1 Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265 28.2 API Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265 28.3 Few Users . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265 28.4 Small Ecosystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 28.5 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 29 Upgrading to Newer Releases 267 29.1 Version 0.10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267 29.2 Version 0.9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268 29.3 Version 0.8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268 29.4 Version 0.7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268 29.5 Version 0.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 29.6 Version 0.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 29.7 Version 0.4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 29.8 Version 0.3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 30 Flask Changelog 275 30.1 Version 1.0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 30.2 Version 0.10.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 30.3 Version 0.10.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276 30.4 Version 0.10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276 30.5 Version 0.9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278 30.6 Version 0.8.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 30.7 Version 0.8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 30.8 Version 0.7.3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 30.9 Version 0.7.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 30.10 Version 0.7.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 30.11 Version 0.7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 30.12 Version 0.6.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 30.13 Version 0.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 30.14 Version 0.5.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 30.15 Version 0.5.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 30.16 Version 0.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284 30.17 Version 0.4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284 30.18 Version 0.3.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284 30.19 Version 0.3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 vi
- 9. 30.20 Version 0.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 30.21 Version 0.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286 31 License 287 31.1 Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 31.2 General License Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . 288 31.3 Flask License . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 31.4 Flask Artwork License . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289 Index 291 vii
- 10. viii
- 11. Part I USER’S GUIDE This part of the documentation, which is mostly prose, begins with some background information about Flask, then focuses on step-by-step instructions for web develop- ment with Flask. 1
- 12. 2
- 13. CHAPTER ONE FOREWORD Read this before you get started with Flask. This hopefully answers some questions about the purpose and goals of the project, and when you should or should not be using it. 1.1 What does “micro” mean? “Micro” does not mean that your whole web application has to fit into a single Python file, although it certainly can. Nor does it mean that Flask is lacking in functionality. The “micro” in microframework means Flask aims to keep the core simple but exten- sible. Flask won’t make many decisions for you, such as what database to use. Those decisions that it does make, such as what templating engine to use, are easy to change. Everything else is up to you, so that Flask can be everything you need and nothing you don’t. By default, Flask does not include a database abstraction layer, form validation or any- thing else where different libraries already exist that can handle that. Instead, Flask supports extensions to add such functionality to your application as if it was imple- mented in Flask itself. Numerous extensions provide database integration, form val- idation, upload handling, various open authentication technologies, and more. Flask may be “micro”, but it’s ready for production use on a variety of needs. 1.2 Configuration and Conventions Flask has many configuration values, with sensible defaults, and a few conventions when getting started. By convention templates and static files are stored in subdirec- tories within the application’s Python source tree, with the names templates and static respectively. While this can be changed you usually don’t have to, especially when getting started. 3
- 14. 1.3 Growing with Flask Once you have Flask up and running, you’ll find a variety of extensions available in the community to integrate your project for production. The Flask core team reviews extensions and ensures approved extensions do not break with future releases. As your codebase grows, you are free to make the design decisions appropriate for your project. Flask will continue to provide a very simple glue layer to the best that Python has to offer. You can implement advanced patterns in SQLAlchemy or an- other database tool, introduce non-relational data persistence as appropriate, and take advantage of framework-agnostic tools built for WSGI, the Python web interface. Flask includes many hooks to customize its behavior. Should you need more cus- tomization, the Flask class is built for subclassing. If you are interested in that, check out the Becoming Big chapter. If you are curious about the Flask design principles, head over to the section about Design Decisions in Flask. Continue to Installation, the Quickstart, or the Foreword for Experienced Programmers. 4
- 15. CHAPTER TWO FOREWORD FOR EXPERIENCED PROGRAMMERS 2.1 Thread-Locals in Flask One of the design decisions in Flask was that simple tasks should be simple; they should not take a lot of code and yet they should not limit you. Because of that, Flask has few design choices that some people might find surprising or unorthodox. For ex- ample, Flask uses thread-local objects internally so that you don’t have to pass objects around from function to function within a request in order to stay threadsafe. This approach is convenient, but requires a valid request context for dependency injection or when attempting to reuse code which uses a value pegged to the request. The Flask project is honest about thread-locals, does not hide them, and calls out in the code and documentation where they are used. 2.2 Develop for the Web with Caution Always keep security in mind when building web applications. If you write a web application, you are probably allowing users to register and leave their data on your server. The users are entrusting you with data. And even if you are the only user that might leave data in your application, you still want that data to be stored securely. Unfortunately, there are many ways the security of a web application can be com- promised. Flask protects you against one of the most common security problems of modern web applications: cross-site scripting (XSS). Unless you deliberately mark in- secure HTML as secure, Flask and the underlying Jinja2 template engine have you covered. But there are many more ways to cause security problems. The documentation will warn you about aspects of web development that require at- tention to security. Some of these security concerns are far more complex than one might think, and we all sometimes underestimate the likelihood that a vulnerability will be exploited - until a clever attacker figures out a way to exploit our applications. And don’t think that your application is not important enough to attract an attacker. 5
- 16. Depending on the kind of attack, chances are that automated bots are probing for ways to fill your database with spam, links to malicious software, and the like. Flask is no different from any other framework in that you the developer must build with caution, watching for exploits when building to your requirements. 2.3 The Status of Python 3 Currently the Python community is in the process of improving libraries to support the new iteration of the Python programming language. While the situation is greatly improving there are still some issues that make it hard for users to switch over to Python 3 just now. These problems are partially caused by changes in the language that went unreviewed for too long, partially also because we have not quite worked out how the lower- level API should change to account for the Unicode differences in Python 3. We strongly recommend using Python 2.6 and 2.7 with activated Python 3 warnings during development. If you plan on upgrading to Python 3 in the near future we strongly recommend that you read How to write forwards compatible Python code. If you do want to dive into Python 3 already have a look at the Python 3 Support page. Continue to Installation or the Quickstart. 6
- 17. CHAPTER THREE INSTALLATION Flask depends on two external libraries, Werkzeug and Jinja2. Werkzeug is a toolkit for WSGI, the standard Python interface between web applications and a variety of servers for both development and deployment. Jinja2 renders templates. So how do you get all that on your computer quickly? There are many ways you could do that, but the most kick-ass method is virtualenv, so let’s have a look at that first. You will need Python 2.6 or higher to get started, so be sure to have an up-to-date Python 2.x installation. For using Flask with Python 3 have a look at Python 3 Support. 3.1 virtualenv Virtualenv is probably what you want to use during development, and if you have shell access to your production machines, you’ll probably want to use it there, too. What problem does virtualenv solve? If you like Python as much as I do, chances are you want to use it for other projects besides Flask-based web applications. But the more projects you have, the more likely it is that you will be working with different versions of Python itself, or at least different versions of Python libraries. Let’s face it: quite often libraries break backwards compatibility, and it’s unlikely that any serious application will have zero dependencies. So what do you do if two or more of your projects have conflicting dependencies? Virtualenv to the rescue! Virtualenv enables multiple side-by-side installations of Python, one for each project. It doesn’t actually install separate copies of Python, but it does provide a clever way to keep different project environments isolated. Let’s see how virtualenv works. If you are on Mac OS X or Linux, chances are that one of the following two commands will work for you: $ sudo easy_install virtualenv or even better: $ sudo pip install virtualenv 7
- 18. One of these will probably install virtualenv on your system. Maybe it’s even in your package manager. If you use Ubuntu, try: $ sudo apt-get install python-virtualenv If you are on Windows and don’t have the easy_install command, you must install it first. Check the pip and distribute on Windows section for more information about how to do that. Once you have it installed, run the same commands as above, but without the sudo prefix. Once you have virtualenv installed, just fire up a shell and create your own environ- ment. I usually create a project folder and a venv folder within: $ mkdir myproject $ cd myproject $ virtualenv venv New python executable in venv/bin/python Installing distribute............done. Now, whenever you want to work on a project, you only have to activate the corre- sponding environment. On OS X and Linux, do the following: $ . venv/bin/activate If you are a Windows user, the following command is for you: $ venvscriptsactivate Either way, you should now be using your virtualenv (notice how the prompt of your shell has changed to show the active environment). Now you can just enter the following command to get Flask activated in your vir- tualenv: $ pip install Flask A few seconds later and you are good to go. 3.2 System-Wide Installation This is possible as well, though I do not recommend it. Just run pip with root privileges: $ sudo pip install Flask (On Windows systems, run it in a command-prompt window with administrator priv- ileges, and leave out sudo.) 8
- 19. 3.3 Living on the Edge If you want to work with the latest version of Flask, there are two ways: you can either let pip pull in the development version, or you can tell it to operate on a git checkout. Either way, virtualenv is recommended. Get the git checkout in a new virtualenv and run in development mode: $ git clone http://github.com/mitsuhiko/flask.git Initialized empty Git repository in ~/dev/flask/.git/ $ cd flask $ virtualenv venv --distribute New python executable in venv/bin/python Installing distribute............done. $ . venv/bin/activate $ python setup.py develop ... Finished processing dependencies for Flask This will pull in the dependencies and activate the git head as the current version inside the virtualenv. Then all you have to do is run git pull origin to update to the latest version. To just get the development version without git, do this instead: $ mkdir flask $ cd flask $ virtualenv venv --distribute $ . venv/bin/activate New python executable in venv/bin/python Installing distribute............done. $ pip install Flask==dev ... Finished processing dependencies for Flask==dev 3.4 pip and distribute on Windows On Windows, installation of easy_install is a little bit trickier, but still quite easy. The easiest way to do it is to download the distribute_setup.py file and run it. The easiest way to run the file is to open your downloads folder and double-click on the file. Next, add the easy_install command and other Python scripts to the command search path, by adding your Python installation’s Scripts folder to the PATH environment variable. To do that, right-click on the “Computer” icon on the Desktop or in the Start menu, and choose “Properties”. Then click on “Advanced System settings” (in Windows XP, click on the “Advanced” tab instead). Then click on the “Environment variables” button. Finally, double-click on the “Path” variable in the “System vari- ables” section, and add the path of your Python interpreter’s Scripts folder. Be sure to 9
- 20. delimit it from existing values with a semicolon. Assuming you are using Python 2.7 on the default path, add the following value: ;C:Python27Scripts And you are done! To check that it worked, open the Command Prompt and exe- cute easy_install. If you have User Account Control enabled on Windows Vista or Windows 7, it should prompt you for administrator privileges. Now that you have easy_install, you can use it to install pip: > easy_install pip 10
- 21. CHAPTER FOUR QUICKSTART Eager to get started? This page gives a good introduction to Flask. It assumes you already have Flask installed. If you do not, head over to the Installation section. 4.1 A Minimal Application A minimal Flask application looks something like this: from flask import Flask app = Flask(__name__) @app.route(’/’) def hello_world(): return ’Hello World!’ if __name__ == ’__main__’: app.run() Just save it as hello.py (or something similar) and run it with your Python interpreter. Make sure to not call your application flask.py because this would conflict with Flask itself. $ python hello.py * Running on http://127.0.0.1:5000/ Now head over to http://127.0.0.1:5000/, and you should see your hello world greet- ing. So what did that code do? 1. First we imported the Flask class. An instance of this class will be our WSGI application. 2. Next we create an instance of this class. The first argument is the name of the application’s module or package. If you are using a single module (as in this example), you should use __name__ because depending on if it’s started as ap- plication or imported as module the name will be different (’__main__’ versus the actual import name). This is needed so that Flask knows where to look for 11
- 22. templates, static files, and so on. For more information have a look at the Flask documentation. 3. We then use the route() decorator to tell Flask what URL should trigger our function. 4. The function is given a name which is also used to generate URLs for that partic- ular function, and returns the message we want to display in the user’s browser. 5. Finally we use the run() function to run the local server with our application. The if __name__ == ’__main__’: makes sure the server only runs if the script is executed directly from the Python interpreter and not used as an imported module. To stop the server, hit control-C. Externally Visible Server If you run the server you will notice that the server is only accessible from your own computer, not from any other in the network. This is the default because in debugging mode a user of the application can execute arbitrary Python code on your computer. If you have debug disabled or trust the users on your network, you can make the server publicly available simply by changing the call of the run() method to look like this: app.run(host=’0.0.0.0’) This tells your operating system to listen on all public IPs. 4.2 Debug Mode The run() method is nice to start a local development server, but you would have to restart it manually after each change to your code. That is not very nice and Flask can do better. If you enable debug support the server will reload itself on code changes, and it will also provide you with a helpful debugger if things go wrong. There are two ways to enable debugging. Either set that flag on the application object: app.debug = True app.run() Or pass it as a parameter to run: app.run(debug=True) Both methods have the exact same effect. Attention Even though the interactive debugger does not work in forking environments (which makes it nearly impossible to use on production servers), it still allows the execution 12
- 23. of arbitrary code. This makes it a major security risk and therefore it must never be used on production machines. Screenshot of the debugger in action: Have another debugger in mind? See Working with Debuggers. 4.3 Routing Modern web applications have beautiful URLs. This helps people remember the URLs, which is especially handy for applications that are used from mobile devices with slower network connections. If the user can directly go to the desired page with- out having to hit the index page it is more likely they will like the page and come back next time. As you have seen above, the route() decorator is used to bind a function to a URL. Here are some basic examples: @app.route(’/’) def index(): 13
- 24. return ’Index Page’ @app.route(’/hello’) def hello(): return ’Hello World’ But there is more to it! You can make certain parts of the URL dynamic and attach multiple rules to a function. 4.3.1 Variable Rules To add variable parts to a URL you can mark these special sections as <variable_name>. Such a part is then passed as a keyword argument to your function. Optionally a converter can be used by specifying a rule with <converter:variable_name>. Here are some nice examples: @app.route(’/user/<username>’) def show_user_profile(username): # show the user profile for that user return ’User %s’ % username @app.route(’/post/<int:post_id>’) def show_post(post_id): # show the post with the given id, the id is an integer return ’Post %d’ % post_id The following converters exist: int accepts integers float like int but for floating point values path like the default but also accepts slashes Unique URLs / Redirection Behavior Flask’s URL rules are based on Werkzeug’s routing module. The idea behind that module is to ensure beautiful and unique URLs based on precedents laid down by Apache and earlier HTTP servers. Take these two rules: @app.route(’/projects/’) def projects(): return ’The project page’ @app.route(’/about’) def about(): return ’The about page’ Though they look rather similar, they differ in their use of the trailing slash in the URL definition. In the first case, the canonical URL for the projects endpoint has a trailing slash. In that sense, it is similar to a folder on a file system. Accessing it without a trailing slash will cause Flask to redirect to the canonical URL with the trailing slash. 14
- 25. In the second case, however, the URL is defined without a trailing slash, rather like the pathname of a file on UNIX-like systems. Accessing the URL with a trailing slash will produce a 404 “Not Found” error. This behavior allows relative URLs to continue working even if the trailing slash is omitted, consistent with how Apache and other servers work. Also, the URLs will stay unique, which helps search engines avoid indexing the same page twice. 4.3.2 URL Building If it can match URLs, can Flask also generate them? Of course it can. To build a URL to a specific function you can use the url_for() function. It accepts the name of the function as first argument and a number of keyword arguments, each corresponding to the variable part of the URL rule. Unknown variable parts are appended to the URL as query parameters. Here are some examples: >>> from flask import Flask, url_for >>> app = Flask(__name__) >>> @app.route(’/’) ... def index(): pass ... >>> @app.route(’/login’) ... def login(): pass ... >>> @app.route(’/user/<username>’) ... def profile(username): pass ... >>> with app.test_request_context(): ... print url_for(’index’) ... print url_for(’login’) ... print url_for(’login’, next=’/’) ... print url_for(’profile’, username=’John Doe’) ... / /login /login?next=/ /user/John%20Doe (This also uses the test_request_context() method, explained below. It tells Flask to behave as though it is handling a request, even though we are interacting with it through a Python shell. Have a look at the explanation below. Context Locals). Why would you want to build URLs instead of hard-coding them into your templates? There are three good reasons for this: 1. Reversing is often more descriptive than hard-coding the URLs. More impor- tantly, it allows you to change URLs in one go, without having to remember to change URLs all over the place. 15
- 26. 2. URL building will handle escaping of special characters and Unicode data trans- parently for you, so you don’t have to deal with them. 3. If your application is placed outside the URL root (say, in /myapplication instead of /), url_for() will handle that properly for you. 4.3.3 HTTP Methods HTTP (the protocol web applications are speaking) knows different methods for ac- cessing URLs. By default, a route only answers to GET requests, but that can be changed by providing the methods argument to the route() decorator. Here are some examples: @app.route(’/login’, methods=[’GET’, ’POST’]) def login(): if request.method == ’POST’: do_the_login() else: show_the_login_form() If GET is present, HEAD will be added automatically for you. You don’t have to deal with that. It will also make sure that HEAD requests are handled as the HTTP RFC (the document describing the HTTP protocol) demands, so you can completely ignore that part of the HTTP specification. Likewise, as of Flask 0.6, OPTIONS is implemented for you automatically as well. You have no idea what an HTTP method is? Worry not, here is a quick introduction to HTTP methods and why they matter: The HTTP method (also often called “the verb”) tells the server what the clients wants to do with the requested page. The following methods are very common: GET The browser tells the server to just get the information stored on that page and send it. This is probably the most common method. HEAD The browser tells the server to get the information, but it is only interested in the headers, not the content of the page. An application is supposed to handle that as if a GET request was received but to not deliver the actual content. In Flask you don’t have to deal with that at all, the underlying Werkzeug library handles that for you. POST The browser tells the server that it wants to post some new information to that URL and that the server must ensure the data is stored and only stored once. This is how HTML forms usually transmit data to the server. PUT Similar to POST but the server might trigger the store procedure multiple times by overwriting the old values more than once. Now you might be asking why this is useful, but there are some good reasons to do it this way. Consider that the connection is lost during transmission: in this situation a system between the browser and the server might receive the request safely a second time without breaking things. With POST that would not be possible because it must only be triggered once. 16
- 27. DELETE Remove the information at the given location. OPTIONS Provides a quick way for a client to figure out which methods are sup- ported by this URL. Starting with Flask 0.6, this is implemented for you auto- matically. Now the interesting part is that in HTML4 and XHTML1, the only methods a form can submit to the server are GET and POST. But with JavaScript and future HTML standards you can use the other methods as well. Furthermore HTTP has become quite popular lately and browsers are no longer the only clients that are using HTTP. For instance, many revision control systems use it. 4.4 Static Files Dynamic web applications also need static files. That’s usually where the CSS and JavaScript files are coming from. Ideally your web server is configured to serve them for you, but during development Flask can do that as well. Just create a folder called static in your package or next to your module and it will be available at /static on the application. To generate URLs for static files, use the special ’static’ endpoint name: url_for(’static’, filename=’style.css’) The file has to be stored on the filesystem as static/style.css. 4.5 Rendering Templates Generating HTML from within Python is not fun, and actually pretty cumbersome be- cause you have to do the HTML escaping on your own to keep the application secure. Because of that Flask configures the Jinja2 template engine for you automatically. To render a template you can use the render_template() method. All you have to do is provide the name of the template and the variables you want to pass to the template engine as keyword arguments. Here’s a simple example of how to render a template: from flask import render_template @app.route(’/hello/’) @app.route(’/hello/<name>’) def hello(name=None): return render_template(’hello.html’, name=name) Flask will look for templates in the templates folder. So if your application is a module, this folder is next to that module, if it’s a package it’s actually inside your package: Case 1: a module: 17
- 28. /application.py /templates /hello.html Case 2: a package: /application /__init__.py /templates /hello.html For templates you can use the full power of Jinja2 templates. Head over to the official Jinja2 Template Documentation for more information. Here is an example template: <!doctype html> <title>Hello from Flask</title> {% if name %} <h1>Hello {{ name }}!</h1> {% else %} <h1>Hello World!</h1> {% endif %} Inside templates you also have access to the request, session and g 1 objects as well as the get_flashed_messages() function. Templates are especially useful if inheritance is used. If you want to know how that works, head over to the Template Inheritance pattern documentation. Basically tem- plate inheritance makes it possible to keep certain elements on each page (like header, navigation and footer). Automatic escaping is enabled, so if name contains HTML it will be escaped automat- ically. If you can trust a variable and you know that it will be safe HTML (for example because it came from a module that converts wiki markup to HTML) you can mark it as safe by using the Markup class or by using the |safe filter in the template. Head over to the Jinja 2 documentation for more examples. Here is a basic introduction to how the Markup class works: >>> from flask import Markup >>> Markup(’<strong>Hello %s!</strong>’) % ’<blink>hacker</blink>’ Markup(u’<strong>Hello <blink>hacker</blink>!</strong>’) >>> Markup.escape(’<blink>hacker</blink>’) Markup(u’<blink>hacker</blink>’) >>> Markup(’<em>Marked up</em> » HTML’).striptags() u’Marked up xbb HTML’ Changed in version 0.5: Autoescaping is no longer enabled for all templates. The following extensions for templates trigger autoescaping: .html, .htm, .xml, .xhtml. Templates loaded from a string will have autoescaping disabled. 1 Unsure what that g object is? It’s something in which you can store information for your own needs, check the documentation of that object (g) and the Using SQLite 3 with Flask for more information. 18
- 29. 4.6 Accessing Request Data For web applications it’s crucial to react to the data a client sent to the server. In Flask this information is provided by the global request object. If you have some experience with Python you might be wondering how that object can be global and how Flask manages to still be threadsafe. The answer is context locals: 4.6.1 Context Locals Insider Information If you want to understand how that works and how you can implement tests with context locals, read this section, otherwise just skip it. Certain objects in Flask are global objects, but not of the usual kind. These objects are actually proxies to objects that are local to a specific context. What a mouthful. But that is actually quite easy to understand. Imagine the context being the handling thread. A request comes in and the web server decides to spawn a new thread (or something else, the underlying object is capable of dealing with concurrency systems other than threads). When Flask starts its inter- nal request handling it figures out that the current thread is the active context and binds the current application and the WSGI environments to that context (thread). It does that in an intelligent way so that one application can invoke another application without breaking. So what does this mean to you? Basically you can completely ignore that this is the case unless you are doing something like unit testing. You will notice that code which depends on a request object will suddenly break because there is no request object. The solution is creating a request object yourself and binding it to the context. The easiest solution for unit testing is to use the test_request_context() context manager. In combination with the with statement it will bind a test request so that you can interact with it. Here is an example: from flask import request with app.test_request_context(’/hello’, method=’POST’): # now you can do something with the request until the # end of the with block, such as basic assertions: assert request.path == ’/hello’ assert request.method == ’POST’ The other possibility is passing a whole WSGI environment to the request_context() method: from flask import request with app.request_context(environ): assert request.method == ’POST’ 19
- 30. 4.6.2 The Request Object The request object is documented in the API section and we will not cover it here in detail (see request). Here is a broad overview of some of the most common operations. First of all you have to import it from the flask module: from flask import request The current request method is available by using the method attribute. To access form data (data transmitted in a POST or PUT request) you can use the form attribute. Here is a full example of the two attributes mentioned above: @app.route(’/login’, methods=[’POST’, ’GET’]) def login(): error = None if request.method == ’POST’: if valid_login(request.form[’username’], request.form[’password’]): return log_the_user_in(request.form[’username’]) else: error = ’Invalid username/password’ # the code below is executed if the request method # was GET or the credentials were invalid return render_template(’login.html’, error=error) What happens if the key does not exist in the form attribute? In that case a special KeyError is raised. You can catch it like a standard KeyError but if you don’t do that, a HTTP 400 Bad Request error page is shown instead. So for many situations you don’t have to deal with that problem. To access parameters submitted in the URL (?key=value) you can use the args at- tribute: searchword = request.args.get(’key’, ’’) We recommend accessing URL parameters with get or by catching the KeyError be- cause users might change the URL and presenting them a 400 bad request page in that case is not user friendly. For a full list of methods and attributes of the request object, head over to the request documentation. 4.6.3 File Uploads You can handle uploaded files with Flask easily. Just make sure not to forget to set the enctype="multipart/form-data" attribute on your HTML form, otherwise the browser will not transmit your files at all. Uploaded files are stored in memory or at a temporary location on the filesystem. You can access those files by looking at the files attribute on the request object. Each uploaded file is stored in that dictionary. It behaves just like a standard Python file 20
- 31. object, but it also has a save() method that allows you to store that file on the filesys- tem of the server. Here is a simple example showing how that works: from flask import request @app.route(’/upload’, methods=[’GET’, ’POST’]) def upload_file(): if request.method == ’POST’: f = request.files[’the_file’] f.save(’/var/www/uploads/uploaded_file.txt’) ... If you want to know how the file was named on the client before it was uploaded to your application, you can access the filename attribute. However please keep in mind that this value can be forged so never ever trust that value. If you want to use the file- name of the client to store the file on the server, pass it through the secure_filename() function that Werkzeug provides for you: from flask import request from werkzeug import secure_filename @app.route(’/upload’, methods=[’GET’, ’POST’]) def upload_file(): if request.method == ’POST’: f = request.files[’the_file’] f.save(’/var/www/uploads/’ + secure_filename(f.filename)) ... For some better examples, checkout the Uploading Files pattern. 4.6.4 Cookies To access cookies you can use the cookies attribute. To set cookies you can use the set_cookie method of response objects. The cookies attribute of request objects is a dictionary with all the cookies the client transmits. If you want to use sessions, do not use the cookies directly but instead use the Sessions in Flask that add some security on top of cookies for you. Reading cookies: from flask import request @app.route(’/’) def index(): username = request.cookies.get(’username’) # use cookies.get(key) instead of cookies[key] to not get a # KeyError if the cookie is missing. Storing cookies: 21
- 32. from flask import make_response @app.route(’/’) def index(): resp = make_response(render_template(...)) resp.set_cookie(’username’, ’the username’) return resp Note that cookies are set on response objects. Since you normally just return strings from the view functions Flask will convert them into response objects for you. If you explicitly want to do that you can use the make_response() function and then modify it. Sometimes you might want to set a cookie at a point where the response object does not exist yet. This is possible by utilizing the Deferred Request Callbacks pattern. For this also see About Responses. 4.7 Redirects and Errors To redirect a user to another endpoint, use the redirect() function; to abort a request early with an error code, use the abort() function: from flask import abort, redirect, url_for @app.route(’/’) def index(): return redirect(url_for(’login’)) @app.route(’/login’) def login(): abort(401) this_is_never_executed() This is a rather pointless example because a user will be redirected from the index to a page they cannot access (401 means access denied) but it shows how that works. By default a black and white error page is shown for each error code. If you want to customize the error page, you can use the errorhandler() decorator: from flask import render_template @app.errorhandler(404) def page_not_found(error): return render_template(’page_not_found.html’), 404 Note the 404 after the render_template() call. This tells Flask that the status code of that page should be 404 which means not found. By default 200 is assumed which translates to: all went well. 22
- 33. 4.8 About Responses The return value from a view function is automatically converted into a response ob- ject for you. If the return value is a string it’s converted into a response object with the string as response body, a 200 OK status code and a text/html mimetype. The logic that Flask applies to converting return values into response objects is as follows: 1. If a response object of the correct type is returned it’s directly returned from the view. 2. If it’s a string, a response object is created with that data and the default param- eters. 3. If a tuple is returned the items in the tuple can provide extra information. Such tuples have to be in the form (response, status, headers) or (response, headers) where at least one item has to be in the tuple. The status value will over- ride the status code and headers can be a list or dictionary of additional header values. 4. If none of that works, Flask will assume the return value is a valid WSGI appli- cation and convert that into a response object. If you want to get hold of the resulting response object inside the view you can use the make_response() function. Imagine you have a view like this: @app.errorhandler(404) def not_found(error): return render_template(’error.html’), 404 You just need to wrap the return expression with make_response() and get the re- sponse object to modify it, then return it: @app.errorhandler(404) def not_found(error): resp = make_response(render_template(’error.html’), 404) resp.headers[’X-Something’] = ’A value’ return resp 4.9 Sessions In addition to the request object there is also a second object called session which allows you to store information specific to a user from one request to the next. This is implemented on top of cookies for you and signs the cookies cryptographically. What this means is that the user could look at the contents of your cookie but not modify it, unless they know the secret key used for signing. In order to use sessions you have to set a secret key. Here is how sessions work: 23
- 34. from flask import Flask, session, redirect, url_for, escape, request app = Flask(__name__) @app.route(’/’) def index(): if ’username’ in session: return ’Logged in as %s’ % escape(session[’username’]) return ’You are not logged in’ @app.route(’/login’, methods=[’GET’, ’POST’]) def login(): if request.method == ’POST’: session[’username’] = request.form[’username’] return redirect(url_for(’index’)) return ’’’ <form action="" method="post"> <p><input type=text name=username> <p><input type=submit value=Login> </form> ’’’ @app.route(’/logout’) def logout(): # remove the username from the session if it’s there session.pop(’username’, None) return redirect(url_for(’index’)) # set the secret key. keep this really secret: app.secret_key = ’A0Zr98j/3yX R~XHH!jmN]LWX/,?RT’ The escape() mentioned here does escaping for you if you are not using the template engine (as in this example). How to generate good secret keys The problem with random is that it’s hard to judge what is truly random. And a secret key should be as random as possible. Your operating system has ways to generate pretty random stuff based on a cryptographic random generator which can be used to get such a key: >>> import os >>> os.urandom(24) ’xfd{Hxe5<x95xf9xe3x96.5xd1x01O<!xd5xa2xa0x9fR"xa1xa8’ Just take that thing and copy/paste it into your code and you’re done. A note on cookie-based sessions: Flask will take the values you put into the session object and serialize them into a cookie. If you are finding some values do not per- sist across requests, cookies are indeed enabled, and you are not getting a clear error message, check the size of the cookie in your page responses compared to the size 24
- 35. supported by web browsers. 4.10 Message Flashing Good applications and user interfaces are all about feedback. If the user does not get enough feedback they will probably end up hating the application. Flask provides a really simple way to give feedback to a user with the flashing system. The flashing system basically makes it possible to record a message at the end of a request and access it on the next (and only the next) request. This is usually combined with a layout template to expose the message. To flash a message use the flash() method, to get hold of the messages you can use get_flashed_messages() which is also available in the templates. Check out the Mes- sage Flashing for a full example. 4.11 Logging New in version 0.3. Sometimes you might be in a situation where you deal with data that should be correct, but actually is not. For example you may have some client-side code that sends an HTTP request to the server but it’s obviously malformed. This might be caused by a user tampering with the data, or the client code failing. Most of the time it’s okay to reply with 400 Bad Request in that situation, but sometimes that won’t do and the code has to continue working. You may still want to log that something fishy happened. This is where loggers come in handy. As of Flask 0.3 a logger is preconfigured for you to use. Here are some example log calls: app.logger.debug(’A value for debugging’) app.logger.warning(’A warning occurred (%d apples)’, 42) app.logger.error(’An error occurred’) The attached logger is a standard logging Logger, so head over to the official logging documentation for more information. 4.12 Hooking in WSGI Middlewares If you want to add a WSGI middleware to your application you can wrap the internal WSGI application. For example if you want to use one of the middlewares from the Werkzeug package to work around bugs in lighttpd, you can do it like this: from werkzeug.contrib.fixers import LighttpdCGIRootFix app.wsgi_app = LighttpdCGIRootFix(app.wsgi_app) 25
- 36. 4.13 Deploying to a Web Server Ready to deploy your new Flask app? To wrap up the quickstart, you can immediately deploy to a hosted platform, all of which offer a free plan for small projects: • Deploying Flask on Heroku • Deploying WSGI on dotCloud with Flask-specific notes Other places where you can host your Flask app: • Deploying Flask on Webfaction • Deploying Flask on Google App Engine • Sharing your Localhost Server with Localtunnel If you manage your own hosts and would like to host yourself, see the chapter on Deployment Options. 26
- 37. CHAPTER FIVE TUTORIAL You want to develop an application with Python and Flask? Here you have the chance to learn that by example. In this tutorial we will create a simple microblog application. It only supports one user that can create text-only entries and there are no feeds or comments, but it still features everything you need to get started. We will use Flask and SQLite as database which comes out of the box with Python, so there is nothing else you need. If you want the full sourcecode in advance or for comparison, check out the example source. 5.1 Introducing Flaskr We will call our blogging application flaskr here, feel free to choose a less web-2.0-ish name ;) Basically we want it to do the following things: 1. let the user sign in and out with credentials specified in the configuration. Only one user is supported. 2. when the user is logged in they can add new entries to the page consisting of a text-only title and some HTML for the text. This HTML is not sanitized because we trust the user here. 3. the page shows all entries so far in reverse order (newest on top) and the user can add new ones from there if logged in. We will be using SQLite3 directly for that application because it’s good enough for an application of that size. For larger applications however it makes a lot of sense to use SQLAlchemy that handles database connections in a more intelligent way, allows you to target different relational databases at once and more. You might also want to consider one of the popular NoSQL databases if your data is more suited for those. Here a screenshot from the final application: 27
- 38. Continue with Step 0: Creating The Folders. 5.2 Step 0: Creating The Folders Before we get started, let’s create the folders needed for this application: /flaskr /static /templates The flaskr folder is not a python package, but just something where we drop our files. We will then put our database schema as well as main module into this folder. It is done in the following way. The files inside the static folder are available to users of the application via HTTP. This is the place where css and javascript files go. Inside the templates folder Flask will look for Jinja2 templates. The templates you create later in the tutorial will go in this directory. Continue with Step 1: Database Schema. 28
- 39. 5.3 Step 1: Database Schema First we want to create the database schema. Only a single table is needed for this ap- plication and we only want to support SQLite so creating the database schema is quite easy. Just put the following contents into a file named schema.sql in the just created flaskr folder: drop table if exists entries; create table entries ( id integer primary key autoincrement, title text not null, text text not null ); This schema consists of a single table called entries and each row in this table has an id, a title and a text. The id is an automatically incrementing integer and a primary key, the other two are strings that must not be null. Continue with Step 2: Application Setup Code. 5.4 Step 2: Application Setup Code Now that we have the schema in place we can create the application module. Let’s call it flaskr.py. We will place this file inside the flask folder. We will begin by adding the imports we need and by adding the config section. For small applications, it is possible to drop the configuration directly into the module, and this is what we will be doing here. However a cleaner solution would be to create a separate .ini or .py file and load that or import the values from there. First we add the imports in flaskr.py: # all the imports import sqlite3 from flask import Flask, request, session, g, redirect, url_for, abort, render_template, flash Next we can create our actual application and initialize it with the config from the same file, in flaskr.py: # create our little application :) app = Flask(__name__) app.config.from_object(__name__) # Load default config and override config from an environment variable app.config.update(dict( DATABASE=os.path.join(app.root_path, ’flaskr.db’), DEBUG=True, SECRET_KEY=’development key’, USERNAME=’admin’, 29
- 40. PASSWORD=’default’ )) app.config.from_envvar(’FLASKR_SETTINGS’, silent=True) The Config object works similar to a dictionary so we can update it with new values. Database Path Operating systems know the concept of a current working directory for each process. Unfortunately you cannot depend on this in web applications because you might have more than one application in the same process. For this reason the app.root_path attribute can be used to get the path to the applica- tion. Together with the os.path module files can then easily be found. In this example we place the database right next to it. For a real-work application it’s recommended to use Instance Folders instead. Usually, it is a good idea to load a separate, environment specific configuration file. Flask allows you to import multiple configurations and it will use the setting defined in the last import. This enables robust configuration setups. from_envvar() can help achieve this. app.config.from_envvar(‘FLASKR_SETTINGS’, silent=True) Simply define the environment variable FLASKR_SETTINGS that points to a config file to be loaded. The silent switch just tells Flask to not complain if no such environment key is set. In addition to that you can use the from_object() method on the config object and provide it with an import name of a module. Flask will the initialize the variable from that module. Note that in all cases only variable names that are uppercase are considered. The SECRET_KEY is needed to keep the client-side sessions secure. Choose that key wisely and as hard to guess and complex as possible. The debug flag enables or dis- ables the interactive debugger. Never leave debug mode activated in a production system, because it will allow users to execute code on the server! We will also add a method that allows for easily connecting to the specified database. This can be used to open a connection on request and also from the interactive Python shell or a script. This will come in handy later. We create a simple database connection through SQLite and then tell it to use the sqlite3.Row object to represent rows. This allows us to treat the rows as if they were dictionaries instead of tuples. def connect_db(): """Connects to the specific database.""" rv = sqlite3.connect(app.config[’DATABASE’]) rv.row_factory = sqlite3.Row return rv Finally we just add a line to the bottom of the file that fires up the server if we want to run that file as a standalone application: 30
- 41. if __name__ == ’__main__’: app.run() With that out of the way you should be able to start up the application without prob- lems. Do this with the following command: python flaskr.py You will see a message telling you that server has started along with the address at which you can access it. When you head over to the server in your browser you will get an 404 page not found error because we don’t have any views yet. But we will focus on that a little later. First we should get the database working. Externally Visible Server Want your server to be publicly available? Check out the externally visible server section for more information. Continue with Step 3: Database Connections. 5.5 Step 3: Database Connections We have created a function for establishing a database connection with connect_db but by itself that’s not particularly useful. Creating and closing database connections all the time is very inefficient, so we want to keep it around for longer. Because database connections encapsulate a transaction we also need to make sure that only one request at the time uses the connection. So how can we elegantly do that with Flask? This is where the application context comes into play. So let’s start there. Flask provides us with two contexts: the application context and the request context. For the time being all you have to know is that there are special variables that use these. For instance the request variable is the request object associated with the current re- quest, whereas g is a general purpose variable associated with the current application context. We will go into the details of this a bit later. For the time being all you have to know is that you can store information safely on the g object. So when do you put it on there? To do that you can make a helper function. The first time the function is called it will create a database connection for the current context and successive calls will return the already established connection: def get_db(): """Opens a new database connection if there is none yet for the current application context. """ if not hasattr(g, ’sqlite_db’): 31
- 42. g.sqlite_db = connect_db() return g.sqlite_db So now we know how to connect, but how do we properly disconnect? For that flask provides us with the teardown_appcontext() decorator. It’s executed every time the application context tears down: @app.teardown_appcontext def close_db(error): """Closes the database again at the end of the request.""" if hasattr(g, ’sqlite_db’): g.sqlite_db.close() Functions marked with teardown_appcontext() are called every time the app context tears down. So what does this mean? Essentially the app context is created before the request comes in and is destroyed (teared down) whenever the request finishes. A teardown can happen because of two reasons: either everything went well (the error parameter will be None) or an exception happend in which case the error is passed to the teardown function. Curious about what these contexts mean? Have a look at the The Application Context documentation to learn more. Continue to Step 4: Creating The Database. Hint: Where do I put this code? If you’ve been following along in this tutorial, you might be wondering where to put the code from this step and the next. A logical place is to group these module-level functions together, and put your new get_db and close_db functions below your ex- isting connect_db function (following the tutorial line-by-line). If you need a moment to find your bearings, take a look at how the example source is organized. In Flask, you can put all of your application code into a single Python module. You don’t have to, and if your app grows larger, it’s a good idea not to. 5.6 Step 4: Creating The Database As outlined earlier, Flaskr is a database powered application, and more precisely, it is an application powered by a relational database system. Such systems need a schema that tells them how to store that information. So before starting the server for the first time it’s important to create that schema. Such a schema can be created by piping the schema.sql file into the sqlite3 command as follows: sqlite3 /tmp/flaskr.db < schema.sql The downside of this is that it requires the sqlite3 command to be installed which is not necessarily the case on every system. This also require that we provide the path 32
- 43. to the database which can introduce errors. It’s a good idea to add a function that initializes the database for you to the application. To do this we can create a function called init_db that initializes the database. Let me show you the code first. Just add this function below the connect_db function in flaskr.py: def init_db(): with app.app_context(): db = get_db() with app.open_resource(’schema.sql’, mode=’r’) as f: db.cursor().executescript(f.read()) db.commit() So what’s happening here? Remember how we learned last chapter that the applica- tion context is created every time a request comes in? Here we don’t have a request yet, so we need to create the application context by hand. Without an application con- text the g object does not know yet to which application it becomes as there could be more than one! The with app.app_context() statement establishes the application context for us. In the body of the with statement the g object will be associated with app. At the end of the with statement the association is released and all teardown functions are executed. This means that our database connection is disconnected after the commit. The open_resource() method of the application object is a convenient helper function that will open a resource that the application provides. This function opens a file from the resource location (your flaskr folder) and allows you to read from it. We are using this here to execute a script on the database connection. The connection object provided by SQLite can give us a cursor object. On that cur- sor there is a method to execute a complete script. Finally we only have to commit the changes. SQLite 3 and other transactional databases will not commit unless you explicitly tell it to. Now it is possible to create a database by starting up a Python shell and importing and calling that function: >>> from flaskr import init_db >>> init_db() Troubleshooting If you get an exception later that a table cannot be found check that you did call the init_db function and that your table names are correct (singular vs. plural for example). Continue with Step 5: The View Functions 33
- 44. 5.7 Step 5: The View Functions Now that the database connections are working we can start writing the view func- tions. We will need four of them: 5.7.1 Show Entries This view shows all the entries stored in the database. It listens on the root of the application and will select title and text from the database. The one with the highest id (the newest entry) will be on top. The rows returned from the cursor look a bit like tuples because we are using the sqlite3.Row row factory. The view function will pass the entries as dicts to the show_entries.html template and return the rendered one: @app.route(’/’) def show_entries(): db = get_db() cur = db.execute(’select title, text from entries order by id desc’) entries = cur.fetchall() return render_template(’show_entries.html’, entries=entries) 5.7.2 Add New Entry This view lets the user add new entries if they are logged in. This only responds to POST requests, the actual form is shown on the show_entries page. If everything worked out well we will flash() an information message to the next request and redi- rect back to the show_entries page: @app.route(’/add’, methods=[’POST’]) def add_entry(): if not session.get(’logged_in’): abort(401) db = get_db() db.execute(’insert into entries (title, text) values (?, ?)’, [request.form[’title’], request.form[’text’]]) db.commit() flash(’New entry was successfully posted’) return redirect(url_for(’show_entries’)) Note that we check that the user is logged in here (the logged_in key is present in the session and True). Security Note Be sure to use question marks when building SQL statements, as done in the example above. Otherwise, your app will be vulnerable to SQL injection when you use string formatting to build SQL statements. See Using SQLite 3 with Flask for more. 34
- 45. 5.7.3 Login and Logout These functions are used to sign the user in and out. Login checks the username and password against the ones from the configuration and sets the logged_in key in the ses- sion. If the user logged in successfully, that key is set to True, and the user is redirected back to the show_entries page. In addition, a message is flashed that informs the user that he or she was logged in successfully. If an error occurred, the template is notified about that, and the user is asked again: @app.route(’/login’, methods=[’GET’, ’POST’]) def login(): error = None if request.method == ’POST’: if request.form[’username’] != app.config[’USERNAME’]: error = ’Invalid username’ elif request.form[’password’] != app.config[’PASSWORD’]: error = ’Invalid password’ else: session[’logged_in’] = True flash(’You were logged in’) return redirect(url_for(’show_entries’)) return render_template(’login.html’, error=error) The logout function, on the other hand, removes that key from the session again. We use a neat trick here: if you use the pop() method of the dict and pass a second param- eter to it (the default), the method will delete the key from the dictionary if present or do nothing when that key is not in there. This is helpful because now we don’t have to check if the user was logged in. @app.route(’/logout’) def logout(): session.pop(’logged_in’, None) flash(’You were logged out’) return redirect(url_for(’show_entries’)) Continue with Step 6: The Templates. 5.8 Step 6: The Templates Now we should start working on the templates. If we request the URLs now we would only get an exception that Flask cannot find the templates. The templates are using Jinja2 syntax and have autoescaping enabled by default. This means that unless you mark a value in the code with Markup or with the |safe filter in the template, Jinja2 will ensure that special characters such as < or > are escaped with their XML equivalents. We are also using template inheritance which makes it possible to reuse the layout of the website in all pages. Put the following templates into the templates folder: 35
