Split Date/Time picker values from IME processing since date/time related form values have been com…

Split Date/Time picker values from IME processing since date/time related form values have been completly split from text related fields in WebKit and keyboard interaction is no longer needed for those elements.


BUG=143540
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=177405

[Miguel Garcia, 2013-01-10 15:27:44]

Adding 

bulach and ollilan as overall reviewers
jschuh@ for security review of the new IPC message
joi@ for Owner approval of the gypi files and the new interface in
public/browser/date_time_chooser.h

[bulach, 2013-01-10 15:58:35]

great stuff, thanks miguel! mostly nits and some naming suggestions, specially
for the android bits.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
File content/browser/android/date_time_chooser_android.cc (right):

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:20: // Internal
communicator class, talks back to the renderer and communicate
how about "Updates date/time via IPC to the RenderView"?

and if you make it an inner class of "DateTimeChooserAndroid", this could be
just "IPCSender" ? note that RenderViewHostObserver is already a IPC::Listener
and IPC::Sender, so we could re-use the term rather than introducing a new
"communicator" concept.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:47: 
nit: remove extra \n

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:54: // static
since this is for DateTimeChooser rather than DateTimeChooseAndroid, move it up
before the ctor above..

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:56: // Initialize platform
specific type values date/time value forms.
I suppose 56-62 could be moved to 106, after registration?

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:68: int textInputTypeDate,
int textInputTypeDateTime,
nit: hacker_style

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:81: if (communicator_)
nit: at this point, communicator_ has got to be set, right? so just call it
directly (ditto for 86 below)

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
File content/browser/android/date_time_chooser_android.h (right):

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:13: namespace content {
nit: add a \n

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:14: class RenderViewHost;
nit: sort order

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:16: class
DateTimeCommunicator;
nit: move this declaration inside the private block below so it doesn't pollute
the content namespace, and see more comments re: this name.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:18: // Glue between renderer
and platform specific browser implementations
hmm, this is not really a glue for the renderer, perhaps:

// Android implementation for DateTimeChooser dialogs.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:24: 
nit:
  // DateTimeChooser implementation:

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:38: int textInputTypeDate,
int textInputTypeDateTime,
nit: hacker_style rather than webkitStyle

https://codereview.chromium.org/11783088/diff/2001/content/browser/web_conten...
File content/browser/web_contents/web_contents_impl.cc (right):

https://codereview.chromium.org/11783088/diff/2001/content/browser/web_conten...
content/browser/web_contents/web_contents_impl.cc:1197:
date_time_chooser_.reset(DateTimeChooser::Create());
nit: unindent

https://codereview.chromium.org/11783088/diff/2001/content/browser/web_conten...
content/browser/web_contents/web_contents_impl.cc:2283: if (date_time_chooser_
!= NULL)
can this ever be null?

https://codereview.chromium.org/11783088/diff/2001/content/common/view_messag...
File content/common/view_messages.h (right):

https://codereview.chromium.org/11783088/diff/2001/content/common/view_messag...
content/common/view_messages.h:1988: int, /* dialog type */
nit: I guess it fits aligned with the parens? also, put the "," after the
comment

https://codereview.chromium.org/11783088/diff/2001/content/public/android/jav...
File
content/public/android/java/src/org/chromium/content/browser/DateTimeChooserAndroid.java
(right):

https://codereview.chromium.org/11783088/diff/2001/content/public/android/jav...
content/public/android/java/src/org/chromium/content/browser/DateTimeChooserAndroid.java:1:
package org.chromium.content.browser;
nit: missing Copyright header

https://codereview.chromium.org/11783088/diff/2001/content/public/android/jav...
content/public/android/java/src/org/chromium/content/browser/DateTimeChooserAndroid.java:12:
public class DateTimeChooserAndroid {
nit: make the class and its ctor package-protected (they're only called by
native, right?

