SAP NetWeaver Gateway - RFC & BOR Generators

SAP NetWeaver Gateway
RFC & BOR Generators (Read Only)
May 2012

A Branded Service provided by SAP Customer Solution Adoption

Objectives

At the end of this module, you will be able to:
 Create a read-only Gateway Service based on a Remote Function
Module or an Object in the Business Object Repository (BOR)
 Expose a Gateway Service for consumption
 Test your Gateway Service from a browser

© 2012 SAP AG. All rights reserved. 2

Agenda

 Remote Function Calls and Business Object Repository Overview
 Understanding Gateway Data and Consumption Models
 Developing Gateway Service based on a Remote Function Module or an Object
in the Business Object Repository
1. Create a Gateway Data Model in the ABAP Development Workbench
2. Map your selected RFMs to Gateway Service operations using the Gateway Model
Generator
3. Define the interface of the Gateway Service Operation based on the fields in the RFM
4. Generate the Gateway Data Model
5. Associate the Gateway Data Model with a Gateway Consumption Model
6. Associate the Gateway Service with a System Alias
7. Test your Gateway Service in a browser


Agenda

Generator


Overview – Remote Function Modules

A Remote Function Module (RFM) is any
ABAP Function Module that has had its
Remote-Enabled flag switched on.
Activating this flag makes the function
module accessible from outside the SAP
system using the Remote Function Call
(RFC) protocol.

Important!
All SAP Business APIs (BAPIs) are
remote enabled function modules, but
not all remote enabled function modules
are BAPIs.
A BAPI is a remote enabled function
module that has been specifically written
to implement the method of a Business
Object.

Overview – Business Object Repository (BOR)

The BOR arranges BAPIs into
Business Objects
Objects contain methods, events, key
fields, and attributes
 Methods are usually mapped to BAPIs
 Key fields are mandatory fields used to
identify an object instance
 Events are used to notify others of some
business event
The BOR is accessed using transaction
SWO2 (Letter “O”, not zero). Here, you
can create your own BOR objects
Existing BOR objects can be extended
using transaction SWO1


Agenda

Generator


Gateway Data and Consumption Models

The Gateway Modelling Wizard will
generate a Data Model
A generated Data Model encapsulates
the functionality to be exposed as a
Gateway Service.
Transaction SE80 allows you to create a
Gateway Data Model; however, such an
object is not immediately usable as a
Gateway Service.
In order to expose a Gateway Data
Model to the outside world, you must first
create a Gateway Consumption Model
object which is then associated with the
Data Model.


Agenda

Generator


Gateway Data Models Based on BOR Objects - Overview

Gateway Data Models can be created
based on RFMs or BOR Objects.
Gateway Objects are created using the ABAP
Development Workbench (txn SE80)
Once the Gateway Object is created, you are
taken to the Gateway Object Modeller
(txn /IWFND/GWO_GEN)
The Gateway Object Modeller is where your
service is created.
Select a BOR Object or RFM by performing a
search.
Each RFM or BOR Object method is then
mapped to a service operation.


Agenda

Generator


Step 1) Create A Gateway Data Model – 1/2

In SE80, select “GW Data Model” and
enter a new object name
After confirming that you want to create a
new object, you will see the pop-up shown
on the right.
The object type is “PS” (meaning Public
Solution)
You should then select “Generate from
Data Source Object” and leave the “OData
Channel” checkbox switched on.

Switching this checkbox off will cause
the generated model to use the older
Generic Channel object model


Step 1) Create A Gateway Data Model – 2/2

Now you must indicate from which type of data object the Gateway Service is
to be generated
After entering a description, set the Data Source Type to “2 Remote Function Calls”
and then enter the System Alias of the relevant backend system
SE80 will now take you directly to the Gateway Modelling Wizard.


Agenda

Generator


Step 2) Map RFM to Gateway Service Operation – 1/2

Select the RFM by performing a search
1) Press the Binoculars icon on the tool bar and search for your required RFM
2) Enter the search pattern
3) Select the required RFM


Step 2) Map RFM to Gateway Service Operation – 2/2

Now your selected RFM must be mapped to a Gateway Service Operation
This is where you decide which of the Create, Read, Update, Delete or Query
operations this RFM will implement


Agenda

Generator


Step 3.1) Define the QUERY Operation’s Interface

The fields in the RFM interface define the fields that can be made available in
the Gateway Service Operation.
IMPORTANT!
Just because a field is present in the RFM’s interface does not mean it will be
visible in the OData interface! You must take care to ensure that mandatory input
fields are either supplied with a constant value or are visible in the OData interface.