- 46. 5.8.1 layout.html This template contains the HTML skeleton, the header and a link to log in (or log out if the user was already logged in). It also displays the flashed messages if there are any. The {% block body %} block can be replaced by a block of the same name (body) in a child template. The session dict is available in the template as well and you can use that to check if the user is logged in or not. Note that in Jinja you can access missing attributes and items of objects / dicts which makes the following code work, even if there is no ’logged_in’ key in the session: <!doctype html> <title>Flaskr</title> <link rel=stylesheet type=text/css href="{{ url_for(’static’, filename=’style.css’) }}"> <div class=page> <h1>Flaskr</h1> <div class=metanav> {% if not session.logged_in %} <a href="{{ url_for(’login’) }}">log in</a> {% else %} <a href="{{ url_for(’logout’) }}">log out</a> {% endif %} </div> {% for message in get_flashed_messages() %} <div class=flash>{{ message }}</div> {% endfor %} {% block body %}{% endblock %} </div> 5.8.2 show_entries.html This template extends the layout.html template from above to display the mes- sages. Note that the for loop iterates over the messages we passed in with the render_template() function. We also tell the form to submit to your add_entry function and use POST as HTTP method: {% extends "layout.html" %} {% block body %} {% if session.logged_in %} <form action="{{ url_for(’add_entry’) }}" method=post class=add-entry> <dl> <dt>Title: <dd><input type=text size=30 name=title> <dt>Text: <dd><textarea name=text rows=5 cols=40></textarea> <dd><input type=submit value=Share> </dl> </form> {% endif %} 36
- 47. <ul class=entries> {% for entry in entries %} <li><h2>{{ entry.title }}</h2>{{ entry.text|safe }} {% else %} <li><em>Unbelievable. No entries here so far</em> {% endfor %} </ul> {% endblock %} 5.8.3 login.html Finally the login template which basically just displays a form to allow the user to login: {% extends "layout.html" %} {% block body %} <h2>Login</h2> {% if error %}<p class=error><strong>Error:</strong> {{ error }}{% endif %} <form action="{{ url_for(’login’) }}" method=post> <dl> <dt>Username: <dd><input type=text name=username> <dt>Password: <dd><input type=password name=password> <dd><input type=submit value=Login> </dl> </form> {% endblock %} Continue with Step 7: Adding Style. 5.9 Step 7: Adding Style Now that everything else works, it’s time to add some style to the application. Just create a stylesheet called style.css in the static folder we created before: body { font-family: sans-serif; background: #eee; } a, h1, h2 { color: #377ba8; } h1, h2 { font-family: ’Georgia’, serif; margin: 0; } h1 { border-bottom: 2px solid #eee; } h2 { font-size: 1.2em; } .page { margin: 2em auto; width: 35em; border: 5px solid #ccc; padding: 0.8em; background: white; } .entries { list-style: none; margin: 0; padding: 0; } .entries li { margin: 0.8em 1.2em; } .entries li h2 { margin-left: -1em; } .add-entry { font-size: 0.9em; border-bottom: 1px solid #ccc; } 37
- 48. .add-entry dl { font-weight: bold; } .metanav { text-align: right; font-size: 0.8em; padding: 0.3em; margin-bottom: 1em; background: #fafafa; } .flash { background: #cee5F5; padding: 0.5em; border: 1px solid #aacbe2; } .error { background: #f0d6d6; padding: 0.5em; } Continue with Bonus: Testing the Application. 5.10 Bonus: Testing the Application Now that you have finished the application and everything works as expected, it’s probably not a bad idea to add automated tests to simplify modifications in the future. The application above is used as a basic example of how to perform unittesting in the Testing Flask Applications section of the documentation. Go there to see how easy it is to test Flask applications. 38
- 49. CHAPTER SIX TEMPLATES Flask leverages Jinja2 as template engine. You are obviously free to use a different tem- plate engine, but you still have to install Jinja2 to run Flask itself. This requirement is necessary to enable rich extensions. An extension can depend on Jinja2 being present. This section only gives a very quick introduction into how Jinja2 is integrated into Flask. If you want information on the template engine’s syntax itself, head over to the official Jinja2 Template Documentation for more information. 6.1 Jinja Setup Unless customized, Jinja2 is configured by Flask as follows: • autoescaping is enabled for all templates ending in .html, .htm, .xml as well as .xhtml • a template has the ability to opt in/out autoescaping with the {% autoescape %} tag. • Flask inserts a couple of global functions and helpers into the Jinja2 context, ad- ditionally to the values that are present by default. 6.2 Standard Context The following global variables are available within Jinja2 templates by default: config The current configuration object (flask.config) New in version 0.6. Changed in version 0.10: This is now always available, even in imported tem- plates. request The current request object (flask.request). This variable is unavailable if the template was rendered without an active request context. 39
- 50. session The current session object (flask.session). This variable is unavailable if the template was rendered without an active request context. g The request-bound object for global variables (flask.g). This variable is unavail- able if the template was rendered without an active request context. url_for() The flask.url_for() function. get_flashed_messages() The flask.get_flashed_messages() function. The Jinja Context Behavior These variables are added to the context of variables, they are not global variables. The difference is that by default these will not show up in the context of imported templates. This is partially caused by performance considerations, partially to keep things explicit. What does this mean for you? If you have a macro you want to import, that needs to access the request object you have two possibilities: 1. you explicitly pass the request to the macro as parameter, or the attribute of the request object you are interested in. 2. you import the macro “with context”. Importing with context looks like this: {% from ’_helpers.html’ import my_macro with context %} 6.3 Standard Filters These filters are available in Jinja2 additionally to the filters provided by Jinja2 itself: tojson() This function converts the given object into JSON representation. This is for example very helpful if you try to generate JavaScript on the fly. Note that inside script tags no escaping must take place, so make sure to disable escaping with |safe before Flask 0.10 if you intend to use it inside script tags: <script type=text/javascript> doSomethingWith({{ user.username|tojson|safe }}); </script> 40
- 51. 6.4 Controlling Autoescaping Autoescaping is the concept of automatically escaping special characters of you. Spe- cial characters in the sense of HTML (or XML, and thus XHTML) are &, >, <, " as well as ’. Because these characters carry specific meanings in documents on their own you have to replace them by so called “entities” if you want to use them for text. Not doing so would not only cause user frustration by the inability to use these characters in text, but can also lead to security problems. (see Cross-Site Scripting (XSS)) Sometimes however you will need to disable autoescaping in templates. This can be the case if you want to explicitly inject HTML into pages, for example if they come from a system that generate secure HTML like a markdown to HTML converter. There are three ways to accomplish that: • In the Python code, wrap the HTML string in a Markup object before passing it to the template. This is in general the recommended way. • Inside the template, use the |safe filter to explicitly mark a string as safe HTML ({{ myvariable|safe }}) • Temporarily disable the autoescape system altogether. To disable the autoescape system in templates, you can use the {% autoescape %} block: {% autoescape false %} <p>autoescaping is disabled here <p>{{ will_not_be_escaped }} {% endautoescape %} Whenever you do this, please be very cautious about the variables you are using in this block. 6.5 Registering Filters If you want to register your own filters in Jinja2 you have two ways to do that. You can either put them by hand into the jinja_env of the application or use the template_filter() decorator. The two following examples work the same and both reverse an object: @app.template_filter(’reverse’) def reverse_filter(s): return s[::-1] def reverse_filter(s): return s[::-1] app.jinja_env.filters[’reverse’] = reverse_filter 41
- 52. In case of the decorator the argument is optional if you want to use the function name as name of the filter. Once registered, you can use the filter in your templates in the same way as Jinja2’s builtin filters, for example if you have a Python list in context called mylist: {% for x in mylist | reverse %} {% endfor %} 6.6 Context Processors To inject new variables automatically into the context of a template, context processors exist in Flask. Context processors run before the template is rendered and have the ability to inject new values into the template context. A context processor is a function that returns a dictionary. The keys and values of this dictionary are then merged with the template context, for all templates in the app: @app.context_processor def inject_user(): return dict(user=g.user) The context processor above makes a variable called user available in the template with the value of g.user. This example is not very interesting because g is available in templates anyways, but it gives an idea how this works. Variables are not limited to values; a context processor can also make functions avail- able to templates (since Python allows passing around functions): @app.context_processor def utility_processor(): def format_price(amount, currency=u’ ’): return u’{0:.2f}{1}’.format(amount, currency) return dict(format_price=format_price) The context processor above makes the format_price function available to all templates: {{ format_price(0.33) }} You could also build format_price as a template filter (see Registering Filters), but this demonstrates how to pass functions in a context processor. 42
- 53. CHAPTER SEVEN TESTING FLASK APPLICATIONS Something that is untested is broken. The origin of this quote is unknown and while it is not entirely correct, it is also not far from the truth. Untested applications make it hard to improve existing code and developers of untested applications tend to become pretty paranoid. If an application has automated tests, you can safely make changes and instantly know if anything breaks. Flask provides a way to test your application by exposing the Werkzeug test Client and handling the context locals for you. You can then use that with your favourite testing solution. In this documentation we will use the unittest package that comes pre-installed with Python. 7.1 The Application First, we need an application to test; we will use the application from the Tutorial. If you don’t have that application yet, get the sources from the examples. 7.2 The Testing Skeleton In order to test the application, we add a second module (flaskr_tests.py) and create a unittest skeleton there: import os import flaskr import unittest import tempfile class FlaskrTestCase(unittest.TestCase): def setUp(self): self.db_fd, flaskr.app.config[’DATABASE’] = tempfile.mkstemp() flaskr.app.config[’TESTING’] = True self.app = flaskr.app.test_client() 43
- 54. flaskr.init_db() def tearDown(self): os.close(self.db_fd) os.unlink(flaskr.app.config[’DATABASE’]) if __name__ == ’__main__’: unittest.main() The code in the setUp() method creates a new test client and initializes a new database. This function is called before each individual test function is run. To delete the database after the test, we close the file and remove it from the filesystem in the tearDown() method. Additionally during setup the TESTING config flag is activated. What it does is disabling the error catching during request handling so that you get better error reports when performing test requests against the application. This test client will give us a simple interface to the application. We can trigger test requests to the application, and the client will also keep track of cookies for us. Because SQLite3 is filesystem-based we can easily use the tempfile module to create a temporary database and initialize it. The mkstemp() function does two things for us: it returns a low-level file handle and a random file name, the latter we use as database name. We just have to keep the db_fd around so that we can use the os.close() func- tion to close the file. If we now run the test suite, we should see the following output: $ python flaskr_tests.py ---------------------------------------------------------------------- Ran 0 tests in 0.000s OK Even though it did not run any actual tests, we already know that our flaskr applica- tion is syntactically valid, otherwise the import would have died with an exception. 7.3 The First Test Now it’s time to start testing the functionality of the application. Let’s check that the application shows “No entries here so far” if we access the root of the application (/). To do this, we add a new test method to our class, like this: class FlaskrTestCase(unittest.TestCase): def setUp(self): self.db_fd, flaskr.app.config[’DATABASE’] = tempfile.mkstemp() self.app = flaskr.app.test_client() flaskr.init_db() 44
- 55. def tearDown(self): os.close(self.db_fd) os.unlink(flaskr.app.config[’DATABASE’]) def test_empty_db(self): rv = self.app.get(’/’) assert ’No entries here so far’ in rv.data Notice that our test functions begin with the word test; this allows unittest to auto- matically identify the method as a test to run. By using self.app.get we can send an HTTP GET request to the application with the given path. The return value will be a response_class object. We can now use the data attribute to inspect the return value (as string) from the application. In this case, we ensure that ’No entries here so far’ is part of the output. Run it again and you should see one passing test: $ python flaskr_tests.py . ---------------------------------------------------------------------- Ran 1 test in 0.034s OK 7.4 Logging In and Out The majority of the functionality of our application is only available for the adminis- trative user, so we need a way to log our test client in and out of the application. To do this, we fire some requests to the login and logout pages with the required form data (username and password). And because the login and logout pages redirect, we tell the client to follow_redirects. Add the following two methods to your FlaskrTestCase class: def login(self, username, password): return self.app.post(’/login’, data=dict( username=username, password=password ), follow_redirects=True) def logout(self): return self.app.get(’/logout’, follow_redirects=True) Now we can easily test that logging in and out works and that it fails with invalid credentials. Add this new test to the class: def test_login_logout(self): rv = self.login(’admin’, ’default’) assert ’You were logged in’ in rv.data 45
- 56. rv = self.logout() assert ’You were logged out’ in rv.data rv = self.login(’adminx’, ’default’) assert ’Invalid username’ in rv.data rv = self.login(’admin’, ’defaultx’) assert ’Invalid password’ in rv.data 7.5 Test Adding Messages We should also test that adding messages works. Add a new test method like this: def test_messages(self): self.login(’admin’, ’default’) rv = self.app.post(’/add’, data=dict( title=’<Hello>’, text=’<strong>HTML</strong> allowed here’ ), follow_redirects=True) assert ’No entries here so far’ not in rv.data assert ’<Hello>’ in rv.data assert ’<strong>HTML</strong> allowed here’ in rv.data Here we check that HTML is allowed in the text but not in the title, which is the in- tended behavior. Running that should now give us three passing tests: $ python flaskr_tests.py ... ---------------------------------------------------------------------- Ran 3 tests in 0.332s OK For more complex tests with headers and status codes, check out the MiniTwit Exam- ple from the sources which contains a larger test suite. 7.6 Other Testing Tricks Besides using the test client as shown above, there is also the test_request_context() method that can be used in combination with the with statement to activate a request context temporarily. With this you can access the request, g and session objects like in view functions. Here is a full example that demonstrates this approach: app = flask.Flask(__name__) with app.test_request_context(’/?name=Peter’): assert flask.request.path == ’/’ assert flask.request.args[’name’] == ’Peter’ 46
- 57. All the other objects that are context bound can be used in the same way. If you want to test your application with different configurations and there does not seem to be a good way to do that, consider switching to application factories (see Application Factories). Note however that if you are using a test request context, the before_request() functions are not automatically called same for after_request() functions. How- ever teardown_request() functions are indeed executed when the test request context leaves the with block. If you do want the before_request() functions to be called as well, you need to call preprocess_request() yourself: app = flask.Flask(__name__) with app.test_request_context(’/?name=Peter’): app.preprocess_request() ... This can be necessary to open database connections or something similar depending on how your application was designed. If you want to call the after_request() functions you need to call into process_response() which however requires that you pass it a response object: app = flask.Flask(__name__) with app.test_request_context(’/?name=Peter’): resp = Response(’...’) resp = app.process_response(resp) ... This in general is less useful because at that point you can directly start using the test client. 7.7 Faking Resources and Context New in version 0.10. A very common pattern is to store user authorization information and database con- nections on the application context or the flask.g object. The general pattern for this is to put the object on there on first usage and then to remove it on a teardown. Imagine for instance this code to get the current user: def get_user(): user = getattr(g, ’user’, None) if user is None: user = fetch_current_user_from_database() g.user = user return user 47
- 58. For a test it would be nice to override this user from the outside without hav- ing to change some code. This can trivially be accomplished with hooking the flask.appcontext_pushed signal: from contextlib import contextmanager from flask import appcontext_pushed @contextmanager def user_set(app, user): def handler(sender, **kwargs): g.user = user with appcontext_pushed.connected_to(handler, app): yield And then to use it: from flask import json, jsonify @app.route(’/users/me’) def users_me(): return jsonify(username=g.user.username) with user_set(app, my_user): with app.test_client() as c: resp = c.get(’/users/me’) data = json.loads(resp.data) self.assert_equal(data[’username’], my_user.username) 7.8 Keeping the Context Around New in version 0.4. Sometimes it is helpful to trigger a regular request but still keep the context around for a little longer so that additional introspection can happen. With Flask 0.4 this is possible by using the test_client() with a with block: app = flask.Flask(__name__) with app.test_client() as c: rv = c.get(’/?tequila=42’) assert request.args[’tequila’] == ’42’ If you were to use just the test_client() without the with block, the assert would fail with an error because request is no longer available (because you are trying to use it outside of the actual request). 48
- 59. 7.9 Accessing and Modifying Sessions New in version 0.8. Sometimes it can be very helpful to access or modify the sessions from the test client. Generally there are two ways for this. If you just want to ensure that a session has certain keys set to certain values you can just keep the context around and access flask.session: with app.test_client() as c: rv = c.get(’/’) assert flask.session[’foo’] == 42 This however does not make it possible to also modify the session or to access the ses- sion before a request was fired. Starting with Flask 0.8 we provide a so called “session transaction” which simulates the appropriate calls to open a session in the context of the test client and to modify it. At the end of the transaction the session is stored. This works independently of the session backend used: with app.test_client() as c: with c.session_transaction() as sess: sess[’a_key’] = ’a value’ # once this is reached the session was stored Note that in this case you have to use the sess object instead of the flask.session proxy. The object however itself will provide the same interface. 49
- 60. 50
- 61. CHAPTER EIGHT LOGGING APPLICATION ERRORS New in version 0.3. Applications fail, servers fail. Sooner or later you will see an exception in production. Even if your code is 100% correct, you will still see exceptions from time to time. Why? Because everything else involved will fail. Here some situations where perfectly fine code can lead to server errors: • the client terminated the request early and the application was still reading from the incoming data. • the database server was overloaded and could not handle the query. • a filesystem is full • a harddrive crashed • a backend server overloaded • a programming error in a library you are using • network connection of the server to another system failed. And that’s just a small sample of issues you could be facing. So how do we deal with that sort of problem? By default if your application runs in production mode, Flask will display a very simple page for you and log the exception to the logger. But there is more you can do, and we will cover some better setups to deal with errors. 8.1 Error Mails If the application runs in production mode (which it will do on your server) you won’t see any log messages by default. Why is that? Flask tries to be a zero-configuration framework. Where should it drop the logs for you if there is no configuration? Guess- ing is not a good idea because chances are, the place it guessed is not the place where the user has permission to create a logfile. Also, for most small applications nobody will look at the logs anyways. In fact, I promise you right now that if you configure a logfile for the application errors you will never look at it except for debugging an issue when a user reported it for you. 51
- 62. What you want instead is a mail the second the exception happened. Then you get an alert and you can do something about it. Flask uses the Python builtin logging system, and it can actually send you mails for errors which is probably what you want. Here is how you can configure the Flask logger to send you mails for exceptions: ADMINS = [’[email protected]’] if not app.debug: import logging from logging.handlers import SMTPHandler mail_handler = SMTPHandler(’127.0.0.1’, ’[email protected]’, ADMINS, ’YourApplication Failed’) mail_handler.setLevel(logging.ERROR) app.logger.addHandler(mail_handler) So what just happened? We created a new SMTPHandler that will send mails with the mail server listening on 127.0.0.1 to all the ADMINS from the address server- [email protected] with the subject “YourApplication Failed”. If your mail server re- quires credentials, these can also be provided. For that check out the documentation for the SMTPHandler. We also tell the handler to only send errors and more critical messages. Because we certainly don’t want to get a mail for warnings or other useless logs that might happen during request handling. Before you run that in production, please also look at Controlling the Log Format to put more information into that error mail. That will save you from a lot of frustration. 8.2 Logging to a File Even if you get mails, you probably also want to log warnings. It’s a good idea to keep as much information around that might be required to debug a problem. Please note that Flask itself will not issue any warnings in the core system, so it’s your responsi- bility to warn in the code if something seems odd. There are a couple of handlers provided by the logging system out of the box but not all of them are useful for basic error logging. The most interesting are probably the following: • FileHandler - logs messages to a file on the filesystem. • RotatingFileHandler - logs messages to a file on the filesystem and will rotate after a certain number of messages. • NTEventLogHandler - will log to the system event log of a Windows system. If you are deploying on a Windows box, this is what you want to use. • SysLogHandler - sends logs to a UNIX syslog. 52
- 63. Once you picked your log handler, do like you did with the SMTP handler above, just make sure to use a lower setting (I would recommend WARNING): if not app.debug: import logging from themodule import TheHandlerYouWant file_handler = TheHandlerYouWant(...) file_handler.setLevel(logging.WARNING) app.logger.addHandler(file_handler) 8.3 Controlling the Log Format By default a handler will only write the message string into a file or send you that message as mail. A log record stores more information, and it makes a lot of sense to configure your logger to also contain that information so that you have a better idea of why that error happened, and more importantly, where it did. A formatter can be instantiated with a format string. Note that tracebacks are ap- pended to the log entry automatically. You don’t have to do that in the log formatter format string. Here some example setups: 8.3.1 Email from logging import Formatter mail_handler.setFormatter(Formatter(’’’ Message type: %(levelname)s Location: %(pathname)s:%(lineno)d Module: %(module)s Function: %(funcName)s Time: %(asctime)s Message: %(message)s ’’’)) 8.3.2 File logging from logging import Formatter file_handler.setFormatter(Formatter( ’%(asctime)s %(levelname)s: %(message)s ’ ’[in %(pathname)s:%(lineno)d]’ )) 53
- 64. 8.3.3 Complex Log Formatting Here is a list of useful formatting variables for the format string. Note that this list is not complete, consult the official documentation of the logging package for a full list. Format Description %(levelname)s Text logging level for the message (’DEBUG’, ’INFO’, ’WARNING’, ’ERROR’, ’CRITICAL’). %(pathname)s Full pathname of the source file where the logging call was issued (if available). %(filename)s Filename portion of pathname. %(module)s Module (name portion of filename). %(funcName)s Name of function containing the logging call. %(lineno)d Source line number where the logging call was issued (if avail- able). %(asctime)s Human-readable time when the LogRecord‘ was created. By de- fault this is of the form "2003-07-08 16:49:45,896" (the num- bers after the comma are millisecond portion of the time). This can be changed by subclassing the formatter and overriding the formatTime() method. %(message)s The logged message, computed as msg % args If you want to further customize the formatting, you can subclass the formatter. The formatter has three interesting methods: format(): handles the actual formatting. It is passed a LogRecord object and has to return the formatted string. formatTime(): called for asctime formatting. If you want a different time format you can override this method. formatException() called for exception formatting. It is passed an exc_info tuple and has to return a string. The default is usually fine, you don’t have to override it. For more information, head over to the official documentation. 8.4 Other Libraries So far we only configured the logger your application created itself. Other libraries might log themselves as well. For example, SQLAlchemy uses logging heavily in its core. While there is a method to configure all loggers at once in the logging package, I would not recommend using it. There might be a situation in which you want to have multiple separate applications running side by side in the same Python interpreter and then it becomes impossible to have different logging setups for those. Instead, I would recommend figuring out which loggers you are interested in, getting the loggers with the getLogger() function and iterating over them to attach handlers: from logging import getLogger loggers = [app.logger, getLogger(’sqlalchemy’), 54
- 65. getLogger(’otherlibrary’)] for logger in loggers: logger.addHandler(mail_handler) logger.addHandler(file_handler) 55
- 66. 56
- 67. CHAPTER NINE DEBUGGING APPLICATION ERRORS For production applications, configure your application with logging and notifications as described in Logging Application Errors. This section provides pointers when de- bugging deployment configuration and digging deeper with a full-featured Python debugger. 9.1 When in Doubt, Run Manually Having problems getting your application configured for production? If you have shell access to your host, verify that you can run your application manually from the shell in the deployment environment. Be sure to run under the same user account as the configured deployment to troubleshoot permission issues. You can use Flask’s builtin development server with debug=True on your production host, which is helpful in catching configuration issues, but be sure to do this temporarily in a controlled environment. Do not run in production with debug=True. 9.2 Working with Debuggers To dig deeper, possibly to trace code execution, Flask provides a debugger out of the box (see Debug Mode). If you would like to use another Python debugger, note that debuggers interfere with each other. You have to set some options in order to use your favorite debugger: • debug - whether to enable debug mode and catch exceptinos • use_debugger - whether to use the internal Flask debugger • use_reloader - whether to reload and fork the process on exception debug must be True (i.e., exceptions must be caught) in order for the other two options to have any value. If you’re using Aptana/Eclipse for debugging you’ll need to set both use_debugger and use_reloader to False. 57
- 68. A possible useful pattern for configuration is to set the following in your config.yaml (change the block as appropriate for your application, of course): FLASK: DEBUG: True DEBUG_WITH_APTANA: True Then in your application’s entry-point (main.py), you could have something like: if __name__ == "__main__": # To allow aptana to receive errors, set use_debugger=False app = create_app(config="config.yaml") if app.debug: use_debugger = True try: # Disable Flask’s debugger if external debugger is requested use_debugger = not(app.config.get(’DEBUG_WITH_APTANA’)) except: pass app.run(use_debugger=use_debugger, debug=app.debug, use_reloader=use_debugger, host=’0.0.0.0’) 58
- 69. CHAPTER TEN CONFIGURATION HANDLING New in version 0.3. Applications need some kind of configuration. There are different settings you might want to change depending on the application environment like toggling the debug mode, setting the secret key, and other such environment-specific things. The way Flask is designed usually requires the configuration to be available when the application starts up. You can hardcode the configuration in the code, which for many small applications is not actually that bad, but there are better ways. Independent of how you load your config, there is a config object available which holds the loaded configuration values: The config attribute of the Flask object. This is the place where Flask itself puts certain configuration values and also where exten- sions can put their configuration values. But this is also where you can have your own configuration. 10.1 Configuration Basics The config is actually a subclass of a dictionary and can be modified just like any dictionary: app = Flask(__name__) app.config[’DEBUG’] = True Certain configuration values are also forwarded to the Flask object so you can read and write them from there: app.debug = True To update multiple keys at once you can use the dict.update() method: app.config.update( DEBUG=True, SECRET_KEY=’...’ ) 59
- 70. 10.2 Builtin Configuration Values The following configuration values are used internally by Flask: 60
- 71. DEBUG enable/disable debug mode TESTING enable/disable testing mode PROPAGATE_EXCEPTIONS explicitly enable or disable the propagation of exceptions. If not set or explicitly set to None this is implicitly true if either TESTING or DE- BUG is true. PRESERVE_CONTEXT_ON_EXCEPTION By default if the application is in debug mode the request context is not popped on excep- tions to enable debuggers to introspect the data. This can be disabled by this key. You can also use this setting to force-enable it for non debug execution which might be useful to debug production applications (but also very risky). SECRET_KEY the secret key SESSION_COOKIE_NAME the name of the session cookie SESSION_COOKIE_DOMAIN the domain for the session cookie. If this is not set, the cookie will be valid for all subdomains of SERVER_NAME. SESSION_COOKIE_PATH the path for the session cookie. If this is not set the cookie will be valid for all of APPLICATION_ROOT or if that is not set for ’/’. SESSION_COOKIE_HTTPONLY controls if the cookie should be set with the httponly flag. Defaults to True. SESSION_COOKIE_SECURE controls if the cookie should be set with the secure flag. Defaults to False. PERMANENT_SESSION_LIFETIME the lifetime of a permanent session as datetime.timedelta object. Starting with Flask 0.8 this can also be an integer represent- ing seconds. SESSION_REFRESH_EACH_REQUEST this flag controls how permanent sessions are refreshed. If set to True (which is the de- fault) then the cookie is refreshed each request which automatically bumps the lifetime. If set to False a set-cookie header is only sent if the session is modified. Non permanent sessions are not affected by this. USE_X_SENDFILE enable/disable x-sendfile LOGGER_NAME the name of the logger SERVER_NAME the name and port number of the server. Required for subdomain support (e.g.: ’myapp.dev:5000’) Note that localhost does not support subdomains so setting this to “lo- calhost” does not help. Setting a SERVER_NAME also by default enables URL generation with- out a request context but with an application context. APPLICATION_ROOT If the application does not occupy a whole do- main or subdomain this can be set to the path where the application is configured to live. This is for session cookie as path value. If do- mains are used, this should be None. MAX_CONTENT_LENGTH If set to a value in bytes, Flask will reject in- 61
- 72. More on SERVER_NAME The SERVER_NAME key is used for the subdomain support. Because Flask cannot guess the subdomain part without the knowledge of the actual server name, this is required if you want to work with subdomains. This is also used for the session cookie. Please keep in mind that not only Flask has the problem of not knowing what sub- domains are, your web browser does as well. Most modern web browsers will not allow cross-subdomain cookies to be set on a server name without dots in it. So if your server name is ’localhost’ you will not be able to set a cookie for ’localhost’ and every subdomain of it. Please chose a different server name in that case, like ’myapplication.local’ and add this name + the subdomains you want to use into your host config or setup a local bind. New in version 0.4: LOGGER_NAME New in version 0.5: SERVER_NAME New in version 0.6: MAX_CONTENT_LENGTH New in version 0.7: PROPAGATE_EXCEPTIONS, PRESERVE_CONTEXT_ON_EXCEPTION New in version 0.8: TRAP_BAD_REQUEST_ERRORS, TRAP_HTTP_EXCEPTIONS, APPLICATION_ROOT, SESSION_COOKIE_DOMAIN, SESSION_COOKIE_PATH, SESSION_COOKIE_HTTPONLY, SESSION_COOKIE_SECURE New in version 0.9: PREFERRED_URL_SCHEME New in version 0.10: JSON_AS_ASCII, JSON_SORT_KEYS, JSONIFY_PRETTYPRINT_REGULAR New in version 1.0: SESSION_REFRESH_EACH_REQUEST 10.3 Configuring from Files Configuration becomes more useful if you can store it in a separate file, ideally located outside the actual application package. This makes packaging and distributing your application possible via various package handling tools (Deploying with Distribute) and finally modifying the configuration file afterwards. So a common pattern is this: app = Flask(__name__) app.config.from_object(’yourapplication.default_settings’) app.config.from_envvar(’YOURAPPLICATION_SETTINGS’) This first loads the configuration from the yourapplication.default_settings module and then overrides the values with the contents of the file the YOURAPPLICATION_SETTINGS environment variable points to. This environment variable can be set on Linux or OS X with the export command in the shell before starting the server: 62
- 73. $ export YOURAPPLICATION_SETTINGS=/path/to/settings.cfg $ python run-app.py * Running on http://127.0.0.1:5000/ * Restarting with reloader... On Windows systems use the set builtin instead: >set YOURAPPLICATION_SETTINGS=pathtosettings.cfg The configuration files themselves are actual Python files. Only values in uppercase are actually stored in the config object later on. So make sure to use uppercase letters for your config keys. Here is an example of a configuration file: # Example configuration DEBUG = False SECRET_KEY = ’?xbf,xb4x8dxa3"<x9cxb0@x0f5xab,wxeex8d$0x13x8b83’ Make sure to load the configuration very early on, so that extensions have the ability to access the configuration when starting up. There are other methods on the config object as well to load from individual files. For a complete reference, read the Config object’s documentation. 10.4 Configuration Best Practices The downside with the approach mentioned earlier is that it makes testing a little harder. There is no single 100% solution for this problem in general, but there are a couple of things you can keep in mind to improve that experience: 1. create your application in a function and register blueprints on it. That way you can create multiple instances of your application with different configurations attached which makes unittesting a lot easier. You can use this to pass in config- uration as needed. 2. Do not write code that needs the configuration at import time. If you limit your- self to request-only accesses to the configuration you can reconfigure the object later on as needed. 10.5 Development / Production Most applications need more than one configuration. There should be at least separate configurations for the production server and the one used during development. The easiest way to handle this is to use a default configuration that is always loaded and part of the version control, and a separate configuration that overrides the values as necessary as mentioned in the example above: 63
- 74. app = Flask(__name__) app.config.from_object(’yourapplication.default_settings’) app.config.from_envvar(’YOURAPPLICATION_SETTINGS’) Then you just have to add a separate config.py file and export YOURAPPLICATION_SETTINGS=/path/to/config.py and you are done. However there are alternative ways as well. For example you could use imports or subclassing. What is very popular in the Django world is to make the import explicit in the config file by adding an from yourapplication.default_settings import * to the top of the file and then overriding the changes by hand. You could also inspect an environment variable like YOURAPPLICATION_MODE and set that to production, development etc and im- port different hardcoded files based on that. An interesting pattern is also to use classes and inheritance for configuration: class Config(object): DEBUG = False TESTING = False DATABASE_URI = ’sqlite://:memory:’ class ProductionConfig(Config): DATABASE_URI = ’mysql://user@localhost/foo’ class DevelopmentConfig(Config): DEBUG = True class TestingConfig(Config): TESTING = True To enable such a config you just have to call into from_object(): app.config.from_object(’configmodule.ProductionConfig’) There are many different ways and it’s up to you how you want to manage your con- figuration files. However here a list of good recommendations: • keep a default configuration in version control. Either populate the config with this default configuration or import it in your own configuration files before overriding values. • use an environment variable to switch between the configurations. This can be done from outside the Python interpreter and makes development and deploy- ment much easier because you can quickly and easily switch between different configs without having to touch the code at all. If you are working often on dif- ferent projects you can even create your own script for sourcing that activates a virtualenv and exports the development configuration for you. • Use a tool like fabric in production to push code and configurations separately to the production server(s). For some details about how to do that, head over to the Deploying with Fabric pattern. 64
- 75. 10.6 Instance Folders New in version 0.8. Flask 0.8 introduces instance folders. Flask for a long time made it possible to refer to paths relative to the application’s folder directly (via Flask.root_path). This was also how many developers loaded configurations stored next to the application. Unfortu- nately however this only works well if applications are not packages in which case the root path refers to the contents of the package. With Flask 0.8 a new attribute was introduced: Flask.instance_path. It refers to a new concept called the “instance folder”. The instance folder is designed to not be under version control and be deployment specific. It’s the perfect place to drop things that either change at runtime or configuration files. You can either explicitly provide the path of the instance folder when creating the Flask application or you can let Flask autodetect the instance folder. For explicit con- figuration use the instance_path parameter: app = Flask(__name__, instance_path=’/path/to/instance/folder’) Please keep in mind that this path must be absolute when provided. If the instance_path parameter is not provided the following default locations are used: • Uninstalled module: /myapp.py /instance • Uninstalled package: /myapp /__init__.py /instance • Installed module or package: $PREFIX/lib/python2.X/site-packages/myapp $PREFIX/var/myapp-instance $PREFIX is the prefix of your Python installation. This can be /usr or the path to your virtualenv. You can print the value of sys.prefix to see what the prefix is set to. Since the config object provided loading of configuration files from relative filenames we made it possible to change the loading via filenames to be relative to the instance path if wanted. The behavior of relative paths in config files can be flipped between “relative to the application root” (the default) to “relative to instance folder” via the instance_relative_config switch to the application constructor: app = Flask(__name__, instance_relative_config=True) 65
- 76. Here is a full example of how to configure Flask to preload the config from a module and then override the config from a file in the config folder if it exists: app = Flask(__name__, instance_relative_config=True) app.config.from_object(’yourapplication.default_settings’) app.config.from_pyfile(’application.cfg’, silent=True) The path to the instance folder can be found via the Flask.instance_path. Flask also provides a shortcut to open a file from the instance folder with Flask.open_instance_resource(). Example usage for both: filename = os.path.join(app.instance_path, ’application.cfg’) with open(filename) as f: config = f.read() # or via open_instance_resource: with app.open_instance_resource(’application.cfg’) as f: config = f.read() 66
- 77. CHAPTER ELEVEN SIGNALS New in version 0.6. Starting with Flask 0.6, there is integrated support for signalling in Flask. This support is provided by the excellent blinker library and will gracefully fall back if it is not available. What are signals? Signals help you decouple applications by sending notifications when actions occur elsewhere in the core framework or another Flask extensions. In short, signals allow certain senders to notify subscribers that something happened. Flask comes with a couple of signals and other extensions might provide more. Also keep in mind that signals are intended to notify subscribers and should not encourage subscribers to modify data. You will notice that there are signals that appear to do the same thing like some of the builtin decorators do (eg: request_started is very similar to before_request()). There are however difference in how they work. The core before_request() handler for example is executed in a specific order and is able to abort the request early by returning a response. In contrast all signal handlers are executed in undefined order and do not modify any data. The big advantage of signals over handlers is that you can safely subscribe to them for the split of a second. These temporary subscriptions are helpful for unittesting for example. Say you want to know what templates were rendered as part of a request: signals allow you to do exactly that. 11.1 Subscribing to Signals To subscribe to a signal, you can use the connect() method of a signal. The first ar- gument is the function that should be called when the signal is emitted, the optional second argument specifies a sender. To unsubscribe from a signal, you can use the disconnect() method. For all core Flask signals, the sender is the application that issued the signal. When you subscribe to a signal, be sure to also provide a sender unless you really want to listen for signals of all applications. This is especially true if you are developing an extension. 67
- 78. Here for example a helper context manager that can be used to figure out in a unittest which templates were rendered and what variables were passed to the template: from flask import template_rendered from contextlib import contextmanager @contextmanager def captured_templates(app): recorded = [] def record(sender, template, context, **extra): recorded.append((template, context)) template_rendered.connect(record, app) try: yield recorded finally: template_rendered.disconnect(record, app) This can now easily be paired with a test client: with captured_templates(app) as templates: rv = app.test_client().get(’/’) assert rv.status_code == 200 assert len(templates) == 1 template, context = templates[0] assert template.name == ’index.html’ assert len(context[’items’]) == 10 Make sure to subscribe with an extra **extra argument so that your calls don’t fail if Flask introduces new arguments to the signals. All the template rendering in the code issued by the application app in the body of the with block will now be recorded in the templates variable. Whenever a template is rendered, the template object as well as context are appended to it. Additionally there is a convenient helper method (connected_to()). that allows you to temporarily subscribe a function to a signal with a context manager on its own. Because the return value of the context manager cannot be specified that way one has to pass the list in as argument: from flask import template_rendered def captured_templates(app, recorded, **extra): def record(sender, template, context): recorded.append((template, context)) return template_rendered.connected_to(record, app) The example above would then look like this: templates = [] with captured_templates(app, templates, **extra): ... template, context = templates[0] 68
- 79. Blinker API Changes The connected_to() method arrived in Blinker with version 1.1. 11.2 Creating Signals If you want to use signals in your own application, you can use the blinker library directly. The most common use case are named signals in a custom Namespace.. This is what is recommended most of the time: from blinker import Namespace my_signals = Namespace() Now you can create new signals like this: model_saved = my_signals.signal(’model-saved’) The name for the signal here makes it unique and also simplifies debugging. You can access the name of the signal with the name attribute. For Extension Developers If you are writing a Flask extension and you want to gracefully degrade for missing blinker installations, you can do so by using the flask.signals.Namespace class. 11.3 Sending Signals If you want to emit a signal, you can do so by calling the send() method. It accepts a sender as first argument and optionally some keyword arguments that are forwarded to the signal subscribers: class Model(object): ... def save(self): model_saved.send(self) Try to always pick a good sender. If you have a class that is emitting a signal, pass self as sender. If you emitting a signal from a random function, you can pass current_app._get_current_object() as sender. Passing Proxies as Senders Never pass current_app as sender to a signal. Use current_app._get_current_object() instead. The reason for this is that current_app is a proxy and not the real application object. 69
- 80. 11.4 Signals and Flask’s Request Context Signals fully support The Request Context when receiving signals. Context-local vari- ables are consistently available between request_started and request_finished, so you can rely on flask.g and others as needed. Note the limitations described in Send- ing Signals and the request_tearing_down signal. 11.5 Decorator Based Signal Subscriptions With Blinker 1.1 you can also easily subscribe to signals by using the new connect_via() decorator: from flask import template_rendered @template_rendered.connect_via(app) def when_template_rendered(sender, template, context, **extra): print ’Template %s is rendered with %s’ % (template.name, context) 11.6 Core Signals The following signals exist in Flask: flask.template_rendered This signal is sent when a template was successfully rendered. The signal is in- voked with the instance of the template as template and the context as dictionary (named context). Example subscriber: def log_template_renders(sender, template, context, **extra): sender.logger.debug(’Rendering template "%s" with context %s’, template.name or ’string template’, context) from flask import template_rendered template_rendered.connect(log_template_renders, app) flask.request_started This signal is sent before any request processing started but when the request context was set up. Because the request context is already bound, the subscriber can access the request with the standard global proxies such as request. Example subscriber: def log_request(sender, **extra): sender.logger.debug(’Request context is set up’) 70
- 81. from flask import request_started request_started.connect(log_request, app) flask.request_finished This signal is sent right before the response is sent to the client. It is passed the response to be sent named response. Example subscriber: def log_response(sender, response, **extra): sender.logger.debug(’Request context is about to close down. ’ ’Response: %s’, response) from flask import request_finished request_finished.connect(log_response, app) flask.got_request_exception This signal is sent when an exception happens during request processing. It is sent before the standard exception handling kicks in and even in debug mode, where no exception handling happens. The exception itself is passed to the sub- scriber as exception. Example subscriber: def log_exception(sender, exception, **extra): sender.logger.debug(’Got exception during processing: %s’, exception) from flask import got_request_exception got_request_exception.connect(log_exception, app) flask.request_tearing_down This signal is sent when the request is tearing down. This is always called, even if an exception is caused. Currently functions listening to this signal are called after the regular teardown handlers, but this is not something you can rely on. Example subscriber: def close_db_connection(sender, **extra): session.close() from flask import request_tearing_down request_tearing_down.connect(close_db_connection, app) As of Flask 0.9, this will also be passed an exc keyword argument that has a reference to the exception that caused the teardown if there was one. flask.appcontext_tearing_down This signal is sent when the app context is tearing down. This is always called, even if an exception is caused. Currently functions listening to this signal are called after the regular teardown handlers, but this is not something you can rely on. Example subscriber: 71
- 82. def close_db_connection(sender, **extra): session.close() from flask import appcontext_tearing_down appcontext_tearing_down.connect(close_db_connection, app) This will also be passed an exc keyword argument that has a reference to the exception that caused the teardown if there was one. flask.appcontext_pushed This signal is sent when an application context is pushed. The sender is the application. This is usually useful for unittests in order to temporarily hook in information. For instance it can be used to set a resource early onto the g object. Example usage: from contextlib import contextmanager from flask import appcontext_pushed @contextmanager def user_set(app, user): def handler(sender, **kwargs): g.user = user with appcontext_pushed.connected_to(handler, app): yield And in the testcode: def test_user_me(self): with user_set(app, ’john’): c = app.test_client() resp = c.get(’/users/me’) assert resp.data == ’username=john’ New in version 0.10. flask.appcontext_popped This signal is sent when an application context is popped. The sender is the application. This usually falls in line with the appcontext_tearing_down signal. New in version 0.10. flask.message_flashed This signal is sent when the application is flashing a message. The messages is sent as message keyword argument and the category as category. Example subscriber: recorded = [] def record(sender, message, category, **extra): recorded.append((message, category)) from flask import message_flashed message_flashed.connect(record, app) 72
- 83. New in version 0.10. 73
- 84. 74
- 85. CHAPTER TWELVE PLUGGABLE VIEWS New in version 0.7. Flask 0.7 introduces pluggable views inspired by the generic views from Django which are based on classes instead of functions. The main intention is that you can replace parts of the implementations and this way have customizable pluggable views. 12.1 Basic Principle Consider you have a function that loads a list of objects from the database and renders into a template: @app.route(’/users/’) def show_users(page): users = User.query.all() return render_template(’users.html’, users=users) This is simple and flexible, but if you want to provide this view in a generic fashion that can be adapted to other models and templates as well you might want more flex- ibility. This is where pluggable class-based views come into place. As the first step to convert this into a class based view you would do this: from flask.views import View class ShowUsers(View): def dispatch_request(self): users = User.query.all() return render_template(’users.html’, objects=users) app.add_url_rule(’/users/’, view_func=ShowUsers.as_view(’show_users’)) As you can see what you have to do is to create a subclass of flask.views.View and im- plement dispatch_request(). Then we have to convert that class into an actual view function by using the as_view() class method. The string you pass to that function is the name of the endpoint that view will then have. But this by itself is not helpful, so let’s refactor the code a bit: 75
- 86. from flask.views import View class ListView(View): def get_template_name(self): raise NotImplementedError() def render_template(self, context): return render_template(self.get_template_name(), **context) def dispatch_request(self): context = {’objects’: self.get_objects()} return self.render_template(context) class UserView(ListView): def get_template_name(self): return ’users.html’ def get_objects(self): return User.query.all() This of course is not that helpful for such a small example, but it’s good enough to explain the basic principle. When you have a class-based view the question comes up what self points to. The way this works is that whenever the request is dispatched a new instance of the class is created and the dispatch_request() method is called with the parameters from the URL rule. The class itself is instantiated with the parameters passed to the as_view() function. For instance you can write a class like this: class RenderTemplateView(View): def __init__(self, template_name): self.template_name = template_name def dispatch_request(self): return render_template(self.template_name) And then you can register it like this: app.add_url_rule(’/about’, view_func=RenderTemplateView.as_view( ’about_page’, template_name=’about.html’)) 12.2 Method Hints Pluggable views are attached to the application like a regular function by either using route() or better add_url_rule(). That however also means that you would have to provide the names of the HTTP methods the view supports when you attach this. In order to move that information to the class you can provide a methods attribute that has this information: 76
- 87. class MyView(View): methods = [’GET’, ’POST’] def dispatch_request(self): if request.method == ’POST’: ... ... app.add_url_rule(’/myview’, view_func=MyView.as_view(’myview’)) 12.3 Method Based Dispatching For RESTful APIs it’s especially helpful to execute a different function for each HTTP method. With the flask.views.MethodView you can easily do that. Each HTTP method maps to a function with the same name (just in lowercase): from flask.views import MethodView class UserAPI(MethodView): def get(self): users = User.query.all() ... def post(self): user = User.from_form_data(request.form) ... app.add_url_rule(’/users/’, view_func=UserAPI.as_view(’users’)) That way you also don’t have to provide the methods attribute. It’s automatically set based on the methods defined in the class. 12.4 Decorating Views Since the view class itself is not the view function that is added to the routing system it does not make much sense to decorate the class itself. Instead you either have to decorate the return value of as_view() by hand: def user_required(f): """Checks whether user is logged in or raises error 401.""" def decorator(*args, **kwargs): if not g.user: abort(401) return f(*args, **kwargs) return decorator 77