https://codereview.chromium.org/11783088/diff/2001/content/public/browser/dat...
File content/public/browser/date_time_chooser.h (right):

https://codereview.chromium.org/11783088/diff/2001/content/public/browser/dat...
content/public/browser/date_time_chooser.h:10: namespace content {
nit: add a \n and also sort order..

https://codereview.chromium.org/11783088/diff/2001/content/public/browser/dat...
content/public/browser/date_time_chooser.h:14: // Abstraction object for date
time pickers.
// An interface for showing date/time picker dialogs.

https://codereview.chromium.org/11783088/diff/2001/content/public/browser/dat...
content/public/browser/date_time_chooser.h:23: int type, const std::string&
value) = 0;
nit:
private:
 DISALLOW_COPY_AND_ASSIGN(DateTimeChooser);

[Jói, 2013-01-10 18:39:11]

-joi, +jam

John wants to review new APIs that are added.

Cheers,
Jói

[jam, 2013-01-11 01:17:32]

why is there a new header in content/public if it's only used inside content?

[Miguel Garcia, 2013-01-11 14:59:36]

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
File content/browser/android/date_time_chooser_android.cc (right):

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:20: // Internal
communicator class, talks back to the renderer and communicate
On 2013/01/10 15:58:35, bulach wrote:
> how about "Updates date/time via IPC to the RenderView"?
> 
> and if you make it an inner class of "DateTimeChooserAndroid", this could be
> just "IPCSender" ? note that RenderViewHostObserver is already a IPC::Listener
> and IPC::Sender, so we could re-use the term rather than introducing a new
> "communicator" concept.

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:47: 
On 2013/01/10 15:58:35, bulach wrote:
> nit: remove extra \n

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:54: // static
On 2013/01/10 15:58:35, bulach wrote:
> since this is for DateTimeChooser rather than DateTimeChooseAndroid, move it
up
> before the ctor above..

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:56: // Initialize platform
specific type values date/time value forms.
On 2013/01/10 15:58:35, bulach wrote:
> I suppose 56-62 could be moved to 106, after registration?

Great idea!

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:68: int textInputTypeDate,
int textInputTypeDateTime,
On 2013/01/10 15:58:35, bulach wrote:
> nit: hacker_style

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.cc:81: if (communicator_)
On 2013/01/10 15:58:35, bulach wrote:
> nit: at this point, communicator_ has got to be set, right? so just call it
> directly (ditto for 86 below)

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
File content/browser/android/date_time_chooser_android.h (right):

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:13: namespace content {
On 2013/01/10 15:58:35, bulach wrote:
> nit: add a \n

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:14: class RenderViewHost;
On 2013/01/10 15:58:35, bulach wrote:
> nit: sort order

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:16: class
DateTimeCommunicator;
On 2013/01/10 15:58:35, bulach wrote:
> nit: move this declaration inside the private block below so it doesn't
pollute
> the content namespace, and see more comments re: this name.

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:18: // Glue between renderer
and platform specific browser implementations
Changed, this makes more sense as an explanation agreed.

On 2013/01/10 15:58:35, bulach wrote:
> hmm, this is not really a glue for the renderer, perhaps:
> 
> // Android implementation for DateTimeChooser dialogs.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:24: 
On 2013/01/10 15:58:35, bulach wrote:
> nit:
>   // DateTimeChooser implementation:

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/android/da...
content/browser/android/date_time_chooser_android.h:38: int textInputTypeDate,
int textInputTypeDateTime,
On 2013/01/10 15:58:35, bulach wrote:
> nit: hacker_style rather than webkitStyle

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/web_conten...
File content/browser/web_contents/web_contents_impl.cc (right):

https://codereview.chromium.org/11783088/diff/2001/content/browser/web_conten...
content/browser/web_contents/web_contents_impl.cc:1197:
date_time_chooser_.reset(DateTimeChooser::Create());
On 2013/01/10 15:58:35, bulach wrote:
> nit: unindent

Done.

https://codereview.chromium.org/11783088/diff/2001/content/browser/web_conten...
content/browser/web_contents/web_contents_impl.cc:2283: if (date_time_chooser_
!= NULL)
On 2013/01/10 15:58:35, bulach wrote:
> can this ever be null?