Step 3.2) Add Mandatory Fields to the Interface

Any mandatory fields in the RFM interface must be made visible in the OData
interface; therefore, mapping routes must be defined for these fields.
This action adds the mandatory field(s) to the OData operation’s interface.
If this is not completed, the operation will be unusable because data cannot be
supplied to a mandatory input field.


Step 3.2) Define Constant Values for Mandatory Fields

Alternatively, a mandatory input field might only require a static value.
Supply a constant value for the mandatory input field. Remember that all non-
numeric values must be specified within single quotes.
This field will now not appear in the OData interface, but will always be supplied
with the static value defined here.


Step 3.3) Remember Which Fields Hold Key Values

The QUERY operation will output a list of banks. The user can then select an
individual bank from this list to see its details.
We must now take care to remember which fields contain the key data on the
output side of the QUERY operation, because these fields must then be used to
supply the key data required by the subsequent READ operation.
(If we were creating a Gateway Service based on a BOR object, this step would be performed automatically)


Step 3.4) Define a Model Key

Each Data Model must have one or
more fields defined to act as keys.
For instance, when you perform a
QUERY, you receive multiple Entities in
an Entity Set. We cannot then perform a
subsequent READ operation unless we
first know from where the key values can
be derived.

Once the key fields have been defined,
the values they contain can be passed
from one operation to another.


Step 3.4) Map RFM Fields to the Model Key

Once the Model Key fields have been defined, we map the RFM key fields.
This allows key data to be passed from one operation to the next.
1) At least one field in the OData interface must be flagged as a key field.
2) For the RFM assigned to the QUERY operation, all its mandatory input fields must
be visible in the OData operation’s interface
3) On the output side of the QUERY RFM, the fields that will supply key data should be
mapped to the OData key field(s)
4) The mandatory input fields of the READ operation are now supplied by the OData
key fields

Flag OData Mandatory input
interface field field of RFM
as a key
Key field on
Key fields output side
supply data to of RFM
READ operation


Agenda

Generator


Step 2) A QUERY Operation On Its Own Is Not Sufficient

The QUERY operation has now been
created by mapping it to the function
module BAPI_BANK_GETLIST.

However, the OData specification
requires that all services implement at
least a QUERY and a READ operation.
Therefore, we must also create a READ
operation.


Step 2) Creating a READ Operation (Summary)

1) Map the required RFM to the READ operation
2) Define the fields from which this operation will obtain its key data
3) Any 0..1 cardinality output structures need to be flattened (i.e. fields that belong
to data structures should be remapped to the Gateway Service’s root node)


Step 2.1) Define the Fields that Will Supply Key Data

For the RFM mapped to the READ operation, select
each of its mandatory input fields and change their
mapping route to point to the correct Model Key field.
Due to our previous mapping assignments, the input
fields of the READ RFM will now be populated with
data coming from the output side of the QUERY
operation.


Step 2.2) Flattening the Interface – 1/3

The RFM mapped to the READ operation exposes the bank’s address in a data
structure called BANK_ADDRESS. If we left these fields within the data structure, we
would need to issue a second HTTP request to obtain the data.
To remove the need for a second HTTP request, we must flatten the BANK_ADDRESS
structure by changing the mapping route of each field.

These fields have been
mapped to the root node

These fields have
yet to be mapped



As before, right-click on each field in the BANK_ADDRESS structure (this selects the
field but avoids deselecting the checkbox), then click “Change Mapping Route”.

Select the root node of the Data Model
and press OK.

This process must be repeated for each
field in BANK_ADDRESS.



Once the BANK_ADDRESS structure has been flattened, the data it produces can then
be supplied to other operations such as UPDATE or CREATE

All fields have now
been mapped


Agenda

Generator


Step 4) Generate the Data Model Runtime Classes

Once the mapping process is complete, the Data Model must be generated.

Generating the Data Model creates
the classes that are used at runtime

A list of all generated classes will be
displayed


Agenda

Generator


Step 5) Gateway Consumption Model – Overview

A GW Consumption Model is used to
expose your GW Data Model to the
outside world
Consumption Models are created in the
ABAP Developer Workbench
One or more GW Data Models can be
bound to the GW Consumption Model.
Also, one GW Data Model can be
exposed through multiple GW
Consumption Models.

An ICF node is created under the path
<default_host>/sap/opu/odata/sap/


Step 5) Gateway Consumption Model – 1/3

Create a GW Consumption Model in the
ABAP Developer Workbench
First create a new Consumption Model

Enter the Service’s Technical Name, being careful to provide a meaningful name.
Whatever name you enter here becomes the service name seen in the URL.