- 88. view = user_required(UserAPI.as_view(’users’)) app.add_url_rule(’/users/’, view_func=view) Starting with Flask 0.8 there is also an alternative way where you can specify a list of decorators to apply in the class declaration: class UserAPI(MethodView): decorators = [user_required] Due to the implicit self from the caller’s perspective you cannot use regular view dec- orators on the individual methods of the view however, keep this in mind. 12.5 Method Views for APIs Web APIs are often working very closely with HTTP verbs so it makes a lot of sense to implement such an API based on the MethodView. That said, you will notice that the API will require different URL rules that go to the same method view most of the time. For instance consider that you are exposing a user object on the web: URL Method Description /users/ GET Gives a list of all users /users/ POST Creates a new user /users/<id> GET Shows a single user /users/<id> PUT Updates a single user /users/<id> DELETE Deletes a single user So how would you go about doing that with the MethodView? The trick is to take advantage of the fact that you can provide multiple rules to the same view. Let’s assume for the moment the view would look like this: class UserAPI(MethodView): def get(self, user_id): if user_id is None: # return a list of users pass else: # expose a single user pass def post(self): # create a new user pass def delete(self, user_id): # delete a single user pass def put(self, user_id): 78
- 89. # update a single user pass So how do we hook this up with the routing system? By adding two rules and explic- itly mentioning the methods for each: user_view = UserAPI.as_view(’user_api’) app.add_url_rule(’/users/’, defaults={’user_id’: None}, view_func=user_view, methods=[’GET’,]) app.add_url_rule(’/users/’, view_func=user_view, methods=[’POST’,]) app.add_url_rule(’/users/<int:user_id>’, view_func=user_view, methods=[’GET’, ’PUT’, ’DELETE’]) If you have a lot of APIs that look similar you can refactor that registration code: def register_api(view, endpoint, url, pk=’id’, pk_type=’int’): view_func = view.as_view(endpoint) app.add_url_rule(url, defaults={pk: None}, view_func=view_func, methods=[’GET’,]) app.add_url_rule(url, view_func=view_func, methods=[’POST’,]) app.add_url_rule(’%s<%s:%s>’ % (url, pk_type, pk), view_func=view_func, methods=[’GET’, ’PUT’, ’DELETE’]) register_api(UserAPI, ’user_api’, ’/users/’, pk=’user_id’) 79
- 90. 80
- 91. CHAPTER THIRTEEN THE APPLICATION CONTEXT New in version 0.9. One of the design ideas behind Flask is that there are two different “states” in which code is executed. The application setup state in which the application implicitly is on the module level. It starts when the Flask object is instantiated, and it implicitly ends when the first request comes in. While the application is in this state a few assumptions are true: • the programmer can modify the application object safely. • no request handling happened so far • you have to have a reference to the application object in order to modify it, there is no magic proxy that can give you a reference to the application object you’re currently creating or modifying. In contrast, during request handling, a couple of other rules exist: • while a request is active, the context local objects (flask.request and others) point to the current request. • any code can get hold of these objects at any time. There is a third state which is sitting in between a little bit. Sometimes you are dealing with an application in a way that is similar to how you interact with applications during request handling just that there is no request active. Consider for instance that you’re sitting in an interactive Python shell and interacting with the application, or a command line application. The application context is what powers the current_app context local. 13.1 Purpose of the Application Context The main reason for the application’s context existence is that in the past a bunch of functionality was attached to the request context in lack of a better solution. Since one of the pillar’s of Flask’s design is that you can have more than one application in the same Python process. 81
- 92. So how does the code find the “right” application? In the past we recommended pass- ing applications around explicitly, but that caused issues with libraries that were not designed with that in mind. A common workaround for that problem was to use the current_app proxy later on, which was bound to the current request’s application reference. Since however creat- ing such a request context is an unnecessarily expensive operation in case there is no request around, the application context was introduced. 13.2 Creating an Application Context To make an application context there are two ways. The first one is the implicit one: whenever a request context is pushed, an application context will be created alongside if this is necessary. As a result of that, you can ignore the existence of the application context unless you need it. The second way is the explicit way using the app_context() method: from flask import Flask, current_app app = Flask(__name__) with app.app_context(): # within this block, current_app points to app. print current_app.name The application context is also used by the url_for() function in case a SERVER_NAME was configured. This allows you to generate URLs even in the absence of a request. 13.3 Locality of the Context The application context is created and destroyed as necessary. It never moves between threads and it will not be shared between requests. As such it is the perfect place to store database connection information and other things. The internal stack object is called flask._app_ctx_stack. Extensions are free to store additional information on the topmost level, assuming they pick a sufficiently unique name and should put their information there, instead of on the flask.g object which is reserved for user code. For more information about that, see Flask Extension Development. 13.4 Context Usage The context is typically used to cache resources on there that need to be created on a per-request or usage case. For instance database connects are destined to go there. When storing things on the application context unique names should be chosen as this is a place that is shared between Flask applications and extensions. 82
- 93. The most common usage is to split resource management into two parts: 1. an implicit resource caching on the context. 2. a context teardown based resource deallocation. Generally there would be a get_X() function that creates resource X if it does not ex- ist yet and otherwise returns the same resource, and a teardown_X() function that is registered as teardown handler. This is an example that connects to a database: import sqlite3 from flask import g def get_db(): db = getattr(g, ’_database’, None) if db is None: db = g._database = connect_to_database() return db @app.teardown_appcontext def teardown_db(exception): db = getattr(g, ’_database’, None) if db is not None: db.close() The first time get_db() is called the connection will be established. To make this im- plicit a LocalProxy can be used: from werkzeug.local import LocalProxy db = LocalProxy(get_db) That way a user can directly access db which internally calls get_db(). 83
- 94. 84
- 95. CHAPTER FOURTEEN THE REQUEST CONTEXT This document describes the behavior in Flask 0.7 which is mostly in line with the old behavior but has some small, subtle differences. It is recommended that you read the The Application Context chapter first. 14.1 Diving into Context Locals Say you have a utility function that returns the URL the user should be redirected to. Imagine it would always redirect to the URL’s next parameter or the HTTP referrer or the index page: from flask import request, url_for def redirect_url(): return request.args.get(’next’) or request.referrer or url_for(’index’) As you can see, it accesses the request object. If you try to run this from a plain Python shell, this is the exception you will see: >>> redirect_url() Traceback (most recent call last): File "<stdin>", line 1, in <module> AttributeError: ’NoneType’ object has no attribute ’request’ That makes a lot of sense because we currently do not have a request we could access. So we have to make a request and bind it to the current context. The test_request_context method can create us a RequestContext: >>> ctx = app.test_request_context(’/?next=http://example.com/’) This context can be used in two ways. Either with the with statement or by calling the push() and pop() methods: >>> ctx.push() 85
- 96. From that point onwards you can work with the request object: >>> redirect_url() u’http://example.com/’ Until you call pop: >>> ctx.pop() Because the request context is internally maintained as a stack you can push and pop multiple times. This is very handy to implement things like internal redirects. For more information of how to utilize the request context from the interactive Python shell, head over to the Working with the Shell chapter. 14.2 How the Context Works If you look into how the Flask WSGI application internally works, you will find a piece of code that looks very much like this: def wsgi_app(self, environ): with self.request_context(environ): try: response = self.full_dispatch_request() except Exception, e: response = self.make_response(self.handle_exception(e)) return response(environ, start_response) The method request_context() returns a new RequestContext object and uses it in combination with the with statement to bind the context. Everything that is called from the same thread from this point onwards until the end of the with statement will have access to the request globals (flask.request and others). The request context internally works like a stack: The topmost level on the stack is the current active request. push() adds the context to the stack on the very top, pop() removes it from the stack again. On popping the application’s teardown_request() functions are also executed. Another thing of note is that the request context will automatically also create an appli- cation context when it’s pushed and there is no application context for that application so far. 14.3 Callbacks and Errors What happens if an error occurs in Flask during request processing? This particular behavior changed in 0.7 because we wanted to make it easier to understand what is actually happening. The new behavior is quite simple: 86
- 97. 1. Before each request, before_request() functions are executed. If one of these functions return a response, the other functions are no longer called. In any case however the return value is treated as a replacement for the view’s return value. 2. If the before_request() functions did not return a response, the regular request handling kicks in and the view function that was matched has the chance to return a response. 3. The return value of the view is then converted into an actual response object and handed over to the after_request() functions which have the chance to replace it or modify it in place. 4. At the end of the request the teardown_request() functions are executed. This always happens, even in case of an unhandled exception down the road or if a before-request handler was not executed yet or at all (for example in test envi- ronments sometimes you might want to not execute before-request callbacks). Now what happens on errors? In production mode if an exception is not caught, the 500 internal server handler is called. In development mode however the exception is not further processed and bubbles up to the WSGI server. That way things like the interactive debugger can provide helpful debug information. An important change in 0.7 is that the internal server error is now no longer post processed by the after request callbacks and after request callbacks are no longer guar- anteed to be executed. This way the internal dispatching code looks cleaner and is easier to customize and understand. The new teardown functions are supposed to be used as a replacement for things that absolutely need to happen at the end of request. 14.4 Teardown Callbacks The teardown callbacks are special callbacks in that they are executed at a different point. Strictly speaking they are independent of the actual request handling as they are bound to the lifecycle of the RequestContext object. When the request context is popped, the teardown_request() functions are called. This is important to know if the life of the request context is prolonged by using the test client in a with statement or when using the request context from the command line: with app.test_client() as client: resp = client.get(’/foo’) # the teardown functions are still not called at that point # even though the response ended and you have the response # object in your hand # only when the code reaches this point the teardown functions # are called. Alternatively the same thing happens if another # request was triggered from the test client 87
- 98. It’s easy to see the behavior from the command line: >>> app = Flask(__name__) >>> @app.teardown_request ... def teardown_request(exception=None): ... print ’this runs after request’ ... >>> ctx = app.test_request_context() >>> ctx.push() >>> ctx.pop() this runs after request >>> Keep in mind that teardown callbacks are always executed, even if before-request call- backs were not executed yet but an exception happened. Certain parts of the test sys- tem might also temporarily create a request context without calling the before-request handlers. Make sure to write your teardown-request handlers in a way that they will never fail. 14.5 Notes On Proxies Some of the objects provided by Flask are proxies to other objects. The reason behind this is that these proxies are shared between threads and they have to dispatch to the actual object bound to a thread behind the scenes as necessary. Most of the time you don’t have to care about that, but there are some exceptions where it is good to know that this object is an actual proxy: • The proxy objects do not fake their inherited types, so if you want to perform actual instance checks, you have to do that on the instance that is being proxied (see _get_current_object below). • if the object reference is important (so for example for sending Signals) If you need to get access to the underlying object that is proxied, you can use the _get_current_object() method: app = current_app._get_current_object() my_signal.send(app) 14.6 Context Preservation on Error If an error occurs or not, at the end of the request the request context is popped and all data associated with it is destroyed. During development however that can be problematic as you might want to have the information around for a longer time in case an exception occurred. In Flask 0.6 and earlier in debug mode, if an exception occurred, the request context was not popped so that the interactive debugger can still provide you with important information. 88
- 99. Starting with Flask 0.7 you have finer control over that behavior by setting the PRESERVE_CONTEXT_ON_EXCEPTION configuration variable. By default it’s linked to the setting of DEBUG. If the application is in debug mode the context is preserved, in pro- duction mode it’s not. Do not force activate PRESERVE_CONTEXT_ON_EXCEPTION in production mode as it will cause your application to leak memory on exceptions. However it can be useful during development to get the same error preserving behavior as in development mode when attempting to debug an error that only occurs under production settings. 89
- 100. 90
- 101. CHAPTER FIFTEEN MODULAR APPLICATIONS WITH BLUEPRINTS New in version 0.7. Flask uses a concept of blueprints for making application components and supporting common patterns within an application or across applications. Blueprints can greatly simplify how large applications work and provide a central means for Flask extensions to register operations on applications. A Blueprint object works similarly to a Flask application object, but it is not actually an application. Rather it is a blueprint of how to construct or extend an application. 15.1 Why Blueprints? Blueprints in Flask are intended for these cases: • Factor an application into a set of blueprints. This is ideal for larger applications; a project could instantiate an application object, initialize several extensions, and register a collection of blueprints. • Register a blueprint on an application at a URL prefix and/or subdomain. Pa- rameters in the URL prefix/subdomain become common view arguments (with defaults) across all view functions in the blueprint. • Register a blueprint multiple times on an application with different URL rules. • Provide template filters, static files, templates, and other utilities through blueprints. A blueprint does not have to implement applications or view func- tions. • Register a blueprint on an application for any of these cases when initializing a Flask extension. A blueprint in Flask is not a pluggable app because it is not actually an application – it’s a set of operations which can be registered on an application, even multiple times. Why not have multiple application objects? You can do that (see Application Dispatching), but your applications will have separate configs and will be managed at the WSGI layer. 91
- 102. Blueprints instead provide separation at the Flask level, share application config, and can change an application object as necessary with being registered. The downside is that you cannot unregister a blueprint once an application was created without having to destroy the whole application object. 15.2 The Concept of Blueprints The basic concept of blueprints is that they record operations to execute when reg- istered on an application. Flask associates view functions with blueprints when dis- patching requests and generating URLs from one endpoint to another. 15.3 My First Blueprint This is what a very basic blueprint looks like. In this case we want to implement a blueprint that does simple rendering of static templates: from flask import Blueprint, render_template, abort from jinja2 import TemplateNotFound simple_page = Blueprint(’simple_page’, __name__, template_folder=’templates’) @simple_page.route(’/’, defaults={’page’: ’index’}) @simple_page.route(’/<page>’) def show(page): try: return render_template(’pages/%s.html’ % page) except TemplateNotFound: abort(404) When you bind a function with the help of the @simple_page.route decorator the blueprint will record the intention of registering the function show on the application when it’s later registered. Additionally it will prefix the endpoint of the function with the name of the blueprint which was given to the Blueprint constructor (in this case also simple_page). 15.4 Registering Blueprints So how do you register that blueprint? Like this: from flask import Flask from yourapplication.simple_page import simple_page app = Flask(__name__) app.register_blueprint(simple_page) 92
- 103. If you check the rules registered on the application, you will find these: [<Rule ’/static/<filename>’ (HEAD, OPTIONS, GET) -> static>, <Rule ’/<page>’ (HEAD, OPTIONS, GET) -> simple_page.show>, <Rule ’/’ (HEAD, OPTIONS, GET) -> simple_page.show>] The first one is obviously from the application ifself for the static files. The other two are for the show function of the simple_page blueprint. As you can see, they are also prefixed with the name of the blueprint and separated by a dot (.). Blueprints however can also be mounted at different locations: app.register_blueprint(simple_page, url_prefix=’/pages’) And sure enough, these are the generated rules: [<Rule ’/static/<filename>’ (HEAD, OPTIONS, GET) -> static>, <Rule ’/pages/<page>’ (HEAD, OPTIONS, GET) -> simple_page.show>, <Rule ’/pages/’ (HEAD, OPTIONS, GET) -> simple_page.show>] On top of that you can register blueprints multiple times though not every blueprint might respond properly to that. In fact it depends on how the blueprint is imple- mented if it can be mounted more than once. 15.5 Blueprint Resources Blueprints can provide resources as well. Sometimes you might want to introduce a blueprint only for the resources it provides. 15.5.1 Blueprint Resource Folder Like for regular applications, blueprints are considered to be contained in a folder. While multiple blueprints can originate from the same folder, it does not have to be the case and it’s usually not recommended. The folder is inferred from the second argument to Blueprint which is usually __name__. This argument specifies what logical Python module or package corre- sponds to the blueprint. If it points to an actual Python package that package (which is a folder on the filesystem) is the resource folder. If it’s a module, the package the mod- ule is contained in will be the resource folder. You can access the Blueprint.root_path property to see what the resource folder is: >>> simple_page.root_path ’/Users/username/TestProject/yourapplication’ To quickly open sources from this folder you can use the open_resource() function: with simple_page.open_resource(’static/style.css’) as f: code = f.read() 93
- 104. 15.5.2 Static Files A blueprint can expose a folder with static files by providing a path to a folder on the filesystem via the static_folder keyword argument. It can either be an absolute path or one relative to the folder of the blueprint: admin = Blueprint(’admin’, __name__, static_folder=’static’) By default the rightmost part of the path is where it is exposed on the web. Because the folder is called static here it will be available at the location of the blueprint + /static. Say the blueprint is registered for /admin the static folder will be at /admin/static. The endpoint is named blueprint_name.static so you can generate URLs to it like you would do to the static folder of the application: url_for(’admin.static’, filename=’style.css’) 15.5.3 Templates If you want the blueprint to expose templates you can do that by providing the tem- plate_folder parameter to the Blueprint constructor: admin = Blueprint(’admin’, __name__, template_folder=’templates’) As for static files, the path can be absolute or relative to the blueprint resource folder. The template folder is added to the searchpath of templates but with a lower prior- ity than the actual application’s template folder. That way you can easily override templates that a blueprint provides in the actual application. So if you have a blueprint in the folder yourapplication/admin and you want to render the template ’admin/index.html’ and you have provided templates as a template_folder you will have to create a file like this: yourapplication/admin/templates/admin/index.html. 15.6 Building URLs If you want to link from one page to another you can use the url_for() function just like you normally would do just that you prefix the URL endpoint with the name of the blueprint and a dot (.): url_for(’admin.index’) Additionally if you are in a view function of a blueprint or a rendered template and you want to link to another endpoint of the same blueprint, you can use relative redi- rects by prefixing the endpoint with a dot only: 94
- 105. url_for(’.index’) This will link to admin.index for instance in case the current request was dispatched to any other admin blueprint endpoint. 95
- 106. 96
- 107. CHAPTER SIXTEEN FLASK EXTENSIONS Flask extensions extend the functionality of Flask in various different ways. For in- stance they add support for databases and other common tasks. 16.1 Finding Extensions Flask extensions are listed on the Flask Extension Registry and can be downloaded with easy_install or pip. If you add a Flask extension as dependency to your requirements.rst or setup.py file they are usually installed with a simple command or when your application installs. 16.2 Using Extensions Extensions typically have documentation that goes along that shows how to use it. There are no general rules in how extensions are supposed to behave but they are imported from common locations. If you have an extension called Flask-Foo or Foo-Flask it will be always importable from flask.ext.foo: from flask.ext import foo 16.3 Flask Before 0.8 If you are using Flask 0.7 or earlier the flask.ext package will not exist, instead you have to import from flaskext.foo or flask_foo depending on how the extension is distributed. If you want to develop an application that supports Flask 0.7 or earlier you should still import from the flask.ext package. We provide you with a compatibility module that provides this package for older versions of Flask. You can download it from github: flaskext_compat.py And here is how you can use it: 97
- 108. import flaskext_compat flaskext_compat.activate() from flask.ext import foo Once the flaskext_compat module is activated the flask.ext will exist and you can start importing from there. 98
- 109. CHAPTER SEVENTEEN WORKING WITH THE SHELL New in version 0.3. One of the reasons everybody loves Python is the interactive shell. It basically allows you to execute Python commands in real time and immediately get results back. Flask itself does not come with an interactive shell, because it does not require any specific setup upfront, just import your application and start playing around. There are however some handy helpers to make playing around in the shell a more pleasant experience. The main issue with interactive console sessions is that you’re not triggering a request like a browser does which means that g, request and others are not available. But the code you want to test might depend on them, so what can you do? This is where some helper functions come in handy. Keep in mind however that these functions are not only there for interactive shell usage, but also for unittesting and other situations that require a faked request context. Generally it’s recommended that you read the The Request Context chapter of the doc- umentation first. 17.1 Creating a Request Context The easiest way to create a proper request context from the shell is by using the test_request_context method which creates us a RequestContext: >>> ctx = app.test_request_context() Normally you would use the with statement to make this request object active, but in the shell it’s easier to use the push() and pop() methods by hand: >>> ctx.push() From that point onwards you can work with the request object until you call pop: >>> ctx.pop() 99
- 110. 17.2 Firing Before/After Request By just creating a request context, you still don’t have run the code that is normally run before a request. This might result in your database being unavailable if you are connecting to the database in a before-request callback or the current user not being stored on the g object etc. This however can easily be done yourself. Just call preprocess_request(): >>> ctx = app.test_request_context() >>> ctx.push() >>> app.preprocess_request() Keep in mind that the preprocess_request() function might return a response object, in that case just ignore it. To shutdown a request, you need to trick a bit before the after request functions (trig- gered by process_response()) operate on a response object: >>> app.process_response(app.response_class()) <Response 0 bytes [200 OK]> >>> ctx.pop() The functions registered as teardown_request() are automatically called when the context is popped. So this is the perfect place to automatically tear down resources that were needed by the request context (such as database connections). 17.3 Further Improving the Shell Experience If you like the idea of experimenting in a shell, create yourself a module with stuff you want to star import into your interactive session. There you could also define some more helper methods for common things such as initializing the database, dropping tables etc. Just put them into a module (like shelltools and import from there): >>> from shelltools import * 100
- 111. CHAPTER EIGHTEEN PATTERNS FOR FLASK Certain things are common enough that the chances are high you will find them in most web applications. For example quite a lot of applications are using relational databases and user authentication. In that case, chances are they will open a database connection at the beginning of the request and get the information of the currently logged in user. At the end of the request, the database connection is closed again. There are more user contributed snippets and patterns in the Flask Snippet Archives. 18.1 Larger Applications For larger applications it’s a good idea to use a package instead of a module. That is quite simple. Imagine a small application looks like this: /yourapplication /yourapplication.py /static /style.css /templates layout.html index.html login.html ... 18.1.1 Simple Packages To convert that into a larger one, just create a new folder yourapplication inside the exist- ing one and move everything below it. Then rename yourapplication.py to __init__.py. (Make sure to delete all .pyc files first, otherwise things would most likely break) You should then end up with something like that: /yourapplication /yourapplication /__init__.py /static 101
- 112. /style.css /templates layout.html index.html login.html ... But how do you run your application now? The naive python yourapplication/__init__.py will not work. Let’s just say that Python does not want modules in packages to be the startup file. But that is not a big problem, just add a new file called runserver.py next to the inner yourapplication folder with the following contents: from yourapplication import app app.run(debug=True) What did we gain from this? Now we can restructure the application a bit into multiple modules. The only thing you have to remember is the following quick checklist: 1. the Flask application object creation has to be in the __init__.py file. That way each module can import it safely and the __name__ variable will resolve to the correct package. 2. all the view functions (the ones with a route() decorator on top) have to be imported in the __init__.py file. Not the object itself, but the module it is in. Import the view module after the application object is created. Here’s an example __init__.py: from flask import Flask app = Flask(__name__) import yourapplication.views And this is what views.py would look like: from yourapplication import app @app.route(’/’) def index(): return ’Hello World!’ You should then end up with something like that: /yourapplication /runserver.py /yourapplication /__init__.py /views.py /static /style.css /templates layout.html 102
- 113. index.html login.html ... Circular Imports Every Python programmer hates them, and yet we just added some: circular imports (That’s when two modules depend on each other. In this case views.py depends on __init__.py). Be advised that this is a bad idea in general but here it is actually fine. The reason for this is that we are not actually using the views in __init__.py and just ensuring the module is imported and we are doing that at the bottom of the file. There are still some problems with that approach but if you want to use decorators there is no way around that. Check out the Becoming Big section for some inspiration how to deal with that. 18.1.2 Working with Blueprints If you have larger applications it’s recommended to divide them into smaller groups where each group is implemented with the help of a blueprint. For a gentle intro- duction into this topic refer to the Modular Applications with Blueprints chapter of the documentation. 18.2 Application Factories If you are already using packages and blueprints for your application (Modular Appli- cations with Blueprints) there are a couple of really nice ways to further improve the experience. A common pattern is creating the application object when the blueprint is imported. But if you move the creation of this object, into a function, you can then create multiple instances of this and later. So why would you want to do this? 1. Testing. You can have instances of the application with different settings to test every case. 2. Multiple instances. Imagine you want to run different versions of the same ap- plication. Of course you could have multiple instances with different configs set up in your webserver, but if you use factories, you can have multiple instances of the same application running in the same application process which can be handy. So how would you then actually implement that? 18.2.1 Basic Factories The idea is to set up the application in a function. Like this: 103
- 114. def create_app(config_filename): app = Flask(__name__) app.config.from_pyfile(config_filename) from yourapplication.model import db db.init_app(app) from yourapplication.views.admin import admin from yourapplication.views.frontend import frontend app.register_blueprint(admin) app.register_blueprint(frontend) return app The downside is that you cannot use the application object in the blueprints at import time. You can however use it from within a request. How do you get access to the application with the config? Use current_app: from flask import current_app, Blueprint, render_template admin = Blueprint(’admin’, __name__, url_prefix=’/admin’) @admin.route(’/’) def index(): return render_template(current_app.config[’INDEX_TEMPLATE’]) Here we look up the name of a template in the config. Extension objects are not initially bound to an application. Using db.init_app, the app gets configured for the extension. No application-specific state is stored on the extension object, so one extension object can be used for multiple apps. For more information about the design of extensions refer to Flask Extension Development. Your model.py might look like this when using Flask-SQLAlchemy: from flask.ext.sqlalchemy import SQLAlchemy # no app object passed! Instead we use use db.init_app in the factory. db = SQLAlchemy() # create some models 18.2.2 Using Applications So to use such an application you then have to create the application first. Here an example run.py file that runs such an application: from yourapplication import create_app app = create_app(’/path/to/config.cfg’) app.run() 104
- 115. 18.2.3 Factory Improvements The factory function from above is not very clever so far, you can improve it. The following changes are straightforward and possible: 1. make it possible to pass in configuration values for unittests so that you don’t have to create config files on the filesystem 2. call a function from a blueprint when the application is setting up so that you have a place to modify attributes of the application (like hooking in before / after request handlers etc.) 3. Add in WSGI middlewares when the application is creating if necessary. 18.3 Application Dispatching Application dispatching is the process of combining multiple Flask applications on the WSGI level. You can not only combine Flask applications into something larger but any WSGI application. This would even allow you to run a Django and a Flask application in the same interpreter side by side if you want. The usefulness of this depends on how the applications work internally. The fundamental difference from the module approach is that in this case you are run- ning the same or different Flask applications that are entirely isolated from each other. They run different configurations and are dispatched on the WSGI level. 18.3.1 Working with this Document Each of the techniques and examples below results in an application object that can be run with any WSGI server. For production, see Deployment Options. For development, Werkzeug provides a builtin server for development available at werkzeug.serving.run_simple(): from werkzeug.serving import run_simple run_simple(’localhost’, 5000, application, use_reloader=True) Note that run_simple is not intended for use in production. Use a full-blown WSGI server. In order to use the interactive debuggger, debugging must be enabled both on the application and the simple server, here is the “hello world” example with debugging and run_simple: from flask import Flask from werkzeug.serving import run_simple app = Flask(__name__) app.debug = True 105
- 116. @app.route(’/’) def hello_world(): return ’Hello World!’ if __name__ == ’__main__’: run_simple(’localhost’, 5000, app, use_reloader=True, use_debugger=True, use_evalex=True) 18.3.2 Combining Applications If you have entirely separated applications and you want them to work next to each other in the same Python interpreter process you can take advantage of the werkzeug.wsgi.DispatcherMiddleware. The idea here is that each Flask application is a valid WSGI application and they are combined by the dispatcher middleware into a larger one that dispatched based on prefix. For example you could have your main application run on / and your backend inter- face on /backend: from werkzeug.wsgi import DispatcherMiddleware from frontend_app import application as frontend from backend_app import application as backend application = DispatcherMiddleware(frontend, { ’/backend’: backend }) 18.3.3 Dispatch by Subdomain Sometimes you might want to use multiple instances of the same application with different configurations. Assuming the application is created inside a function and you can call that function to instantiate it, that is really easy to implement. In order to develop your application to support creating new instances in functions have a look at the Application Factories pattern. A very common example would be creating applications per subdomain. For instance you configure your webserver to dispatch all requests for all subdomains to your application and you then use the subdomain information to create user-specific in- stances. Once you have your server set up to listen on all subdomains you can use a very simple WSGI application to do the dynamic application creation. The perfect level for abstraction in that regard is the WSGI layer. You write your own WSGI application that looks at the request that comes and delegates it to your Flask application. If that application does not exist yet, it is dynamically created and remem- bered: from threading import Lock 106
- 117. class SubdomainDispatcher(object): def __init__(self, domain, create_app): self.domain = domain self.create_app = create_app self.lock = Lock() self.instances = {} def get_application(self, host): host = host.split(’:’)[0] assert host.endswith(self.domain), ’Configuration error’ subdomain = host[:-len(self.domain)].rstrip(’.’) with self.lock: app = self.instances.get(subdomain) if app is None: app = self.create_app(subdomain) self.instances[subdomain] = app return app def __call__(self, environ, start_response): app = self.get_application(environ[’HTTP_HOST’]) return app(environ, start_response) This dispatcher can then be used like this: from myapplication import create_app, get_user_for_subdomain from werkzeug.exceptions import NotFound def make_app(subdomain): user = get_user_for_subdomain(subdomain) if user is None: # if there is no user for that subdomain we still have # to return a WSGI application that handles that request. # We can then just return the NotFound() exception as # application which will render a default 404 page. # You might also redirect the user to the main page then return NotFound() # otherwise create the application for the specific user return create_app(user) application = SubdomainDispatcher(’example.com’, make_app) 18.3.4 Dispatch by Path Dispatching by a path on the URL is very similar. Instead of looking at the Host header to figure out the subdomain one simply looks at the request path up to the first slash: 107
- 118. from threading import Lock from werkzeug.wsgi import pop_path_info, peek_path_info class PathDispatcher(object): def __init__(self, default_app, create_app): self.default_app = default_app self.create_app = create_app self.lock = Lock() self.instances = {} def get_application(self, prefix): with self.lock: app = self.instances.get(prefix) if app is None: app = self.create_app(prefix) if app is not None: self.instances[prefix] = app return app def __call__(self, environ, start_response): app = self.get_application(peek_path_info(environ)) if app is not None: pop_path_info(environ) else: app = self.default_app return app(environ, start_response) The big difference between this and the subdomain one is that this one falls back to another application if the creator function returns None: from myapplication import create_app, default_app, get_user_for_prefix def make_app(prefix): user = get_user_for_prefix(prefix) if user is not None: return create_app(user) application = PathDispatcher(default_app, make_app) 18.4 Implementing API Exceptions It’s very common to implement RESTful APIs on top of Flask. One of the first thing that developers run into is the realization that the builtin exceptions are not expressive enough for APIs and that the content type of text/html they are emitting is not very useful for API consumers. The better solution than using abort to signal errors for invalid API usage is to im- plement your own exception type and install an error handler for it that produces the 108
- 119. errors in the format the user is expecting. 18.4.1 Simple Exception Class The basic idea is to introduce a new exception that can take a proper human readable message, a status code for the error and some optional payload to give more context for the error. This is a simple example: from flask import jsonify class InvalidUsage(Exception): status_code = 400 def __init__(self, message, status_code=None, payload=None): Exception.__init__(self) self.message = message if status_code is not None: self.status_code = status_code self.payload = payload def to_dict(self): rv = dict(self.payload or ()) rv[’message’] = self.message return rv A view can now raise that exception with an error message. Additionally some extra payload can be provided as a dictionary through the payload parameter. 18.4.2 Registering an Error Handler At that point views can raise that error, but it would immediately result in an internal server error. The reason for this is that there is no handler registered for this error class. That however is easy to add: @app.errorhandler(InvalidUsage) def handle_invalid_usage(error): response = jsonify(error.to_dict()) response.status_code = error.status_code return response 18.4.3 Usage in Views Here is how a view can use that functionality: 109
- 120. @app.route(’/foo’) def get_foo(): raise InvalidUsage(’This view is gone’, status_code=410) 18.5 Using URL Processors New in version 0.7. Flask 0.7 introduces the concept of URL processors. The idea is that you might have a bunch of resources with common parts in the URL that you don’t always explicitly want to provide. For instance you might have a bunch of URLs that have the language code in it but you don’t want to have to handle it in every single function yourself. URL processors are especially helpful when combined with blueprints. We will handle both application specific URL processors here as well as blueprint specifics. 18.5.1 Internationalized Application URLs Consider an application like this: from flask import Flask, g app = Flask(__name__) @app.route(’/<lang_code>/’) def index(lang_code): g.lang_code = lang_code ... @app.route(’/<lang_code>/about’) def about(lang_code): g.lang_code = lang_code ... This is an awful lot of repetition as you have to handle the language code setting on the g object yourself in every single function. Sure, a decorator could be used to simplify this, but if you want to generate URLs from one function to another you would have to still provide the language code explicitly which can be annoying. For the latter, this is where url_defaults() functions come in. They can automatically inject values into a call for url_for() automatically. The code below checks if the language code is not yet in the dictionary of URL values and if the endpoint wants a value named ’lang_code’: @app.url_defaults def add_language_code(endpoint, values): if ’lang_code’ in values or not g.lang_code: return 110
- 121. if app.url_map.is_endpoint_expecting(endpoint, ’lang_code’): values[’lang_code’] = g.lang_code The method is_endpoint_expecting() of the URL map can be used to figure out if it would make sense to provide a language code for the given endpoint. The reverse of that function are url_value_preprocessor()s. They are executed right after the request was matched and can execute code based on the URL values. The idea is that they pull information out of the values dictionary and put it somewhere else: @app.url_value_preprocessor def pull_lang_code(endpoint, values): g.lang_code = values.pop(’lang_code’, None) That way you no longer have to do the lang_code assignment to g in every function. You can further improve that by writing your own decorator that prefixes URLs with the language code, but the more beautiful solution is using a blueprint. Once the ’lang_code’ is popped from the values dictionary and it will no longer be forwarded to the view function reducing the code to this: from flask import Flask, g app = Flask(__name__) @app.url_defaults def add_language_code(endpoint, values): if ’lang_code’ in values or not g.lang_code: return if app.url_map.is_endpoint_expecting(endpoint, ’lang_code’): values[’lang_code’] = g.lang_code @app.url_value_preprocessor def pull_lang_code(endpoint, values): g.lang_code = values.pop(’lang_code’, None) @app.route(’/<lang_code>/’) def index(): ... @app.route(’/<lang_code>/about’) def about(): ... 18.5.2 Internationalized Blueprint URLs Because blueprints can automatically prefix all URLs with a common string it’s easy to automatically do that for every function. Furthermore blueprints can have per- blueprint URL processors which removes a whole lot of logic from the url_defaults() 111
- 122. function because it no longer has to check if the URL is really interested in a ’lang_code’ parameter: from flask import Blueprint, g bp = Blueprint(’frontend’, __name__, url_prefix=’/<lang_code>’) @bp.url_defaults def add_language_code(endpoint, values): values.setdefault(’lang_code’, g.lang_code) @bp.url_value_preprocessor def pull_lang_code(endpoint, values): g.lang_code = values.pop(’lang_code’) @bp.route(’/’) def index(): ... @bp.route(’/about’) def about(): ... 18.6 Deploying with Distribute distribute, formerly setuptools, is an extension library that is commonly used to (like the name says) distribute Python libraries and extensions. It extends distutils, a basic module installation system shipped with Python to also support various more com- plex constructs that make larger applications easier to distribute: • support for dependencies: a library or application can declare a list of other libraries it depends on which will be installed automatically for you. • package registry: setuptools registers your package with your Python installa- tion. This makes it possible to query information provided by one package from another package. The best known feature of this system is the entry point sup- port which allows one package to declare an “entry point” another package can hook into to extend the other package. • installation manager: easy_install, which comes with distribute can install other libraries for you. You can also use pip which sooner or later will replace easy_install which does more than just installing packages for you. Flask itself, and all the libraries you can find on the cheeseshop are distributed with either distribute, the older setuptools or distutils. In this case we assume your application is called yourapplication.py and you are not using a module, but a package. Distributing resources with standard modules is not supported by distribute so we will not bother with it. If you have not yet converted 112
- 123. your application into a package, head over to the Larger Applications pattern to see how this can be done. A working deployment with distribute is the first step into more complex and more automated deployment scenarios. If you want to fully automate the process, also read the Deploying with Fabric chapter. 18.6.1 Basic Setup Script Because you have Flask running, you either have setuptools or distribute available on your system anyways. If you do not, fear not, there is a script to install it for you: distribute_setup.py. Just download and run with your Python interpreter. Standard disclaimer applies: you better use a virtualenv. Your setup code always goes into a file named setup.py next to your application. The name of the file is only convention, but because everybody will look for a file with that name, you better not change it. Yes, even if you are using distribute, you are importing from a package called setuptools. distribute is fully backwards compatible with setuptools, so it also uses the same import name. A basic setup.py file for a Flask application looks like this: from setuptools import setup setup( name=’Your Application’, version=’1.0’, long_description=__doc__, packages=[’yourapplication’], include_package_data=True, zip_safe=False, install_requires=[’Flask’] ) Please keep in mind that you have to list subpackages explicitly. If you want distribute to lookup the packages for you automatically, you can use the find_packages function: from setuptools import setup, find_packages setup( ... packages=find_packages() ) Most parameters to the setup function should be self explanatory, include_package_data and zip_safe might not be. include_package_data tells distribute to look for a MANI- FEST.in file and install all the entries that match as package data. We will use this to distribute the static files and templates along with the Python module (see Distribut- ing Resources). The zip_safe flag can be used to force or prevent zip Archive creation. 113
- 124. In general you probably don’t want your packages to be installed as zip files because some tools do not support them and they make debugging a lot harder. 18.6.2 Distributing Resources If you try to install the package you just created, you will notice that folders like static or templates are not installed for you. The reason for this is that distribute does not know which files to add for you. What you should do, is to create a MANIFEST.in file next to your setup.py file. This file lists all the files that should be added to your tarball: recursive-include yourapplication/templates * recursive-include yourapplication/static * Don’t forget that even if you enlist them in your MANIFEST.in file, they won’t be installed for you unless you set the include_package_data parameter of the setup function to True! 18.6.3 Declaring Dependencies Dependencies are declared in the install_requires parameter as list. Each item in that list is the name of a package that should be pulled from PyPI on installation. By default it will always use the most recent version, but you can also provide minimum and maximum version requirements. Here some examples: install_requires=[ ’Flask>=0.2’, ’SQLAlchemy>=0.6’, ’BrokenPackage>=0.7,<=1.0’ ] I mentioned earlier that dependencies are pulled from PyPI. What if you want to de- pend on a package that cannot be found on PyPI and won’t be because it is an internal package you don’t want to share with anyone? Just still do as if there was a PyPI en- try for it and provide a list of alternative locations where distribute should look for tarballs: dependency_links=[’http://example.com/yourfiles’] Make sure that page has a directory listing and the links on the page are pointing to the actual tarballs with their correct filenames as this is how distribute will find the files. If you have an internal company server that contains the packages, provide the URL to that server there. 18.6.4 Installing / Developing To install your application (ideally into a virtualenv) just run the setup.py script with the install parameter. It will install your application into the virtualenv’s site-packages folder and also download and install all dependencies: 114
- 125. $ python setup.py install If you are developing on the package and also want the requirements to be installed, you can use the develop command instead: $ python setup.py develop This has the advantage of just installing a link to the site-packages folder instead of copying the data over. You can then continue to work on the code without having to run install again after each change. 18.7 Deploying with Fabric Fabric is a tool for Python similar to Makefiles but with the ability to execute com- mands on a remote server. In combination with a properly set up Python package (Larger Applications) and a good concept for configurations (Configuration Handling) it is very easy to deploy Flask applications to external servers. Before we get started, here a quick checklist of things we have to ensure upfront: • Fabric 1.0 has to be installed locally. This tutorial assumes the latest version of Fabric. • The application already has to be a package and requires a working setup.py file (Deploying with Distribute). • In the following example we are using mod_wsgi for the remote servers. You can of course use your own favourite server there, but for this example we chose Apache + mod_wsgi because it’s very easy to setup and has a simple way to reload applications without root access. 18.7.1 Creating the first Fabfile A fabfile is what controls what Fabric executes. It is named fabfile.py and executed by the fab command. All the functions defined in that file will show up as fab subcom- mands. They are executed on one or more hosts. These hosts can be defined either in the fabfile or on the command line. In this case we will add them to the fabfile. This is a basic first example that has the ability to upload the current sourcecode to the server and install it into a pre-existing virtual environment: from fabric.api import * # the user to use for the remote commands env.user = ’appuser’ # the servers where the commands are executed env.hosts = [’server1.example.com’, ’server2.example.com’] def pack(): 115
- 126. # create a new source distribution as tarball local(’python setup.py sdist --formats=gztar’, capture=False) def deploy(): # figure out the release name and version dist = local(’python setup.py --fullname’, capture=True).strip() # upload the source tarball to the temporary folder on the server put(’dist/%s.tar.gz’ % dist, ’/tmp/yourapplication.tar.gz’) # create a place where we can unzip the tarball, then enter # that directory and unzip it run(’mkdir /tmp/yourapplication’) with cd(’/tmp/yourapplication’): run(’tar xzf /tmp/yourapplication.tar.gz’) # now setup the package with our virtual environment’s # python interpreter run(’/var/www/yourapplication/env/bin/python setup.py install’) # now that all is set up, delete the folder again run(’rm -rf /tmp/yourapplication /tmp/yourapplication.tar.gz’) # and finally touch the .wsgi file so that mod_wsgi triggers # a reload of the application run(’touch /var/www/yourapplication.wsgi’) The example above is well documented and should be straightforward. Here a recap of the most common commands fabric provides: • run - executes a command on a remote server • local - executes a command on the local machine • put - uploads a file to the remote server • cd - changes the directory on the serverside. This has to be used in combination with the with statement. 18.7.2 Running Fabfiles Now how do you execute that fabfile? You use the fab command. To deploy the current version of the code on the remote server you would use this command: $ fab pack deploy However this requires that our server already has the /var/www/yourapplication folder created and /var/www/yourapplication/env to be a virtual environment. Fur- thermore are we not creating the configuration or .wsgi file on the server. So how do we bootstrap a new server into our infrastructure? This now depends on the number of servers we want to set up. If we just have one application server (which the majority of applications will have), creating a command in the fabfile for this is overkill. But obviously you can do that. In that case you would probably call it setup or bootstrap and then pass the servername explicitly on the command line: 116
- 127. $ fab -H newserver.example.com bootstrap To setup a new server you would roughly do these steps: 1. Create the directory structure in /var/www: $ mkdir /var/www/yourapplication $ cd /var/www/yourapplication $ virtualenv --distribute env 2. Upload a new application.wsgi file to the server and the configuration file for the application (eg: application.cfg) 3. Create a new Apache config for yourapplication and activate it. Make sure to activate watching for changes of the .wsgi file so that we can automatically reload the application by touching it. (See mod_wsgi (Apache) for more information) So now the question is, where do the application.wsgi and application.cfg files come from? 18.7.3 The WSGI File The WSGI file has to import the application and also to set an environment variable so that the application knows where to look for the config. This is a short example that does exactly that: import os os.environ[’YOURAPPLICATION_CONFIG’] = ’/var/www/yourapplication/application.cfg’ from yourapplication import app The application itself then has to initialize itself like this to look for the config at that environment variable: app = Flask(__name__) app.config.from_object(’yourapplication.default_config’) app.config.from_envvar(’YOURAPPLICATION_CONFIG’) This approach is explained in detail in the Configuration Handling section of the docu- mentation. 18.7.4 The Configuration File Now as mentioned above, the application will find the correct configuration file by looking up the YOURAPPLICATION_CONFIG environment variable. So we have to put the configuration in a place where the application will able to find it. Configura- tion files have the unfriendly quality of being different on all computers, so you do not version them usually. A popular approach is to store configuration files for different servers in a sep- arate version control repository and check them out on all servers. Then sym- 117
- 128. link the file that is active for the server into the location where it’s expected (eg: /var/www/yourapplication). Either way, in our case here we only expect one or two servers and we can upload them ahead of time by hand. 18.7.5 First Deployment Now we can do our first deployment. We have set up the servers so that they have their virtual environments and activated apache configs. Now we can pack up the application and deploy it: $ fab pack deploy Fabric will now connect to all servers and run the commands as written down in the fabfile. First it will execute pack so that we have our tarball ready and then it will execute deploy and upload the source code to all servers and install it there. Thanks to the setup.py file we will automatically pull in the required libraries into our virtual environment. 18.7.6 Next Steps From that point onwards there is so much that can be done to make deployment actu- ally fun: • Create a bootstrap command that initializes new servers. It could initialize a new virtual environment, setup apache appropriately etc. • Put configuration files into a separate version control repository and symlink the active configs into place. • You could also put your application code into a repository and check out the latest version on the server and then install. That way you can also easily go back to older versions. • hook in testing functionality so that you can deploy to an external server and run the testsuite. Working with Fabric is fun and you will notice that it’s quite magical to type fab deploy and see your application being deployed automatically to one or more remote servers. 18.8 Using SQLite 3 with Flask In Flask you can implement the opening of database connections on demand and clos- ing it when the context dies (usually at the end of the request) easily. Here is a simple example of how you can use SQLite 3 with Flask: 118