Well not right now since this call is compiled out for other platforms but it
seems nice to be defensive here.

https://codereview.chromium.org/11783088/diff/2001/content/common/view_messag...
File content/common/view_messages.h (right):

https://codereview.chromium.org/11783088/diff/2001/content/common/view_messag...
content/common/view_messages.h:1988: int, /* dialog type */
On 2013/01/10 15:58:35, bulach wrote:
> nit: I guess it fits aligned with the parens? also, put the "," after the
> comment

Done.

https://codereview.chromium.org/11783088/diff/2001/content/public/android/jav...
File
content/public/android/java/src/org/chromium/content/browser/DateTimeChooserAndroid.java
(right):

https://codereview.chromium.org/11783088/diff/2001/content/public/android/jav...
content/public/android/java/src/org/chromium/content/browser/DateTimeChooserAndroid.java:1:
package org.chromium.content.browser;
On 2013/01/10 15:58:35, bulach wrote:
> nit: missing Copyright header

Done.

https://codereview.chromium.org/11783088/diff/2001/content/public/android/jav...
content/public/android/java/src/org/chromium/content/browser/DateTimeChooserAndroid.java:12:
public class DateTimeChooserAndroid {
Correct. Done

On 2013/01/10 15:58:35, bulach wrote:
> nit: make the class and its ctor package-protected (they're only called by
> native, right?

https://codereview.chromium.org/11783088/diff/2001/content/public/browser/dat...
File content/public/browser/date_time_chooser.h (right):

https://codereview.chromium.org/11783088/diff/2001/content/public/browser/dat...
content/public/browser/date_time_chooser.h:10: namespace content {
On 2013/01/10 15:58:35, bulach wrote:
> nit: add a \n and also sort order..

Done.

https://codereview.chromium.org/11783088/diff/2001/content/public/browser/dat...
content/public/browser/date_time_chooser.h:14: // Abstraction object for date
time pickers.
On 2013/01/10 15:58:35, bulach wrote:
> // An interface for showing date/time picker dialogs.

Done.

https://codereview.chromium.org/11783088/diff/2001/content/public/browser/dat...
content/public/browser/date_time_chooser.h:23: int type, const std::string&
value) = 0;
On 2013/01/10 15:58:35, bulach wrote:
> nit:
> private:
>  DISALLOW_COPY_AND_ASSIGN(DateTimeChooser);

Done.

[Miguel Garcia, 2013-01-11 15:00:37]

On 2013/01/11 01:17:32, John Abd-El-Malek wrote:
> why is there a new header in content/public if it's only used inside content?

That's a good point, I was mimicking what the color_chooser.h does but it seems
we should also move that one out of public. I moved date_time_chooser.h to
content/browser/ now.

[jam, 2013-01-11 23:47:18]

On 2013/01/11 15:00:37, Miguel Garcia wrote:
> On 2013/01/11 01:17:32, John Abd-El-Malek wrote:
> > why is there a new header in content/public if it's only used inside
content?
> 
> That's a good point, I was mimicking what the color_chooser.h does but it
seems
> we should also move that one out of public. I moved date_time_chooser.h to
> content/browser/ now.