Do not uncheck this checkbox!
Your Data Model and Consumption
Model must use the same runtime.



Now assign your existing Data
Model to the Consumption Model

Right click on the Consumption Model

Enter in the Data Model’s
identifier



The Consumption Model and Data
Model are now associated with each
other

Double click on the name of the
consumption model

The URL of the
Gateway Service is
now shown.
However, we’re not quite
ready to test the Gateway
Service!


Agenda

Generator


Step 6) Assign System Alias to Gateway Service

Transaction /IWFND/MAINT_SERVICE in Gateway Hub
This transaction allows you to perform Gateway Service administration


Step 6.1) Locate Your Gateway Service

Your Gateway Service is most easily located by selecting the “External Service
Name” column header and then sorting the list.


Step 6.2) Display Service Properties

Once you have located and then selected your Gateway Service, you will see its
properties displayed in the lower half of the screen.

As can be seen, an active
ICF node has been created
automatically for the service

However, no system
alias is yet defined
Click on “Add System Alias”


Step 6.3) Assign a System Alias

Click “New Entries” and add a system alias for your service

The “Service Document Identifier” is the
full name of your Consumption Model

Enter the system alias for the backend
server that provides your business data

If the Gateway add-ons have been installed in the same system as your backend
Business Suite, then the system alias will typically be called LOCAL; however, this is
only a convention and not a rule.


Agenda

Generator


Step 7) Test Your Gateway Service – Overview

Testing a read-only Gateway Service can be done from a Browser
The QUERY and READ operations of any Gateway Service use the HTTP GET
method, and therefore can be tested from a browser.


Step 7.1) Invoke Gateway Service From a Browser

Once you have saved your changes, you can test your Gateway Service by
pressing the “Call Browser” button. This will display the “Service Document”


Step 7.2) The Gateway Service Document – 1/2

The URL shown in SE80 for the Consumption Model points you to the OData
Service Document.
The Service Document is the starting point for Gateway Service consumption


Step 7.2) The Gateway Service Document – 2/2

The Service Document lists the collections provided by this Gateway Service
Each collection is identified by a relative URL.
The collection’s relative URL should be appended to the service’s base URL


Step 7.3) Testing the QUERY Operation – 1/4

To obtain an OData collection, append the collection’s relative URL to the
service’s base URL
In our case, we want to see a list of banks, so we construct the URL and issue this
from a browser.
Be careful! Except for the Gateway Service part of the name, OData URLs are
case-sensitive.
So let’s run the constructed URL...

Hmmm, that’s not what we expected...



We have not defined a constant value for the mandatory input field BANK_CTRY
and neither has a value been supplied by the client.
We can fix this by taking one of two approaches here:
(Less flexible) Hardcode a constant value in the service definition, or
(More flexible) Add a parameter to the URL to supply the missing value



The most flexible option is to supply the missing parameter in the URL
The URL query string syntax for filtering an OData collection is as follows:

?$filter=<field_name> <op> '<value>'

“?” means “start ofYou must include a space or %20
the query string” field on which wethe operator Non-numeric
The on either side of wish value.
Comparison
“$filter” is thefilter (case-sensitive!)
to Operator such as in single quotes
values must be
OData command eq or ne

Important: As of Gateway 2.0 SP03, operators must be in lower case



All data that matches the $filter command is now supplied to the client
The QUERY operation returns 0..n <atom:entry>s in an <atom:feed> element

?$filter=bank_ctry eq 'US'


Step 7.4) Testing the READ Operation – 1/2

Each <atom:entry> in an OData Collection contains an <atom:id> element
This element contains the URL to perform the READ operation for that single entry


Step 7.4) Testing the READ Operation – 2/2

The READ operation returns 0..1 entries in an <atom:entry> element
If the entity cannot be found, then the <atom:content> element will be empty

If found, the business data
for the entry being READ
will be contained within the
<atom:content> element


Summary

You should now be able to:
 Create a read-only Gateway Service based on a Remote Function
Module or an Object in the Business Object Repository (BOR)
 Expose a Gateway Service for consumption
 Test your Gateway Service from a browser


SAP NetWeaver Gateway - RFC & BOR Generators

Recommended

More Related Content

What's hot (19)

Viewers also liked (20)

Similar to SAP NetWeaver Gateway - RFC & BOR Generators (20)

More from SAP PartnerEdge program for Application Development (20)

SAP NetWeaver Gateway - RFC & BOR Generators