- 129. import sqlite3 from flask import g DATABASE = ’/path/to/database.db’ def get_db(): db = getattr(g, ’_database’, None) if db is None: db = g._database = connect_to_database() return db @app.teardown_appcontext def close_connection(exception): db = getattr(g, ’_database’, None) if db is not None: db.close() All the application needs to do in order to now use the database is having an active application context (which is always true if there is an request in flight) or to create an application context itself. At that point the get_db function can be used to get the cur- rent database connection. Whenever the context is destroyed the database connection will be terminated. Note: if you use Flask 0.9 or older you need to use flask._app_ctx_stack.top instead of g as the flask.g object was bound to the request and not application context. Example: @app.route(’/’) def index(): cur = get_db().cursor() ... Note: Please keep in mind that the teardown request and appcontext functions are al- ways executed, even if a before-request handler failed or was never executed. Because of this we have to make sure here that the database is there before we close it. 18.8.1 Connect on Demand The upside of this approach (connecting on first use) is that this will only opening the connection if truly necessary. If you want to use this code outside a request context you can use it in a Python shell by opening the application context by hand: with app.app_context(): # now you can use get_db() 119
- 130. 18.8.2 Easy Querying Now in each request handling function you can access g.db to get the current open database connection. To simplify working with SQLite, a row factory function is use- ful. It is executed for every result returned from the database to convert the result. For instance in order to get dictionaries instead of tuples this could be inserted into get_db: def make_dicts(cursor, row): return dict((cursor.description[idx][0], value) for idx, value in enumerate(row)) db.row_factory = make_dicts Or even simpler: db.row_factory = sqlite3.Row Additionally it is a good idea to provide a query function that combines getting the cursor, executing and fetching the results: def query_db(query, args=(), one=False): cur = get_db().execute(query, args) rv = cur.fetchall() cur.close() return (rv[0] if rv else None) if one else rv This handy little function in combination with a row factory makes working with the database much more pleasant than it is by just using the raw cursor and connection objects. Here is how you can use it: for user in query_db(’select * from users’): print user[’username’], ’has the id’, user[’user_id’] Or if you just want a single result: user = query_db(’select * from users where username = ?’, [the_username], one=True) if user is None: print ’No such user’ else: print the_username, ’has the id’, user[’user_id’] To pass variable parts to the SQL statement, use a question mark in the statement and pass in the arguments as a list. Never directly add them to the SQL statement with string formatting because this makes it possible to attack the application using SQL Injections. 120
- 131. 18.8.3 Initial Schemas Relational databases need schemas, so applications often ship a schema.sql file that creates the database. It’s a good idea to provide a function that creates the database based on that schema. This function can do that for you: def init_db(): with app.app_context(): db = get_db() with app.open_resource(’schema.sql’, mode=’r’) as f: db.cursor().executescript(f.read()) db.commit() You can then create such a database from the python shell: >>> from yourapplication import init_db >>> init_db() 18.9 SQLAlchemy in Flask Many people prefer SQLAlchemy for database access. In this case it’s encouraged to use a package instead of a module for your flask application and drop the models into a separate module (Larger Applications). While that is not necessary, it makes a lot of sense. There are four very common ways to use SQLAlchemy. I will outline each of them here: 18.9.1 Flask-SQLAlchemy Extension Because SQLAlchemy is a common database abstraction layer and object relational mapper that requires a little bit of configuration effort, there is a Flask extension that handles that for you. This is recommended if you want to get started quickly. You can download Flask-SQLAlchemy from PyPI. 18.9.2 Declarative The declarative extension in SQLAlchemy is the most recent method of using SQLAlchemy. It allows you to define tables and models in one go, similar to how Django works. In addition to the following text I recommend the official documenta- tion on the declarative extension. Here the example database.py module for your application: from sqlalchemy import create_engine from sqlalchemy.orm import scoped_session, sessionmaker 121
- 132. from sqlalchemy.ext.declarative import declarative_base engine = create_engine(’sqlite:////tmp/test.db’, convert_unicode=True) db_session = scoped_session(sessionmaker(autocommit=False, autoflush=False, bind=engine)) Base = declarative_base() Base.query = db_session.query_property() def init_db(): # import all modules here that might define models so that # they will be registered properly on the metadata. Otherwise # you will have to import them first before calling init_db() import yourapplication.models Base.metadata.create_all(bind=engine) To define your models, just subclass the Base class that was created by the code above. If you are wondering why we don’t have to care about threads here (like we did in the SQLite3 example above with the g object): that’s because SQLAlchemy does that for us already with the scoped_session. To use SQLAlchemy in a declarative way with your application, you just have to put the following code into your application module. Flask will automatically remove database sessions at the end of the request or when the application shuts down: from yourapplication.database import db_session @app.teardown_appcontext def shutdown_session(exception=None): db_session.remove() Here is an example model (put this into models.py, e.g.): from sqlalchemy import Column, Integer, String from yourapplication.database import Base class User(Base): __tablename__ = ’users’ id = Column(Integer, primary_key=True) name = Column(String(50), unique=True) email = Column(String(120), unique=True) def __init__(self, name=None, email=None): self.name = name self.email = email def __repr__(self): return ’<User %r>’ % (self.name) To create the database you can use the init_db function: 122
- 133. >>> from yourapplication.database import init_db >>> init_db() You can insert entries into the database like this: >>> from yourapplication.database import db_session >>> from yourapplication.models import User >>> u = User(’admin’, ’admin@localhost’) >>> db_session.add(u) >>> db_session.commit() Querying is simple as well: >>> User.query.all() [<User u’admin’>] >>> User.query.filter(User.name == ’admin’).first() <User u’admin’> 18.9.3 Manual Object Relational Mapping Manual object relational mapping has a few upsides and a few downsides versus the declarative approach from above. The main difference is that you define tables and classes separately and map them together. It’s more flexible but a little more to type. In general it works like the declarative approach, so make sure to also split up your application into multiple modules in a package. Here is an example database.py module for your application: from sqlalchemy import create_engine, MetaData from sqlalchemy.orm import scoped_session, sessionmaker engine = create_engine(’sqlite:////tmp/test.db’, convert_unicode=True) metadata = MetaData() db_session = scoped_session(sessionmaker(autocommit=False, autoflush=False, bind=engine)) def init_db(): metadata.create_all(bind=engine) As for the declarative approach you need to close the session after each request or application context shutdown. Put this into your application module: from yourapplication.database import db_session @app.teardown_appcontext def shutdown_session(exception=None): db_session.remove() Here is an example table and model (put this into models.py): 123
- 134. from sqlalchemy import Table, Column, Integer, String from sqlalchemy.orm import mapper from yourapplication.database import metadata, db_session class User(object): query = db_session.query_property() def __init__(self, name=None, email=None): self.name = name self.email = email def __repr__(self): return ’<User %r>’ % (self.name) users = Table(’users’, metadata, Column(’id’, Integer, primary_key=True), Column(’name’, String(50), unique=True), Column(’email’, String(120), unique=True) ) mapper(User, users) Querying and inserting works exactly the same as in the example above. 18.9.4 SQL Abstraction Layer If you just want to use the database system (and SQL) abstraction layer you basically only need the engine: from sqlalchemy import create_engine, MetaData engine = create_engine(’sqlite:////tmp/test.db’, convert_unicode=True) metadata = MetaData(bind=engine) Then you can either declare the tables in your code like in the examples above, or automatically load them: users = Table(’users’, metadata, autoload=True) To insert data you can use the insert method. We have to get a connection first so that we can use a transaction: >>> con = engine.connect() >>> con.execute(users.insert(), name=’admin’, email=’admin@localhost’) SQLAlchemy will automatically commit for us. To query your database, you use the engine directly or use a connection: >>> users.select(users.c.id == 1).execute().first() (1, u’admin’, u’admin@localhost’) 124
- 135. These results are also dict-like tuples: >>> r = users.select(users.c.id == 1).execute().first() >>> r[’name’] u’admin’ You can also pass strings of SQL statements to the execute() method: >>> engine.execute(’select * from users where id = :1’, [1]).first() (1, u’admin’, u’admin@localhost’) For more information about SQLAlchemy, head over to the website. 18.10 Uploading Files Ah yes, the good old problem of file uploads. The basic idea of file uploads is actually quite simple. It basically works like this: 1. A <form> tag is marked with enctype=multipart/form-data and an <input type=file> is placed in that form. 2. The application accesses the file from the files dictionary on the request object. 3. use the save() method of the file to save the file permanently somewhere on the filesystem. 18.10.1 A Gentle Introduction Let’s start with a very basic application that uploads a file to a specific upload folder and displays a file to the user. Let’s look at the bootstrapping code for our application: import os from flask import Flask, request, redirect, url_for from werkzeug.utils import secure_filename UPLOAD_FOLDER = ’/path/to/the/uploads’ ALLOWED_EXTENSIONS = set([’txt’, ’pdf’, ’png’, ’jpg’, ’jpeg’, ’gif’]) app = Flask(__name__) app.config[’UPLOAD_FOLDER’] = UPLOAD_FOLDER So first we need a couple of imports. Most should be straightforward, the werkzeug.secure_filename() is explained a little bit later. The UPLOAD_FOLDER is where we will store the uploaded files and the ALLOWED_EXTENSIONS is the set of allowed file extensions. Then we add a URL rule by hand to the application. Now usually we’re not doing that, so why here? The reasons is that we want the webserver (or our development server) to serve these files for us and so we only need a rule to generate the URL to these files. 125
- 136. Why do we limit the extensions that are allowed? You probably don’t want your users to be able to upload everything there if the server is directly sending out the data to the client. That way you can make sure that users are not able to upload HTML files that would cause XSS problems (see Cross-Site Scripting (XSS)). Also make sure to disallow .php files if the server executes them, but who has PHP installed on his server, right? :) Next the functions that check if an extension is valid and that uploads the file and redirects the user to the URL for the uploaded file: def allowed_file(filename): return ’.’ in filename and filename.rsplit(’.’, 1)[1] in ALLOWED_EXTENSIONS @app.route(’/’, methods=[’GET’, ’POST’]) def upload_file(): if request.method == ’POST’: file = request.files[’file’] if file and allowed_file(file.filename): filename = secure_filename(file.filename) file.save(os.path.join(app.config[’UPLOAD_FOLDER’], filename)) return redirect(url_for(’uploaded_file’, filename=filename)) return ’’’ <!doctype html> <title>Upload new File</title> <h1>Upload new File</h1> <form action="" method=post enctype=multipart/form-data> <p><input type=file name=file> <input type=submit value=Upload> </form> ’’’ So what does that secure_filename() function actually do? Now the problem is that there is that principle called “never trust user input”. This is also true for the filename of an uploaded file. All submitted form data can be forged, and filenames can be dan- gerous. For the moment just remember: always use that function to secure a filename before storing it directly on the filesystem. Information for the Pros So you’re interested in what that secure_filename() function does and what the prob- lem is if you’re not using it? So just imagine someone would send the following infor- mation as filename to your application: filename = "../../../../home/username/.bashrc" Assuming the number of ../ is correct and you would join this with the UP- LOAD_FOLDER the user might have the ability to modify a file on the server’s filesys- tem he or she should not modify. This does require some knowledge about how the application looks like, but trust me, hackers are patient :) Now let’s look how that function works: 126
- 137. >>> secure_filename(’../../../../home/username/.bashrc’) ’home_username_.bashrc’ Now one last thing is missing: the serving of the uploaded files. As of Flask 0.5 we can use a function that does that for us: from flask import send_from_directory @app.route(’/uploads/<filename>’) def uploaded_file(filename): return send_from_directory(app.config[’UPLOAD_FOLDER’], filename) Alternatively you can register uploaded_file as build_only rule and use the SharedDataMiddleware. This also works with older versions of Flask: from werkzeug import SharedDataMiddleware app.add_url_rule(’/uploads/<filename>’, ’uploaded_file’, build_only=True) app.wsgi_app = SharedDataMiddleware(app.wsgi_app, { ’/uploads’: app.config[’UPLOAD_FOLDER’] }) If you now run the application everything should work as expected. 18.10.2 Improving Uploads New in version 0.6. So how exactly does Flask handle uploads? Well it will store them in the webserver’s memory if the files are reasonable small otherwise in a temporary location (as returned by tempfile.gettempdir()). But how do you specify the maximum file size after which an upload is aborted? By default Flask will happily accept file uploads to an un- limited amount of memory, but you can limit that by setting the MAX_CONTENT_LENGTH config key: from flask import Flask, Request app = Flask(__name__) app.config[’MAX_CONTENT_LENGTH’] = 16 * 1024 * 1024 The code above will limited the maximum allowed payload to 16 megabytes. If a larger file is transmitted, Flask will raise an RequestEntityTooLarge exception. This feature was added in Flask 0.6 but can be achieved in older versions as well by subclassing the request object. For more information on that consult the Werkzeug documentation on file handling. 127
- 138. 18.10.3 Upload Progress Bars A while ago many developers had the idea to read the incoming file in small chunks and store the upload progress in the database to be able to poll the progress with JavaScript from the client. Long story short: the client asks the server every 5 seconds how much it has transmitted already. Do you realize the irony? The client is asking for something it should already know. Now there are better solutions to that work faster and more reliable. The web changed a lot lately and you can use HTML5, Java, Silverlight or Flash to get a nicer uploading experience on the client side. Look at the following libraries for some nice examples how to do that: • Plupload - HTML5, Java, Flash • SWFUpload - Flash • JumpLoader - Java 18.10.4 An Easier Solution Because the common pattern for file uploads exists almost unchanged in all applica- tions dealing with uploads, there is a Flask extension called Flask-Uploads that imple- ments a full fledged upload mechanism with white and blacklisting of extensions and more. 18.11 Caching When your application runs slow, throw some caches in. Well, at least it’s the easiest way to speed up things. What does a cache do? Say you have a function that takes some time to complete but the results would still be good enough if they were 5 min- utes old. So then the idea is that you actually put the result of that calculation into a cache for some time. Flask itself does not provide caching for you, but Werkzeug, one of the libraries it is based on, has some very basic cache support. It supports multiple cache backends, normally you want to use a memcached server. 18.11.1 Setting up a Cache You create a cache object once and keep it around, similar to how Flask objects are created. If you are using the development server you can create a SimpleCache object, that one is a simple cache that keeps the item stored in the memory of the Python interpreter: from werkzeug.contrib.cache import SimpleCache cache = SimpleCache() 128
- 139. If you want to use memcached, make sure to have one of the memcache modules supported (you get them from PyPI) and a memcached server running somewhere. This is how you connect to such an memcached server then: from werkzeug.contrib.cache import MemcachedCache cache = MemcachedCache([’127.0.0.1:11211’]) If you are using App Engine, you can connect to the App Engine memcache server easily: from werkzeug.contrib.cache import GAEMemcachedCache cache = GAEMemcachedCache() 18.11.2 Using a Cache Now how can one use such a cache? There are two very important operations: get() and set(). This is how to use them: To get an item from the cache call get() with a string as key name. If something is in the cache, it is returned. Otherwise that function will return None: rv = cache.get(’my-item’) To add items to the cache, use the set() method instead. The first argument is the key and the second the value that should be set. Also a timeout can be provided after which the cache will automatically remove item. Here a full example how this looks like normally: def get_my_item(): rv = cache.get(’my-item’) if rv is None: rv = calculate_value() cache.set(’my-item’, rv, timeout=5 * 60) return rv 18.12 View Decorators Python has a really interesting feature called function decorators. This allow some really neat things for web applications. Because each view in Flask is a function dec- orators can be used to inject additional functionality to one or more functions. The route() decorator is the one you probably used already. But there are use cases for implementing your own decorator. For instance, imagine you have a view that should only be used by people that are logged in to. If a user goes to the site and is not logged in, they should be redirected to the login page. This is a good example of a use case where a decorator is an excellent solution. 129
- 140. 18.12.1 Login Required Decorator So let’s implement such a decorator. A decorator is a function that returns a function. Pretty simple actually. The only thing you have to keep in mind when implementing something like this is to update the __name__, __module__ and some other attributes of a function. This is often forgotten, but you don’t have to do that by hand, there is a function for that that is used like a decorator (functools.wraps()). This example assumes that the login page is called ’login’ and that the current user is stored as g.user and None if there is no-one logged in: from functools import wraps from flask import g, request, redirect, url_for def login_required(f): @wraps(f) def decorated_function(*args, **kwargs): if g.user is None: return redirect(url_for(’login’, next=request.url)) return f(*args, **kwargs) return decorated_function So how would you use that decorator now? Apply it as innermost decorator to a view function. When applying further decorators, always remember that the route() decorator is the outermost: @app.route(’/secret_page’) @login_required def secret_page(): pass 18.12.2 Caching Decorator Imagine you have a view function that does an expensive calculation and because of that you would like to cache the generated results for a certain amount of time. A decorator would be nice for that. We’re assuming you have set up a cache like mentioned in Caching. Here an example cache function. It generates the cache key from a specific prefix (ac- tually a format string) and the current path of the request. Notice that we are using a function that first creates the decorator that then decorates the function. Sounds aw- ful? Unfortunately it is a little bit more complex, but the code should still be straight- forward to read. The decorated function will then work as follows 1. get the unique cache key for the current request base on the current path. 2. get the value for that key from the cache. If the cache returned something we will return that value. 130
- 141. 3. otherwise the original function is called and the return value is stored in the cache for the timeout provided (by default 5 minutes). Here the code: from functools import wraps from flask import request def cached(timeout=5 * 60, key=’view/%s’): def decorator(f): @wraps(f) def decorated_function(*args, **kwargs): cache_key = key % request.path rv = cache.get(cache_key) if rv is not None: return rv rv = f(*args, **kwargs) cache.set(cache_key, rv, timeout=timeout) return rv return decorated_function return decorator Notice that this assumes an instantiated cache object is available, see Caching for more information. 18.12.3 Templating Decorator A common pattern invented by the TurboGears guys a while back is a templating decorator. The idea of that decorator is that you return a dictionary with the values passed to the template from the view function and the template is automatically ren- dered. With that, the following three examples do exactly the same: @app.route(’/’) def index(): return render_template(’index.html’, value=42) @app.route(’/’) @templated(’index.html’) def index(): return dict(value=42) @app.route(’/’) @templated() def index(): return dict(value=42) As you can see, if no template name is provided it will use the endpoint of the URL map with dots converted to slashes + ’.html’. Otherwise the provided template name is used. When the decorated function returns, the dictionary returned is passed to the template rendering function. If None is returned, an empty dictionary is assumed, if 131
- 142. something else than a dictionary is returned we return it from the function unchanged. That way you can still use the redirect function or return simple strings. Here the code for that decorator: from functools import wraps from flask import request def templated(template=None): def decorator(f): @wraps(f) def decorated_function(*args, **kwargs): template_name = template if template_name is None: template_name = request.endpoint .replace(’.’, ’/’) + ’.html’ ctx = f(*args, **kwargs) if ctx is None: ctx = {} elif not isinstance(ctx, dict): return ctx return render_template(template_name, **ctx) return decorated_function return decorator 18.12.4 Endpoint Decorator When you want to use the werkzeug routing system for more flexibility you need to map the endpoint as defined in the Rule to a view function. This is possible with this decorator. For example: from flask import Flask from werkzeug.routing import Rule app = Flask(__name__) app.url_map.add(Rule(’/’, endpoint=’index’)) @app.endpoint(’index’) def my_index(): return "Hello world" 18.13 Form Validation with WTForms When you have to work with form data submitted by a browser view code quickly becomes very hard to read. There are libraries out there designed to make this process easier to manage. One of them is WTForms which we will handle here. If you find yourself in the situation of having many forms, you might want to give it a try. 132
- 143. When you are working with WTForms you have to define your forms as classes first. I recommend breaking up the application into multiple modules (Larger Applications) for that and adding a separate module for the forms. Getting most of WTForms with an Extension The Flask-WTF extension expands on this pattern and adds a few handful little helpers that make working with forms and Flask more fun. You can get it from PyPI. 18.13.1 The Forms This is an example form for a typical registration page: from wtforms import Form, BooleanField, TextField, PasswordField, validators class RegistrationForm(Form): username = TextField(’Username’, [validators.Length(min=4, max=25)]) email = TextField(’Email Address’, [validators.Length(min=6, max=35)]) password = PasswordField(’New Password’, [ validators.Required(), validators.EqualTo(’confirm’, message=’Passwords must match’) ]) confirm = PasswordField(’Repeat Password’) accept_tos = BooleanField(’I accept the TOS’, [validators.Required()]) 18.13.2 In the View In the view function, the usage of this form looks like this: @app.route(’/register’, methods=[’GET’, ’POST’]) def register(): form = RegistrationForm(request.form) if request.method == ’POST’ and form.validate(): user = User(form.username.data, form.email.data, form.password.data) db_session.add(user) flash(’Thanks for registering’) return redirect(url_for(’login’)) return render_template(’register.html’, form=form) Notice that we are implying that the view is using SQLAlchemy here (SQLAlchemy in Flask) but this is no requirement of course. Adapt the code as necessary. Things to remember: 1. create the form from the request form value if the data is submitted via the HTTP POST method and args if the data is submitted as GET. 2. to validate the data, call the validate() method which will return True if the data validates, False otherwise. 133
- 144. 3. to access individual values from the form, access form.<NAME>.data. 18.13.3 Forms in Templates Now to the template side. When you pass the form to the templates you can easily render them there. Look at the following example template to see how easy this is. WTForms does half the form generation for us already. To make it even nicer, we can write a macro that renders a field with label and a list of errors if there are any. Here’s an example _formhelpers.html template with such a macro: {% macro render_field(field) %} <dt>{{ field.label }} <dd>{{ field(**kwargs)|safe }} {% if field.errors %} <ul class=errors> {% for error in field.errors %} <li>{{ error }}</li> {% endfor %} </ul> {% endif %} </dd> {% endmacro %} This macro accepts a couple of keyword arguments that are forwarded to WTForm’s field function that renders the field for us. The keyword arguments will be in- serted as HTML attributes. So for example you can call render_field(form.username, class=’username’) to add a class to the input element. Note that WTForms returns standard Python unicode strings, so we have to tell Jinja2 that this data is already HTML escaped with the |safe filter. Here the register.html template for the function we used above which takes advantage of the _formhelpers.html template: {% from "_formhelpers.html" import render_field %} <form method=post action="/register"> <dl> {{ render_field(form.username) }} {{ render_field(form.email) }} {{ render_field(form.password) }} {{ render_field(form.confirm) }} {{ render_field(form.accept_tos) }} </dl> <p><input type=submit value=Register> </form> For more information about WTForms, head over to the WTForms website. 134
- 145. 18.14 Template Inheritance The most powerful part of Jinja is template inheritance. Template inheritance allows you to build a base “skeleton” template that contains all the common elements of your site and defines blocks that child templates can override. Sounds complicated but is very basic. It’s easiest to understand it by starting with an example. 18.14.1 Base Template This template, which we’ll call layout.html, defines a simple HTML skeleton docu- ment that you might use for a simple two-column page. It’s the job of “child” tem- plates to fill the empty blocks with content: <!doctype html> <html> <head> {% block head %} <link rel="stylesheet" href="{{ url_for(’static’, filename=’style.css’) }}"> <title>{% block title %}{% endblock %} - My Webpage</title> {% endblock %} </head> <body> <div id="content">{% block content %}{% endblock %}</div> <div id="footer"> {% block footer %} © Copyright 2010 by <a href="http://domain.invalid/">you</a>. {% endblock %} </div> </body> </html> In this example, the {% block %} tags define four blocks that child templates can fill in. All the block tag does is tell the template engine that a child template may override those portions of the template. 18.14.2 Child Template A child template might look like this: {% extends "layout.html" %} {% block title %}Index{% endblock %} {% block head %} {{ super() }} <style type="text/css"> .important { color: #336699; } </style> 135
- 146. {% endblock %} {% block content %} <h1>Index</h1> <p class="important"> Welcome on my awesome homepage. {% endblock %} The {% extends %} tag is the key here. It tells the template engine that this template “extends” another template. When the template system evaluates this template, first it locates the parent. The extends tag must be the first tag in the template. To render the contents of a block defined in the parent template, use {{ super() }}. 18.15 Message Flashing Good applications and user interfaces are all about feedback. If the user does not get enough feedback they will probably end up hating the application. Flask provides a really simple way to give feedback to a user with the flashing system. The flashing system basically makes it possible to record a message at the end of a request and access it next request and only next request. This is usually combined with a layout template that does this. 18.15.1 Simple Flashing So here is a full example: from flask import Flask, flash, redirect, render_template, request, url_for app = Flask(__name__) app.secret_key = ’some_secret’ @app.route(’/’) def index(): return render_template(’index.html’) @app.route(’/login’, methods=[’GET’, ’POST’]) def login(): error = None if request.method == ’POST’: if request.form[’username’] != ’admin’ or request.form[’password’] != ’secret’: error = ’Invalid credentials’ else: flash(’You were successfully logged in’) return redirect(url_for(’index’)) return render_template(’login.html’, error=error) 136
- 147. if __name__ == "__main__": app.run() And here the layout.html template which does the magic: <!doctype html> <title>My Application</title> {% with messages = get_flashed_messages() %} {% if messages %} <ul class=flashes> {% for message in messages %} <li>{{ message }}</li> {% endfor %} </ul> {% endif %} {% endwith %} {% block body %}{% endblock %} And here the index.html template: {% extends "layout.html" %} {% block body %} <h1>Overview</h1> <p>Do you want to <a href="{{ url_for(’login’) }}">log in?</a> {% endblock %} And of course the login template: {% extends "layout.html" %} {% block body %} <h1>Login</h1> {% if error %} <p class=error><strong>Error:</strong> {{ error }} {% endif %} <form action="" method=post> <dl> <dt>Username: <dd><input type=text name=username value="{{ request.form.username }}"> <dt>Password: <dd><input type=password name=password> </dl> <p><input type=submit value=Login> </form> {% endblock %} 18.15.2 Flashing With Categories New in version 0.3. 137
- 148. It is also possible to provide categories when flashing a message. The default cate- gory if nothing is provided is ’message’. Alternative categories can be used to give the user better feedback. For example error messages could be displayed with a red background. To flash a message with a different category, just use the second argument to the flash() function: flash(u’Invalid password provided’, ’error’) Inside the template you then have to tell the get_flashed_messages() function to also return the categories. The loop looks slightly different in that situation then: {% with messages = get_flashed_messages(with_categories=true) %} {% if messages %} <ul class=flashes> {% for category, message in messages %} <li class="{{ category }}">{{ message }}</li> {% endfor %} </ul> {% endif %} {% endwith %} This is just one example of how to render these flashed messages. One might also use the category to add a prefix such as <strong>Error:</strong> to the message. 18.15.3 Filtering Flash Messages New in version 0.9. Optionally you can pass a list of categories which filters the results of get_flashed_messages(). This is useful if you wish to render each category in a sepa- rate block. {% with errors = get_flashed_messages(category_filter=["error"]) %} {% if errors %} <div class="alert-message block-message error"> <a class="close" href="#">×</a> <ul> {%- for msg in errors %} <li>{{ msg }}</li> {% endfor -%} </ul> </div> {% endif %} {% endwith %} 138
- 149. 18.16 AJAX with jQuery jQuery is a small JavaScript library commonly used to simplify working with the DOM and JavaScript in general. It is the perfect tool to make web applications more dynamic by exchanging JSON between server and client. JSON itself is a very lightweight transport format, very similar to how Python primi- tives (numbers, strings, dicts and lists) look like which is widely supported and very easy to parse. It became popular a few years ago and quickly replaced XML as trans- port format in web applications. 18.16.1 Loading jQuery In order to use jQuery, you have to download it first and place it in the static folder of your application and then ensure it’s loaded. Ideally you have a layout template that is used for all pages where you just have to add a script statement to the bottom of your <body> to load jQuery: <script type=text/javascript src="{{ url_for(’static’, filename=’jquery.js’) }}"></script> Another method is using Google’s AJAX Libraries API to load jQuery: <script src="//ajax.googleapis.com/ajax/libs/jquery/1.9.1/jquery.min.js"></script> <script>window.jQuery || document.write(’<script src="{{ url_for(’static’, filename=’jquery.js’) }}">x3C/script>’)</script> In this case you have to put jQuery into your static folder as a fallback, but it will first try to load it directly from Google. This has the advantage that your website will probably load faster for users if they went to at least one other website before using the same jQuery version from Google because it will already be in the browser cache. 18.16.2 Where is My Site? Do you know where your application is? If you are developing the answer is quite simple: it’s on localhost port something and directly on the root of that server. But what if you later decide to move your application to a different location? For example to http://example.com/myapp? On the server side this never was a problem because we were using the handy url_for() function that could answer that question for us, but if we are using jQuery we should not hardcode the path to the application but make that dynamic, so how can we do that? A simple method would be to add a script tag to our page that sets a global variable to the prefix to the root of the application. Something like this: <script type=text/javascript> $SCRIPT_ROOT = {{ request.script_root|tojson|safe }}; </script> 139
- 150. The |safe is necessary in Flask before 0.10 so that Jinja does not escape the JSON encoded string with HTML rules. Usually this would be necessary, but we are inside a script block here where different rules apply. Information for Pros In HTML the script tag is declared CDATA which means that entities will not be parsed. Everything until </script> is handled as script. This also means that there must never be any </ between the script tags. |tojson is kind enough to do the right thing here and escape slashes for you ({{ "</script>"|tojson|safe }} is rendered as "</script>"). In Flask 0.10 it goes a step further and escapes all HTML tags with unicode escapes. This makes it possible for Flask to automatically mark the result as HTML safe. 18.16.3 JSON View Functions Now let’s create a server side function that accepts two URL arguments of numbers which should be added together and then sent back to the application in a JSON object. This is a really ridiculous example and is something you usually would do on the client side alone, but a simple example that shows how you would use jQuery and Flask nonetheless: from flask import Flask, jsonify, render_template, request app = Flask(__name__) @app.route(’/_add_numbers’) def add_numbers(): a = request.args.get(’a’, 0, type=int) b = request.args.get(’b’, 0, type=int) return jsonify(result=a + b) @app.route(’/’) def index(): return render_template(’index.html’) As you can see I also added an index method here that renders a template. This tem- plate will load jQuery as above and have a little form we can add two numbers and a link to trigger the function on the server side. Note that we are using the get() method here which will never fail. If the key is missing a default value (here 0) is returned. Furthermore it can convert values to a specific type (like in our case int). This is especially handy for code that is triggered by a script (APIs, JavaScript etc.) because you don’t need special error reporting in that case. 140
- 151. 18.16.4 The HTML Your index.html template either has to extend a layout.html template with jQuery loaded and the $SCRIPT_ROOT variable set, or do that on the top. Here’s the HTML code needed for our little application (index.html). Notice that we also drop the script directly into the HTML here. It is usually a better idea to have that in a separate script file: <script type=text/javascript> $(function() { $(’a#calculate’).bind(’click’, function() { $.getJSON($SCRIPT_ROOT + ’/_add_numbers’, { a: $(’input[name="a"]’).val(), b: $(’input[name="b"]’).val() }, function(data) { $("#result").text(data.result); }); return false; }); }); </script> <h1>jQuery Example</h1> <p><input type=text size=5 name=a> + <input type=text size=5 name=b> = <span id=result>?</span> <p><a href=# id=calculate>calculate server side</a> I won’t go into detail here about how jQuery works, just a very quick explanation of the little bit of code above: 1. $(function() { ... }) specifies code that should run once the browser is done loading the basic parts of the page. 2. $(’selector’) selects an element and lets you operate on it. 3. element.bind(’event’, func) specifies a function that should run when the user clicked on the element. If that function returns false, the default behavior will not kick in (in this case, navigate to the # URL). 4. $.getJSON(url, data, func) sends a GET request to url and will send the con- tents of the data object as query parameters. Once the data arrived, it will call the given function with the return value as argument. Note that we can use the $SCRIPT_ROOT variable here that we set earlier. If you don’t get the whole picture, download the sourcecode for this example from github. 18.17 Custom Error Pages Flask comes with a handy abort() function that aborts a request with an HTTP error code early. It will also provide a plain black and white error page for you with a basic 141
- 152. description, but nothing fancy. Depending on the error code it is less or more likely for the user to actually see such an error. 18.17.1 Common Error Codes The following error codes are some that are often displayed to the user, even if the application behaves correctly: 404 Not Found The good old “chap, you made a mistake typing that URL” message. So common that even novices to the internet know that 404 means: damn, the thing I was looking for is not there. It’s a very good idea to make sure there is actually something useful on a 404 page, at least a link back to the index. 403 Forbidden If you have some kind of access control on your website, you will have to send a 403 code for disallowed resources. So make sure the user is not lost when they try to access a forbidden resource. 410 Gone Did you know that there the “404 Not Found” has a brother named “410 Gone”? Few people actually implement that, but the idea is that resources that previously existed and got deleted answer with 410 instead of 404. If you are not deleting documents permanently from the database but just mark them as deleted, do the user a favour and use the 410 code instead and display a message that what they were looking for was deleted for all eternity. 500 Internal Server Error Usually happens on programming errors or if the server is overloaded. A terrible good idea to have a nice page there, because your appli- cation will fail sooner or later (see also: Logging Application Errors). 18.17.2 Error Handlers An error handler is a function, just like a view function, but it is called when an error happens and is passed that error. The error is most likely a HTTPException, but in one case it can be a different error: a handler for internal server errors will be passed other exception instances as well if they are uncaught. An error handler is registered with the errorhandler() decorator and the error code of the exception. Keep in mind that Flask will not set the error code for you, so make sure to also provide the HTTP status code when returning a response. Here an example implementation for a “404 Page Not Found” exception: from flask import render_template @app.errorhandler(404) def page_not_found(e): return render_template(’404.html’), 404 An example template might be this: 142
- 153. {% extends "layout.html" %} {% block title %}Page Not Found{% endblock %} {% block body %} <h1>Page Not Found</h1> <p>What you were looking for is just not there. <p><a href="{{ url_for(’index’) }}">go somewhere nice</a> {% endblock %} 18.18 Lazily Loading Views Flask is usually used with the decorators. Decorators are simple and you have the URL right next to the function that is called for that specific URL. However there is a downside to this approach: it means all your code that uses decorators has to be imported upfront or Flask will never actually find your function. This can be a problem if your application has to import quick. It might have to do that on systems like Google’s App Engine or other systems. So if you suddenly notice that your application outgrows this approach you can fall back to a centralized URL mapping. The system that enables having a central URL map is the add_url_rule() function. Instead of using decorators, you have a file that sets up the application with all URLs. 18.18.1 Converting to Centralized URL Map Imagine the current application looks somewhat like this: from flask import Flask app = Flask(__name__) @app.route(’/’) def index(): pass @app.route(’/user/<username>’) def user(username): pass Then the centralized approach you would have one file with the views (views.py) but without any decorator: def index(): pass def user(username): pass And then a file that sets up an application which maps the functions to URLs: 143
- 154. from flask import Flask from yourapplication import views app = Flask(__name__) app.add_url_rule(’/’, view_func=views.index) app.add_url_rule(’/user/<username>’, view_func=views.user) 18.18.2 Loading Late So far we only split up the views and the routing, but the module is still loaded up- front. The trick to actually load the view function as needed. This can be accomplished with a helper class that behaves just like a function but internally imports the real func- tion on first use: from werkzeug import import_string, cached_property class LazyView(object): def __init__(self, import_name): self.__module__, self.__name__ = import_name.rsplit(’.’, 1) self.import_name = import_name @cached_property def view(self): return import_string(self.import_name) def __call__(self, *args, **kwargs): return self.view(*args, **kwargs) What’s important here is is that __module__ and __name__ are properly set. This is used by Flask internally to figure out how to name the URL rules in case you don’t provide a name for the rule yourself. Then you can define your central place to combine the views like this: from flask import Flask from yourapplication.helpers import LazyView app = Flask(__name__) app.add_url_rule(’/’, view_func=LazyView(’yourapplication.views.index’)) app.add_url_rule(’/user/<username>’, view_func=LazyView(’yourapplication.views.user’)) You can further optimize this in terms of amount of keystrokes needed to write this by having a function that calls into add_url_rule() by prefixing a string with the project name and a dot, and by wrapping view_func in a LazyView as needed: def url(url_rule, import_name, **options): view = LazyView(’yourapplication.’ + import_name) app.add_url_rule(url_rule, view_func=view, **options) 144
- 155. url(’/’, ’views.index’) url(’/user/<username>’, ’views.user’) One thing to keep in mind is that before and after request handlers have to be in a file that is imported upfront to work properly on the first request. The same goes for any kind of remaining decorator. 18.19 MongoKit in Flask Using a document database rather than a full DBMS gets more common these days. This pattern shows how to use MongoKit, a document mapper library, to integrate with MongoDB. This pattern requires a running MongoDB server and the MongoKit library installed. There are two very common ways to use MongoKit. I will outline each of them here: 18.19.1 Declarative The default behavior of MongoKit is the declarative one that is based on common ideas from Django or the SQLAlchemy declarative extension. Here an example app.py module for your application: from flask import Flask from mongokit import Connection, Document # configuration MONGODB_HOST = ’localhost’ MONGODB_PORT = 27017 # create the little application object app = Flask(__name__) app.config.from_object(__name__) # connect to the database connection = Connection(app.config[’MONGODB_HOST’], app.config[’MONGODB_PORT’]) To define your models, just subclass the Document class that is imported from Mon- goKit. If you’ve seen the SQLAlchemy pattern you may wonder why we do not have a session and even do not define a init_db function here. On the one hand, MongoKit does not have something like a session. This sometimes makes it more to type but also makes it blazingly fast. On the other hand, MongoDB is schemaless. This means you can modify the data structure from one insert query to the next without any prob- lem. MongoKit is just schemaless too, but implements some validation to ensure data integrity. Here is an example document (put this also into app.py, e.g.): 145
- 156. def max_length(length): def validate(value): if len(value) <= length: return True raise Exception(’%s must be at most %s characters long’ % length) return validate class User(Document): structure = { ’name’: unicode, ’email’: unicode, } validators = { ’name’: max_length(50), ’email’: max_length(120) } use_dot_notation = True def __repr__(self): return ’<User %r>’ % (self.name) # register the User document with our current connection connection.register([User]) This example shows you how to define your schema (named structure), a valida- tor for the maximum character length and uses a special MongoKit feature called use_dot_notation. Per default MongoKit behaves like a python dictionary but with use_dot_notation set to True you can use your documents like you use models in nearly any other ORM by using dots to separate between attributes. You can insert entries into the database like this: >>> from yourapplication.database import connection >>> from yourapplication.models import User >>> collection = connection[’test’].users >>> user = collection.User() >>> user[’name’] = u’admin’ >>> user[’email’] = u’admin@localhost’ >>> user.save() Note that MongoKit is kinda strict with used column types, you must not use a com- mon str type for either name or email but unicode. Querying is simple as well: >>> list(collection.User.find()) [<User u’admin’>] >>> collection.User.find_one({’name’: u’admin’}) <User u’admin’> 146
- 157. 18.19.2 PyMongo Compatibility Layer If you just want to use PyMongo, you can do that with MongoKit as well. You may use this process if you need the best performance to get. Note that this example does not show how to couple it with Flask, see the above MongoKit code for examples: from MongoKit import Connection connection = Connection() To insert data you can use the insert method. We have to get a collection first, this is somewhat the same as a table in the SQL world. >>> collection = connection[’test’].users >>> user = {’name’: u’admin’, ’email’: u’admin@localhost’} >>> collection.insert(user) MongoKit will automatically commit for us. To query your database, you use the collection directly: >>> list(collection.find()) [{u’_id’: ObjectId(’4c271729e13823182f000000’), u’name’: u’admin’, u’email’: u’admin@localhost’} >>> collection.find_one({’name’: u’admin’}) {u’_id’: ObjectId(’4c271729e13823182f000000’), u’name’: u’admin’, u’email’: u’admin@localhost’} These results are also dict-like objects: >>> r = collection.find_one({’name’: u’admin’}) >>> r[’email’] u’admin@localhost’ For more information about MongoKit, head over to the website. 18.20 Adding a favicon A “favicon” is an icon used by browsers for tabs and bookmarks. This helps to distin- guish your website and to give it a unique brand. A common question is how to add a favicon to a flask application. First, of course, you need an icon. It should be 16 × 16 pixels and in the ICO file format. This is not a requirement but a de-facto standard supported by all relevant browsers. Put the icon in your static directory as favicon.ico. Now, to get browsers to find your icon, the correct way is to add a link tag in your HTML. So, for example: <link rel="shortcut icon" href="{{ url_for(’static’, filename=’favicon.ico’) }}"> That’s all you need for most browsers, however some really old ones do not support this standard. The old de-facto standard is to serve this file, with this name, at the website root. If your application is not mounted at the root path of the domain you 147
- 158. either need to configure the webserver to serve the icon at the root or if you can’t do that you’re out of luck. If however your application is the root you can simply route a redirect: app.add_url_rule(’/favicon.ico’, redirect_to=url_for(’static’, filename=’favicon.ico’)) If you want to save the extra redirect request you can also write a view using send_from_directory(): import os from flask import send_from_directory @app.route(’/favicon.ico’) def favicon(): return send_from_directory(os.path.join(app.root_path, ’static’), ’favicon.ico’, mimetype=’image/vnd.microsoft.icon’) We can leave out the explicit mimetype and it will be guessed, but we may as well specify it to avoid the extra guessing, as it will always be the same. The above will serve the icon via your application and if possible it’s better to config- ure your dedicated web server to serve it; refer to the webserver’s documentation. 18.20.1 See also • The Favicon article on Wikipedia 18.21 Streaming Contents Sometimes you want to send an enormous amount of data to the client, much more than you want to keep in memory. When you are generating the data on the fly though, how do you send that back to the client without the roundtrip to the filesystem? The answer is by using generators and direct responses. 18.21.1 Basic Usage This is a basic view function that generates a lot of CSV data on the fly. The trick is to have an inner function that uses a generator to generate data and to then invoke that function and pass it to a response object: from flask import Response @app.route(’/large.csv’) def generate_large_csv(): def generate(): for row in iter_all_rows(): 148
- 159. yield ’,’.join(row) + ’n’ return Response(generate(), mimetype=’text/csv’) Each yield expression is directly sent to the browser. Note though that some WSGI middlewares might break streaming, so be careful there in debug environments with profilers and other things you might have enabled. 18.21.2 Streaming from Templates The Jinja2 template engine also supports rendering templates piece by piece. This functionality is not directly exposed by Flask because it is quite uncommon, but you can easily do it yourself: from flask import Response def stream_template(template_name, **context): app.update_template_context(context) t = app.jinja_env.get_template(template_name) rv = t.stream(context) rv.enable_buffering(5) return rv @app.route(’/my-large-page.html’) def render_large_template(): rows = iter_all_rows() return Response(stream_template(’the_template.html’, rows=rows)) The trick here is to get the template object from the Jinja2 environment on the appli- cation and to call stream() instead of render() which returns a stream object instead of a string. Since we’re bypassing the Flask template render functions and using the template object itself we have to make sure to update the render context ourselves by calling update_template_context(). The template is then evaluated as the stream is iterated over. Since each time you do a yield the server will flush the content to the client you might want to buffer up a few items in the template which you can do with rv.enable_buffering(size). 5 is a sane default. 18.21.3 Streaming with Context New in version 0.9. Note that when you stream data, the request context is already gone the moment the function executes. Flask 0.9 provides you with a helper that can keep the request context around during the execution of the generator: from flask import stream_with_context, request, Response @app.route(’/stream’) def streamed_response(): 149
- 160. def generate(): yield ’Hello ’ yield request.args[’name’] yield ’!’ return Response(stream_with_context(generate())) Without the stream_with_context() function you would get a RuntimeError at that point. 18.22 Deferred Request Callbacks One of the design principles of Flask is that response objects are created and passed down a chain of potential callbacks that can modify them or replace them. When the request handling starts, there is no response object yet. It is created as necessary either by a view function or by some other component in the system. But what happens if you want to modify the response at a point where the response does not exist yet? A common example for that would be a before-request function that wants to set a cookie on the response object. One way is to avoid the situation. Very often that is possible. For instance you can try to move that logic into an after-request callback instead. Sometimes however moving that code there is just not a very pleasant experience or makes code look very awk- ward. As an alternative possibility you can attach a bunch of callback functions to the g object and call them at the end of the request. This way you can defer code execution from anywhere in the application. 18.22.1 The Decorator The following decorator is the key. It registers a function on a list on the g object: from flask import g def after_this_request(f): if not hasattr(g, ’after_request_callbacks’): g.after_request_callbacks = [] g.after_request_callbacks.append(f) return f 18.22.2 Calling the Deferred Now you can use the after_this_request decorator to mark a function to be called at the end of the request. But we still need to call them. For this the following function needs to be registered as after_request() callback: 150
- 161. @app.after_request def call_after_request_callbacks(response): for callback in getattr(g, ’after_request_callbacks’, ()): callback(response) return response 18.22.3 A Practical Example Now we can easily at any point in time register a function to be called at the end of this particular request. For example you can remember the current language of the user in a cookie in the before-request function: from flask import request @app.before_request def detect_user_language(): language = request.cookies.get(’user_lang’) if language is None: language = guess_language_from_request() @after_this_request def remember_language(response): response.set_cookie(’user_lang’, language) g.language = language 18.23 Adding HTTP Method Overrides Some HTTP proxies do not support arbitrary HTTP methods or newer HTTP methods (such as PATCH). In that case it’s possible to “proxy” HTTP methods through another HTTP method in total violation of the protocol. The way this works is by letting the client do an HTTP POST request and set the X-HTTP-Method-Override header and set the value to the intended HTTP method (such as PATCH). This can easily be accomplished with an HTTP middleware: class HTTPMethodOverrideMiddleware(object): allowed_methods = frozenset([ ’GET’, ’HEAD’, ’POST’, ’DELETE’, ’PUT’, ’PATCH’, ’OPTIONS’ ]) bodyless_methods = frozenset([’GET’, ’HEAD’, ’OPTIONS’, ’DELETE’]) 151
- 162. def __init__(self, app): self.app = app def __call__(self, environ, start_response): method = environ.get(’HTTP_X_HTTP_METHOD_OVERRIDE’, ’’).upper() if method in self.allowed_methods: method = method.encode(’ascii’, ’replace’) environ[’REQUEST_METHOD’] = method if method in self.bodyless_methods: environ[’CONTENT_LENGTH’] = ’0’ return self.app(environ, start_response) To use this with Flask this is all that is necessary: from flask import Flask app = Flask(__name__) app.wsgi_app = HTTPMethodOverrideMiddleware(app.wsgi_app) 18.24 Request Content Checksums Various pieces of code can consume the request data and preprocess it. For instance JSON data ends up on the request object already read and processed, form data ends up there as well but goes through a different code path. This seems inconvenient when you want to calculate the checksum of the incoming request data. This is necessary sometimes for some APIs. Fortunately this is however very simple to change by wrapping the input stream. The following example calculates the SHA1 checksum of the incoming data as it gets read and stores it in the WSGI environment: import hashlib class ChecksumCalcStream(object): def __init__(self, stream): self._stream = stream self._hash = hashlib.sha1() def read(self, bytes): rv = self._stream.read(bytes) self._hash.update(rv) return rv def readline(self, size_hint): rv = self._stream.readline(size_hint) self._hash.update(rv) return rv 152