That's because color_chooser.h is used from chrome

https://codereview.chromium.org/11783088/diff/12001/content/browser/date_time...
File content/browser/date_time_chooser.h (right):

https://codereview.chromium.org/11783088/diff/12001/content/browser/date_time...
content/browser/date_time_chooser.h:1: // Copyright (c) 2013 The Chromium
Authors. All rights reserved.
why is there an interface now? i.e. instead of just the implementation only?

https://codereview.chromium.org/11783088/diff/12001/content/browser/web_conte...
File content/browser/web_contents/web_contents_impl.cc (right):

https://codereview.chromium.org/11783088/diff/12001/content/browser/web_conte...
content/browser/web_contents/web_contents_impl.cc:723: #if defined(OS_ANDROID)
nit: can you move this ifdef to the bottom of IPC_MESSAGE_HANDLER section?
that's what we've been trying to do in other places (move ifdef section to the
bottom). it makes things more readable.

[Miguel Garcia, 2013-01-14 16:39:39]

Thanks for the review John

https://codereview.chromium.org/11783088/diff/12001/content/browser/date_time...
File content/browser/date_time_chooser.h (right):

https://codereview.chromium.org/11783088/diff/12001/content/browser/date_time...
content/browser/date_time_chooser.h:1: // Copyright (c) 2013 The Chromium
Authors. All rights reserved.
On 2013/01/11 23:47:18, John Abd-El-Malek wrote:
> why is there an interface now? i.e. instead of just the implementation only?


Well, WebContentsImpl holds a member of this class and it did not look right to
add a platform specific implementation directly here (even if it's ifdef-ed for
android). Or perhaps are you asking to have empty implementations of the methods
instead of pure virtual ones? I will change the pure virtual method to an empty
implementation just in case.

https://codereview.chromium.org/11783088/diff/12001/content/browser/web_conte...
File content/browser/web_contents/web_contents_impl.cc (right):

https://codereview.chromium.org/11783088/diff/12001/content/browser/web_conte...
content/browser/web_contents/web_contents_impl.cc:723: #if defined(OS_ANDROID)
On 2013/01/11 23:47:18, John Abd-El-Malek wrote:
> nit: can you move this ifdef to the bottom of IPC_MESSAGE_HANDLER section?
> that's what we've been trying to do in other places (move ifdef section to the
> bottom). it makes things more readable.

Done.

[jam, 2013-01-14 17:48:17]

https://codereview.chromium.org/11783088/diff/12001/content/browser/date_time...
File content/browser/date_time_chooser.h (right):

https://codereview.chromium.org/11783088/diff/12001/content/browser/date_time...
content/browser/date_time_chooser.h:1: // Copyright (c) 2013 The Chromium
Authors. All rights reserved.
On 2013/01/14 16:39:40, Miguel Garcia wrote:
> On 2013/01/11 23:47:18, John Abd-El-Malek wrote:
> > why is there an interface now? i.e. instead of just the implementation only?
> 
> 
> Well, WebContentsImpl holds a member of this class and it did not look right
to
> add a platform specific implementation directly here (even if it's ifdef-ed
for
> android).

I don't understand the aversion :) I think if we don't need an interface, better
to have less layers.

> Or perhaps are you asking to have empty implementations of the methods
> instead of pure virtual ones? I will change the pure virtual method to an
empty
> implementation just in case.
>

[Miguel Garcia, 2013-01-14 18:46:38]

https://codereview.chromium.org/11783088/diff/12001/content/browser/date_time...
File content/browser/date_time_chooser.h (right):

https://codereview.chromium.org/11783088/diff/12001/content/browser/date_time...
content/browser/date_time_chooser.h:1: // Copyright (c) 2013 The Chromium
Authors. All rights reserved.
Well if you are ok with it I am ok with it :) I have also ifdef-ed the import on
web_contents_impl since it is now platform specific.


On 2013/01/14 17:48:17, John Abd-El-Malek wrote:
> On 2013/01/14 16:39:40, Miguel Garcia wrote:
> > On 2013/01/11 23:47:18, John Abd-El-Malek wrote:
> > > why is there an interface now? i.e. instead of just the implementation
only?
> > 
> > 
> > Well, WebContentsImpl holds a member of this class and it did not look right
> to
> > add a platform specific implementation directly here (even if it's ifdef-ed
> for
> > android).
> 
> I don't understand the aversion :) I think if we don't need an interface,
better
> to have less layers.
> 
> > Or perhaps are you asking to have empty implementations of the methods
> > instead of pure virtual ones? I will change the pure virtual method to an
> empty
> > implementation just in case.
> > 
>

[Miguel Garcia, 2013-01-14 18:56:59]

Justin, can you rubber stamp the new IPC message?

On 2013/01/14 18:46:38, Miguel Garcia wrote:
>
https://codereview.chromium.org/11783088/diff/12001/content/browser/date_time...
> File content/browser/date_time_chooser.h (right):
> 
>
https://codereview.chromium.org/11783088/diff/12001/content/browser/date_time...
> content/browser/date_time_chooser.h:1: // Copyright (c) 2013 The Chromium
> Authors. All rights reserved.
> Well if you are ok with it I am ok with it :) I have also ifdef-ed the import
on
> web_contents_impl since it is now platform specific.
> 
> 
> On 2013/01/14 17:48:17, John Abd-El-Malek wrote:
> > On 2013/01/14 16:39:40, Miguel Garcia wrote:
> > > On 2013/01/11 23:47:18, John Abd-El-Malek wrote:
> > > > why is there an interface now? i.e. instead of just the implementation
> only?
> > > 
> > > 
> > > Well, WebContentsImpl holds a member of this class and it did not look
right
> > to
> > > add a platform specific implementation directly here (even if it's
ifdef-ed
> > for
> > > android).
> > 
> > I don't understand the aversion :) I think if we don't need an interface,
> better
> > to have less layers.
> > 
> > > Or perhaps are you asking to have empty implementations of the methods
> > > instead of pure virtual ones? I will change the pure virtual method to an
> > empty
> > > implementation just in case.
> > > 
> >

[jschuh, 2013-01-14 21:06:45]

@palmer is really more the Android security guy, so I'd rather defer to him.

[jam, 2013-01-14 21:10:55]

lgtm

https://codereview.chromium.org/11783088/diff/24005/content/common/view_messa...
File content/common/view_messages.h (right):

https://codereview.chromium.org/11783088/diff/24005/content/common/view_messa...
content/common/view_messages.h:1995: int /* dialog type */,
nit: here and below, the convention is to put the parameter name in the
dispatching function inside the comment (so type, and value)

https://codereview.chromium.org/11783088/diff/24005/content/content_browser.gypi
File content/content_browser.gypi (right):

https://codereview.chromium.org/11783088/diff/24005/content/content_browser.g...
content/content_browser.gypi:215:
'browser/android/date_time_chooser_android.cc',
nit: isn't having _android in the filename redundant for files in an android
directory?

[palmer, 2013-01-14 23:53:35]

https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
File content/browser/android/date_time_chooser_android.cc (right):

https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
content/browser/android/date_time_chooser_android.cc:68: JNIEnv* env, jobject,
jstring text) {
My security and correctness instinct is to receive an int/long of (milli)seconds
since epoch here, or a DateTime object, and generate the string representation
in the body of this function. (After all, you have to convert the representation
to string16 anyway.)

See related comment in the IPC messages file.

https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
content/browser/android/date_time_chooser_android.cc:82:
base::android::ScopedJavaLocalRef<jstring>  java_value =
NIT: Two spaces between type and identifier.

https://codereview.chromium.org/11783088/diff/24005/content/common/view_messa...
File content/common/view_messages.h (right):

https://codereview.chromium.org/11783088/diff/24005/content/common/view_messa...
content/common/view_messages.h:1996: std::string /* initial value */)
I'd rather send plain old data (e.g. an int of seconds since the epoch), or a
specific Date/Time/DateTime object, rather than a pre-baked *representation* of
the value. From the sender side of the code, we don't know how the recipient is
going to interpret/interpolate this value — what if it's a (cross-site
scriptable!) WebView someday, instead of a native Android GUI widget? What if
the string is formatted for the wrong locale? That kind of thing.