- 163. def generate_checksum(request): env = request.environ stream = ChecksumCalcStream(env[’wsgi.input’]) env[’wsgi.input’] = stream return stream._hash To use this, all you need to do is to hook the calculating stream in before the request starts consuming data. (Eg: be careful accessing request.form or anything of that nature. before_request_handlers for instance should be careful not to access it). Example usage: @app.route(’/special-api’, methods=[’POST’]) def special_api(): hash = generate_checksum(request) # Accessing this parses the input stream files = request.files # At this point the hash is fully constructed. checksum = hash.hexdigest() return ’Hash was: %s’ % checksum 18.25 Celery Based Background Tasks Celery is a task queue for Python with batteries included. It used to have a Flask inte- gration but it became unnecessary after some restructuring of the internals of Celery with Version 3. This guide fills in the blanks in how to properly use Celery with Flask but assumes that you generally already read the First Steps with Celery guide in the official Celery documentation. 18.25.1 Installing Celery Celery is on the Python Package Index (PyPI), so it can be installed with standard Python tools like pip or easy_install: $ pip install celery 18.25.2 Configuring Celery The first thing you need is a Celery instance, this is called the celery application. It serves the same purpose as the Flask object in Flask, just for Celery. Since this instance is used as the entry-point for everything you want to do in Celery, like creating tasks and managing workers, it must be possible for other modules to import it. For instance you can place this in a tasks module. While you can use Celery without any reconfiguration with Flask, it becomes a bit nicer by subclassing tasks and adding 153
- 164. support for Flask’s application contexts and hooking it up with the Flask configura- tion. This is all that is necessary to properly integrate Celery with Flask: from celery import Celery def make_celery(app): celery = Celery(app.import_name, broker=app.config[’CELERY_BROKER_URL’]) celery.conf.update(app.config) TaskBase = celery.Task class ContextTask(TaskBase): abstract = True def __call__(self, *args, **kwargs): with app.app_context(): return TaskBase.__call__(self, *args, **kwargs) celery.Task = ContextTask return celery The function creates a new Celery object, configures it with the broker from the ap- plication config, updates the rest of the Celery config from the Flask config and then creates a subclass of the task that wraps the task execution in an application context. 18.25.3 Minimal Example With what we have above this is the minimal example of using Celery with Flask: from flask import Flask flask_app = Flask(__name__) flask_app.config.update( CELERY_BROKER_URL=’redis://localhost:6379’, CELERY_RESULT_BACKEND=’redis://localhost:6379’ ) celery = make_celery(flask_app) @celery.task() def add_together(a, b): return a + b This task can now be called in the background: >>> result = add_together.delay(23, 42) >>> result.wait() 65 154
- 165. 18.25.4 Running the Celery Worker Now if you jumped in and already executed the above code you will be disappointed to learn that your .wait() will never actually return. That’s because you also need to run celery. You can do that by running celery as a worker: $ celery -A your_application.celery worker The your_application string has to point to your application’s package or module that creates the celery object. 155
- 166. 156
- 167. CHAPTER NINETEEN DEPLOYMENT OPTIONS Depending on what you have available there are multiple ways to run Flask applica- tions. You can use the builtin server during development, but you should use a full deployment option for production applications. (Do not use the builtin development server in production.) Several options are available and documented here. If you have a different WSGI server look up the server documentation about how to use a WSGI app with it. Just remember that your Flask application object is the actual WSGI application. For hosted options to get up and running quickly, see Deploying to a Web Server in the Quickstart. 19.1 mod_wsgi (Apache) If you are using the Apache webserver, consider using mod_wsgi. Watch Out Please make sure in advance that any app.run() calls you might have in your applica- tion file are inside an if __name__ == ’__main__’: block or moved to a separate file. Just make sure it’s not called because this will always start a local WSGI server which we do not want if we deploy that application to mod_wsgi. 19.1.1 Installing mod_wsgi If you don’t have mod_wsgi installed yet you have to either install it using a package manager or compile it yourself. The mod_wsgi installation instructions cover source installations on UNIX systems. If you are using Ubuntu/Debian you can apt-get it and activate it as follows: # apt-get install libapache2-mod-wsgi On FreeBSD install mod_wsgi by compiling the www/mod_wsgi port or by using pkg_add: 157
- 168. # pkg_add -r mod_wsgi If you are using pkgsrc you can install mod_wsgi by compiling the www/ap2-wsgi pack- age. If you encounter segfaulting child processes after the first apache reload you can safely ignore them. Just restart the server. 19.1.2 Creating a .wsgi file To run your application you need a yourapplication.wsgi file. This file contains the code mod_wsgi is executing on startup to get the application object. The object called appli- cation in that file is then used as application. For most applications the following file should be sufficient: from yourapplication import app as application If you don’t have a factory function for application creation but a singleton instance you can directly import that one as application. Store that file somewhere that you will find it again (e.g.: /var/www/yourapplication) and make sure that yourapplication and all the libraries that are in use are on the python load path. If you don’t want to install it system wide consider using a virtual python instance. Keep in mind that you will have to actually install your application into the virtualenv as well. Alternatively there is the option to just patch the path in the .wsgi file before the import: import sys sys.path.insert(0, ’/path/to/the/application’) 19.1.3 Configuring Apache The last thing you have to do is to create an Apache configuration file for your ap- plication. In this example we are telling mod_wsgi to execute the application under a different user for security reasons: <VirtualHost *> ServerName example.com WSGIDaemonProcess yourapplication user=user1 group=group1 threads=5 WSGIScriptAlias / /var/www/yourapplication/yourapplication.wsgi <Directory /var/www/yourapplication> WSGIProcessGroup yourapplication WSGIApplicationGroup %{GLOBAL} Order deny,allow Allow from all 158
- 169. </Directory> </VirtualHost> Note: WSGIDaemonProcess isn’t implemented in Windows and Apache will refuse to run with the above configuration. On a Windows system, eliminate those lines: <VirtualHost *> ServerName example.com WSGIScriptAlias / C:yourdiryourapp.wsgi <Directory C:yourdir> Order deny,allow Allow from all </Directory> </VirtualHost> For more information consult the mod_wsgi wiki. 19.1.4 Troubleshooting If your application does not run, follow this guide to troubleshoot: Problem: application does not run, errorlog shows SystemExit ignored You have a app.run() call in your application file that is not guarded by an if __name__ == ’__main__’: condition. Either remove that run() call from the file and move it into a separate run.py file or put it into such an if block. Problem: application gives permission errors Probably caused by your application running as the wrong user. Make sure the folders the application needs access to have the proper privileges set and the application runs as the correct user (user and group parameter to the WSGIDaemonProcess directive) Problem: application dies with an error on print Keep in mind that mod_wsgi dis- allows doing anything with sys.stdout and sys.stderr. You can disable this protection from the config by setting the WSGIRestrictStdout to off: WSGIRestrictStdout Off Alternatively you can also replace the standard out in the .wsgi file with a differ- ent stream: import sys sys.stdout = sys.stderr Problem: accessing resources gives IO errors Your application probably is a single .py file you symlinked into the site-packages folder. Please be aware that this does not work, instead you either have to put the folder into the pythonpath the file is stored in, or convert your application into a package. The reason for this is that for non-installed packages, the module filename is used to locate the resources and for symlinks the wrong filename is picked up. 159
- 170. 19.1.5 Support for Automatic Reloading To help deployment tools you can activate support for automatic reloading. Whenever something changes the .wsgi file, mod_wsgi will reload all the daemon processes for us. For that, just add the following directive to your Directory section: WSGIScriptReloading On 19.1.6 Working with Virtual Environments Virtual environments have the advantage that they never install the required depen- dencies system wide so you have a better control over what is used where. If you want to use a virtual environment with mod_wsgi you have to modify your .wsgi file slightly. Add the following lines to the top of your .wsgi file: activate_this = ’/path/to/env/bin/activate_this.py’ execfile(activate_this, dict(__file__=activate_this)) This sets up the load paths according to the settings of the virtual environment. Keep in mind that the path has to be absolute. 19.2 Standalone WSGI Containers There are popular servers written in Python that contain WSGI applications and serve HTTP. These servers stand alone when they run; you can proxy to them from your web server. Note the section on Proxy Setups if you run into issues. 19.2.1 Gunicorn Gunicorn ‘Green Unicorn’ is a WSGI HTTP Server for UNIX. It’s a pre-fork worker model ported from Ruby’s Unicorn project. It supports both eventlet and greenlet. Running a Flask application on this server is quite simple: gunicorn myproject:app Gunicorn provides many command-line options – see gunicorn -h. For example, to run a Flask application with 4 worker processes (-w 4) binding to localhost port 4000 (-b 127.0.0.1:4000): gunicorn -w 4 -b 127.0.0.1:4000 myproject:app 160
- 171. 19.2.2 Tornado Tornado is an open source version of the scalable, non-blocking web server and tools that power FriendFeed. Because it is non-blocking and uses epoll, it can handle thou- sands of simultaneous standing connections, which means it is ideal for real-time web services. Integrating this service with Flask is straightforward: from tornado.wsgi import WSGIContainer from tornado.httpserver import HTTPServer from tornado.ioloop import IOLoop from yourapplication import app http_server = HTTPServer(WSGIContainer(app)) http_server.listen(5000) IOLoop.instance().start() 19.2.3 Gevent Gevent is a coroutine-based Python networking library that uses greenlet to provide a high-level synchronous API on top of libevent event loop: from gevent.wsgi import WSGIServer from yourapplication import app http_server = WSGIServer((’’, 5000), app) http_server.serve_forever() 19.2.4 Twisted Web Twisted Web is the web server shipped with Twisted, a mature, non-blocking event- driven networking library. Twisted Web comes with a standard WSGI container which can be controlled from the command line using the twistd utility: twistd web --wsgi myproject.app This example will run a Flask application called app from a module named myproject. Twisted Web supports many flags and options, and the twistd utility does as well; see twistd -h and twistd web -h for more information. For example, to run a Twisted Web server in the foreground, on port 8080, with an application from myproject: twistd -n web --port 8080 --wsgi myproject.app 19.2.5 Proxy Setups If you deploy your application using one of these servers behind an HTTP proxy you will need to rewrite a few headers in order for the application to work. The 161
- 172. two problematic values in the WSGI environment usually are REMOTE_ADDR and HTTP_HOST. You can configure your httpd to pass these headers, or you can fix them in middleware. Werkzeug ships a fixer that will solve some common setups, but you might want to write your own WSGI middleware for specific setups. Here’s a simple nginx configuration which proxies to an application served on local- host at port 8000, setting appropriate headers: server { listen 80; server_name _; access_log /var/log/nginx/access.log; error_log /var/log/nginx/error.log; location / { proxy_pass http://127.0.0.1:8000/; proxy_redirect off; proxy_set_header Host $host; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; } } If your httpd is not providing these headers, the most common setup invokes the host being set from X-Forwarded-Host and the remote address from X-Forwarded-For: from werkzeug.contrib.fixers import ProxyFix app.wsgi_app = ProxyFix(app.wsgi_app) Trusting Headers Please keep in mind that it is a security issue to use such a middleware in a non-proxy setup because it will blindly trust the incoming headers which might be forged by malicious clients. If you want to rewrite the headers from another header, you might want to use a fixer like this: class CustomProxyFix(object): def __init__(self, app): self.app = app def __call__(self, environ, start_response): host = environ.get(’HTTP_X_FHOST’, ’’) if host: environ[’HTTP_HOST’] = host return self.app(environ, start_response) 162
- 173. app.wsgi_app = CustomProxyFix(app.wsgi_app) 19.3 uWSGI uWSGI is a deployment option on servers like nginx, lighttpd, and cherokee; see FastCGI and Standalone WSGI Containers for other options. To use your WSGI appli- cation with uWSGI protocol you will need a uWSGI server first. uWSGI is both a protocol and an application server; the application server can serve uWSGI, FastCGI, and HTTP protocols. The most popular uWSGI server is uwsgi, which we will use for this guide. Make sure to have it installed to follow along. Watch Out Please make sure in advance that any app.run() calls you might have in your applica- tion file are inside an if __name__ == ’__main__’: block or moved to a separate file. Just make sure it’s not called because this will always start a local WSGI server which we do not want if we deploy that application to uWSGI. 19.3.1 Starting your app with uwsgi uwsgi is designed to operate on WSGI callables found in python modules. Given a flask application in myapp.py, use the following command: $ uwsgi -s /tmp/uwsgi.sock --module myapp --callable app Or, if you prefer: $ uwsgi -s /tmp/uwsgi.sock -w myapp:app 19.3.2 Configuring nginx A basic flask uWSGI configuration for nginx looks like this: location = /yourapplication { rewrite ^ /yourapplication/; } location /yourapplication { try_files $uri @yourapplication; } location @yourapplication { include uwsgi_params; uwsgi_param SCRIPT_NAME /yourapplication; uwsgi_modifier1 30; uwsgi_pass unix:/tmp/uwsgi.sock; } 163
- 174. This configuration binds the application to /yourapplication. If you want to have it in the URL root it’s a bit simpler because you don’t have to tell it the WSGI SCRIPT_NAME or set the uwsgi modifier to make use of it: location / { try_files $uri @yourapplication; } location @yourapplication { include uwsgi_params; uwsgi_pass unix:/tmp/uwsgi.sock; } 19.4 FastCGI FastCGI is a deployment option on servers like nginx, lighttpd, and cherokee; see uWSGI and Standalone WSGI Containers for other options. To use your WSGI applica- tion with any of them you will need a FastCGI server first. The most popular one is flup which we will use for this guide. Make sure to have it installed to follow along. Watch Out Please make sure in advance that any app.run() calls you might have in your applica- tion file are inside an if __name__ == ’__main__’: block or moved to a separate file. Just make sure it’s not called because this will always start a local WSGI server which we do not want if we deploy that application to FastCGI. 19.4.1 Creating a .fcgi file First you need to create the FastCGI server file. Let’s call it yourapplication.fcgi: #!/usr/bin/python from flup.server.fcgi import WSGIServer from yourapplication import app if __name__ == ’__main__’: WSGIServer(app).run() This is enough for Apache to work, however nginx and older versions of lighttpd need a socket to be explicitly passed to communicate with the FastCGI server. For that to work you need to pass the path to the socket to the WSGIServer: WSGIServer(application, bindAddress=’/path/to/fcgi.sock’).run() The path has to be the exact same path you define in the server config. Save the yourapplication.fcgi file somewhere you will find it again. It makes sense to have that in /var/www/yourapplication or something similar. Make sure to set the executable bit on that file so that the servers can execute it: 164
- 175. # chmod +x /var/www/yourapplication/yourapplication.fcgi 19.4.2 Configuring Apache The example above is good enough for a basic Apache deployment but your .fcgi file will appear in your application URL e.g. example.com/yourapplication.fcgi/news/. There are few ways to configure your application so that yourapplication.fcgi does not appear in the URL. A preferable way is to use the ScriptAlias and SetHandler config- uration directives to route requests to the FastCGI server. The following example uses FastCgiServer to start 5 instances of the application which will handle all incoming requests: LoadModule fastcgi_module /usr/lib64/httpd/modules/mod_fastcgi.so FastCgiServer /var/www/html/yourapplication/app.fcgi -idle-timeout 300 -processes 5 <VirtualHost *> ServerName webapp1.mydomain.com DocumentRoot /var/www/html/yourapplication AddHandler fastcgi-script fcgi ScriptAlias / /var/www/html/yourapplication/app.fcgi/ <Location /> SetHandler fastcgi-script </Location> </VirtualHost> These processes will be managed by Apache. If you’re using an standalone FastCGI server, you can use the FastCgiExternalServer directive instead. Note that in the fol- lowing the path is not real, it’s simply used as an identifier to other directives such as AliasMatch: FastCgiServer /var/www/html/yourapplication -host 127.0.0.1:3000 If you cannot set ScriptAlias, for example on an shared web host, you can use WSGI middleware to remove yourapplication.fcgi from the URLs. Set .htaccess: <IfModule mod_fcgid.c> AddHandler fcgid-script .fcgi <Files ~ (.fcgi)> SetHandler fcgid-script Options +FollowSymLinks +ExecCGI </Files> </IfModule> <IfModule mod_rewrite.c> Options +FollowSymlinks RewriteEngine On RewriteBase / 165
- 176. RewriteCond %{REQUEST_FILENAME} !-f RewriteRule ^(.*)$ yourapplication.fcgi/$1 [QSA,L] </IfModule> Set yourapplication.fcgi: #!/usr/bin/python #: optional path to your local python site-packages folder import sys sys.path.insert(0, ’<your_local_path>/lib/python2.6/site-packages’) from flup.server.fcgi import WSGIServer from yourapplication import app class ScriptNameStripper(object): def __init__(self, app): self.app = app def __call__(self, environ, start_response): environ[’SCRIPT_NAME’] = ’’ return self.app(environ, start_response) app = ScriptNameStripper(app) if __name__ == ’__main__’: WSGIServer(app).run() 19.4.3 Configuring lighttpd A basic FastCGI configuration for lighttpd looks like that: fastcgi.server = ("/yourapplication.fcgi" => (( "socket" => "/tmp/yourapplication-fcgi.sock", "bin-path" => "/var/www/yourapplication/yourapplication.fcgi", "check-local" => "disable", "max-procs" => 1 )) ) alias.url = ( "/static/" => "/path/to/your/static" ) url.rewrite-once = ( "^(/static($|/.*))$" => "$1", "^(/.*)$" => "/yourapplication.fcgi$1" ) Remember to enable the FastCGI, alias and rewrite modules. This configuration binds 166
- 177. the application to /yourapplication. If you want the application to work in the URL root you have to work around a lighttpd bug with the LighttpdCGIRootFix middleware. Make sure to apply it only if you are mounting the application the URL root. Also, see the Lighty docs for more information on FastCGI and Python (note that explicitly passing a socket to run() is no longer necessary). 19.4.4 Configuring nginx Installing FastCGI applications on nginx is a bit different because by default no FastCGI parameters are forwarded. A basic flask FastCGI configuration for nginx looks like this: location = /yourapplication { rewrite ^ /yourapplication/ last; } location /yourapplication { try_files $uri @yourapplication; } location @yourapplication { include fastcgi_params; fastcgi_split_path_info ^(/yourapplication)(.*)$; fastcgi_param PATH_INFO $fastcgi_path_info; fastcgi_param SCRIPT_NAME $fastcgi_script_name; fastcgi_pass unix:/tmp/yourapplication-fcgi.sock; } This configuration binds the application to /yourapplication. If you want to have it in the URL root it’s a bit simpler because you don’t have to figure out how to calculate PATH_INFO and SCRIPT_NAME: location / { try_files $uri @yourapplication; } location @yourapplication { include fastcgi_params; fastcgi_param PATH_INFO $fastcgi_script_name; fastcgi_param SCRIPT_NAME ""; fastcgi_pass unix:/tmp/yourapplication-fcgi.sock; } 19.4.5 Running FastCGI Processes Since Nginx and others do not load FastCGI apps, you have to do it by yourself. Super- visor can manage FastCGI processes. You can look around for other FastCGI process managers or write a script to run your .fcgi file at boot, e.g. using a SysV init.d script. For a temporary solution, you can always run the .fcgi script inside GNU screen. See man screen for details, and note that this is a manual solution which does not persist across system restart: $ screen $ /var/www/yourapplication/yourapplication.fcgi 167
- 178. 19.4.6 Debugging FastCGI deployments tend to be hard to debug on most webservers. Very often the only thing the server log tells you is something along the lines of “premature end of headers”. In order to debug the application the only thing that can really give you ideas why it breaks is switching to the correct user and executing the application by hand. This example assumes your application is called application.fcgi and that your web- server user is www-data: $ su www-data $ cd /var/www/yourapplication $ python application.fcgi Traceback (most recent call last): File "yourapplication.fcgi", line 4, in <module> ImportError: No module named yourapplication In this case the error seems to be “yourapplication” not being on the python path. Common problems are: • Relative paths being used. Don’t rely on the current working directory • The code depending on environment variables that are not set by the web server. • Different python interpreters being used. 19.5 CGI If all other deployment methods do not work, CGI will work for sure. CGI is sup- ported by all major servers but usually has a sub-optimal performance. This is also the way you can use a Flask application on Google’s App Engine, where execution happens in a CGI-like environment. Watch Out Please make sure in advance that any app.run() calls you might have in your applica- tion file are inside an if __name__ == ’__main__’: block or moved to a separate file. Just make sure it’s not called because this will always start a local WSGI server which we do not want if we deploy that application to CGI / app engine. With CGI, you will also have to make sure that your code does not contain any print statements, or that sys.stdout is overridden by something that doesn’t write into the HTTP response. 19.5.1 Creating a .cgi file First you need to create the CGI application file. Let’s call it yourapplication.cgi: 168
- 179. #!/usr/bin/python from wsgiref.handlers import CGIHandler from yourapplication import app CGIHandler().run(app) 19.5.2 Server Setup Usually there are two ways to configure the server. Either just copy the .cgi into a cgi-bin (and use mod_rewrite or something similar to rewrite the URL) or let the server point to the file directly. In Apache for example you can put something like this into the config: ScriptAlias /app /path/to/the/application.cgi On shared webhosting, though, you might not have access to your Apache config. In this case, a file called .htaccess, sitting in the public directory you want your app to be available, works too but the ScriptAlias directive won’t work in that case: RewriteEngine On RewriteCond %{REQUEST_FILENAME} !-f # Don’t interfere with static files RewriteRule ^(.*)$ /path/to/the/application.cgi/$1 [L] For more information consult the documentation of your webserver. 169
- 180. 170
- 181. CHAPTER TWENTY BECOMING BIG Here are your options when growing your codebase or scaling your application. 20.1 Read the Source. Flask started in part to demonstrate how to build your own framework on top of ex- isting well-used tools Werkzeug (WSGI) and Jinja (templating), and as it developed, it became useful to a wide audience. As you grow your codebase, don’t just use Flask – understand it. Read the source. Flask’s code is written to be read; it’s documenta- tion published so you can use its internal APIs. Flask sticks to documented APIs in upstream libraries, and documents its internal utilities so that you can find the hook points needed for your project. 20.2 Hook. Extend. The API docs are full of available overrides, hook points, and Signals. You can provide custom classes for things like the request and response objects. Dig deeper on the APIs you use, and look for the customizations which are available out of the box in a Flask release. Look for ways in which your project can be refactored into a collection of utilities and Flask extensions. Explore the many extensions in the community, and look for patterns to build your own extensions if you do not find the tools you need. 20.3 Subclass. The Flask class has many methods designed for subclassing. You can quickly add or customize behavior by subclassing Flask (see the linked method docs) and using that subclass wherever you instantiate an application class. This works well with Applica- tion Factories. 171
- 182. 20.4 Wrap with middleware. The Application Dispatching chapter shows in detail how to apply middleware. You can introduce WSGI middleware to wrap your Flask instances and introduce fixes and changes at the layer between your Flask application and your HTTP server. Werkzeug includes several middlewares. 20.5 Fork. If none of the above options work, fork Flask. The majority of code of Flask is within Werkzeug and Jinja2. These libraries do the majority of the work. Flask is just the paste that glues those together. For every project there is the point where the underlying framework gets in the way (due to assumptions the original developers had). This is natural because if this would not be the case, the framework would be a very complex system to begin with which causes a steep learning curve and a lot of user frustration. This is not unique to Flask. Many people use patched and modified versions of their framework to counter shortcomings. This idea is also reflected in the license of Flask. You don’t have to contribute any changes back if you decide to modify the framework. The downside of forking is of course that Flask extensions will most likely break be- cause the new framework has a different import name. Furthermore integrating up- stream changes can be a complex process, depending on the number of changes. Be- cause of that, forking should be the very last resort. 20.6 Scale like a pro. For many web applications the complexity of the code is less an issue than the scaling for the number of users or data entries expected. Flask by itself is only limited in terms of scaling by your application code, the data store you want to use and the Python implementation and webserver you are running on. Scaling well means for example that if you double the amount of servers you get about twice the performance. Scaling bad means that if you add a new server the application won’t perform any better or would not even support a second server. There is only one limiting factor regarding scaling in Flask which are the context local proxies. They depend on context which in Flask is defined as being either a thread, process or greenlet. If your server uses some kind of concurrency that is not based on threads or greenlets, Flask will no longer be able to support these global proxies. However the majority of servers are using either threads, greenlets or separate pro- cesses to achieve concurrency which are all methods well supported by the underlying Werkzeug library. 172
- 183. 20.7 Discuss with the community. The Flask developers keep the framework accessible to users with codebases big and small. If you find an obstacle in your way, caused by Flask, don’t hesitate to contact the developers on the mailinglist or IRC channel. The best way for the Flask and Flask extension developers to improve the tools for larger applications is getting feedback from users. 173
- 184. 174
- 185. Part II API REFERENCE If you are looking for information on a specific function, class or method, this part of the documentation is for you. 175
- 186. 176
- 187. CHAPTER TWENTYONE API This part of the documentation covers all the interfaces of Flask. For parts where Flask depends on external libraries, we document the most important right here and provide links to the canonical documentation. 21.1 Application Object class flask.Flask(import_name, static_path=None, static_url_path=None, static_folder=’static’, template_folder=’templates’, in- stance_path=None, instance_relative_config=False) The flask object implements a WSGI application and acts as the central object. It is passed the name of the module or package of the application. Once it is created it will act as a central registry for the view functions, the URL rules, template configuration and much more. The name of the package is used to resolve resources from inside the package or the folder the module is contained in depending on if the package parameter resolves to an actual python package (a folder with an __init__.py file inside) or a standard module (just a .py file). For more information about resource loading, see open_resource(). Usually you create a Flask instance in your main module or in the __init__.py file of your package like this: from flask import Flask app = Flask(__name__) About the First Parameter The idea of the first parameter is to give Flask an idea what belongs to your application. This name is used to find resources on the file system, can be used by extensions to improve debugging information and a lot more. So it’s important what you provide there. If you are using a single module, __name__ is always the correct value. If you however are using a package, it’s usually recommended to hardcode the name of your package there. 177
- 188. For example if your application is defined in yourapplication/app.py you should create it with one of the two versions below: app = Flask(’yourapplication’) app = Flask(__name__.split(’.’)[0]) Why is that? The application will work even with __name__, thanks to how re- sources are looked up. However it will make debugging more painful. Certain extensions can make assumptions based on the import name of your application. For example the Flask-SQLAlchemy extension will look for the code in your ap- plication that triggered an SQL query in debug mode. If the import name is not properly set up, that debugging information is lost. (For example it would only pick up SQL queries in yourapplication.app and not yourapplication.views.frontend) New in version 0.7: The static_url_path, static_folder, and template_folder parame- ters were added. New in version 0.8: The instance_path and instance_relative_config parameters were added. Parameters • import_name – the name of the application package • static_url_path – can be used to specify a different path for the static files on the web. Defaults to the name of the static_folder folder. • static_folder – the folder with static files that should be served at static_url_path. Defaults to the ’static’ folder in the root path of the application. • template_folder – the folder that contains the templates that should be used by the application. Defaults to ’templates’ folder in the root path of the application. • instance_path – An alternative instance path for the applica- tion. By default the folder ’instance’ next to the package or module is assumed to be the instance path. • instance_relative_config – if set to True relative filenames for loading the config are assumed to be relative to the instance path instead of the application root. add_template_filter(*args, **kwargs) Register a custom template filter. Works exactly like the template_filter() decorator. Parameters name – the optional name of the filter, otherwise the function name will be used. add_template_global(*args, **kwargs) Register a custom template global function. Works exactly like the template_global() decorator. 178
- 189. New in version 0.10. Parameters name – the optional name of the global function, oth- erwise the function name will be used. add_template_test(*args, **kwargs) Register a custom template test. Works exactly like the template_test() decorator. New in version 0.10. Parameters name – the optional name of the test, otherwise the function name will be used. add_url_rule(*args, **kwargs) Connects a URL rule. Works exactly like the route() decorator. If a view_func is provided it will be registered with the endpoint. Basically this example: @app.route(’/’) def index(): pass Is equivalent to the following: def index(): pass app.add_url_rule(’/’, ’index’, index) If the view_func is not provided you will need to connect the endpoint to a view function like so: app.view_functions[’index’] = index Internally route() invokes add_url_rule() so if you want to customize the behavior via subclassing you only need to change this method. For more information refer to URL Route Registrations. Changed in version 0.2: view_func parameter added. Changed in version 0.6: OPTIONS is added automatically as method. Parameters • rule – the URL rule as string • endpoint – the endpoint for the registered URL rule. Flask itself assumes the name of the view function as endpoint • view_func – the function to call when serving a request to the provided endpoint • options – the options to be forwarded to the underlying Rule object. A change to Werkzeug is handling of method options. methods is a list of methods this rule should be limited to 179
- 190. (GET, POST etc.). By default a rule just listens for GET (and implicitly HEAD). Starting with Flask 0.6, OPTIONS is im- plicitly added and handled by the standard request handling. after_request(*args, **kwargs) Register a function to be run after each request. Your function must take one parameter, a response_class object and return a new response object or the same (see process_response()). As of Flask 0.7 this function might not be executed at the end of the request in case an unhandled exception occurred. after_request_funcs = None A dictionary with lists of functions that should be called after each request. The key of the dictionary is the name of the blueprint this function is active for, None for all requests. This can for example be used to open database connections or getting hold of the currently logged in user. To register a function here, use the after_request() decorator. app_context() Binds the application only. For as long as the application is bound to the current context the flask.current_app points to that application. An appli- cation context is automatically created when a request context is pushed if necessary. Example usage: with app.app_context(): ... New in version 0.9. app_ctx_globals_class The class that is used for the g instance. Example use cases for a custom class: 1.Store arbitrary attributes on flask.g. 2.Add a property for lazy per-request database connectors. 3.Return None instead of AttributeError on expected attributes. 4.Raise exception if an unexpected attr is set, a “controlled” flask.g. In Flask 0.9 this property was called request_globals_class but it was changed in 0.10 to app_ctx_globals_class because the flask.g object is not applica- tion context scoped. New in version 0.10. alias of _AppCtxGlobals auto_find_instance_path() Tries to locate the instance path if it was not provided to the constructor of 180
- 191. the application class. It will basically calculate the path to a folder named instance next to your main file or the package. New in version 0.8. before_first_request(*args, **kwargs) Registers a function to be run before the first request to this instance of the application. New in version 0.8. before_first_request_funcs = None A lists of functions that should be called at the beginning of the first request to this instance. To register a function here, use the before_first_request() decorator. New in version 0.8. before_request(*args, **kwargs) Registers a function to run before each request. before_request_funcs = None A dictionary with lists of functions that should be called at the beginning of the request. The key of the dictionary is the name of the blueprint this function is active for, None for all requests. This can for example be used to open database connections or getting hold of the currently logged in user. To register a function here, use the before_request() decorator. blueprints = None all the attached blueprints in a dictionary by name. Blueprints can be at- tached multiple times so this dictionary does not tell you how often they got attached. New in version 0.7. config = None The configuration dictionary as Config. This behaves exactly like a regular dictionary but supports additional methods to load a config from files. context_processor(*args, **kwargs) Registers a template context processor function. create_global_jinja_loader() Creates the loader for the Jinja2 environment. Can be used to override just the loader and keeping the rest unchanged. It’s discouraged to override this function. Instead one should override the jinja_loader() function instead. The global loader dispatches between the loaders of the application and the individual blueprints. New in version 0.7. create_jinja_environment() Creates the Jinja2 environment based on jinja_options and select_jinja_autoescape(). Since 0.7 this also adds the Jinja2 globals 181
- 192. and filters after initialization. Override this function to customize the behavior. New in version 0.5. create_url_adapter(request) Creates a URL adapter for the given request. The URL adapter is created at a point where the request context is not yet set up so the request is passed explicitly. New in version 0.6. Changed in version 0.9: This can now also be called without a request object when the URL adapter is created for the application context. debug The debug flag. Set this to True to enable debugging of the application. In debug mode the debugger will kick in when an unhandled exception occurs and the integrated server will automatically reload the application if changes in the code are detected. This attribute can also be configured from the config with the DEBUG con- figuration key. Defaults to False. debug_log_format = ‘——————————————————————————–n%(levelna The logging format used for the debug logger. This is only used when the application is in debug mode, otherwise the attached logging handler does the formatting. New in version 0.3. default_config = ImmutableDict({‘JSON_AS_ASCII’: True, ‘USE_X_SENDFILE’: False, ‘SE Default configuration parameters. dispatch_request() Does the request dispatching. Matches the URL and returns the return value of the view or error handler. This does not have to be a response object. In order to convert the return value to a proper response object, call make_response(). Changed in version 0.7: This no longer does the exception handling, this code was moved to the new full_dispatch_request(). do_teardown_appcontext(exc=None) Called when an application context is popped. This works pretty much the same as do_teardown_request() but for the application context. New in version 0.9. do_teardown_request(exc=None) Called after the actual request dispatching and will call every as teardown_request() decorated function. This is not actually called by the Flask object itself but is always triggered when the request context is popped. That way we have a tighter control over certain resources under testing environments. 182
- 193. Changed in version 0.9: Added the exc argument. Previously this was al- ways using the current exception information. enable_modules = True Enable the deprecated module support? This is active by default in 0.7 but will be changed to False in 0.8. With Flask 1.0 modules will be removed in favor of Blueprints endpoint(*args, **kwargs) A decorator to register a function as an endpoint. Example: @app.endpoint(’example.endpoint’) def example(): return "example" Parameters endpoint – the name of the endpoint error_handler_spec = None A dictionary of all registered error handlers. The key is None for error handlers active on the application, otherwise the key is the name of the blueprint. Each key points to another dictionary where the key is the status code of the http exception. The special key None points to a list of tuples where the first item is the class for the instance check and the second the error handler function. To register a error handler, use the errorhandler() decorator. errorhandler(*args, **kwargs) A decorator that is used to register a function give a given error code. Ex- ample: @app.errorhandler(404) def page_not_found(error): return ’This page does not exist’, 404 You can also register handlers for arbitrary exceptions: @app.errorhandler(DatabaseError) def special_exception_handler(error): return ’Database connection failed’, 500 You can also register a function as error handler without using the errorhandler() decorator. The following example is equivalent to the one above: def page_not_found(error): return ’This page does not exist’, 404 app.error_handler_spec[None][404] = page_not_found Setting error handlers via assignments to error_handler_spec however is discouraged as it requires fiddling with nested dictionaries and the special case for arbitrary exception types. 183
- 194. The first None refers to the active blueprint. If the error handler should be application wide None shall be used. New in version 0.7: One can now additionally also register custom excep- tion types that do not necessarily have to be a subclass of the HTTPException class. Parameters code – the code as integer for the handler extensions = None a place where extensions can store application specific state. For exam- ple this is where an extension could store database engines and similar things. For backwards compatibility extensions should register themselves like this: if not hasattr(app, ’extensions’): app.extensions = {} app.extensions[’extensionname’] = SomeObject() The key must match the name of the extension module. For example in case of a “Flask-Foo” extension in flask_foo, the key would be ’foo’. New in version 0.7. full_dispatch_request() Dispatches the request and on top of that performs request pre and post- processing as well as HTTP exception catching and error handling. New in version 0.7. get_send_file_max_age(filename) Provides default cache_timeout for the send_file() functions. By default, this function returns SEND_FILE_MAX_AGE_DEFAULT from the con- figuration of current_app. Static file functions such as send_from_directory() use this function, and send_file() calls this function on current_app when the given cache_timeout is None. If a cache_timeout is given in send_file(), that timeout is used; otherwise, this method is called. This allows subclasses to change the behavior when sending files based on the filename. For example, to set the cache timeout for .js files to 60 seconds: class MyFlask(flask.Flask): def get_send_file_max_age(self, name): if name.lower().endswith(’.js’): return 60 return flask.Flask.get_send_file_max_age(self, name) New in version 0.9. got_first_request This attribute is set to True if the application started handling the first re- quest. 184
- 195. New in version 0.8. handle_exception(e) Default exception handling that kicks in when an exception occurs that is not caught. In debug mode the exception will be re-raised immediately, oth- erwise it is logged and the handler for a 500 internal server error is used. If no such handler exists, a default 500 internal server error message is dis- played. New in version 0.3. handle_http_exception(e) Handles an HTTP exception. By default this will invoke the registered error handlers and fall back to returning the exception as response. New in version 0.3. handle_url_build_error(error, endpoint, values) Handle BuildError on url_for(). handle_user_exception(e) This method is called whenever an exception occurs that should be handled. A special case are HTTPExceptions which are forwarded by this function to the handle_http_exception() method. This function will either return a response value or reraise the exception with the same traceback. New in version 0.7. has_static_folder This is True if the package bound object’s container has a folder named ’static’. New in version 0.5. init_jinja_globals() Deprecated. Used to initialize the Jinja2 globals. New in version 0.5. Changed in version 0.7: This method is deprecated with 0.7. Override create_jinja_environment() instead. inject_url_defaults(endpoint, values) Injects the URL defaults for the given endpoint directly into the values dic- tionary passed. This is used internally and automatically called on URL building. New in version 0.7. instance_path = None Holds the path to the instance folder. New in version 0.8. jinja_env The Jinja2 environment used to load templates. 185
- 196. jinja_loader The Jinja loader for this package bound object. New in version 0.5. jinja_options = ImmutableDict({‘extensions’: [’jinja2.ext.autoescape’, ‘jinja2.ext.with_’]}) Options that are passed directly to the Jinja2 environment. json_decoder The JSON decoder class to use. Defaults to JSONDecoder. New in version 0.10. alias of JSONDecoder json_encoder The JSON encoder class to use. Defaults to JSONEncoder. New in version 0.10. alias of JSONEncoder log_exception(exc_info) Logs an exception. This is called by handle_exception() if debugging is disabled and right before the handler is called. The default implementation logs the exception as error on the logger. New in version 0.8. logger A logging.Logger object for this application. The default configuration is to log to stderr if the application is in debug mode. This logger can be used to (surprise) log messages. Here some examples: app.logger.debug(’A value for debugging’) app.logger.warning(’A warning occurred (%d apples)’, 42) app.logger.error(’An error occurred’) New in version 0.3. logger_name The name of the logger to use. By default the logger name is the package name passed to the constructor. New in version 0.4. make_config(instance_relative=False) Used to create the config attribute by the Flask constructor. The in- stance_relative parameter is passed in from the constructor of Flask (there named instance_relative_config) and indicates if the config should be relative to the instance path or the root path of the application. New in version 0.8. make_default_options_response() This method is called to create the default OPTIONS response. This can be 186
- 197. changed through subclassing to change the default behavior of OPTIONS responses. New in version 0.7. make_null_session() Creates a new instance of a missing session. Instead of overriding this method we recommend replacing the session_interface. New in version 0.7. make_response(rv) Converts the return value from a view function to a real response object that is an instance of response_class. The following types are allowed for rv: response_class the object is returned unchanged str a response object is created with the string as body unicode a response object is created with the string encoded to utf-8 as body a WSGI function the function is called as WSGI application and buffered as response object tuple A tuple in the form (response, status, headers) or (response, headers) where response is any of the types defined here, status is a string or an inte- ger and headers is a list or a dictionary with header values. Parameters rv – the return value from the view function Changed in version 0.9: Previously a tuple was interpreted as the arguments for the response object. name The name of the application. This is usually the import name with the dif- ference that it’s guessed from the run file if the import name is main. This name is used as a display name when Flask needs the name of the applica- tion. It can be set and overridden to change the value. New in version 0.8. open_instance_resource(resource, mode=’rb’) Opens a resource from the application’s instance folder (instance_path). Otherwise works like open_resource(). Instance resources can also be opened for writing. Parameters • resource – the name of the resource. To access resources within subfolders use forward slashes as separator. • mode – resource file opening mode, default is ‘rb’. 187
- 198. open_resource(resource, mode=’rb’) Opens a resource from the application’s resource folder. To see how this works, consider the following folder structure: /myapplication.py /schema.sql /static /style.css /templates /layout.html /index.html If you want to open the schema.sql file you would do the following: with app.open_resource(’schema.sql’) as f: contents = f.read() do_something_with(contents) Parameters • resource – the name of the resource. To access resources within subfolders use forward slashes as separator. • mode – resource file opening mode, default is ‘rb’. open_session(request) Creates or opens a new session. Default implementation stores all session data in a signed cookie. This requires that the secret_key is set. Instead of overriding this method we recommend replacing the session_interface. Parameters request – an instance of request_class. permanent_session_lifetime A timedelta which is used to set the expiration date of a permanent session. The default is 31 days which makes a permanent session survive for roughly one month. This attribute can also be configured from the config with the PERMANENT_SESSION_LIFETIME configuration key. Defaults to timedelta(days=31) preprocess_request() Called before the actual request dispatching and will call every as before_request() decorated function. If any of these function returns a value it’s handled as if it was the return value from the view and further request handling is stopped. This also triggers the url_value_processor() functions before the actual before_request() functions are called. preserve_context_on_exception Returns the value of the PRESERVE_CONTEXT_ON_EXCEPTION configu- ration value in case it’s set, otherwise a sensible default is returned. 188
- 199. New in version 0.7. process_response(response) Can be overridden in order to modify the response object before it’s sent to the WSGI server. By default this will call all the after_request() decorated functions. Changed in version 0.5: As of Flask 0.5 the functions registered for after request execution are called in reverse order of registration. Parameters response – a response_class object. Returns a new response object or the same, has to be an instance of response_class. propagate_exceptions Returns the value of the PROPAGATE_EXCEPTIONS configuration value in case it’s set, otherwise a sensible default is returned. New in version 0.7. register_blueprint(*args, **kwargs) Registers a blueprint on the application. New in version 0.7. register_error_handler(code_or_exception, f) Alternative error attach function to the errorhandler() decorator that is more straightforward to use for non decorator usage. New in version 0.7. register_module(module, **options) Registers a module with this application. The keyword argument of this function are the same as the ones for the constructor of the Module class and will override the values of the module if provided. Changed in version 0.7: The module system was deprecated in favor for the blueprint system. request_class The class that is used for request objects. See Request for more information. alias of Request request_context(environ) Creates a RequestContext from the given environment and binds it to the current context. This must be used in combination with the with statement because the request is only bound to the current context for the duration of the with block. Example usage: with app.request_context(environ): do_something_with(request) 189
- 200. The object returned can also be used without the with statement which is useful for working in the shell. The example above is doing exactly the same as this code: ctx = app.request_context(environ) ctx.push() try: do_something_with(request) finally: ctx.pop() Changed in version 0.3: Added support for non-with statement usage and with statement is now passed the ctx object. Parameters environ – a WSGI environment response_class The class that is used for response objects. See Response for more informa- tion. alias of Response route(rule, **options) A decorator that is used to register a view function for a given URL rule. This does the same thing as add_url_rule() but is intended for decorator usage: @app.route(’/’) def index(): return ’Hello World’ For more information refer to URL Route Registrations. Parameters • rule – the URL rule as string • endpoint – the endpoint for the registered URL rule. Flask itself assumes the name of the view function as endpoint • options – the options to be forwarded to the underlying Rule object. A change to Werkzeug is handling of method options. methods is a list of methods this rule should be limited to (GET, POST etc.). By default a rule just listens for GET (and implicitly HEAD). Starting with Flask 0.6, OPTIONS is im- plicitly added and handled by the standard request handling. run(host=None, port=None, debug=None, **options) Runs the application on a local development server. If the debug flag is set the server will automatically reload for code changes and show a debugger in case an exception happened. If you want to run the application in debug mode, but disable the code ex- ecution on the interactive debugger, you can pass use_evalex=False as pa- 190
- 201. rameter. This will keep the debugger’s traceback screen active, but disable code execution. Keep in Mind Flask will suppress any server error with a generic error page unless it is in debug mode. As such to enable just the interactive debugger with- out the code reloading, you have to invoke run() with debug=True and use_reloader=False. Setting use_debugger to True without being in debug mode won’t catch any exceptions because there won’t be any to catch. Changed in version 0.10: The default port is now picked from the SERVER_NAME variable. Parameters • host – the hostname to listen on. Set this to ’0.0.0.0’ to have the server available externally as well. Defaults to ’127.0.0.1’. • port – the port of the webserver. Defaults to 5000 or the port defined in the SERVER_NAME config variable if present. • debug – if given, enable or disable debug mode. See debug. • options – the options to be forwarded to the underlying Werkzeug server. See werkzeug.serving.run_simple() for more information. save_session(session, response) Saves the session if it needs updates. For the default implementation, check open_session(). Instead of overriding this method we recommend replac- ing the session_interface. Parameters • session – the session to be saved (a SecureCookie object) • response – an instance of response_class secret_key If a secret key is set, cryptographic components can use this to sign cookies and other things. Set this to a complex random value when you want to use the secure cookie for instance. This attribute can also be configured from the config with the SECRET_KEY configuration key. Defaults to None. select_jinja_autoescape(filename) Returns True if autoescaping should be active for the given template name. New in version 0.5. send_static_file(filename) Function used internally to send static files from the static folder to the 191
- 202. browser. New in version 0.5. session_cookie_name The secure cookie uses this for the name of the session cookie. This attribute can also be configured from the config with the SES- SION_COOKIE_NAME configuration key. Defaults to ’session’ session_interface = <flask.sessions.SecureCookieSessionInterface object at 0x10ff89f50> the session interface to use. By default an instance of SecureCookieSessionInterface is used here. New in version 0.8. should_ignore_error(error) This is called to figure out if an error should be ignored or not as far as the teardown system is concerned. If this function returns True then the teardown handlers will not be passed the error. New in version 0.10. teardown_appcontext(*args, **kwargs) Registers a function to be called when the application context ends. These functions are typically also called when the request context is popped. Example: ctx = app.app_context() ctx.push() ... ctx.pop() When ctx.pop() is executed in the above example, the teardown functions are called just before the app context moves from the stack of active con- texts. This becomes relevant if you are using such constructs in tests. Since a request context typically also manages an application context it would also be called when you pop a request context. When a teardown function was called because of an exception it will be passed an error object. New in version 0.9. teardown_appcontext_funcs = None A list of functions that are called when the application context is destroyed. Since the application context is also torn down if the request ends this is the place to store code that disconnects from databases. New in version 0.9. teardown_request(*args, **kwargs) Register a function to be run at the end of each request, regardless of 192
- 203. whether there was an exception or not. These functions are executed when the request context is popped, even if not an actual request was performed. Example: ctx = app.test_request_context() ctx.push() ... ctx.pop() When ctx.pop() is executed in the above example, the teardown functions are called just before the request context moves from the stack of active con- texts. This becomes relevant if you are using such constructs in tests. Generally teardown functions must take every necessary step to avoid that they will fail. If they do execute code that might fail they will have to sur- round the execution of these code by try/except statements and log occur- ring errors. When a teardown function was called because of a exception it will be passed an error object. Debug Note In debug mode Flask will not tear down a request on an exception immediately. Instead if will keep it alive so that the interactive de- bugger can still access it. This behavior can be controlled by the PRESERVE_CONTEXT_ON_EXCEPTION configuration variable. teardown_request_funcs = None A dictionary with lists of functions that are called after each request, even if an exception has occurred. The key of the dictionary is the name of the blueprint this function is active for, None for all requests. These func- tions are not allowed to modify the request, and their return values are ig- nored. If an exception occurred while processing the request, it gets passed to each teardown_request function. To register a function here, use the teardown_request() decorator. New in version 0.7. template_context_processors = None A dictionary with list of functions that are called without argument to pop- ulate the template context. The key of the dictionary is the name of the blueprint this function is active for, None for all requests. Each returns a dictionary that the template context is updated with. To register a function here, use the context_processor() decorator. template_filter(*args, **kwargs) A decorator that is used to register custom template filter. You can specify a name for the filter, otherwise the function name will be used. Example: 193
- 204. @app.template_filter() def reverse(s): return s[::-1] Parameters name – the optional name of the filter, otherwise the function name will be used. template_global(*args, **kwargs) A decorator that is used to register a custom template global function. You can specify a name for the global function, otherwise the function name will be used. Example: @app.template_global() def double(n): return 2 * n New in version 0.10. Parameters name – the optional name of the global function, oth- erwise the function name will be used. template_test(*args, **kwargs) A decorator that is used to register custom template test. You can specify a name for the test, otherwise the function name will be used. Example: @app.template_test() def is_prime(n): if n == 2: return True for i in range(2, int(math.ceil(math.sqrt(n))) + 1): if n % i == 0: return False return True New in version 0.10. Parameters name – the optional name of the test, otherwise the function name will be used. test_client(use_cookies=True) Creates a test client for this application. For information about unit testing head over to Testing Flask Applications. Note that if you are testing for assertions or exceptions in your application code, you must set app.testing = True in order for the exceptions to prop- agate to the test client. Otherwise, the exception will be handled by the application (not visible to the test client) and the only indication of an As- sertionError or other exception will be a 500 status code response to the test client. See the testing attribute. For example: app.testing = True client = app.test_client() 194
- 205. The test client can be used in a with block to defer the closing down of the context until the end of the with block. This is useful if you want to access the context locals for testing: with app.test_client() as c: rv = c.get(’/?vodka=42’) assert request.args[’vodka’] == ’42’ See FlaskClient for more information. Changed in version 0.4: added support for with block usage for the client. New in version 0.7: The use_cookies parameter was added as well as the ability to override the client to be used by setting the test_client_class attribute. test_client_class = None the test client that is used with when test_client is used. New in version 0.7. test_request_context(*args, **kwargs) Creates a WSGI environment from the given values (see werkzeug.test.EnvironBuilder for more information, this function ac- cepts the same arguments). testing The testing flag. Set this to True to enable the test mode of Flask extensions (and in the future probably also Flask itself). For example this might activate unittest helpers that have an additional runtime cost which should not be enabled by default. If this is enabled and PROPAGATE_EXCEPTIONS is not changed from the default it’s implicitly enabled. This attribute can also be configured from the config with the TESTING con- figuration key. Defaults to False. trap_http_exception(e) Checks if an HTTP exception should be trapped or not. By default this will return False for all exceptions except for a bad request key er- ror if TRAP_BAD_REQUEST_ERRORS is set to True. It also returns True if TRAP_HTTP_EXCEPTIONS is set to True. This is called for all HTTP exceptions raised by a view function. If it returns True for any exception the error handler for this exception is not called and it shows up as regular exception in the traceback. This is helpful for debug- ging implicitly raised HTTP exceptions. New in version 0.8. update_template_context(context) Update the template context with some commonly used variables. This in- jects request, session, config and g into the template context as well as every- thing template context processors want to inject. Note that the as of Flask 195
- 206. 0.6, the original values in the context will not be overridden if a context processor decides to return a value with the same key. Parameters context – the context as a dictionary that is updated in place to add extra variables. url_build_error_handlers = None A list of functions that are called when url_for() raises a BuildError. Each function registered here is called with error, endpoint and values. If a function returns None or raises a BuildError the next function is tried. New in version 0.9. url_default_functions = None A dictionary with lists of functions that can be used as URL value prepro- cessors. The key None here is used for application wide callbacks, otherwise the key is the name of the blueprint. Each of these functions has the chance to modify the dictionary of URL values before they are used as the key- word arguments of the view function. For each function registered this one should also provide a url_defaults() function that adds the parameters automatically again that were removed that way. New in version 0.7. url_defaults(*args, **kwargs) Callback function for URL defaults for all view functions of the applica- tion. It’s called with the endpoint and values and should update the values passed in place. url_map = None The Map for this instance. You can use this to change the routing converters after the class was created but before any routes are connected. Example: from werkzeug.routing import BaseConverter class ListConverter(BaseConverter): def to_python(self, value): return value.split(’,’) def to_url(self, values): return ’,’.join(BaseConverter.to_url(value) for value in values) app = Flask(__name__) app.url_map.converters[’list’] = ListConverter url_rule_class The rule object to use for URL rules created. This is used by add_url_rule(). Defaults to werkzeug.routing.Rule. New in version 0.7. alias of Rule url_value_preprocessor(*args, **kwargs) 196
- 207. Registers a function as URL value preprocessor for all view functions of the application. It’s called before the view functions are called and can modify the url values provided. url_value_preprocessors = None A dictionary with lists of functions that can be used as URL value processor functions. Whenever a URL is built these functions are called to modify the dictionary of values in place. The key None here is used for application wide callbacks, otherwise the key is the name of the blueprint. Each of these functions has the chance to modify the dictionary New in version 0.7. use_x_sendfile Enable this if you want to use the X-Sendfile feature. Keep in mind that the server has to support this. This only affects files sent with the send_file() method. New in version 0.2. This attribute can also be configured from the config with the USE_X_SENDFILE configuration key. Defaults to False. view_functions = None A dictionary of all view functions registered. The keys will be function names which are also used to generate URLs and the values are the function objects themselves. To register a view function, use the route() decorator. wsgi_app(environ, start_response) The actual WSGI application. This is not implemented in __call__ so that middlewares can be applied without losing a reference to the class. So in- stead of doing this: app = MyMiddleware(app) It’s a better idea to do this instead: app.wsgi_app = MyMiddleware(app.wsgi_app) Then you still have the original application object around and can continue to call methods on it. Changed in version 0.7: The behavior of the before and after request call- backs was changed under error conditions and a new callback was added that will always execute at the end of the request, independent on if an error occurred or not. See Callbacks and Errors. Parameters • environ – a WSGI environment • start_response – a callable accepting a status code, a list of headers and an optional exception context to start the re- sponse 197
- 208. 21.2 Blueprint Objects class flask.Blueprint(name, import_name, static_folder=None, static_url_path=None, template_folder=None, url_prefix=None, subdomain=None, url_defaults=None) Represents a blueprint. A blueprint is an object that records functions that will be called with the BlueprintSetupState later to register functions or other things on the main application. See Modular Applications with Blueprints for more infor- mation. New in version 0.7. add_app_template_filter(f, name=None) Register a custom template filter, available application wide. Like Flask.add_template_filter() but for a blueprint. Works exactly like the app_template_filter() decorator. Parameters name – the optional name of the filter, otherwise the function name will be used. add_app_template_global(f, name=None) Register a custom template global, available application wide. Like Flask.add_template_global() but for a blueprint. Works exactly like the app_template_global() decorator. New in version 0.10. Parameters name – the optional name of the global, otherwise the function name will be used. add_app_template_test(f, name=None) Register a custom template test, available application wide. Like Flask.add_template_test() but for a blueprint. Works exactly like the app_template_test() decorator. New in version 0.10. Parameters name – the optional name of the test, otherwise the function name will be used. add_url_rule(rule, endpoint=None, view_func=None, **options) Like Flask.add_url_rule() but for a blueprint. The endpoint for the url_for() function is prefixed with the name of the blueprint. after_app_request(f) Like Flask.after_request() but for a blueprint. Such a function is executed after each request, even if outside of the blueprint. after_request(f) Like Flask.after_request() but for a blueprint. This function is only exe- cuted after each request that is handled by a function of that blueprint. app_context_processor(f) 198
- 209. Like Flask.context_processor() but for a blueprint. Such a function is ex- ecuted each request, even if outside of the blueprint. app_errorhandler(code) Like Flask.errorhandler() but for a blueprint. This handler is used for all requests, even if outside of the blueprint. app_template_filter(name=None) Register a custom template filter, available application wide. Like Flask.template_filter() but for a blueprint. Parameters name – the optional name of the filter, otherwise the function name will be used. app_template_global(name=None) Register a custom template global, available application wide. Like Flask.template_global() but for a blueprint. New in version 0.10. Parameters name – the optional name of the global, otherwise the function name will be used. app_template_test(name=None) Register a custom template test, available application wide. Like Flask.template_test() but for a blueprint. New in version 0.10. Parameters name – the optional name of the test, otherwise the function name will be used. app_url_defaults(f) Same as url_defaults() but application wide. app_url_value_preprocessor(f) Same as url_value_preprocessor() but application wide. before_app_first_request(f) Like Flask.before_first_request(). Such a function is executed before the first request to the application. before_app_request(f) Like Flask.before_request(). Such a function is executed before each re- quest, even if outside of a blueprint. before_request(f) Like Flask.before_request() but for a blueprint. This function is only exe- cuted before each request that is handled by a function of that blueprint. context_processor(f) Like Flask.context_processor() but for a blueprint. This function is only executed for requests handled by a blueprint. endpoint(endpoint) Like Flask.endpoint() but for a blueprint. This does not prefix the end- 199
- 210. point with the blueprint name, this has to be done explicitly by the user of this method. If the endpoint is prefixed with a . it will be registered to the current blueprint, otherwise it’s an application independent endpoint. errorhandler(code_or_exception) Registers an error handler that becomes active for this blueprint only. Please be aware that routing does not happen local to a blueprint so an error han- dler for 404 usually is not handled by a blueprint unless it is caused inside a view function. Another special case is the 500 internal server error which is always looked up from the application. Otherwise works as the errorhandler() decorator of the Flask object. get_send_file_max_age(filename) Provides default cache_timeout for the send_file() functions. By default, this function returns SEND_FILE_MAX_AGE_DEFAULT from the con- figuration of current_app. Static file functions such as send_from_directory() use this function, and send_file() calls this function on current_app when the given cache_timeout is None. If a cache_timeout is given in send_file(), that timeout is used; otherwise, this method is called. This allows subclasses to change the behavior when sending files based on the filename. For example, to set the cache timeout for .js files to 60 seconds: class MyFlask(flask.Flask): def get_send_file_max_age(self, name): if name.lower().endswith(’.js’): return 60 return flask.Flask.get_send_file_max_age(self, name) New in version 0.9. has_static_folder This is True if the package bound object’s container has a folder named ’static’. New in version 0.5. jinja_loader The Jinja loader for this package bound object. New in version 0.5. make_setup_state(app, options, first_registration=False) Creates an instance of BlueprintSetupState() object that is later passed to the register callback functions. Subclasses can override this to return a sub- class of the setup state. open_resource(resource, mode=’rb’) Opens a resource from the application’s resource folder. To see how this works, consider the following folder structure: 200
- 211. /myapplication.py /schema.sql /static /style.css /templates /layout.html /index.html If you want to open the schema.sql file you would do the following: with app.open_resource(’schema.sql’) as f: contents = f.read() do_something_with(contents) Parameters • resource – the name of the resource. To access resources within subfolders use forward slashes as separator. • mode – resource file opening mode, default is ‘rb’. record(func) Registers a function that is called when the blueprint is registered on the application. This function is called with the state as argument as returned by the make_setup_state() method. record_once(func) Works like record() but wraps the function in another function that will ensure the function is only called once. If the blueprint is registered a second time on the application, the function passed is not called. register(app, options, first_registration=False) Called by Flask.register_blueprint() to register a blueprint on the ap- plication. This can be overridden to customize the register behavior. Key- word arguments from register_blueprint() are directly forwarded to this method in the options dictionary. route(rule, **options) Like Flask.route() but for a blueprint. The endpoint for the url_for() function is prefixed with the name of the blueprint. send_static_file(filename) Function used internally to send static files from the static folder to the browser. New in version 0.5. teardown_app_request(f) Like Flask.teardown_request() but for a blueprint. Such a function is exe- cuted when tearing down each request, even if outside of the blueprint. teardown_request(f) Like Flask.teardown_request() but for a blueprint. This function is 201
- 212. only executed when tearing down requests handled by a function of that blueprint. Teardown request functions are executed when the request con- text is popped, even when no actual request was performed. url_defaults(f) Callback function for URL defaults for this blueprint. It’s called with the endpoint and values and should update the values passed in place. url_value_preprocessor(f) Registers a function as URL value preprocessor for this blueprint. It’s called before the view functions are called and can modify the url values provided. 21.3 Incoming Request Data class flask.Request(environ, populate_request=True, shallow=False) The request object used by default in Flask. Remembers the matched endpoint and view arguments. It is what ends up as request. If you want to replace the request object used you can subclass this and set request_class to your subclass. The request object is a Request subclass and provides all of the attributes Werkzeug defines plus a few Flask specific ones. form A MultiDict with the parsed form data from POST or PUT requests. Please keep in mind that file uploads will not end up here, but instead in the files attribute. args A MultiDict with the parsed contents of the query string. (The part in the URL after the question mark). values A CombinedMultiDict with the contents of both form and args. cookies A dict with the contents of all cookies transmitted with the request. stream If the incoming form data was not encoded with a known mimetype the data is stored unmodified in this stream for consumption. Most of the time it is a better idea to use data which will give you that data as a string. The stream only returns the data once. headers The incoming request headers as a dictionary like object. data Contains the incoming request data as string in case it came with a mime- type Flask does not handle. 202
- 213. files A MultiDict with files uploaded as part of a POST or PUT request. Each file is stored as FileStorage object. It basically behaves like a standard file object you know from Python, with the difference that it also has a save() function that can store the file on the filesystem. environ The underlying WSGI environment. method The current request method (POST, GET etc.) path script_root url base_url url_root Provides different ways to look at the current URL. Imagine your applica- tion is listening on the following URL: http://www.example.com/myapplication And a user requests the following URL: http://www.example.com/myapplication/page.html?x=y In this case the values of the above mentioned attributes would be the fol- lowing: path /page.html script_root /myapplication base_url http://www.example.com/myapplication/page.html url http://www.example.com/myapplication/page.html?x=y url_root http://www.example.com/myapplication/ is_xhr True if the request was triggered via a JavaScript XMLHttpRequest. This only works with libraries that support the X-Requested-With header and set it to XMLHttpRequest. Libraries that do that are prototype, jQuery and Mochikit and probably some more. blueprint The name of the current blueprint endpoint The endpoint that matched the request. This in combination with view_args can be used to reconstruct the same or a modified URL. If an exception happened when matching, this will be None. get_json(force=False, silent=False, cache=True) Parses the incoming JSON request data and returns it. If parsing fails the 203
- 214. on_json_loading_failed() method on the request object will be invoked. By default this function will only load the json data if the mimetype is application/json but this can be overriden by the force parameter. Parameters • force – if set to True the mimetype is ignored. • silent – if set to True this method will fail silently and return None. • cache – if set to True the parsed JSON data is remembered on the request. is_json Indicates if this request is JSON or not. By default a request is con- sidered to include JSON data if the mimetype is application/json or application/*+json. New in version 0.11. json If the mimetype is application/json this will contain the parsed JSON data. Otherwise this will be None. The get_json() method should be used instead. max_content_length Read-only view of the MAX_CONTENT_LENGTH config key. module The name of the current module if the request was dispatched to an actual module. This is deprecated functionality, use blueprints instead. on_json_loading_failed(e) Called if decoding of the JSON data failed. The return value of this method is used by get_json() when an error occurred. The default implementation just raises a BadRequest exception. Changed in version 0.10: Removed buggy previous behavior of generating a random JSON response. If you want that behavior back you can trivially add it by subclassing. New in version 0.8. routing_exception = None if matching the URL failed, this is the exception that will be raised / was raised as part of the request handling. This is usually a NotFound exception or something similar. url_rule = None the internal URL rule that matched the request. This can be useful to in- spect which methods are allowed for the URL from a before/after handler (request.url_rule.methods) etc. New in version 0.6. 204
- 215. view_args = None a dict of view arguments that matched the request. If an exception hap- pened when matching, this will be None. class flask.request To access incoming request data, you can use the global request object. Flask parses incoming request data for you and gives you access to it through that global object. Internally Flask makes sure that you always get the correct data for the active thread if you are in a multithreaded environment. This is a proxy. See Notes On Proxies for more information. The request object is an instance of a Request subclass and provides all of the attributes Werkzeug defines. This just shows a quick overview of the most im- portant ones. 21.4 Response Objects class flask.Response(response=None, status=None, headers=None, mime- type=None, content_type=None, direct_passthrough=False) The response object that is used by default in Flask. Works like the response object from Werkzeug but is set to have an HTML mimetype by default. Quite often you don’t have to create this object yourself because make_response() will take care of that for you. If you want to replace the response object used you can subclass this and set response_class to your subclass. headers A Headers object representing the response headers. status A string with a response status. status_code The response status as integer. data A descriptor that calls get_data() and set_data(). This should not be used and will eventually get deprecated. mimetype The mimetype (content type without charset etc.) set_cookie(key, value=’‘, max_age=None, expires=None, path=’/’, do- main=None, secure=None, httponly=False) Sets a cookie. The parameters are the same as in the cookie Morsel object in the Python standard library but it accepts unicode data, too. Parameters • key – the key (name) of the cookie to be set. 205
- 216. • value – the value of the cookie. • max_age – should be a number of seconds, or None (default) if the cookie should last only as long as the client’s browser session. • expires – should be a datetime object or UNIX timestamp. • domain – if you want to set a cross-domain cookie. For exam- ple, domain=".example.com" will set a cookie that is readable by the domain www.example.com, foo.example.com etc. Oth- erwise, a cookie will only be readable by the domain that set it. • path – limits the cookie to a given path, per default it will span the whole domain. 21.5 Sessions If you have the Flask.secret_key set you can use sessions in Flask applications. A session basically makes it possible to remember information from one request to an- other. The way Flask does this is by using a signed cookie. So the user can look at the session contents, but not modify it unless they know the secret key, so make sure to set that to something complex and unguessable. To access the current session you can use the session object: class flask.session The session object works pretty much like an ordinary dict, with the difference that it keeps track on modifications. This is a proxy. See Notes On Proxies for more information. The following attributes are interesting: new True if the session is new, False otherwise. modified True if the session object detected a modification. Be advised that modifica- tions on mutable structures are not picked up automatically, in that situation you have to explicitly set the attribute to True yourself. Here an example: # this change is not picked up because a mutable object (here # a list) is changed. session[’objects’].append(42) # so mark it as modified yourself session.modified = True permanent If set to True the session lives for permanent_session_lifetime seconds. The 206
- 217. default is 31 days. If set to False (which is the default) the session will be deleted when the user closes the browser. 21.6 Session Interface New in version 0.8. The session interface provides a simple way to replace the session implementation that Flask is using. class flask.sessions.SessionInterface The basic interface you have to implement in order to replace the default session interface which uses werkzeug’s securecookie implementation. The only meth- ods you have to implement are open_session() and save_session(), the others have useful defaults which you don’t need to change. The session object returned by the open_session() method has to provide a dic- tionary like interface plus the properties and methods from the SessionMixin. We recommend just subclassing a dict and adding that mixin: class Session(dict, SessionMixin): pass If open_session() returns None Flask will call into make_null_session() to create a session that acts as replacement if the session support cannot work because some requirement is not fulfilled. The default NullSession class that is created will complain that the secret key was not set. To replace the session interface on an application all you have to do is to assign flask.Flask.session_interface: app = Flask(__name__) app.session_interface = MySessionInterface() New in version 0.8. get_cookie_domain(app) Helpful helper method that returns the cookie domain that should be used for the session cookie if session cookies are used. get_cookie_httponly(app) Returns True if the session cookie should be httponly. This currently just returns the value of the SESSION_COOKIE_HTTPONLY config var. get_cookie_path(app) Returns the path for which the cookie should be valid. The default imple- mentation uses the value from the SESSION_COOKIE_PATH‘‘ config var if it’s set, and falls back to APPLICATION_ROOT or uses / if it’s None. get_cookie_secure(app) Returns True if the cookie should be secure. This currently just returns the value of the SESSION_COOKIE_SECURE setting. 207
- 218. get_expiration_time(app, session) A helper method that returns an expiration date for the session or None if the session is linked to the browser session. The default implementation returns now + the permanent session lifetime configured on the application. is_null_session(obj) Checks if a given object is a null session. Null sessions are not asked to be saved. This checks if the object is an instance of null_session_class by default. make_null_session(app) Creates a null session which acts as a replacement object if the real session support could not be loaded due to a configuration error. This mainly aids the user experience because the job of the null session is to still support lookup without complaining but modifications are answered with a helpful error message of what failed. This creates an instance of null_session_class by default. null_session_class make_null_session() will look here for the class that should be created when a null session is requested. Likewise the is_null_session() method will perform a typecheck against this type. alias of NullSession open_session(app, request) This method has to be implemented and must either return None in case the loading failed because of a configuration error or an instance of a ses- sion object which implements a dictionary like interface + the methods and attributes on SessionMixin. pickle_based = False A flag that indicates if the session interface is pickle based. This can be used by flask extensions to make a decision in regards to how to deal with the session object. New in version 0.10. save_session(app, session, response) This is called for actual sessions returned by open_session() at the end of the request. This is still called during a request context so if you absolutely need access to the request you can do that. should_set_cookie(app, session) Indicates weather a cookie should be set now or not. This is used by session backends to figure out if they should emit a set-cookie header or not. The default behavior is controlled by the SESSION_REFRESH_EACH_REQUEST config variable. If it’s set to False then a cookie is only set if the session is modified, if set to True it’s always set if the session is permanent. This check is usually skipped if sessions get deleted. 208
- 219. New in version 1.0. class flask.sessions.SecureCookieSessionInterface The default session interface that stores sessions in signed cookies through the itsdangerous module. digest_method() the hash function to use for the signature. The default is sha1 key_derivation = ‘hmac’ the name of the itsdangerous supported key derivation. The default is hmac. salt = ‘cookie-session’ the salt that should be applied on top of the secret key for the signing of cookie based sessions. serializer = <flask.sessions.TaggedJSONSerializer object at 0x10ff895d0> A python serializer for the payload. The default is a compact JSON derived serializer with support for some extra Python types such as datetime objects or tuples. session_class alias of SecureCookieSession class flask.sessions.SecureCookieSession(initial=None) Baseclass for sessions based on signed cookies. class flask.sessions.NullSession(initial=None) Class used to generate nicer error messages if sessions are not available. Will still allow read-only access to the empty session but fail on setting. class flask.sessions.SessionMixin Expands a basic dictionary with an accessors that are expected by Flask exten- sions and users for the session. modified = True for some backends this will always be True, but some backends will default this to false and detect changes in the dictionary for as long as changes do not happen on mutable structures in the session. The default mixin imple- mentation just hardcodes True in. new = False some session backends can tell you if a session is new, but that is not nec- essarily guaranteed. Use with caution. The default mixin implementation just hardcodes False in. permanent this reflects the ’_permanent’ key in the dict. flask.sessions.session_json_serializer = <flask.sessions.TaggedJSONSerializer object at 0x10 A customized JSON serializer that supports a few extra types that we take for granted when serializing (tuples, markup objects, datetime). 209
- 220. This object provides dumping and loading methods similar to simplejson but it also tags certain builtin Python objects that commonly appear in sessions. Cur- rently the following extended values are supported in the JSON it dumps: •Markup objects •UUID objects •datetime objects •tuples Notice The PERMANENT_SESSION_LIFETIME config key can also be an integer starting with Flask 0.8. Either catch this down yourself or use the permanent_session_lifetime attribute on the app which converts the result to an integer automatically. 21.7 Test Client class flask.testing.FlaskClient(application, response_wrapper=None, use_cookies=True, al- low_subdomain_redirects=False) Works like a regular Werkzeug test client but has some knowledge about how Flask works to defer the cleanup of the request context stack to the end of a with body when used in a with statement. For general information about how to use this class refer to werkzeug.test.Client. Basic usage is outlined in the Testing Flask Applications chapter. session_transaction(*args, **kwds) When used in combination with a with statement this opens a session trans- action. This can be used to modify the session that the test client uses. Once the with block is left the session is stored back. with client.session_transaction() as session: session[’value’] = 42 Internally this is implemented by going through a temporary test request context and since session handling could depend on request variables this function accepts the same arguments as test_request_context() which are directly passed through. 21.8 Application Globals To share data that is valid for one request only from one function to another, a global variable is not good enough because it would break in threaded environments. Flask provides you with a special object that ensures it is only valid for the active request and that will return different values for each request. In a nutshell: it does the right thing, like it does for request and session. 210
- 221. flask.g Just store on this whatever you want. For example a database connection or the user that is currently logged in. Starting with Flask 0.10 this is stored on the application context and no longer on the request context which means it becomes available if only the application context is bound and not yet a request. This is especially useful when combined with the Faking Resources and Context pattern for testing. Additionally as of 0.10 you can use the get() method to get an attribute or None (or the second argument) if it’s not set. These two usages are now equivalent: user = getattr(flask.g, ’user’, None) user = flask.g.get(’user’, None) It’s now also possible to use the in operator on it to see if an attribute is defined and it yields all keys on iteration. This is a proxy. See Notes On Proxies for more information. 21.9 Useful Functions and Classes flask.current_app Points to the application handling the request. This is useful for extensions that want to support multiple applications running side by side. This is powered by the application context and not by the request context, so you can change the value of this proxy by using the app_context() method. This is a proxy. See Notes On Proxies for more information. flask.has_request_context() If you have code that wants to test if a request context is there or not this function can be used. For instance, you may want to take advantage of request informa- tion if the request object is available, but fail silently if it is unavailable. class User(db.Model): def __init__(self, username, remote_addr=None): self.username = username if remote_addr is None and has_request_context(): remote_addr = request.remote_addr self.remote_addr = remote_addr Alternatively you can also just test any of the context bound objects (such as request or g for truthness): class User(db.Model): def __init__(self, username, remote_addr=None): self.username = username if remote_addr is None and request: 211
- 222. remote_addr = request.remote_addr self.remote_addr = remote_addr New in version 0.7. flask.copy_current_request_context(f) A helper function that decorates a function to retain the current request context. This is useful when working with greenlets. The moment the function is deco- rated a copy of the request context is created and then pushed when the function is called. Example: import gevent from flask import copy_current_request_context @app.route(’/’) def index(): @copy_current_request_context def do_some_work(): # do some work here, it can access flask.request like you # would otherwise in the view function. ... gevent.spawn(do_some_work) return ’Regular response’ New in version 0.10. flask.has_app_context() Works like has_request_context() but for the application context. You can also just do a boolean check on the current_app object instead. New in version 0.9. flask.url_for(endpoint, **values) Generates a URL to the given endpoint with the method provided. Variable arguments that are unknown to the target endpoint are appended to the generated URL as query arguments. If the value of a query argument is None, the whole pair is skipped. In case blueprints are active you can shortcut references to the same blueprint by prefixing the local endpoint with a dot (.). This will reference the index function local to the current blueprint: url_for(’.index’) For more information, head over to the Quickstart. To integrate applications, Flask has a hook to intercept URL build errors through Flask.build_error_handler. The url_for function results in a BuildError when the current app does not have a URL for the given endpoint and values. When it does, the current_app calls its build_error_handler if it is not None, which can return a string to use as the result of url_for (instead of url_for‘s default to raise the BuildError exception) or re-raise the exception. An example: 212
- 223. def external_url_handler(error, endpoint, **values): "Looks up an external URL when ‘url_for‘ cannot build a URL." # This is an example of hooking the build_error_handler. # Here, lookup_url is some utility function you’ve built # which looks up the endpoint in some external URL registry. url = lookup_url(endpoint, **values) if url is None: # External lookup did not have a URL. # Re-raise the BuildError, in context of original traceback. exc_type, exc_value, tb = sys.exc_info() if exc_value is error: raise exc_type, exc_value, tb else: raise error # url_for will use this result, instead of raising BuildError. return url app.build_error_handler = external_url_handler Here, error is the instance of BuildError, and endpoint and **values are the argu- ments passed into url_for. Note that this is for building URLs outside the current application, and not for handling 404 NotFound errors. New in version 0.10: The _scheme parameter was added. New in version 0.9: The _anchor and _method parameters were added. New in version 0.9: Calls Flask.handle_build_error() on BuildError. Parameters • endpoint – the endpoint of the URL (name of the function) • values – the variable arguments of the URL rule • _external – if set to True, an absolute URL is generated. Server address can be changed via SERVER_NAME configuration variable which defaults to localhost. • _scheme – a string specifying the desired URL scheme. The _external parameter must be set to True or a ValueError is raised. • _anchor – if provided this is added as anchor to the URL. • _method – if provided this explicitly specifies an HTTP method. flask.abort(code) Raises an HTTPException for the given status code. For example to abort request handling with a page not found exception, you would call abort(404). Parameters code – the HTTP error code. flask.redirect(location, code=302) Return a response object (a WSGI application) that, if called, redirects the client 213
- 224. to the target location. Supported codes are 301, 302, 303, 305, and 307. 300 is not supported because it’s not a real redirect and 304 because it’s the answer for a request with a request with defined If-Modified-Since headers. New in version 0.6: The location can now be a unicode string that is encoded using the iri_to_uri() function. Parameters • location – the location the response should redirect to. • code – the redirect status code. defaults to 302. flask.make_response(*args) Sometimes it is necessary to set additional headers in a view. Because views do not have to return response objects but can return a value that is converted into a response object by Flask itself, it becomes tricky to add headers to it. This function can be called instead of using a return and you will get a response object which you can use to attach headers. If view looked like this and you want to add a new header: def index(): return render_template(’index.html’, foo=42) You can now do something like this: def index(): response = make_response(render_template(’index.html’, foo=42)) response.headers[’X-Parachutes’] = ’parachutes are cool’ return response This function accepts the very same arguments you can return from a view func- tion. This for example creates a response with a 404 error code: response = make_response(render_template(’not_found.html’), 404) The other use case of this function is to force the return value of a view function into a response which is helpful with view decorators: response = make_response(view_function()) response.headers[’X-Parachutes’] = ’parachutes are cool’ Internally this function does the following things: •if no arguments are passed, it creates a new response argument •if one argument is passed, flask.Flask.make_response() is invoked with it. •if more than one argument is passed, the arguments are passed to the flask.Flask.make_response() function as tuple. New in version 0.6. 214
- 225. flask.after_this_request(f) Executes a function after this request. This is useful to modify response objects. The function is passed the response object and has to return the same or a new one. Example: @app.route(’/’) def index(): @after_this_request def add_header(response): response.headers[’X-Foo’] = ’Parachute’ return response return ’Hello World!’ This is more useful if a function other than the view function wants to modify a response. For instance think of a decorator that wants to add some headers without converting the return value into a response object. New in version 0.9. flask.send_file(filename_or_fp, mimetype=None, as_attachment=False, attach- ment_filename=None, add_etags=True, cache_timeout=None, con- ditional=False) Sends the contents of a file to the client. This will use the most efficient method available and configured. By default it will try to use the WSGI server’s file_wrapper support. Alternatively you can set the application’s use_x_sendfile attribute to True to directly emit an X-Sendfile header. This how- ever requires support of the underlying webserver for X-Sendfile. By default it will try to guess the mimetype for you, but you can also explicitly provide one. For extra security you probably want to send certain files as at- tachment (HTML for instance). The mimetype guessing requires a filename or an attachment_filename to be provided. Please never pass filenames to this function from user sources without checking them first. Something like this is usually sufficient to avoid security problems: if ’..’ in filename or filename.startswith(’/’): abort(404) New in version 0.2. New in version 0.5: The add_etags, cache_timeout and conditional parameters were added. The default behavior is now to attach etags. Changed in version 0.7: mimetype guessing and etag support for file objects was deprecated because it was unreliable. Pass a filename if you are able to, otherwise attach an etag yourself. This functionality will be removed in Flask 1.0 Changed in version 0.9: cache_timeout pulls its default from application config, when None. Parameters 215
- 226. • filename_or_fp – the filename of the file to send. This is rel- ative to the root_path if a relative path is specified. Alterna- tively a file object might be provided in which case X-Sendfile might not work and fall back to the traditional method. Make sure that the file pointer is positioned at the start of data to send before calling send_file(). • mimetype – the mimetype of the file if provided, otherwise auto detection happens. • as_attachment – set to True if you want to send this file with a Content-Disposition: attachment header. • attachment_filename – the filename for the attachment if it dif- fers from the file’s filename. • add_etags – set to False to disable attaching of etags. • conditional – set to True to enable conditional responses. • cache_timeout – the timeout in seconds for the headers. When None (default), this value is set by get_send_file_max_age() of current_app. flask.send_from_directory(directory, filename, **options) Send a file from a given directory with send_file(). This is a secure way to quickly expose static files from an upload folder or something similar. Example usage: @app.route(’/uploads/<path:filename>’) def download_file(filename): return send_from_directory(app.config[’UPLOAD_FOLDER’], filename, as_attachment=True) Sending files and Performance It is strongly recommended to activate either X-Sendfile support in your web- server or (if no authentication happens) to tell the webserver to serve files for the given path on its own without calling into the web application for improved performance. New in version 0.5. Parameters • directory – the directory where all the files are stored. • filename – the filename relative to that directory to download. • options – optional keyword arguments that are directly for- warded to send_file(). flask.safe_join(directory, filename) Safely join directory and filename. 216
- 227. Example usage: @app.route(’/wiki/<path:filename>’) def wiki_page(filename): filename = safe_join(app.config[’WIKI_FOLDER’], filename) with open(filename, ’rb’) as fd: content = fd.read() # Read and process the file content... Parameters • directory – the base directory. • filename – the untrusted filename relative to that directory. Raises NotFound if the resulting path would fall out of directory. flask.escape(s) Convert the characters &, <, >, ‘ and ” in string s to HTML-safe sequences. Use this if you need to display text that might contain such characters in HTML. Marks return value as markup string. class flask.Markup Marks a string as being safe for inclusion in HTML/XML output without need- ing to be escaped. This implements the __html__ interface a couple of frame- works and web applications use. Markup is a direct subclass of unicode and pro- vides all the methods of unicode just that it escapes arguments passed and always returns Markup. The escape function returns markup objects so that double escaping can’t happen. The constructor of the Markup class can be used for three different things: When passed an unicode object it’s assumed to be safe, when passed an object with an HTML representation (has an __html__ method) that representation is used, otherwise the object passed is converted into a unicode string and then assumed to be safe: >>> Markup("Hello <em>World</em>!") Markup(u’Hello <em>World</em>!’) >>> class Foo(object): ... def __html__(self): ... return ’<a href="#">foo</a>’ ... >>> Markup(Foo()) Markup(u’<a href="#">foo</a>’) If you want object passed being always treated as unsafe you can use the escape() classmethod to create a Markup object: >>> Markup.escape("Hello <em>World</em>!") Markup(u’Hello <em>World</em>!’) Operations on a markup string are markup aware which means that all argu- ments are passed through the escape() function: 217
- 228. >>> em = Markup("<em>%s</em>") >>> em % "foo & bar" Markup(u’<em>foo & bar</em>’) >>> strong = Markup("<strong>%(text)s</strong>") >>> strong % {’text’: ’<blink>hacker here</blink>’} Markup(u’<strong><blink>hacker here</blink></strong>’) >>> Markup("<em>Hello</em> ") + "<foo>" Markup(u’<em>Hello</em> <foo>’) classmethod escape(s) Escape the string. Works like escape() with the difference that for sub- classes of Markup this function would return the correct subclass. striptags() Unescape markup into an unicode string and strip all tags. This also re- solves known HTML4 and XHTML entities. Whitespace is normalized to one: >>> Markup("Main » <em>About</em>").striptags() u’Main xbb About’ unescape() Unescape markup again into an unicode string. This also resolves known HTML4 and XHTML entities: >>> Markup("Main » <em>About</em>").unescape() u’Main xbb <em>About</em>’ 21.10 Message Flashing flask.flash(message, category=’message’) Flashes a message to the next request. In order to remove the flashed mes- sage from the session and to display it to the user, the template has to call get_flashed_messages(). Changed in version 0.3: category parameter added. Parameters • message – the message to be flashed. • category – the category for the message. The following values are recommended: ’message’ for any kind of message, ’error’ for errors, ’info’ for information messages and ’warning’ for warnings. However any kind of string can be used as category. flask.get_flashed_messages(with_categories=False, category_filter=[]) Pulls all flashed messages from the session and returns them. Further calls in the same request to the function will return the same messages. By default just the messages are returned, but when with_categories is set to True, the return value will be a list of tuples in the form (category, message) instead. 218
- 229. Filter the flashed messages to one or more categories by providing those cate- gories in category_filter. This allows rendering categories in separate html blocks. The with_categories and category_filter arguments are distinct: •with_categories controls whether categories are returned with message text (True gives a tuple, where False gives just the message text). •category_filter filters the messages down to only those matching the pro- vided categories. See Message Flashing for examples. Changed in version 0.3: with_categories parameter added. Changed in version 0.9: category_filter parameter added. Parameters • with_categories – set to True to also receive categories. • category_filter – whitelist of categories to limit return values 21.11 JSON Support Flask uses simplejson for the JSON implementation. Since simplejson is provided both by the standard library as well as extension Flask will try simplejson first and then fall back to the stdlib json module. On top of that it will delegate access to the current application’s JSON encoders and decoders for easier customization. So for starters instead of doing: try: import simplejson as json except ImportError: import json You can instead just do this: from flask import json For usage examples, read the json documentation in the standard library. The follow- ing extensions are by default applied to the stdlib’s JSON module: 1. datetime objects are serialized as RFC 822 strings. 2. Any object with an __html__ method (like Markup) will have that method called and then the return value is serialized as string. The htmlsafe_dumps() function of this json module is also available as filter called |tojson in Jinja2. Note that inside script tags no escaping must take place, so make sure to disable escaping with |safe if you intend to use it inside script tags unless you are using Flask 0.10 which implies that: 219
- 230. <script type=text/javascript> doSomethingWith({{ user.username|tojson|safe }}); </script> Auto-Sort JSON Keys The configuration variable JSON_SORT_KEYS (Configuration Handling) can be set to false to stop Flask from auto-sorting keys. By default sorting is enabled and outside of the app context sorting is turned on. Notice that disabling key sorting can cause issues when using content based HTTP caches and Python’s hash randomization feature. flask.json.jsonify(*args, **kwargs) Creates a Response with the JSON representation of the given arguments with an application/json mimetype. The arguments to this function are the same as to the dict constructor. Example usage: from flask import jsonify @app.route(’/_get_current_user’) def get_current_user(): return jsonify(username=g.user.username, email=g.user.email, id=g.user.id) This will send a JSON response like this to the browser: { "username": "admin", "email": "admin@localhost", "id": 42 } For security reasons only objects are supported toplevel. For more information about this, have a look at JSON Security. This function’s response will be pretty printed if it was not requested with X-Requested-With: XMLHttpRequest to simplify debugging unless the JSONIFY_PRETTYPRINT_REGULAR config parameter is set to false. New in version 0.2. flask.json.dumps(obj, **kwargs) Serialize obj to a JSON formatted str by using the application’s configured en- coder (json_encoder) if there is an application on the stack. This function can return unicode strings or ascii-only bytestrings by default which coerce into unicode strings automatically. That behavior by default is con- trolled by the JSON_AS_ASCII configuration variable and can be overriden by the simplejson ensure_ascii parameter. 220
- 231. flask.json.dump(obj, fp, **kwargs) Like dumps() but writes into a file object. flask.json.loads(s, **kwargs) Unserialize a JSON object from a string s by using the application’s configured decoder (json_decoder) if there is an application on the stack. flask.json.load(fp, **kwargs) Like loads() but reads from a file object. class flask.json.JSONEncoder(skipkeys=False, ensure_ascii=True, check_circular=True, allow_nan=True, sort_keys=False, indent=None, separators=None, encoding=’utf-8’, default=None) The default Flask JSON encoder. This one extends the default simplejson encoder by also supporting datetime objects, UUID as well as Markup objects which are serialized as RFC 822 datetime strings (same as the HTTP date format). In order to support more data types override the default() method. default(o) Implement this method in a subclass such that it returns a serializable object for o, or calls the base implementation (to raise a TypeError). For example, to support arbitrary iterators, you could implement default like this: def default(self, o): try: iterable = iter(o) except TypeError: pass else: return list(iterable) return JSONEncoder.default(self, o) class flask.json.JSONDecoder(encoding=None, object_hook=None, parse_float=None, parse_int=None, parse_constant=None, strict=True, ob- ject_pairs_hook=None) The default JSON decoder. This one does not change the behavior from the de- fault simplejson encoder. Consult the json documentation for more informa- tion. This decoder is not only used for the load functions of this module but also Request. 21.12 Template Rendering flask.render_template(template_name_or_list, **context) Renders a template from the template folder with the given context. Parameters 221
- 232. • template_name_or_list – the name of the template to be ren- dered, or an iterable with template names the first one existing will be rendered • context – the variables that should be available in the context of the template. flask.render_template_string(source, **context) Renders a template from the given template source string with the given context. Parameters • source – the sourcecode of the template to be rendered • context – the variables that should be available in the context of the template. flask.get_template_attribute(template_name, attribute) Loads a macro (or variable) a template exports. This can be used to invoke a macro from within Python code. If you for example have a template named _cider.html with the following contents: {% macro hello(name) %}Hello {{ name }}!{% endmacro %} You can access this from Python code like this: hello = get_template_attribute(’_cider.html’, ’hello’) return hello(’World’) New in version 0.2. Parameters • template_name – the name of the template • attribute – the name of the variable of macro to access 21.13 Configuration class flask.Config(root_path, defaults=None) Works exactly like a dict but provides ways to fill it from files or special dictio- naries. There are two common patterns to populate the config. Either you can fill the config from a config file: app.config.from_pyfile(’yourconfig.cfg’) Or alternatively you can define the configuration options in the module that calls from_object() or provide an import path to a module that should be loaded. It is also possible to tell it to use the same module and with that provide the configuration values just before the call: 222
- 233. DEBUG = True SECRET_KEY = ’development key’ app.config.from_object(__name__) In both cases (loading from any Python file or loading from modules), only up- percase keys are added to the config. This makes it possible to use lowercase values in the config file for temporary values that are not added to the config or to define the config keys in the same file that implements the application. Probably the most interesting way to load configurations is from an environment variable pointing to a file: app.config.from_envvar(’YOURAPPLICATION_SETTINGS’) In this case before launching the application you have to set this environment variable to the file you want to use. On Linux and OS X use the export statement: export YOURAPPLICATION_SETTINGS=’/path/to/config/file’ On windows use set instead. Parameters • root_path – path to which files are read relative from. When the config object is created by the application, this is the appli- cation’s root_path. • defaults – an optional dictionary of default values from_envvar(variable_name, silent=False) Loads a configuration from an environment variable pointing to a configu- ration file. This is basically just a shortcut with nicer error messages for this line of code: app.config.from_pyfile(os.environ[’YOURAPPLICATION_SETTINGS’]) Parameters • variable_name – name of the environment variable • silent – set to True if you want silent failure for missing files. Returns bool. True if able to load config, False otherwise. from_object(obj) Updates the values from the given object. An object can be of one of the following two types: •a string: in this case the object with that name will be imported •an actual object reference: that object is used directly Objects are usually either modules or classes. 223
- 234. Just the uppercase variables in that object are stored in the config. Example usage: app.config.from_object(’yourapplication.default_config’) from yourapplication import default_config app.config.from_object(default_config) You should not use this function to load the actual configuration but rather configuration defaults. The actual config should be loaded with from_pyfile() and ideally from a location not within the package because the package might be installed system wide. Parameters obj – an import name or object from_pyfile(filename, silent=False) Updates the values in the config from a Python file. This function behaves as if the file was imported as module with the from_object() function. Parameters • filename – the filename of the config. This can either be an absolute filename or a filename relative to the root path. • silent – set to True if you want silent failure for missing files. New in version 0.7: silent parameter. 21.14 Extensions flask.ext This module acts as redirect import module to Flask extensions. It was added in 0.8 as the canonical way to import Flask extensions and makes it possible for us to have more flexibility in how we distribute extensions. If you want to use an extension named “Flask-Foo” you would import it from ext as follows: from flask.ext import foo New in version 0.8. 21.15 Stream Helpers flask.stream_with_context(generator_or_function) Request contexts disappear when the response is started on the server. This is done for efficiency reasons and to make it less likely to encounter memory leaks with badly written WSGI middlewares. The downside is that if you are using streamed responses, the generator cannot access request bound information any more. 224