So, send the specific type you *mean*. That also has the benefit of bounding the
size of the sent object, which is important in IPC.

[Miguel Garcia, 2013-01-16 17:31:40]

This actually makes a lot of sense, we already do the parsing of the string on
the java side anyway  (see InputDialogContainer.java) so porting it and do it
earlier on the C++ side before sending it out via IPC makes a lot of sense.

If at all possible however I'd like to do this on a different cl. This would be
a non trivial change that will make this cl much bigger and submitting as is
will not fundamentally  anything right now since the date/time information is
already sent  as a string in IPC (in the TextInputStateChanged). More over, once
I change it I'd like to do so in both directions (so we would also change the
ReplaceDateTime format as well).


On 2013/01/14 23:53:35, Chris P. wrote:
>
https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
> File content/browser/android/date_time_chooser_android.cc (right):
> 
>
https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
> content/browser/android/date_time_chooser_android.cc:68: JNIEnv* env, jobject,
> jstring text) {
> My security and correctness instinct is to receive an int/long of
(milli)seconds
> since epoch here, or a DateTime object, and generate the string representation
> in the body of this function. (After all, you have to convert the
representation
> to string16 anyway.)
> 
> See related comment in the IPC messages file.
> 
>
https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
> content/browser/android/date_time_chooser_android.cc:82:
> base::android::ScopedJavaLocalRef<jstring>  java_value =
> NIT: Two spaces between type and identifier.
> 
>
https://codereview.chromium.org/11783088/diff/24005/content/common/view_messa...
> File content/common/view_messages.h (right):
> 
>
https://codereview.chromium.org/11783088/diff/24005/content/common/view_messa...
> content/common/view_messages.h:1996: std::string /* initial value */)
> I'd rather send plain old data (e.g. an int of seconds since the epoch), or a
> specific Date/Time/DateTime object, rather than a pre-baked *representation*
of
> the value. From the sender side of the code, we don't know how the recipient
is
> going to interpret/interpolate this value — what if it's a (cross-site
> scriptable!) WebView someday, instead of a native Android GUI widget? What if
> the string is formatted for the wrong locale? That kind of thing.
> 
> So, send the specific type you *mean*. That also has the benefit of bounding
the
> size of the sent object, which is important in IPC.

[palmer, 2013-01-16 19:58:39]

> If at all possible however I'd like to do this on a different cl.

What would be the timeline for that second CL? If it's soon, I'll provisionally
approve this one. Please include me as a reviewer on the second CL, too.

[miguelg, 2013-01-16 20:12:43]

I have started it already and will be sending it for review this week. I
will certainly include you so you can review the new measages
On 16 Jan 2013 19:58, <[email protected]> wrote:

> If at all possible however I'd like to do this on a different cl.
>>
>
> What would be the timeline for that second CL? If it's soon, I'll
> provisionally
> approve this one. Please include me as a reviewer on the second CL, too.
>
>
https://codereview.chromium.**org/11783088/<https://codereview.chromium.org/1...
>

[palmer, 2013-01-17 01:39:35]

> I have started it already and will be sending it for review this week. I
> will certainly include you so you can review the new measages

Awesome, thanks. LGTM. (Sorry I didn't see this earlier! got busy with stuff.)