- 235. This function however can help you keep the context around for longer: from flask import stream_with_context, request, Response @app.route(’/stream’) def streamed_response(): @stream_with_context def generate(): yield ’Hello ’ yield request.args[’name’] yield ’!’ return Response(generate()) Alternatively it can also be used around a specific generator: from flask import stream_with_context, request, Response @app.route(’/stream’) def streamed_response(): def generate(): yield ’Hello ’ yield request.args[’name’] yield ’!’ return Response(stream_with_context(generate())) New in version 0.9. 21.16 Useful Internals class flask.ctx.RequestContext(app, environ, request=None) The request context contains all request relevant information. It is created at the beginning of the request and pushed to the _request_ctx_stack and removed at the end of it. It will create the URL adapter and request object for the WSGI environment provided. Do not attempt to use this class directly, instead use test_request_context() and request_context() to create this object. When the request context is popped, it will evaluate all the functions registered on the application for teardown execution (teardown_request()). The request context is automatically popped at the end of the request for you. In debug mode the request context is kept around if exceptions happen so that interactive debuggers have a chance to introspect the data. With 0.4 this can also be forced for requests that did not fail and outside of DEBUG mode. By setting ’flask._preserve_context’ to True on the WSGI environment the context will not pop itself at the end of the request. This is used by the test_client() for example to implement the deferred cleanup functionality. You might find this helpful for unittests where you need the information from the context local around for a little longer. Make sure to properly pop() the stack 225
- 236. yourself in that situation, otherwise your unittests will leak memory. copy() Creates a copy of this request context with the same request object. This can be used to move a request context to a different greenlet. Because the actual request object is the same this cannot be used to move a request context to a different thread unless access to the request object is locked. New in version 0.10. match_request() Can be overridden by a subclass to hook into the matching of the request. pop(exc=None) Pops the request context and unbinds it by doing that. This will also trigger the execution of functions registered by the teardown_request() decorator. Changed in version 0.9: Added the exc argument. push() Binds the request context to the current context. flask._request_ctx_stack The internal LocalStack that is used to implement all the context local objects used in Flask. This is a documented instance and can be used by extensions and application code but the use is discouraged in general. The following attributes are always present on each layer of the stack: app the active Flask application. url_adapter the URL adapter that was used to match the request. request the current request object. session the active session object. g an object with all the attributes of the flask.g object. flashes an internal cache for the flashed messages. Example usage: from flask import _request_ctx_stack def get_session(): ctx = _request_ctx_stack.top if ctx is not None: return ctx.session class flask.ctx.AppContext(app) The application context binds an application object implicitly to the current thread or greenlet, similar to how the RequestContext binds request informa- tion. The application context is also implicitly created if a request context is created but the application is not on top of the individual application context. 226
- 237. pop(exc=None) Pops the app context. push() Binds the app context to the current context. flask._app_ctx_stack Works similar to the request context but only binds the application. This is mainly there for extensions to store data. New in version 0.9. class flask.blueprints.BlueprintSetupState(blueprint, app, options, first_registration) Temporary holder object for registering a blueprint with the application. An in- stance of this class is created by the make_setup_state() method and later passed to all register callback functions. add_url_rule(rule, endpoint=None, view_func=None, **options) A helper method to register a rule (and optionally a view function) to the application. The endpoint is automatically prefixed with the blueprint’s name. app = None a reference to the current application blueprint = None a reference to the blueprint that created this setup state. first_registration = None as blueprints can be registered multiple times with the application and not everything wants to be registered multiple times on it, this attribute can be used to figure out if the blueprint was registered in the past already. options = None a dictionary with all options that were passed to the register_blueprint() method. subdomain = None The subdomain that the blueprint should be active for, None otherwise. url_defaults = None A dictionary with URL defaults that is added to each and every URL that was defined with the blueprint. url_prefix = None The prefix that should be used for all URLs defined on the blueprint. 21.17 Signals New in version 0.6. 227
- 238. flask.signals_available True if the signaling system is available. This is the case when blinker is installed. flask.template_rendered This signal is sent when a template was successfully rendered. The signal is in- voked with the instance of the template as template and the context as dictionary (named context). flask.request_started This signal is sent before any request processing started but when the request context was set up. Because the request context is already bound, the subscriber can access the request with the standard global proxies such as request. flask.request_finished This signal is sent right before the response is sent to the client. It is passed the response to be sent named response. flask.got_request_exception This signal is sent when an exception happens during request processing. It is sent before the standard exception handling kicks in and even in debug mode, where no exception handling happens. The exception itself is passed to the sub- scriber as exception. flask.request_tearing_down This signal is sent when the application is tearing down the request. This is always called, even if an error happened. An exc keyword argument is passed with the exception that caused the teardown. Changed in version 0.9: The exc parameter was added. flask.appcontext_tearing_down This signal is sent when the application is tearing down the application context. This is always called, even if an error happened. An exc keyword argument is passed with the exception that caused the teardown. The sender is the applica- tion. flask.appcontext_pushed This signal is sent when an application context is pushed. The sender is the application. New in version 0.10. flask.appcontext_popped This signal is sent when an application context is popped. The sender is the application. This usually falls in line with the appcontext_tearing_down signal. New in version 0.10. flask.message_flashed This signal is sent when the application is flashing a message. The messages is sent as message keyword argument and the category as category. New in version 0.10. 228
- 239. class flask.signals.Namespace An alias for blinker.base.Namespace if blinker is available, otherwise a dummy class that creates fake signals. This class is available for Flask extensions that want to provide the same fallback system as Flask itself. signal(name, doc=None) Creates a new signal for this namespace if blinker is available, otherwise returns a fake signal that has a send method that will do nothing but will fail with a RuntimeError for all other operations, including connecting. 21.18 Class-Based Views New in version 0.7. class flask.views.View Alternative way to use view functions. A subclass has to implement dispatch_request() which is called with the view arguments from the URL rout- ing system. If methods is provided the methods do not have to be passed to the add_url_rule() method explicitly: class MyView(View): methods = [’GET’] def dispatch_request(self, name): return ’Hello %s!’ % name app.add_url_rule(’/hello/<name>’, view_func=MyView.as_view(’myview’)) When you want to decorate a pluggable view you will have to either do that when the view function is created (by wrapping the return value of as_view()) or you can use the decorators attribute: class SecretView(View): methods = [’GET’] decorators = [superuser_required] def dispatch_request(self): ... The decorators stored in the decorators list are applied one after another when the view function is created. Note that you can not use the class based decorators since those would decorate the view class and not the generated view function! classmethod as_view(name, *class_args, **class_kwargs) Converts the class into an actual view function that can be used with the routing system. Internally this generates a function on the fly which will in- stantiate the View on each request and call the dispatch_request() method on it. The arguments passed to as_view() are forwarded to the constructor of the class. 229
- 240. decorators = [] The canonical way to decorate class-based views is to decorate the return value of as_view(). However since this moves parts of the logic from the class declaration to the place where it’s hooked into the routing system. You can place one or more decorators in this list and whenever the view function is created the result is automatically decorated. New in version 0.8. dispatch_request() Subclasses have to override this method to implement the actual view func- tion code. This method is called with all the arguments from the URL rule. methods = None A for which methods this pluggable view can handle. class flask.views.MethodView Like a regular class-based view but that dispatches requests to particular meth- ods. For instance if you implement a method called get() it means you will response to ’GET’ requests and the dispatch_request() implementation will au- tomatically forward your request to that. Also options is set for you automati- cally: class CounterAPI(MethodView): def get(self): return session.get(’counter’, 0) def post(self): session[’counter’] = session.get(’counter’, 0) + 1 return ’OK’ app.add_url_rule(’/counter’, view_func=CounterAPI.as_view(’counter’)) 21.19 URL Route Registrations Generally there are three ways to define rules for the routing system: 1. You can use the flask.Flask.route() decorator. 2. You can use the flask.Flask.add_url_rule() function. 3. You can directly access the underlying Werkzeug routing system which is ex- posed as flask.Flask.url_map. Variable parts in the route can be specified with angular brackets (/user/<username>). By default a variable part in the URL accepts any string without a slash however a different converter can be specified as well by using <converter:name>. Variable parts are passed to the view function as keyword arguments. The following converters are available: 230
- 241. string accepts any text without a slash (the default) int accepts integers float like int but for floating point values path like the default but also accepts slashes Here are some examples: @app.route(’/’) def index(): pass @app.route(’/<username>’) def show_user(username): pass @app.route(’/post/<int:post_id>’) def show_post(post_id): pass An important detail to keep in mind is how Flask deals with trailing slashes. The idea is to keep each URL unique so the following rules apply: 1. If a rule ends with a slash and is requested without a slash by the user, the user is automatically redirected to the same page with a trailing slash attached. 2. If a rule does not end with a trailing slash and the user requests the page with a trailing slash, a 404 not found is raised. This is consistent with how web servers deal with static files. This also makes it possi- ble to use relative link targets safely. You can also define multiple rules for the same function. They have to be unique however. Defaults can also be specified. Here for example is a definition for a URL that accepts an optional page: @app.route(’/users/’, defaults={’page’: 1}) @app.route(’/users/page/<int:page>’) def show_users(page): pass This specifies that /users/ will be the URL for page one and /users/page/N will be the URL for page N. Here are the parameters that route() and add_url_rule() accept. The only difference is that with the route parameter the view function is defined with the decorator instead of the view_func parameter. 231
- 242. rule the URL rule as string end- point the endpoint for the registered URL rule. Flask itself assumes that the name of the view function is the name of the endpoint if not explicitly stated. view_functhe function to call when serving a request to the provided endpoint. If this is not provided one can specify the function later by storing it in the view_functions dictionary with the endpoint as key. de- faults A dictionary with defaults for this rule. See the example above for how defaults work. sub- do- main specifies the rule for the subdomain in case subdomain matching is in use. If not specified the default subdomain is assumed. **op- tions the options to be forwarded to the underlying Rule object. A change to Werkzeug is handling of method options. methods is a list of methods this rule should be limited to (GET, POST etc.). By default a rule just listens for GET (and implicitly HEAD). Starting with Flask 0.6, OPTIONS is implicitly added and handled by the standard request handling. They have to be specified as keyword arguments. 21.20 View Function Options For internal usage the view functions can have some attributes attached to customize behavior the view function would normally not have control over. The following at- tributes can be provided optionally to either override some defaults to add_url_rule() or general behavior: • __name__: The name of a function is by default used as endpoint. If endpoint is provided explicitly this value is used. Additionally this will be prefixed with the name of the blueprint by default which cannot be customized from the function itself. • methods: If methods are not provided when the URL rule is added, Flask will look on the view function object itself is an methods attribute exists. If it does, it will pull the information for the methods from there. • provide_automatic_options: if this attribute is set Flask will either force enable or disable the automatic implementation of the HTTP OPTIONS response. This can be useful when working with decorators that want to customize the OPTIONS response on a per-view basis. • required_methods: if this attribute is set, Flask will always add these methods when registering a URL rule even if the methods were explicitly overridden in the route() call. Full example: def index(): if request.method == ’OPTIONS’: # custom options handling here ... 232
- 243. return ’Hello World!’ index.provide_automatic_options = False index.methods = [’GET’, ’OPTIONS’] app.add_url_rule(’/’, index) New in version 0.8: The provide_automatic_options functionality was added. 233
- 244. 234
- 245. Part III ADDITIONAL NOTES Design notes, legal information and changelog are here for the interested. 235
- 246. 236
- 247. CHAPTER TWENTYTWO DESIGN DECISIONS IN FLASK If you are curious why Flask does certain things the way it does and not differently, this section is for you. This should give you an idea about some of the design decisions that may appear arbitrary and surprising at first, especially in direct comparison with other frameworks. 22.1 The Explicit Application Object A Python web application based on WSGI has to have one central callable object that implements the actual application. In Flask this is an instance of the Flask class. Each Flask application has to create an instance of this class itself and pass it the name of the module, but why can’t Flask do that itself? Without such an explicit application object the following code: from flask import Flask app = Flask(__name__) @app.route(’/’) def index(): return ’Hello World!’ Would look like this instead: from hypothetical_flask import route @route(’/’) def index(): return ’Hello World!’ There are three major reasons for this. The most important one is that implicit appli- cation objects require that there may only be one instance at the time. There are ways to fake multiple applications with a single application object, like maintaining a stack of applications, but this causes some problems I won’t outline here in detail. Now the question is: when does a microframework need more than one application at the same time? A good example for this is unittesting. When you want to test something 237
- 248. it can be very helpful to create a minimal application to test specific behavior. When the application object is deleted everything it allocated will be freed again. Another thing that becomes possible when you have an explicit object lying around in your code is that you can subclass the base class (Flask) to alter specific behavior. This would not be possible without hacks if the object were created ahead of time for you based on a class that is not exposed to you. But there is another very important reason why Flask depends on an explicit instantia- tion of that class: the package name. Whenever you create a Flask instance you usually pass it __name__ as package name. Flask depends on that information to properly load resources relative to your module. With Python’s outstanding support for reflection it can then access the package to figure out where the templates and static files are stored (see open_resource()). Now obviously there are frameworks around that do not need any configuration and will still be able to load templates relative to your application module. But they have to use the current working directory for that, which is a very unreliable way to determine where the application is. The current working directory is process-wide and if you are running multiple applications in one process (which could happen in a webserver without you knowing) the paths will be off. Worse: many web- servers do not set the working directory to the directory of your application but to the document root which does not have to be the same folder. The third reason is “explicit is better than implicit”. That object is your WSGI ap- plication, you don’t have to remember anything else. If you want to apply a WSGI middleware, just wrap it and you’re done (though there are better ways to do that so that you do not lose the reference to the application object wsgi_app()). Furthermore this design makes it possible to use a factory function to create the appli- cation which is very helpful for unittesting and similar things (Application Factories). 22.2 The Routing System Flask uses the Werkzeug routing system which was designed to automatically order routes by complexity. This means that you can declare routes in arbitrary order and they will still work as expected. This is a requirement if you want to properly imple- ment decorator based routing since decorators could be fired in undefined order when the application is split into multiple modules. Another design decision with the Werkzeug routing system is that routes in Werkzeug try to ensure that URLs are unique. Werkzeug will go quite far with that in that it will automatically redirect to a canonical URL if a route is ambiguous. 22.3 One Template Engine Flask decides on one template engine: Jinja2. Why doesn’t Flask have a pluggable tem- plate engine interface? You can obviously use a different template engine, but Flask will still configure Jinja2 for you. While that limitation that Jinja2 is always configured 238
- 249. will probably go away, the decision to bundle one template engine and use that will not. Template engines are like programming languages and each of those engines has a certain understanding about how things work. On the surface they all work the same: you tell the engine to evaluate a template with a set of variables and take the return value as string. But that’s about where similarities end. Jinja2 for example has an extensive filter sys- tem, a certain way to do template inheritance, support for reusable blocks (macros) that can be used from inside templates and also from Python code, uses Unicode for all operations, supports iterative template rendering, configurable syntax and more. On the other hand an engine like Genshi is based on XML stream evaluation, template inheritance by taking the availability of XPath into account and more. Mako on the other hand treats templates similar to Python modules. When it comes to connecting a template engine with an application or framework there is more than just rendering templates. For instance, Flask uses Jinja2’s extensive autoescaping support. Also it provides ways to access macros from Jinja2 templates. A template abstraction layer that would not take the unique features of the template engines away is a science on its own and a too large undertaking for a microframework like Flask. Furthermore extensions can then easily depend on one template language being present. You can easily use your own templating language, but an extension could still depend on Jinja itself. 22.4 Micro with Dependencies Why does Flask call itself a microframework and yet it depends on two libraries (namely Werkzeug and Jinja2). Why shouldn’t it? If we look over to the Ruby side of web development there we have a protocol very similar to WSGI. Just that it’s called Rack there, but besides that it looks very much like a WSGI rendition for Ruby. But nearly all applications in Ruby land do not work with Rack directly, but on top of a library with the same name. This Rack library has two equivalents in Python: WebOb (formerly Paste) and Werkzeug. Paste is still around but from my understanding it’s sort of deprecated in favour of WebOb. The development of WebOb and Werkzeug started side by side with similar ideas in mind: be a good implementation of WSGI for other applications to take advantage. Flask is a framework that takes advantage of the work already done by Werkzeug to properly interface WSGI (which can be a complex task at times). Thanks to recent developments in the Python package infrastructure, packages with dependencies are no longer an issue and there are very few reasons against having libraries that depend on others. 239
- 250. 22.5 Thread Locals Flask uses thread local objects (context local objects in fact, they support greenlet con- texts as well) for request, session and an extra object you can put your own things on (g). Why is that and isn’t that a bad idea? Yes it is usually not such a bright idea to use thread locals. They cause troubles for servers that are not based on the concept of threads and make large applications harder to maintain. However Flask is just not designed for large applications or asynchronous servers. Flask wants to make it quick and easy to write a traditional web application. Also see the Becoming Big section of the documentation for some inspiration for larger applications based on Flask. 22.6 What Flask is, What Flask is Not Flask will never have a database layer. It will not have a form library or anything else in that direction. Flask itself just bridges to Werkzeug to implement a proper WSGI application and to Jinja2 to handle templating. It also binds to a few common standard library packages such as logging. Everything else is up for extensions. Why is this the case? Because people have different preferences and requirements and Flask could not meet those if it would force any of this into the core. The majority of web applications will need a template engine in some sort. However not every application needs a SQL database. The idea of Flask is to build a good foundation for all applications. Everything else is up to you or extensions. 240
- 251. CHAPTER TWENTYTHREE HTML/XHTML FAQ The Flask documentation and example applications are using HTML5. You may no- tice that in many situations, when end tags are optional they are not used, so that the HTML is cleaner and faster to load. Because there is much confusion about HTML and XHTML among developers, this document tries to answer some of the major ques- tions. 23.1 History of XHTML For a while, it appeared that HTML was about to be replaced by XHTML. However, barely any websites on the Internet are actual XHTML (which is HTML processed us- ing XML rules). There are a couple of major reasons why this is the case. One of them is Internet Explorer’s lack of proper XHTML support. The XHTML spec states that XHTML must be served with the MIME type application/xhtml+xml, but Internet Explorer refuses to read files with that MIME type. While it is relatively easy to con- figure Web servers to serve XHTML properly, few people do. This is likely because properly using XHTML can be quite painful. One of the most important causes of pain is XML’s draconian (strict and ruthless) er- ror handling. When an XML parsing error is encountered, the browser is supposed to show the user an ugly error message, instead of attempting to recover from the error and display what it can. Most of the (X)HTML generation on the web is based on non-XML template engines (such as Jinja, the one used in Flask) which do not protect you from accidentally creating invalid XHTML. There are XML based template en- gines, such as Kid and the popular Genshi, but they often come with a larger runtime overhead and, are not as straightforward to use because they have to obey XML rules. The majority of users, however, assumed they were properly using XHTML. They wrote an XHTML doctype at the top of the document and self-closed all the necessary tags (<br> becomes <br/> or <br></br> in XHTML). However, even if the document properly validates as XHTML, what really determines XHTML/HTML processing in browsers is the MIME type, which as said before is often not set properly. So the valid XHTML was being treated as invalid HTML. XHTML also changed the way JavaScript is used. To properly work with XHTML, pro- grammers have to use the namespaced DOM interface with the XHTML namespace 241
- 252. to query for HTML elements. 23.2 History of HTML5 Development of the HTML5 specification was started in 2004 under the name “Web Applications 1.0” by the Web Hypertext Application Technology Working Group, or WHATWG (which was formed by the major browser vendors Apple, Mozilla, and Opera) with the goal of writing a new and improved HTML specification, based on existing browser behavior instead of unrealistic and backwards-incompatible specifi- cations. For example, in HTML4 <title/Hello/ theoretically parses exactly the same as <title>Hello</title>. However, since people were using XHTML-like tags along the lines of <link />, browser vendors implemented the XHTML syntax over the syntax defined by the specification. In 2007, the specification was adopted as the basis of a new HTML specification under the umbrella of the W3C, known as HTML5. Currently, it appears that XHTML is losing traction, as the XHTML 2 working group has been disbanded and HTML5 is being implemented by all major browser vendors. 23.3 HTML versus XHTML The following table gives you a quick overview of features available in HTML 4.01, XHTML 1.1 and HTML5. (XHTML 1.0 is not included, as it was superseded by XHTML 1.1 and the barely-used XHTML5.) HTML4.01 XHTML1.1 HTML5 <tag/value/ == <tag>value</tag> 1 <br/> supported 2 <script/> supported should be served as text/html 3 should be served as application/xhtml+xml strict error handling inline SVG inline MathML <video> tag <audio> tag New semantic tags like <article> 1This is an obscure feature inherited from SGML. It is usually not supported by browsers, for reasons detailed above. 242
- 253. 23.4 What does “strict” mean? HTML5 has strictly defined parsing rules, but it also specifies exactly how a browser should react to parsing errors - unlike XHTML, which simply states parsing should abort. Some people are confused by apparently invalid syntax that still generates the expected results (for example, missing end tags or unquoted attribute values). Some of these work because of the lenient error handling most browsers use when they encounter a markup error, others are actually specified. The following constructs are optional in HTML5 by standard, but have to be supported by browsers: • Wrapping the document in an <html> tag • Wrapping header elements in <head> or the body elements in <body> • Closing the <p>, <li>, <dt>, <dd>, <tr>, <td>, <th>, <tbody>, <thead>, or <tfoot> tags. • Quoting attributes, so long as they contain no whitespace or special characters (like <, >, ’, or "). • Requiring boolean attributes to have a value. This means the following page in HTML5 is perfectly valid: <!doctype html> <title>Hello HTML5</title> <div class=header> <h1>Hello HTML5</h1> <p class=tagline>HTML5 is awesome </div> <ul class=nav> <li><a href=/https/www.slideshare.net/index>Index</a> <li><a href=/https/www.slideshare.net/downloads>Downloads</a> <li><a href=/https/www.slideshare.net/about>About</a> </ul> <div class=body> <h2>HTML5 is probably the future</h2> <p> There might be some other things around but in terms of browser vendor support, HTML5 is hard to beat. <dl> <dt>Key 1 <dd>Value 1 <dt>Key 2 <dd>Value 2 </dl> </div> 2This is for compatibility with server code that generates XHTML for tags such as <br>. It should not be used in new code. 3XHTML 1.0 is the last XHTML standard that allows to be served as text/html for backwards com- patibility reasons. 243
- 254. 23.5 New technologies in HTML5 HTML5 adds many new features that make Web applications easier to write and to use. • The <audio> and <video> tags provide a way to embed audio and video without complicated add-ons like QuickTime or Flash. • Semantic elements like <article>, <header>, <nav>, and <time> that make con- tent easier to understand. • The <canvas> tag, which supports a powerful drawing API, reducing the need for server-generated images to present data graphically. • New form control types like <input type="date"> that allow user agents to make entering and validating values easier. • Advanced JavaScript APIs like Web Storage, Web Workers, Web Sockets, geolo- cation, and offline applications. Many other features have been added, as well. A good guide to new features in HTML5 is Mark Pilgrim’s soon-to-be-published book, Dive Into HTML5. Not all of them are supported in browsers yet, however, so use caution. 23.6 What should be used? Currently, the answer is HTML5. There are very few reasons to use XHTML consider- ing the latest developments in Web browsers. To summarize the reasons given above: • Internet Explorer (which, sadly, currently leads in market share) has poor sup- port for XHTML. • Many JavaScript libraries also do not support XHTML, due to the more compli- cated namespacing API it requires. • HTML5 adds several new features, including semantic tags and the long-awaited <audio> and <video> tags. • It has the support of most browser vendors behind it. • It is much easier to write, and more compact. For most applications, it is undoubtedly better to use HTML5 than XHTML. 244
- 255. CHAPTER TWENTYFOUR SECURITY CONSIDERATIONS Web applications usually face all kinds of security problems and it’s very hard to get everything right. Flask tries to solve a few of these things for you, but there are a couple more you have to take care of yourself. 24.1 Cross-Site Scripting (XSS) Cross site scripting is the concept of injecting arbitrary HTML (and with it JavaScript) into the context of a website. To remedy this, developers have to properly escape text so that it cannot include arbitrary HTML tags. For more information on that have a look at the Wikipedia article on Cross-Site Scripting. Flask configures Jinja2 to automatically escape all values unless explicitly told other- wise. This should rule out all XSS problems caused in templates, but there are still other places where you have to be careful: • generating HTML without the help of Jinja2 • calling Markup on data submitted by users • sending out HTML from uploaded files, never do that, use the Content- Disposition: attachment header to prevent that problem. • sending out textfiles from uploaded files. Some browsers are using content-type guessing based on the first few bytes so users could trick a browser to execute HTML. Another thing that is very important are unquoted attributes. While Jinja2 can protect you from XSS issues by escaping HTML, there is one thing it cannot protect you from: XSS by attribute injection. To counter this possible attack vector, be sure to always quote your attributes with either double or single quotes when using Jinja expressions in them: <a href="{{ href }}">the text</a> Why is this necessary? Because if you would not be doing that, an attacker could easily inject custom JavaScript handlers. For example an attacker could inject this piece of HTML+JavaScript: 245
- 256. onmouseover=alert(document.cookie) When the user would then move with the mouse over the link, the cookie would be presented to the user in an alert window. But instead of showing the cookie to the user, a good attacker might also execute any other JavaScript code. In combination with CSS injections the attacker might even make the element fill out the entire page so that the user would just have to have the mouse anywhere on the page to trigger the attack. 24.2 Cross-Site Request Forgery (CSRF) Another big problem is CSRF. This is a very complex topic and I won’t outline it here in detail just mention what it is and how to theoretically prevent it. If your authentication information is stored in cookies, you have implicit state man- agement. The state of “being logged in” is controlled by a cookie, and that cookie is sent with each request to a page. Unfortunately that includes requests triggered by 3rd party sites. If you don’t keep that in mind, some people might be able to trick your application’s users with social engineering to do stupid things without them knowing. Say you have a specific URL that, when you sent POST requests to will delete a user’s profile (say http://example.com/user/delete). If an attacker now creates a page that sends a post request to that page with some JavaScript they just has to trick some users to load that page and their profiles will end up being deleted. Imagine you were to run Facebook with millions of concurrent users and someone would send out links to images of little kittens. When users would go to that page, their profiles would get deleted while they are looking at images of fluffy cats. How can you prevent that? Basically for each request that modifies content on the server you would have to either use a one-time token and store that in the cookie and also transmit it with the form data. After receiving the data on the server again, you would then have to compare the two tokens and ensure they are equal. Why does Flask not do that for you? The ideal place for this to happen is the form validation framework, which does not exist in Flask. 24.3 JSON Security ECMAScript 5 Changes Starting with ECMAScript 5 the behavior of literals changed. Now they are not con- structed with the constructor of Array and others, but with the builtin constructor of Array which closes this particular attack vector. JSON itself is a high-level serialization format, so there is barely anything that could cause security problems, right? You can’t declare recursive structures that could cause 246
- 257. problems and the only thing that could possibly break are very large responses that can cause some kind of denial of service at the receiver’s side. However there is a catch. Due to how browsers work the CSRF issue comes up with JSON unfortunately. Fortunately there is also a weird part of the JavaScript specifica- tion that can be used to solve that problem easily and Flask is kinda doing that for you by preventing you from doing dangerous stuff. Unfortunately that protection is only there for jsonify() so you are still at risk when using other ways to generate JSON. So what is the issue and how to avoid it? The problem are arrays at top-level in JSON. Imagine you send the following data out in a JSON request. Say that’s exporting the names and email addresses of all your friends for a part of the user interface that is written in JavaScript. Not very uncommon: [ {"username": "admin", "email": "admin@localhost"} ] And it is doing that of course only as long as you are logged in and only for you. And it is doing that for all GET requests to a certain URL, say the URL for that request is http://example.com/api/get_friends.json. So now what happens if a clever hacker is embedding this to his website and social engineers a victim to visiting his site: <script type=text/javascript> var captured = []; var oldArray = Array; function Array() { var obj = this, id = 0, capture = function(value) { obj.__defineSetter__(id++, capture); if (value) captured.push(value); }; capture(); } </script> <script type=text/javascript src=http://example.com/api/get_friends.json></script> <script type=text/javascript> Array = oldArray; // now we have all the data in the captured array. </script> If you know a bit of JavaScript internals you might know that it’s possible to patch constructors and register callbacks for setters. An attacker can use this (like above) to get all the data you exported in your JSON file. The browser will totally ignore the application/json mimetype if text/javascript is defined as content type in the script tag and evaluate that as JavaScript. Because top-level array elements are allowed (albeit useless) and we hooked in our own constructor, after that page loaded the data from the JSON response is in the captured array. 247
- 258. Because it is a syntax error in JavaScript to have an object literal ({...}) toplevel an attacker could not just do a request to an external URL with the script tag to load up the data. So what Flask does is to only allow objects as toplevel elements when using jsonify(). Make sure to do the same when using an ordinary JSON generate function. 248
- 259. CHAPTER TWENTYFIVE UNICODE IN FLASK Flask like Jinja2 and Werkzeug is totally Unicode based when it comes to text. Not only these libraries, also the majority of web related Python libraries that deal with text. If you don’t know Unicode so far, you should probably read The Absolute Min- imum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets. This part of the documentation just tries to cover the very basics so that you have a pleasant experience with Unicode related things. 25.1 Automatic Conversion Flask has a few assumptions about your application (which you can change of course) that give you basic and painless Unicode support: • the encoding for text on your website is UTF-8 • internally you will always use Unicode exclusively for text except for literal strings with only ASCII character points. • encoding and decoding happens whenever you are talking over a protocol that requires bytes to be transmitted. So what does this mean to you? HTTP is based on bytes. Not only the protocol, also the system used to address doc- uments on servers (so called URIs or URLs). However HTML which is usually trans- mitted on top of HTTP supports a large variety of character sets and which ones are used, are transmitted in an HTTP header. To not make this too complex Flask just assumes that if you are sending Unicode out you want it to be UTF-8 encoded. Flask will do the encoding and setting of the appropriate headers for you. The same is true if you are talking to databases with the help of SQLAlchemy or a similar ORM system. Some databases have a protocol that already transmits Unicode and if they do not, SQLAlchemy or your other ORM should take care of that. 249
- 260. 25.2 The Golden Rule So the rule of thumb: if you are not dealing with binary data, work with Unicode. What does working with Unicode in Python 2.x mean? • as long as you are using ASCII charpoints only (basically numbers, some special characters of latin letters without umlauts or anything fancy) you can use regular string literals (’Hello World’). • if you need anything else than ASCII in a string you have to mark this string as Unicode string by prefixing it with a lowercase u. (like u’Hänsel und Gretel’) • if you are using non-Unicode characters in your Python files you have to tell Python which encoding your file uses. Again, I recommend UTF-8 for this purpose. To tell the interpreter your encoding you can put the # -*- coding: utf-8 -*- into the first or second line of your Python source file. • Jinja is configured to decode the template files from UTF-8. So make sure to tell your editor to save the file as UTF-8 there as well. 25.3 Encoding and Decoding Yourself If you are talking with a filesystem or something that is not really based on Unicode you will have to ensure that you decode properly when working with Unicode inter- face. So for example if you want to load a file on the filesystem and embed it into a Jinja2 template you will have to decode it from the encoding of that file. Here the old problem that text files do not specify their encoding comes into play. So do yourself a favour and limit yourself to UTF-8 for text files as well. Anyways. To load such a file with Unicode you can use the built-in str.decode() method: def read_file(filename, charset=’utf-8’): with open(filename, ’r’) as f: return f.read().decode(charset) To go from Unicode into a specific charset such as UTF-8 you can use the unicode.encode() method: def write_file(filename, contents, charset=’utf-8’): with open(filename, ’w’) as f: f.write(contents.encode(charset)) 25.4 Configuring Editors Most editors save as UTF-8 by default nowadays but in case your editor is not config- ured to do this you have to change it. Here some common ways to set your editor to store as UTF-8: 250
- 261. • Vim: put set enc=utf-8 to your .vimrc file. • Emacs: either use an encoding cookie or put this into your .emacs file: (prefer-coding-system ’utf-8) (setq default-buffer-file-coding-system ’utf-8) • Notepad++: 1. Go to Settings -> Preferences ... 2. Select the “New Document/Default Directory” tab 3. Select “UTF-8 without BOM” as encoding It is also recommended to use the Unix newline format, you can select it in the same panel but this is not a requirement. 251
- 262. 252
- 263. CHAPTER TWENTYSIX FLASK EXTENSION DEVELOPMENT Flask, being a microframework, often requires some repetitive steps to get a third party library working. Because very often these steps could be abstracted to support multi- ple projects the Flask Extension Registry was created. If you want to create your own Flask extension for something that does not exist yet, this guide to extension development will help you get your extension running in no time and to feel like users would expect your extension to behave. 26.1 Anatomy of an Extension Extensions are all located in a package called flask_something where “something” is the name of the library you want to bridge. So for example if you plan to add support for a library named simplexml to Flask, you would name your extension’s package flask_simplexml. The name of the actual extension (the human readable name) however would be some- thing like “Flask-SimpleXML”. Make sure to include the name “Flask” somewhere in that name and that you check the capitalization. This is how users can then register dependencies to your extension in their setup.py files. Flask sets up a redirect package called flask.ext where users should import the ex- tensions from. If you for instance have a package called flask_something users would import it as flask.ext.something. This is done to transition from the old namespace packages. See Extension Import Transition for more details. But how do extensions look like themselves? An extension has to ensure that it works with multiple Flask application instances at once. This is a requirement because many people will use patterns like the Application Factories pattern to create their application as needed to aid unittests and to support multiple configurations. Because of that it is crucial that your application supports that kind of behavior. Most importantly the extension must be shipped with a setup.py file and registered on PyPI. Also the development checkout link should work so that people can easily install the development version into their virtualenv without having to download the library by hand. 253
- 264. Flask extensions must be licensed under a BSD, MIT or more liberal license to be able to be enlisted in the Flask Extension Registry. Keep in mind that the Flask Extension Registry is a moderated place and libraries will be reviewed upfront if they behave as required. 26.2 “Hello Flaskext!” So let’s get started with creating such a Flask extension. The extension we want to create here will provide very basic support for SQLite3. First we create the following folder structure: flask-sqlite3/ flask_sqlite3.py LICENSE README Here’s the contents of the most important files: 26.2.1 setup.py The next file that is absolutely required is the setup.py file which is used to install your Flask extension. The following contents are something you can work with: """ Flask-SQLite3 ------------- This is the description for that library """ from setuptools import setup setup( name=’Flask-SQLite3’, version=’1.0’, url=’http://example.com/flask-sqlite3/’, license=’BSD’, author=’Your Name’, author_email=’[email protected]’, description=’Very short description’, long_description=__doc__, py_modules=[’flask_sqlite3’], # if you would be using a package instead use packages instead # of py_modules: # packages=[’flask_sqlite3’], zip_safe=False, include_package_data=True, platforms=’any’, 254
- 265. install_requires=[ ’Flask’ ], classifiers=[ ’Environment :: Web Environment’, ’Intended Audience :: Developers’, ’License :: OSI Approved :: BSD License’, ’Operating System :: OS Independent’, ’Programming Language :: Python’, ’Topic :: Internet :: WWW/HTTP :: Dynamic Content’, ’Topic :: Software Development :: Libraries :: Python Modules’ ] ) That’s a lot of code but you can really just copy/paste that from existing extensions and adapt. 26.2.2 flask_sqlite3.py Now this is where your extension code goes. But how exactly should such an exten- sion look like? What are the best practices? Continue reading for some insight. 26.3 Initializing Extensions Many extensions will need some kind of initialization step. For example, consider an application that’s currently connecting to SQLite like the documentation suggests (Us- ing SQLite 3 with Flask). So how does the extension know the name of the application object? Quite simple: you pass it to it. There are two recommended ways for an extension to initialize: initialization functions: If your extension is called helloworld you might have a function called init_helloworld(app[, extra_args]) that initializes the extension for that application. It could attach before / after handlers etc. classes: Classes work mostly like initialization functions but can later be used to further change the behavior. For an example look at how the OAuth exten- sion works: there is an OAuth object that provides some helper functions like OAuth.remote_app to create a reference to a remote application that uses OAuth. What to use depends on what you have in mind. For the SQLite 3 extension we will use the class-based approach because it will provide users with an object that handles opening and closing database connections. 255
- 266. What’s important about classes is that they encourage to be shared around on mod- ule level. In that case, the object itself must not under any circumstances store any application specific state and must be shareable between different application. 26.4 The Extension Code Here’s the contents of the flask_sqlite3.py for copy/paste: import sqlite3 from flask import current_app # Find the stack on which we want to store the database connection. # Starting with Flask 0.9, the _app_ctx_stack is the correct one, # before that we need to use the _request_ctx_stack. try: from flask import _app_ctx_stack as stack except ImportError: from flask import _request_ctx_stack as stack class SQLite3(object): def __init__(self, app=None): self.app = app if app is not None: self.init_app(app) def init_app(self, app): app.config.setdefault(’SQLITE3_DATABASE’, ’:memory:’) # Use the newstyle teardown_appcontext if it’s available, # otherwise fall back to the request context if hasattr(app, ’teardown_appcontext’): app.teardown_appcontext(self.teardown) else: app.teardown_request(self.teardown) def connect(self): return sqlite3.connect(current_app.config[’SQLITE3_DATABASE’]) def teardown(self, exception): ctx = stack.top if hasattr(ctx, ’sqlite3_db’): ctx.sqlite3_db.close() @property def connection(self): ctx = stack.top if ctx is not None: if not hasattr(ctx, ’sqlite3_db’): 256
- 267. ctx.sqlite3_db = self.connect() return ctx.sqlite3_db So here’s what these lines of code do: 1. The __init__ method takes an optional app object and, if supplied, will call init_app. 2. The init_app method exists so that the SQLite3 object can be instantiated with- out requiring an app object. This method supports the factory pattern for cre- ating applications. The init_app will set the configuration for the database, de- faulting to an in memory database if no configuration is supplied. In addition, the init_app method attaches the teardown handler. It will try to use the new- style app context handler and if it does not exist, falls back to the request context one. 3. Next, we define a connect method that opens a database connection. 4. Finally, we add a connection property that on first access opens the database connection and stores it on the context. This is also the recommended way to handling resources: fetch resources lazily the first time they are used. Note here that we’re attaching our database connection to the top application context via _app_ctx_stack.top. Extensions should use the top context for stor- ing their own information with a sufficiently complex name. Note that we’re falling back to the _request_ctx_stack.top if the application is using an older version of Flask that does not support it. So why did we decide on a class-based approach here? Because using our extension looks something like this: from flask import Flask from flask_sqlite3 import SQLite3 app = Flask(__name__) app.config.from_pyfile(’the-config.cfg’) db = SQLite3(app) You can then use the database from views like this: @app.route(’/’) def show_all(): cur = db.connection.cursor() cur.execute(...) Likewise if you are outside of a request but you are using Flask 0.9 or later with the app context support, you can use the database in the same way: with app.app_context(): cur = db.connection.cursor() cur.execute(...) At the end of the with block the teardown handles will be executed automatically. 257
- 268. Additionally, the init_app method is used to support the factory pattern for creating apps: db = Sqlite3() # Then later on. app = create_app(’the-config.cfg’) db.init_app(app) Keep in mind that supporting this factory pattern for creating apps is required for approved flask extensions (described below). Note on init_app As you noticed, init_app does not assign app to self. This is intentional! Class based Flask extensions must only store the application on the object when the application was passed to the constructor. This tells the extension: I am not interested in using multiple applications. When the extension needs to find the current application and it does not have a refer- ence to it, it must either use the current_app context local or change the API in a way that you can pass the application explicitly. 26.5 Using _app_ctx_stack In the example above, before every request, a sqlite3_db variable is assigned to _app_ctx_stack.top. In a view function, this variable is accessible using the connection property of SQLite3. During the teardown of a request, the sqlite3_db connection is closed. By using this pattern, the same connection to the sqlite3 database is accessible to anything that needs it for the duration of the request. If the _app_ctx_stack does not exist because the user uses an old version of Flask, it is recommended to fall back to _request_ctx_stack which is bound to a request. 26.6 Teardown Behavior This is only relevant if you want to support Flask 0.6 and older Due to the change in Flask 0.7 regarding functions that are run at the end of the request your extension will have to be extra careful there if it wants to continue to support older versions of Flask. The following pattern is a good way to support both: def close_connection(response): ctx = _request_ctx_stack.top ctx.sqlite3_db.close() return response if hasattr(app, ’teardown_request’): 258
- 269. app.teardown_request(close_connection) else: app.after_request(close_connection) Strictly speaking the above code is wrong, because teardown functions are passed the exception and typically don’t return anything. However because the return value is discarded this will just work assuming that the code in between does not touch the passed parameter. 26.7 Learn from Others This documentation only touches the bare minimum for extension development. If you want to learn more, it’s a very good idea to check out existing extensions on the Flask Extension Registry. If you feel lost there is still the mailinglist and the IRC chan- nel to get some ideas for nice looking APIs. Especially if you do something nobody before you did, it might be a very good idea to get some more input. This not only to get an idea about what people might want to have from an extension, but also to avoid having multiple developers working on pretty much the same side by side. Remember: good API design is hard, so introduce your project on the mailinglist, and let other developers give you a helping hand with designing the API. The best Flask extensions are extensions that share common idioms for the API. And this can only work if collaboration happens early. 26.8 Approved Extensions Flask also has the concept of approved extensions. Approved extensions are tested as part of Flask itself to ensure extensions do not break on new releases. These approved extensions are listed on the Flask Extension Registry and marked appropriately. If you want your own extension to be approved you have to follow these guidelines: 0. An approved Flask extension requires a maintainer. In the event an extension author would like to move beyond the project, the project should find a new maintainer including full source hosting transition and PyPI access. If no main- tainer is available, give access to the Flask core team. 1. An approved Flask extension must provide exactly one package or module named flask_extensionname. They might also reside inside a flaskext names- pace packages though this is discouraged now. 2. It must ship a testing suite that can either be invoked with make test or python setup.py test. For test suites invoked with make test the extension has to en- sure that all dependencies for the test are installed automatically. If tests are invoked with python setup.py test, test dependencies can be specified in the setup.py file. The test suite also has to be part of the distribution. 3. APIs of approved extensions will be checked for the following characteristics: 259
- 270. • an approved extension has to support multiple applications running in the same Python process. • it must be possible to use the factory pattern for creating applications. 4. The license must be BSD/MIT/WTFPL licensed. 5. The naming scheme for official extensions is Flask-ExtensionName or ExtensionName-Flask. 6. Approved extensions must define all their dependencies in the setup.py file un- less a dependency cannot be met because it is not available on PyPI. 7. The extension must have documentation that uses one of the two Flask themes for Sphinx documentation. 8. The setup.py description (and thus the PyPI description) has to link to the doc- umentation, website (if there is one) and there must be a link to automatically install the development version (PackageName==dev). 9. The zip_safe flag in the setup script must be set to False, even if the extension would be safe for zipping. 10. An extension currently has to support Python 2.6 as well as Python 2.7 26.9 Extension Import Transition For a while we recommended using namespace packages for Flask extensions. This turned out to be problematic in practice because many different competing namespace package systems exist and pip would automatically switch between different systems and this caused a lot of problems for users. Instead we now recommend naming packages flask_foo instead of the now depre- cated flaskext.foo. Flask 0.8 introduces a redirect import system that lets uses import from flask.ext.foo and it will try flask_foo first and if that fails flaskext.foo. Flask extensions should urge users to import from flask.ext.foo instead of flask_foo or flaskext_foo so that extensions can transition to the new package name without affecting users. 260
- 271. CHAPTER TWENTYSEVEN POCOO STYLEGUIDE The Pocoo styleguide is the styleguide for all Pocoo Projects, including Flask. This styleguide is a requirement for Patches to Flask and a recommendation for Flask ex- tensions. In general the Pocoo Styleguide closely follows PEP 8 with some small differences and extensions. 27.1 General Layout Indentation: 4 real spaces. No tabs, no exceptions. Maximum line length: 79 characters with a soft limit for 84 if absolutely necessary. Try to avoid too nested code by cleverly placing break, continue and return state- ments. Continuing long statements: To continue a statement you can use backslashes in which case you should align the next line with the last dot or equal sign, or indent four spaces: this_is_a_very_long(function_call, ’with many parameters’) .that_returns_an_object_with_an_attribute MyModel.query.filter(MyModel.scalar > 120) .order_by(MyModel.name.desc()) .limit(10) If you break in a statement with parentheses or braces, align to the braces: this_is_a_very_long(function_call, ’with many parameters’, 23, 42, ’and even more’) For lists or tuples with many items, break immediately after the opening brace: items = [ ’this is the first’, ’set of items’, ’with more items’, ’to come in this line’, ’like this’ ] 261
- 272. Blank lines: Top level functions and classes are separated by two lines, everything else by one. Do not use too many blank lines to separate logical segments in code. Example: def hello(name): print ’Hello %s!’ % name def goodbye(name): print ’See you %s.’ % name class MyClass(object): """This is a simple docstring""" def __init__(self, name): self.name = name def get_annoying_name(self): return self.name.upper() + ’!!!!111’ 27.2 Expressions and Statements General whitespace rules: • No whitespace for unary operators that are not words (e.g.: -, ~ etc.) as well on the inner side of parentheses. • Whitespace is placed between binary operators. Good: exp = -1.05 value = (item_value / item_count) * offset / exp value = my_list[index] value = my_dict[’key’] Bad: exp = - 1.05 value = ( item_value / item_count ) * offset / exp value = (item_value/item_count)*offset/exp value=( item_value/item_count ) * offset/exp value = my_list[ index ] value = my_dict [’key’] Yoda statements are a no-go: Never compare constant with variable, always variable with constant: Good: 262
- 273. if method == ’md5’: pass Bad: if ’md5’ == method: pass Comparisons: • against arbitrary types: == and != • against singletons with is and is not (eg: foo is not None) • never compare something with True or False (for example never do foo == False, do not foo instead) Negated containment checks: use foo not in bar instead of not foo in bar Instance checks: isinstance(a, C) instead of type(A) is C, but try to avoid instance checks in general. Check for features. 27.3 Naming Conventions • Class names: CamelCase, with acronyms kept uppercase (HTTPWriter and not HttpWriter) • Variable names: lowercase_with_underscores • Method and function names: lowercase_with_underscores • Constants: UPPERCASE_WITH_UNDERSCORES • precompiled regular expressions: name_re Protected members are prefixed with a single underscore. Double underscores are reserved for mixin classes. On classes with keywords, trailing underscores are appended. Clashes with builtins are allowed and must not be resolved by appending an underline to the variable name. If the function needs to access a shadowed builtin, rebind the builtin to a different name instead. Function and method arguments: • class methods: cls as first parameter • instance methods: self as first parameter • lambdas for properties might have the first parameter replaced with x like in display_name = property(lambda x: x.real_name or x.username) 263
- 274. 27.4 Docstrings Docstring conventions: All docstrings are formatted with reStructuredText as under- stood by Sphinx. Depending on the number of lines in the docstring, they are laid out differently. If it’s just one line, the closing triple quote is on the same line as the opening, otherwise the text is on the same line as the opening quote and the triple quote that closes the string on its own line: def foo(): """This is a simple docstring""" def bar(): """This is a longer docstring with so much information in there that it spans three lines. In this case the closing triple quote is on its own line. """ Module header: The module header consists of an utf-8 encoding declaration (if non ASCII letters are used, but it is recommended all the time) and a standard doc- string: # -*- coding: utf-8 -*- """ package.module ~~~~~~~~~~~~~~ A brief description goes here. :copyright: (c) YEAR by AUTHOR. :license: LICENSE_NAME, see LICENSE_FILE for more details. """ Please keep in mind that proper copyrights and license files are a requirement for approved Flask extensions. 27.5 Comments Rules for comments are similar to docstrings. Both are formatted with reStructured- Text. If a comment is used to document an attribute, put a colon after the opening pound sign (#): class User(object): #: the name of the user as unicode string name = Column(String) #: the sha1 hash of the password + inline salt pw_hash = Column(String) 264
- 275. CHAPTER TWENTYEIGHT PYTHON 3 SUPPORT Flask and all of its dependencies support Python 3 so you can in theory start working on it already. There are however a few things you should be aware of before you start using Python 3 for your next project. 28.1 Requirements If you want to use Flask with Python 3 you will need to use Python 3.3 or higher. 3.2 and older are not supported. In addition to that you need to use the latest and greatest versions of itsdangerous, Jinja2 and Werkzeug. 28.2 API Stability Some of the decisions made in regards to unicode and byte untilization on Python 3 make it hard to write low level code. This mainly affects WSGI middlewares and inter- acting with the WSGI provided information. Werkzeug wraps all that information in high-level helpers but some of those were specifically added for the Python 3 support and are quite new. A lot of the documentation out there on using WSGI leaves out those details as it was written before WSGI was updated to Python 3. While the API for Werkzeug and Flask on Python 2.x should not change much we cannot guarantee that this won’t happen on Python 3. 28.3 Few Users Python 3 currently has less than 1% of the users of Python 2 going by PyPI download stats. As a result many of the problems you will encounter are probably hard to search for on the internet if they are Python 3 specific. 265
- 276. 28.4 Small Ecosystem The majority of the Flask extensions, all of the documentation and the vast majority of the PyPI provided libraries do not support Python 3 yet. Even if you start your project with knowing that all you will need is supported by Python 3 you don’t know what happens six months from now. If you are adventurous you can start porting libraries on your own, but that is nothing for the faint of heart. 28.5 Recommendations Unless you are already familiar with the differences in the versions we recommend sticking to current versions of Python until the ecosystem caught up. The majority of the upgrade pain is in the lower-level libararies like Flask and Werkzeug and not in the actual high-level application code. For instance all of the Flask examples that are in the Flask repository work out of the box on both 2.x and 3.x and did not require a single line of code changed. 266
- 277. CHAPTER TWENTYNINE UPGRADING TO NEWER RELEASES Flask itself is changing like any software is changing over time. Most of the changes are the nice kind, the kind where you don’t have to change anything in your code to profit from a new release. However every once in a while there are changes that do require some changes in your code or there are changes that make it possible for you to improve your own code quality by taking advantage of new features in Flask. This section of the documentation enumerates all the changes in Flask from release to release and how you can change your code to have a painless updating experience. If you want to use the easy_install command to upgrade your Flask installation, make sure to pass it the -U parameter: $ easy_install -U Flask 29.1 Version 0.10 The biggest change going from 0.9 to 0.10 is that the cookie serialization format changed from pickle to a specialized JSON format. This change has been done in order to avoid the damage an attacker can do if the secret key is leaked. When you upgrade you will notice two major changes: all sessions that were issued before the upgrade are invalidated and you can only store a limited amount of types in the session. The new sessions are by design much more restricted to only allow JSON with a few small extensions for tuples and strings with HTML markup. In order to not break people’s sessions it is possible to continue using the old session system by using the Flask-OldSessions extension. Flask also started storing the flask.g object on the application context instead of the request context. This change should be transparent for you but it means that you now can store things on the g object when there is no request context yet but an applica- tion context. The old flask.Flask.request_globals_class attribute was renamed to flask.Flask.app_ctx_globals_class. 267
- 278. 29.2 Version 0.9 The behavior of returning tuples from a function was simplified. If you return a tuple it no longer defines the arguments for the response object you’re creating, it’s now always a tuple in the form (response, status, headers) where at least one item has to be provided. If you depend on the old behavior, you can add it easily by subclassing Flask: class TraditionalFlask(Flask): def make_response(self, rv): if isinstance(rv, tuple): return self.response_class(*rv) return Flask.make_response(self, rv) If you maintain an extension that was using _request_ctx_stack before, please con- sider changing to _app_ctx_stack if it makes sense for your extension. For instance, the app context stack makes sense for extensions which connect to databases. Using the app context stack instead of the request stack will make extensions more readily handle use cases outside of requests. 29.3 Version 0.8 Flask introduced a new session interface system. We also noticed that there was a naming collision between flask.session the module that implements sessions and flask.session which is the global session object. With that introduction we moved the implementation details for the session system into a new module called flask.sessions. If you used the previously undocumented session support we urge you to upgrade. If invalid JSON data was submitted Flask will now raise a BadRequest exception in- stead of letting the default ValueError bubble up. This has the advantage that you no longer have to handle that error to avoid an internal server error showing up for the user. If you were catching this down explicitly in the past as ValueError you will need to change this. Due to a bug in the test client Flask 0.7 did not trigger teardown handlers when the test client was used in a with statement. This was since fixed but might require some changes in your testsuites if you relied on this behavior. 29.4 Version 0.7 In Flask 0.7 we cleaned up the code base internally a lot and did some backwards in- compatible changes that make it easier to implement larger applications with Flask. Because we want to make upgrading as easy as possible we tried to counter the prob- lems arising from these changes by providing a script that can ease the transition. 268
- 279. The script scans your whole application and generates an unified diff with changes it assumes are safe to apply. However as this is an automated tool it won’t be able to find all use cases and it might miss some. We internally spread a lot of deprecation warnings all over the place to make it easy to find pieces of code that it was unable to upgrade. We strongly recommend that you hand review the generated patchfile and only apply the chunks that look good. If you are using git as version control system for your project we recommend apply- ing the patch with path -p1 < patchfile.diff and then using the interactive commit feature to only apply the chunks that look good. To apply the upgrade script do the following: 1. Download the script: flask-07-upgrade.py 2. Run it in the directory of your application: python flask-07-upgrade.py > patchfile.diff 3. Review the generated patchfile. 4. Apply the patch: patch -p1 < patchfile.diff 5. If you were using per-module template folders you need to move some templates around. Previously if you had a folder named templates next to a blueprint named admin the implicit template path automatically was admin/index.html for a template file called templates/index.html. This no longer is the case. Now you need to name the template templates/admin/index.html. The tool will not detect this so you will have to do that on your own. Please note that deprecation warnings are disabled by default starting with Python 2.7. In order to see the deprecation warnings that might be emitted you have to enabled them with the warnings module. If you are working with windows and you lack the patch command line utility you can get it as part of various Unix runtime environments for windows including cygwin, msysgit or ming32. Also source control systems like svn, hg or git have builtin support for applying unified diffs as generated by the tool. Check the manual of your version control system for more information. 29.4.1 Bug in Request Locals Due to a bug in earlier implementations the request local proxies now raise a RuntimeError instead of an AttributeError when they are unbound. If you caught these exceptions with AttributeError before, you should catch them with RuntimeError now. Additionally the send_file() function is now issuing deprecation warnings if you depend on functionality that will be removed in Flask 1.0. Previously it was possible 269
- 280. to use etags and mimetypes when file objects were passed. This was unreliable and caused issues for a few setups. If you get a deprecation warning, make sure to update your application to work with either filenames there or disable etag attaching and attach them yourself. Old code: return send_file(my_file_object) return send_file(my_file_object) New code: return send_file(my_file_object, add_etags=False) 29.4.2 Upgrading to new Teardown Handling We streamlined the behavior of the callbacks for request handling. For things that modify the response the after_request() decorators continue to work as expected, but for things that absolutely must happen at the end of request we introduced the new teardown_request() decorator. Unfortunately that change also made after-request work differently under error conditions. It’s not consistently skipped if exceptions happen whereas previously it might have been called twice to ensure it is executed at the end of the request. If you have database connection code that looks like this: @app.after_request def after_request(response): g.db.close() return response You are now encouraged to use this instead: @app.teardown_request def after_request(exception): if hasattr(g, ’db’): g.db.close() On the upside this change greatly improves the internal code flow and makes it easier to customize the dispatching and error handling. This makes it now a lot easier to write unit tests as you can prevent closing down of database connections for a while. You can take advantage of the fact that the teardown callbacks are called when the re- sponse context is removed from the stack so a test can query the database after request handling: with app.test_client() as client: resp = client.get(’/’) # g.db is still bound if there is such a thing # and here it’s gone 270
- 281. 29.4.3 Manual Error Handler Attaching While it is still possible to attach error handlers to Flask.error_handlers it’s discour- aged to do so and in fact deprecated. In general we no longer recommend custom error handler attaching via assignments to the underlying dictionary due to the more complex internal handling to support arbitrary exception classes and blueprints. See Flask.errorhandler() for more information. The proper upgrade is to change this: app.error_handlers[403] = handle_error Into this: app.register_error_handler(403, handle_error) Alternatively you should just attach the function with a decorator: @app.errorhandler(403) def handle_error(e): ... (Note that register_error_handler() is new in Flask 0.7) 29.4.4 Blueprint Support Blueprints replace the previous concept of “Modules” in Flask. They provide better semantics for various features and work better with large applications. The update script provided should be able to upgrade your applications automatically, but there might be some cases where it fails to upgrade. What changed? • Blueprints need explicit names. Modules had an automatic name guesssing scheme where the shortname for the module was taken from the last part of the import module. The upgrade script tries to guess that name but it might fail as this information could change at runtime. • Blueprints have an inverse behavior for url_for(). Previously .foo told url_for() that it should look for the endpoint foo on the application. Now it means “relative to current module”. The script will inverse all calls to url_for() automatically for you. It will do this in a very eager way so you might end up with some unnecessary leading dots in your code if you’re not using modules. • Blueprints do not automatically provide static folders. They will also no longer automatically export templates from a folder called templates next to their loca- tion however but it can be enabled from the constructor. Same with static files: if you want to continue serving static files you need to tell the constructor explic- itly the path to the static folder (which can be relative to the blueprint’s module path). • Rendering templates was simplified. Now the blueprints can provide template folders which are added to a general template searchpath. This means that you 271
- 282. need to add another subfolder with the blueprint’s name into that folder if you want blueprintname/template.html as the template name. If you continue to use the Module object which is deprecated, Flask will restore the previous behavior as good as possible. However we strongly recommend upgrading to the new blueprints as they provide a lot of useful improvement such as the ability to attach a blueprint multiple times, blueprint specific error handlers and a lot more. 29.5 Version 0.6 Flask 0.6 comes with a backwards incompatible change which affects the order of after- request handlers. Previously they were called in the order of the registration, now they are called in reverse order. This change was made so that Flask behaves more like people expected it to work and how other systems handle request pre- and postpro- cessing. If you depend on the order of execution of post-request functions, be sure to change the order. Another change that breaks backwards compatibility is that context processors will no longer override values passed directly to the template rendering function. If for exam- ple request is as variable passed directly to the template, the default context processor will not override it with the current request object. This makes it easier to extend con- text processors later to inject additional variables without breaking existing template not expecting them. 29.6 Version 0.5 Flask 0.5 is the first release that comes as a Python package instead of a single mod- ule. There were a couple of internal refactoring so if you depend on undocumented internal details you probably have to adapt the imports. The following changes may be relevant to your application: • autoescaping no longer happens for all templates. Instead it is configured to only happen on files ending with .html, .htm, .xml and .xhtml. If you have templates with different extensions you should override the select_jinja_autoescape() method. • Flask no longer supports zipped applications in this release. This functionality might come back in future releases if there is demand for this feature. Removing support for this makes the Flask internal code easier to understand and fixes a couple of small issues that make debugging harder than necessary. • The create_jinja_loader function is gone. If you want to customize the Jinja loader now, use the create_jinja_environment() method instead. 272
- 283. 29.7 Version 0.4 For application developers there are no changes that require changes in your code. In case you are developing on a Flask extension however, and that extension has a unittest-mode you might want to link the activation of that mode to the new TESTING flag. 29.8 Version 0.3 Flask 0.3 introduces configuration support and logging as well as categories for flash- ing messages. All these are features that are 100% backwards compatible but you might want to take advantage of them. 29.8.1 Configuration Support The configuration support makes it easier to write any kind of application that requires some sort of configuration. (Which most likely is the case for any application out there). If you previously had code like this: app.debug = DEBUG app.secret_key = SECRET_KEY You no longer have to do that, instead you can just load a configuration into the config object. How this works is outlined in Configuration Handling. 29.8.2 Logging Integration Flask now configures a logger for you with some basic and useful defaults. If you run your application in production and want to profit from automatic error logging, you might be interested in attaching a proper log handler. Also you can start logging warnings and errors into the logger when appropriately. For more information on that, read Logging Application Errors. 29.8.3 Categories for Flash Messages Flash messages can now have categories attached. This makes it possible to render errors, warnings or regular messages differently for example. This is an opt-in feature because it requires some rethinking in the code. Read all about that in the Message Flashing pattern. 273
- 284. 274
- 285. CHAPTER THIRTY FLASK CHANGELOG Here you can see the full list of changes between each Flask release. 30.1 Version 1.0 (release date to be announced, codename to be selected) • Added SESSION_REFRESH_EACH_REQUEST config key that controls the set-cookie behavior. If set to True a permanent session will be refreshed each request and get their lifetime extended, if set to False it will only be modified if the session ac- tually modifies. Non permanent sessions are not affected by this and will always expire if the browser window closes. • Made Flask support custom JSON mimetypes for incoming data. • Added support for returning tuples in the form (response, headers) from a view function. 30.2 Version 0.10.2 (bugfix release, release date to be announced) • Fixed broken test_appcontext_signals() test case. • Raise an AttributeError in flask.helpers.find_package() with a useful mes- sage explaining why it is raised when a PEP 302 import hook is used without an is_package() method. • Fixed an issue causing exceptions raised before entering a request or app context to be passed to teardown handlers. • Fixed an issue with query parameters getting removed from requests in the test client when absolute URLs were requested. • Made @before_first_request into a decorator as intended. 275
- 286. 30.3 Version 0.10.1 (bugfix release, released on June 14th 2013) • Fixed an issue where |tojson was not quoting single quotes which made the filter not work properly in HTML attributes. Now it’s possible to use that filter in single quoted attributes. This should make using that filter with angular.js easier. • Added support for byte strings back to the session system. This broke compati- bility with the common case of people putting binary data for token verification into the session. • Fixed an issue where registering the same method twice for the same endpoint would trigger an exception incorrectly. 30.4 Version 0.10 Released on June 13nd 2013, codename Limoncello. • Changed default cookie serialization format from pickle to JSON to limit the impact an attacker can do if the secret key leaks. See Version 0.10 for more infor- mation. • Added template_test methods in addition to the already existing template_filter method family. • Added template_global methods in addition to the already existing template_filter method family. • Set the content-length header for x-sendfile. • tojson filter now does not escape script blocks in HTML5 parsers. • tojson used in templates is now safe by default due. This was allowed due to the different escaping behavior. • Flask will now raise an error if you attempt to register a new function on an already used endpoint. • Added wrapper module around simplejson and added default serialization of datetime objects. This allows much easier customization of how JSON is handled by Flask or any Flask extension. • Removed deprecated internal flask.session module alias. Use flask.sessions instead to get the session module. This is not to be confused with flask.session the session proxy. • Templates can now be rendered without request context. The behavior is slightly different as the request, session and g objects will not be available and blueprint’s context processors are not called. 276
- 287. • The config object is now available to the template as a real global and not through a context processor which makes it available even in imported templates by de- fault. • Added an option to generate non-ascii encoded JSON which should result in less bytes being transmitted over the network. It’s disabled by default to not cause confusion with existing libraries that might expect flask.json.dumps to return bytestrings by default. • flask.g is now stored on the app context instead of the request context. • flask.g now gained a get() method for not erroring out on non existing items. • flask.g now can be used with the in operator to see what’s defined and it now is iterable and will yield all attributes stored. • flask.Flask.request_globals_class got renamed to flask.Flask.app_ctx_globals_class which is a better name to what it does since 0.10. • request, session and g are now also added as proxies to the template context which makes them available in imported templates. One has to be very careful with those though because usage outside of macros might cause caching. • Flask will no longer invoke the wrong error handlers if a proxy exception is passed through. • Added a workaround for chrome’s cookies in localhost not working as intended with domain names. • Changed logic for picking defaults for cookie values from sessions to work better with Google Chrome. • Added message_flashed signal that simplifies flashing testing. • Added support for copying of request contexts for better working with greenlets. • Removed custom JSON HTTP exception subclasses. If you were relying on them you can reintroduce them again yourself trivially. Using them however is strongly discouraged as the interface was flawed. • Python requirements changed: requiring Python 2.6 or 2.7 now to prepare for Python 3.3 port. • Changed how the teardown system is informed about exceptions. This is now more reliable in case something handles an exception halfway through the error handling process. • Request context preservation in debug mode now keeps the exception informa- tion around which means that teardown handlers are able to distinguish error from success cases. • Added the JSONIFY_PRETTYPRINT_REGULAR configuration variable. • Flask now orders JSON keys by default to not trash HTTP caches due to different hash seeds between different workers. 277
- 288. • Added appcontext_pushed and appcontext_popped signals. • The builtin run method now takes the SERVER_NAME into account when picking the default port to run on. • Added flask.request.get_json() as a replacement for the old flask.request.json prop- erty. 30.5 Version 0.9 Released on July 1st 2012, codename Campari. • The flask.Request.on_json_loading_failed() now returns a JSON formatted response by default. • The flask.url_for() function now can generate anchors to the generated links. • The flask.url_for() function now can also explicitly generate URL rules spe- cific to a given HTTP method. • Logger now only returns the debug log setting if it was not set explicitly. • Unregister a circular dependency between the WSGI environment and the request object when shutting down the request. This means that environ werkzeug.request will be None after the response was returned to the WSGI server but has the advantage that the garbage collector is not needed on CPython to tear down the request unless the user created circular dependencies them- selves. • Session is now stored after callbacks so that if the session payload is stored in the session you can still modify it in an after request callback. • The flask.Flask class will avoid importing the provided import name if it can (the required first parameter), to benefit tools which build Flask instances pro- grammatically. The Flask class will fall back to using import on systems with custom module hooks, e.g. Google App Engine, or when the import name is inside a zip archive (usually a .egg) prior to Python 2.7. • Blueprints now have a decorator to add custom template filters application wide, flask.Blueprint.app_template_filter(). • The Flask and Blueprint classes now have a non-decorator method for adding custom template filters application wide, flask.Flask.add_template_filter() and flask.Blueprint.add_app_template_filter(). • The flask.get_flashed_messages() function now allows rendering flashed mes- sage categories in separate blocks, through a category_filter argument. • The flask.Flask.run() method now accepts None for host and port arguments, using default values when None. This allows for calling run using configuration values, e.g. app.run(app.config.get(’MYHOST’), app.config.get(’MYPORT’)), with proper behavior whether or not a config file is provided. 278
- 289. • The flask.render_template() method now accepts a either an iterable of tem- plate names or a single template name. Previously, it only accepted a single template name. On an iterable, the first template found is rendered. • Added flask.Flask.app_context() which works very similar to the request con- text but only provides access to the current application. This also adds support for URL generation without an active request context. • View functions can now return a tuple with the first instance being an instance of flask.Response. This allows for returning jsonify(error="error msg"), 400 from a view function. • Flask and Blueprint now provide a get_send_file_max_age() hook for sub- classes to override behavior of serving static files from Flask when using flask.Flask.send_static_file() (used for the default static file handler) and send_file(). This hook is provided a filename, which for example allows chang- ing cache controls by file extension. The default max-age for send_file and static files can be configured through a new SEND_FILE_MAX_AGE_DEFAULT configuration variable, which is used in the default get_send_file_max_age implementation. • Fixed an assumption in sessions implementation which could break message flashing on sessions implementations which use external storage. • Changed the behavior of tuple return values from functions. They are no longer arguments to the response object, they now have a defined meaning. • Added flask.Flask.request_globals_class to allow a specific class to be used on creation of the g instance of each request. • Added required_methods attribute to view functions to force-add methods on reg- istration. • Added flask.after_this_request(). • Added flask.stream_with_context() and the ability to push contexts multiple times without producing unexpected behavior. 30.6 Version 0.8.1 Bugfix release, released on July 1st 2012 • Fixed an issue with the undocumented flask.session module to not work properly on Python 2.5. It should not be used but did cause some problems for package managers. 30.7 Version 0.8 Released on September 29th 2011, codename Rakija 279
- 290. • Refactored session support into a session interface so that the implementation of the sessions can be changed without having to override the Flask class. • Empty session cookies are now deleted properly automatically. • View functions can now opt out of getting the automatic OPTIONS implementa- tion. • HTTP exceptions and Bad Request errors can now be trapped so that they show up normally in the traceback. • Flask in debug mode is now detecting some common problems and tries to warn you about them. • Flask in debug mode will now complain with an assertion error if a view was attached after the first request was handled. This gives earlier feedback when users forget to import view code ahead of time. • Added the ability to register callbacks that are only triggered once at the begin- ning of the first request. (Flask.before_first_request()) • Malformed JSON data will now trigger a bad request HTTP exception instead of a value error which usually would result in a 500 internal server error if not handled. This is a backwards incompatible change. • Applications now not only have a root path where the resources and modules are located but also an instance path which is the designated place to drop files that are modified at runtime (uploads etc.). Also this is conceptionally only instance depending and outside version control so it’s the perfect place to put configura- tion files etc. For more information see Instance Folders. • Added the APPLICATION_ROOT configuration variable. • Implemented session_transaction() to easily modify sessions from the test en- vironment. • Refactored test client internally. The APPLICATION_ROOT configuration variable as well as SERVER_NAME are now properly used by the test client as defaults. • Added flask.views.View.decorators to support simpler decorating of plug- gable (class-based) views. • Fixed an issue where the test client if used with the “with” statement did not trigger the execution of the teardown handlers. • Added finer control over the session cookie parameters. • HEAD requests to a method view now automatically dispatch to the get method if no handler was implemented. • Implemented the virtual flask.ext package to import extensions from. • The context preservation on exceptions is now an integral component of Flask itself and no longer of the test client. This cleaned up some internal logic and lowers the odds of runaway request contexts in unittests. 280
- 291. 30.8 Version 0.7.3 Bugfix release, release date to be decided • Fixed the Jinja2 environment’s list_templates method not returning the correct names when blueprints or modules were involved. 30.9 Version 0.7.2 Bugfix release, released on July 6th 2011 • Fixed an issue with URL processors not properly working on blueprints. 30.10 Version 0.7.1 Bugfix release, released on June 29th 2011 • Added missing future import that broke 2.5 compatibility. • Fixed an infinite redirect issue with blueprints. 30.11 Version 0.7 Released on June 28th 2011, codename Grappa • Added make_default_options_response() which can be used by subclasses to alter the default behavior for OPTIONS responses. • Unbound locals now raise a proper RuntimeError instead of an AttributeError. • Mimetype guessing and etag support based on file objects is now deprecated for flask.send_file() because it was unreliable. Pass filenames instead or attach your own etags and provide a proper mimetype by hand. • Static file handling for modules now requires the name of the static folder to be supplied explicitly. The previous autodetection was not reliable and caused issues on Google’s App Engine. Until 1.0 the old behavior will continue to work but issue dependency warnings. • fixed a problem for Flask to run on jython. • added a PROPAGATE_EXCEPTIONS configuration variable that can be used to flip the setting of exception propagation which previously was linked to DEBUG alone and is now linked to either DEBUG or TESTING. • Flask no longer internally depends on rules being added through the add_url_rule function and can now also accept regular werkzeug rules added to the url map. 281
- 292. • Added an endpoint method to the flask application object which allows one to register a callback to an arbitrary endpoint with a decorator. • Use Last-Modified for static file sending instead of Date which was incorrectly introduced in 0.6. • Added create_jinja_loader to override the loader creation process. • Implemented a silent flag for config.from_pyfile. • Added teardown_request decorator, for functions that should run at the end of a request regardless of whether an exception occurred. Also the behavior for after_request was changed. It’s now no longer executed when an exception is raised. See Upgrading to new Teardown Handling • Implemented flask.has_request_context() • Deprecated init_jinja_globals. Override the create_jinja_environment() method instead to achieve the same functionality. • Added flask.safe_join() • The automatic JSON request data unpacking now looks at the charset mimetype parameter. • Don’t modify the session on flask.get_flashed_messages() if there are no mes- sages in the session. • before_request handlers are now able to abort requests with errors. • it is not possible to define user exception handlers. That way you can provide custom error messages from a central hub for certain errors that might occur dur- ing request processing (for instance database connection errors, timeouts from remote resources etc.). • Blueprints can provide blueprint specific error handlers. • Implemented generic Pluggable Views (class-based views). 30.12 Version 0.6.1 Bugfix release, released on December 31st 2010 • Fixed an issue where the default OPTIONS response was not exposing all valid methods in the Allow header. • Jinja2 template loading syntax now allows ”./” in front of a template load path. Previously this caused issues with module setups. • Fixed an issue where the subdomain setting for modules was ignored for the static folder. • Fixed a security problem that allowed clients to download arbitrary files if the host server was a windows based operating system and the client uses back- slashes to escape the directory the files where exposed from. 282
- 293. 30.13 Version 0.6 Released on July 27th 2010, codename Whisky • after request functions are now called in reverse order of registration. • OPTIONS is now automatically implemented by Flask unless the application explicitly adds ‘OPTIONS’ as method to the URL rule. In this case no automatic OPTIONS handling kicks in. • static rules are now even in place if there is no static folder for the module. This was implemented to aid GAE which will remove the static folder if it’s part of a mapping in the .yml file. • the config is now available in the templates as config. • context processors will no longer override values passed directly to the render function. • added the ability to limit the incoming request data with the new MAX_CONTENT_LENGTH configuration value. • the endpoint for the flask.Module.add_url_rule() method is now optional to be consistent with the function of the same name on the application object. • added a flask.make_response() function that simplifies creating response object instances in views. • added signalling support based on blinker. This feature is currently optional and supposed to be used by extensions and applications. If you want to use it, make sure to have blinker installed. • refactored the way URL adapters are created. This process is now fully customiz- able with the create_url_adapter() method. • modules can now register for a subdomain instead of just an URL prefix. This makes it possible to bind a whole module to a configurable subdomain. 30.14 Version 0.5.2 Bugfix Release, released on July 15th 2010 • fixed another issue with loading templates from directories when modules were used. 30.15 Version 0.5.1 Bugfix Release, released on July 6th 2010 • fixes an issue with template loading from directories when modules where used. 283
- 294. 30.16 Version 0.5 Released on July 6th 2010, codename Calvados • fixed a bug with subdomains that was caused by the inability to specify the server name. The server name can now be set with the SERVER_NAME config key. This key is now also used to set the session cookie cross-subdomain wide. • autoescaping is no longer active for all templates. Instead it is only active for .html, .htm, .xml and .xhtml. Inside templates this behavior can be changed with the autoescape tag. • refactored Flask internally. It now consists of more than a single file. • flask.send_file() now emits etags and has the ability to do conditional re- sponses builtin. • (temporarily) dropped support for zipped applications. This was a rarely used feature and led to some confusing behavior. • added support for per-package template and static-file directories. • removed support for create_jinja_loader which is no longer used in 0.5 due to the improved module support. • added a helper function to expose files from any directory. 30.17 Version 0.4 Released on June 18th 2010, codename Rakia • added the ability to register application wide error handlers from modules. • after_request() handlers are now also invoked if the request dies with an ex- ception and an error handling page kicks in. • test client has not the ability to preserve the request context for a little longer. This can also be used to trigger custom requests that do not pop the request stack for testing. • because the Python standard library caches loggers, the name of the logger is configurable now to better support unittests. • added TESTING switch that can activate unittesting helpers. • the logger switches to DEBUG mode now if debug is enabled. 30.18 Version 0.3.1 Bugfix release, released on May 28th 2010 284
- 295. • fixed a error reporting bug with flask.Config.from_envvar() • removed some unused code from flask • release does no longer include development leftover files (.git folder for themes, built documentation in zip and pdf file and some .pyc files) 30.19 Version 0.3 Released on May 28th 2010, codename Schnaps • added support for categories for flashed messages. • the application now configures a logging.Handler and will log request handling exceptions to that logger when not in debug mode. This makes it possible to receive mails on server errors for example. • added support for context binding that does not require the use of the with state- ment for playing in the console. • the request context is now available within the with statement making it possible to further push the request context or pop it. • added support for configurations. 30.20 Version 0.2 Released on May 12th 2010, codename Jägermeister • various bugfixes • integrated JSON support • added get_template_attribute() helper function. • add_url_rule() can now also register a view function. • refactored internal request dispatching. • server listens on 127.0.0.1 by default now to fix issues with chrome. • added external URL support. • added support for send_file() • module support and internal request handling refactoring to better support pluggable applications. • sessions can be set to be permanent now on a per-session basis. • better error reporting on missing secret keys. • added support for Google Appengine. 285
- 296. 30.21 Version 0.1 First public preview release. 286
- 297. CHAPTER THIRTYONE LICENSE Flask is licensed under a three clause BSD License. It basically means: do whatever you want with it as long as the copyright in Flask sticks around, the conditions are not modified and the disclaimer is present. Furthermore you must not use the names of the authors to promote derivatives of the software without written consent. The full license text can be found below (Flask License). For the documentation and artwork different licenses apply. 31.1 Authors Flask is written and maintained by Armin Ronacher and various contributors: 31.1.1 Development Lead • Armin Ronacher <[email protected]> 31.1.2 Patches and Suggestions • Adam Zapletal • Ali Afshar • Chris Edgemon • Chris Grindstaff • Christopher Grebs • Daniel Neuhäuser • Florent Xicluna • Georg Brandl • Justin Quick • Kenneth Reitz 287
- 298. • Marian Sigler • Matt Campell • Matthew Frazier • Michael van Tellingen • Ron DuPlain • Sebastien Estienne • Simon Sapin • Stephane Wirtel • Thomas Schranz • Zhao Xiaohong • Edmond Burnett 31.2 General License Definitions The following section contains the full license texts for Flask and the documentation. • “AUTHORS” hereby refers to all the authors listed in the Authors section. • The “Flask License” applies to all the sourcecode shipped as part of Flask (Flask itself as well as the examples and the unittests) as well as documentation. • The “Flask Artwork License” applies to the project’s Horn-Logo. 31.3 Flask License Copyright (c) 2014 by Armin Ronacher and contributors. See AUTHORS for more details. Some rights reserved. Redistribution and use in source and binary forms of the software as well as docu- mentation, with or without modification, are permitted provided that the following conditions are met: • Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. • Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. • The names of the contributors may not be used to endorse or promote products derived from this software without specific prior written permission. 288
- 299. THIS SOFTWARE AND DOCUMENTATION IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS “AS IS” AND ANY EXPRESS OR IMPLIED WAR- RANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DIS- CLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PRO- CUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE AND DOCUMENTATION, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 31.4 Flask Artwork License Copyright (c) 2010 by Armin Ronacher. Some rights reserved. This logo or a modified version may be used by anyone to refer to the Flask project, but does not indicate endorsement by the project. Redistribution and use in source (the SVG file) and binary forms (rendered PNG files etc.) of the image, with or without modification, are permitted provided that the fol- lowing conditions are met: • Redistributions of source code must retain the above copyright notice and this list of conditions. • The names of the contributors to the Flask software (see AUTHORS) may not be used to endorse or promote products derived from this software without specific prior written permission. Note: we would appreciate that you make the image a link to http://flask.pocoo.org/ if you use it on a web page. 289
- 300. 290
- 301. INDEX Symbols _app_ctx_stack (in module flask), 227 _request_ctx_stack (in module flask), 226 A abort() (in module flask), 213 add_app_template_filter() (flask.Blueprint method), 198 add_app_template_global() (flask.Blueprint method), 198 add_app_template_test() (flask.Blueprint method), 198 add_template_filter() (flask.Flask method), 178 add_template_global() (flask.Flask method), 178 add_template_test() (flask.Flask method), 179 add_url_rule() (flask.Blueprint method), 198 add_url_rule() (flask.blueprints.BlueprintSetupState method), 227 add_url_rule() (flask.Flask method), 179 after_app_request() (flask.Blueprint method), 198 after_request() (flask.Blueprint method), 198 after_request() (flask.Flask method), 180 after_request_funcs (flask.Flask attribute), 180 after_this_request() (in module flask), 214 app (flask.blueprints.BlueprintSetupState attribute), 227 app_context() (flask.Flask method), 180 app_context_processor() (flask.Blueprint method), 198 app_ctx_globals_class (flask.Flask at- tribute), 180 app_errorhandler() (flask.Blueprint method), 199 app_template_filter() (flask.Blueprint method), 199 app_template_global() (flask.Blueprint method), 199 app_template_test() (flask.Blueprint method), 199 app_url_defaults() (flask.Blueprint method), 199 app_url_value_preprocessor() (flask.Blueprint method), 199 AppContext (class in flask.ctx), 226 appcontext_popped (in module flask), 228 appcontext_pushed (in module flask), 228 appcontext_tearing_down (in module flask), 228 args (flask.Request attribute), 202 as_view() (flask.views.View class method), 229 auto_find_instance_path() (flask.Flask method), 180 B base_url (flask.Request attribute), 203 before_app_first_request() (flask.Blueprint method), 199 before_app_request() (flask.Blueprint method), 199 before_first_request() (flask.Flask method), 181 before_first_request_funcs (flask.Flask at- tribute), 181 before_request() (flask.Blueprint method), 199 before_request() (flask.Flask method), 181 291
- 302. before_request_funcs (flask.Flask at- tribute), 181 Blueprint (class in flask), 198 blueprint (flask.blueprints.BlueprintSetupState attribute), 227 blueprint (flask.Request attribute), 203 blueprints (flask.Flask attribute), 181 BlueprintSetupState (class in flask.blueprints), 227 C Config (class in flask), 222 config (flask.Flask attribute), 181 context_processor() (flask.Blueprint method), 199 context_processor() (flask.Flask method), 181 cookies (flask.Request attribute), 202 copy() (flask.ctx.RequestContext method), 226 copy_current_request_context() (in mod- ule flask), 212 create_global_jinja_loader() (flask.Flask method), 181 create_jinja_environment() (flask.Flask method), 181 create_url_adapter() (flask.Flask method), 182 current_app (in module flask), 211 D data (flask.Request attribute), 202 data (flask.Response attribute), 205 debug (flask.Flask attribute), 182 debug_log_format (flask.Flask attribute), 182 decorators (flask.views.View attribute), 230 default() (flask.json.JSONEncoder method), 221 default_config (flask.Flask attribute), 182 digest_method() (flask.sessions.SecureCookieSessionInterface method), 209 dispatch_request() (flask.Flask method), 182 dispatch_request() (flask.views.View method), 230 do_teardown_appcontext() (flask.Flask method), 182 do_teardown_request() (flask.Flask method), 182 dump() (in module flask.json), 221 dumps() (in module flask.json), 220 E enable_modules (flask.Flask attribute), 183 endpoint (flask.Request attribute), 203 endpoint() (flask.Blueprint method), 199 endpoint() (flask.Flask method), 183 environ (flask.Request attribute), 203 environment variable FLASKR_SETTINGS, 30 YOURAPPLICATION_SETTINGS, 62 error_handler_spec (flask.Flask attribute), 183 errorhandler() (flask.Blueprint method), 200 errorhandler() (flask.Flask method), 183 escape() (flask.Markup class method), 218 escape() (in module flask), 217 extensions (flask.Flask attribute), 184 F files (flask.Request attribute), 202 first_registration (flask.blueprints.BlueprintSetupState attribute), 227 flash() (in module flask), 218 Flask (class in flask), 177 flask (module), 177 flask.ext (in module flask), 224 flask.json (module), 219 flask.signals.Namespace (built-in class), 228 FlaskClient (class in flask.testing), 210 FLASKR_SETTINGS, 30 form (flask.Request attribute), 202 from_envvar() (flask.Config method), 223 from_object() (flask.Config method), 223 from_pyfile() (flask.Config method), 224 full_dispatch_request() (flask.Flask method), 184 G g (in module flask), 210 292
- 303. get_cookie_domain() (flask.sessions.SessionInterface method), 207 get_cookie_httponly() (flask.sessions.SessionInterface method), 207 get_cookie_path() (flask.sessions.SessionInterface method), 207 get_cookie_secure() (flask.sessions.SessionInterface method), 207 get_expiration_time() (flask.sessions.SessionInterface method), 208 get_flashed_messages() (in module flask), 218 get_json() (flask.Request method), 203 get_send_file_max_age() (flask.Blueprint method), 200 get_send_file_max_age() (flask.Flask method), 184 get_template_attribute() (in module flask), 222 got_first_request (flask.Flask attribute), 184 got_request_exception (in module flask), 228 H handle_exception() (flask.Flask method), 185 handle_http_exception() (flask.Flask method), 185 handle_url_build_error() (flask.Flask method), 185 handle_user_exception() (flask.Flask method), 185 has_app_context() (in module flask), 212 has_request_context() (in module flask), 211 has_static_folder (flask.Blueprint at- tribute), 200 has_static_folder (flask.Flask attribute), 185 headers (flask.Request attribute), 202 headers (flask.Response attribute), 205 I init_jinja_globals() (flask.Flask method), 185 inject_url_defaults() (flask.Flask method), 185 instance_path (flask.Flask attribute), 185 is_json (flask.Request attribute), 204 is_null_session() (flask.sessions.SessionInterface method), 208 is_xhr (flask.Request attribute), 203 J jinja_env (flask.Flask attribute), 185 jinja_loader (flask.Blueprint attribute), 200 jinja_loader (flask.Flask attribute), 185 jinja_options (flask.Flask attribute), 186 json (flask.Request attribute), 204 json_decoder (flask.Flask attribute), 186 json_encoder (flask.Flask attribute), 186 JSONDecoder (class in flask.json), 221 JSONEncoder (class in flask.json), 221 jsonify() (in module flask.json), 220 K key_derivation (flask.sessions.SecureCookieSessionInterface attribute), 209 L load() (in module flask.json), 221 loads() (in module flask.json), 221 log_exception() (flask.Flask method), 186 logger (flask.Flask attribute), 186 logger_name (flask.Flask attribute), 186 M make_config() (flask.Flask method), 186 make_default_options_response() (flask.Flask method), 186 make_null_session() (flask.Flask method), 187 make_null_session() (flask.sessions.SessionInterface method), 208 make_response() (flask.Flask method), 187 293
- 304. make_response() (in module flask), 214 make_setup_state() (flask.Blueprint method), 200 Markup (class in flask), 217 match_request() (flask.ctx.RequestContext method), 226 max_content_length (flask.Request at- tribute), 204 message_flashed (in module flask), 228 method (flask.Request attribute), 203 methods (flask.views.View attribute), 230 MethodView (class in flask.views), 230 mimetype (flask.Response attribute), 205 modified (flask.session attribute), 206 modified (flask.sessions.SessionMixin at- tribute), 209 module (flask.Request attribute), 204 N name (flask.Flask attribute), 187 new (flask.session attribute), 206 new (flask.sessions.SessionMixin at- tribute), 209 null_session_class (flask.sessions.SessionInterface attribute), 208 NullSession (class in flask.sessions), 209 O on_json_loading_failed() (flask.Request method), 204 open_instance_resource() (flask.Flask method), 187 open_resource() (flask.Blueprint method), 200 open_resource() (flask.Flask method), 187 open_session() (flask.Flask method), 188 open_session() (flask.sessions.SessionInterface method), 208 options (flask.blueprints.BlueprintSetupState attribute), 227 P path (flask.Request attribute), 203 permanent (flask.session attribute), 206 permanent (flask.sessions.SessionMixin attribute), 209 permanent_session_lifetime (flask.Flask attribute), 188 pickle_based (flask.sessions.SessionInterface attribute), 208 pop() (flask.ctx.AppContext method), 226 pop() (flask.ctx.RequestContext method), 226 preprocess_request() (flask.Flask method), 188 preserve_context_on_exception (flask.Flask attribute), 188 process_response() (flask.Flask method), 189 propagate_exceptions (flask.Flask at- tribute), 189 push() (flask.ctx.AppContext method), 227 push() (flask.ctx.RequestContext method), 226 Python Enhancement Proposals PEP 8, 261 R record() (flask.Blueprint method), 201 record_once() (flask.Blueprint method), 201 redirect() (in module flask), 213 register() (flask.Blueprint method), 201 register_blueprint() (flask.Flask method), 189 register_error_handler() (flask.Flask method), 189 register_module() (flask.Flask method), 189 render_template() (in module flask), 221 render_template_string() (in module flask), 222 Request (class in flask), 202 request (class in flask), 205 request_class (flask.Flask attribute), 189 request_context() (flask.Flask method), 189 request_finished (in module flask), 228 request_started (in module flask), 228 request_tearing_down (in module flask), 228 RequestContext (class in flask.ctx), 225 294
- 305. Response (class in flask), 205 response_class (flask.Flask attribute), 190 RFC RFC 822, 219 route() (flask.Blueprint method), 201 route() (flask.Flask method), 190 routing_exception (flask.Request at- tribute), 204 run() (flask.Flask method), 190 S safe_join() (in module flask), 216 salt (flask.sessions.SecureCookieSessionInterface attribute), 209 save_session() (flask.Flask method), 191 save_session() (flask.sessions.SessionInterface method), 208 script_root (flask.Request attribute), 203 secret_key (flask.Flask attribute), 191 SecureCookieSession (class in flask.sessions), 209 SecureCookieSessionInterface (class in flask.sessions), 209 select_jinja_autoescape() (flask.Flask method), 191 send_file() (in module flask), 215 send_from_directory() (in module flask), 216 send_static_file() (flask.Blueprint method), 201 send_static_file() (flask.Flask method), 191 serializer (flask.sessions.SecureCookieSessionInterface attribute), 209 session (class in flask), 206 session_class (flask.sessions.SecureCookieSessionInterface attribute), 209 session_cookie_name (flask.Flask at- tribute), 192 session_interface (flask.Flask attribute), 192 session_json_serializer (in module flask.sessions), 209 session_transaction() (flask.testing.FlaskClient method), 210 SessionInterface (class in flask.sessions), 207 SessionMixin (class in flask.sessions), 209 set_cookie() (flask.Response method), 205 should_ignore_error() (flask.Flask method), 192 should_set_cookie() (flask.sessions.SessionInterface method), 208 signal() (flask.signals.Namespace method), 229 signals_available (in module flask), 227 status (flask.Response attribute), 205 status_code (flask.Response attribute), 205 stream (flask.Request attribute), 202 stream_with_context() (in module flask), 224 striptags() (flask.Markup method), 218 subdomain (flask.blueprints.BlueprintSetupState attribute), 227 T teardown_app_request() (flask.Blueprint method), 201 teardown_appcontext() (flask.Flask method), 192 teardown_appcontext_funcs (flask.Flask attribute), 192 teardown_request() (flask.Blueprint method), 201 teardown_request() (flask.Flask method), 192 teardown_request_funcs (flask.Flask at- tribute), 193 template_context_processors (flask.Flask attribute), 193 template_filter() (flask.Flask method), 193 template_global() (flask.Flask method), 194 template_rendered (in module flask), 228 template_test() (flask.Flask method), 194 test_client() (flask.Flask method), 194 test_client_class (flask.Flask attribute), 195 test_request_context() (flask.Flask method), 195 testing (flask.Flask attribute), 195 295
- 306. trap_http_exception() (flask.Flask method), 195 U unescape() (flask.Markup method), 218 update_template_context() (flask.Flask method), 195 url (flask.Request attribute), 203 url_build_error_handlers (flask.Flask at- tribute), 196 url_default_functions (flask.Flask at- tribute), 196 url_defaults (flask.blueprints.BlueprintSetupState attribute), 227 url_defaults() (flask.Blueprint method), 202 url_defaults() (flask.Flask method), 196 url_for() (in module flask), 212 url_map (flask.Flask attribute), 196 url_prefix (flask.blueprints.BlueprintSetupState attribute), 227 url_root (flask.Request attribute), 203 url_rule (flask.Request attribute), 204 url_rule_class (flask.Flask attribute), 196 url_value_preprocessor() (flask.Blueprint method), 202 url_value_preprocessor() (flask.Flask method), 196 url_value_preprocessors (flask.Flask at- tribute), 197 use_x_sendfile (flask.Flask attribute), 197 V values (flask.Request attribute), 202 View (class in flask.views), 229 view_args (flask.Request attribute), 204 view_functions (flask.Flask attribute), 197 W wsgi_app() (flask.Flask method), 197 Y YOURAPPLICATION_SETTINGS, 62 296