[bulach, 2013-01-17 10:32:21]

lgtm for android, some nits:

https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
File content/browser/android/date_time_chooser_android.h (right):

https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
content/browser/android/date_time_chooser_android.h:22: virtual
~DateTimeChooserAndroid();
no need for virtual since it's a concrete, final class.

https://codereview.chromium.org/11783088/diff/24005/content/browser/web_conte...
File content/browser/web_contents/web_contents_impl.cc (right):

https://codereview.chromium.org/11783088/diff/24005/content/browser/web_conte...
content/browser/web_contents/web_contents_impl.cc:2286: if (date_time_chooser_
!= NULL)
I think this can't ever be NULL at this point, right?
if so, remove the test..

https://codereview.chromium.org/11783088/diff/24005/content/public/android/ja...
File
content/public/android/java/src/org/chromium/content/browser/DateTimeChooserAndroid.java
(right):

https://codereview.chromium.org/11783088/diff/24005/content/public/android/ja...
content/public/android/java/src/org/chromium/content/browser/DateTimeChooserAndroid.java:21:
DateTimeChooserAndroid(Context context,
nit: private here and maybe on the @CalledByNative methods as well..

[Miguel Garcia, 2013-01-17 12:17:40]

Thanks for the review!

https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
File content/browser/android/date_time_chooser_android.cc (right):

https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
content/browser/android/date_time_chooser_android.cc:82:
base::android::ScopedJavaLocalRef<jstring>  java_value =
On 2013/01/14 23:53:35, Chris P. wrote:
> NIT: Two spaces between type and identifier.

Done.

https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
File content/browser/android/date_time_chooser_android.h (right):

https://codereview.chromium.org/11783088/diff/24005/content/browser/android/d...
content/browser/android/date_time_chooser_android.h:22: virtual
~DateTimeChooserAndroid();
On 2013/01/17 10:32:21, bulach wrote:
> no need for virtual since it's a concrete, final class.

Done.

https://codereview.chromium.org/11783088/diff/24005/content/browser/web_conte...
File content/browser/web_contents/web_contents_impl.cc (right):

https://codereview.chromium.org/11783088/diff/24005/content/browser/web_conte...
content/browser/web_contents/web_contents_impl.cc:2286: if (date_time_chooser_
!= NULL)
On 2013/01/17 10:32:21, bulach wrote:
> I think this can't ever be NULL at this point, right?
> if so, remove the test..

Done.

https://codereview.chromium.org/11783088/diff/24005/content/common/view_messa...
File content/common/view_messages.h (right):

https://codereview.chromium.org/11783088/diff/24005/content/common/view_messa...
content/common/view_messages.h:1995: int /* dialog type */,
On 2013/01/14 21:10:55, John Abd-El-Malek wrote:
> nit: here and below, the convention is to put the parameter name in the
> dispatching function inside the comment (so type, and value)

Done.

https://codereview.chromium.org/11783088/diff/24005/content/common/view_messa...
content/common/view_messages.h:1996: std::string /* initial value */)
As agreed, I'll do this in a separate patch.

On 2013/01/14 23:53:35, Chris P. wrote:
> I'd rather send plain old data (e.g. an int of seconds since the epoch), or a
> specific Date/Time/DateTime object, rather than a pre-baked *representation*
of
> the value. From the sender side of the code, we don't know how the recipient
is
> going to interpret/interpolate this value — what if it's a (cross-site
> scriptable!) WebView someday, instead of a native Android GUI widget? What if
> the string is formatted for the wrong locale? That kind of thing.
> 
> So, send the specific type you *mean*. That also has the benefit of bounding
the
> size of the sent object, which is important in IPC.

https://codereview.chromium.org/11783088/diff/24005/content/public/android/ja...
File
content/public/android/java/src/org/chromium/content/browser/DateTimeChooserAndroid.java
(right):

https://codereview.chromium.org/11783088/diff/24005/content/public/android/ja...
content/public/android/java/src/org/chromium/content/browser/DateTimeChooserAndroid.java:21:
DateTimeChooserAndroid(Context context,
On 2013/01/17 10:32:21, bulach wrote:
> nit: private here and maybe on the @CalledByNative methods as well..

Done.

[commit-bot: I haz the power, 2013-01-17 12:18:01]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/11783088/41001

[commit-bot: I haz the power, 2013-01-17 14:01:12]

Change committed as 177405