Selecting Data Objects
This wizard page will present you with a list of suggested data objects that you might want to use in your application. The accuracy of this list depends on the method you used to discover your data objects. If you have used sample resource URL's, you might see some candidate data objects that you do not need. For example, when using a TopLink/EclipseLink REST resource like this:
this page will show candidate data objects for the relationship link information that is included in the payload:
You can simply uncheck the Select checkbox for these unwanted data objects. You should modify the class names for each data object you select, so you get Java classes generated with valid, nice singular class names.
Check the Persist checkbox for each data object you want to cache the data so you can access it in offline mode. This will instruct AMPA to create a SQLite table for the data object that is populated at runtime based on the REST resources you specify later on to retrieve the data.
Note: if you do not want to persist data objects for offline access, for example because you are using the MAF transport layer caching policy, you can still use AMPA for offline transactions. The transactions are stored in a separate DATA_SYNCH_ACTIONS table and synchronization is independent of the data object tables.
If you have used a RAML file or ADF BC REST Describe URL, the list of candidate data objects is usually quite accurate as it is based on the metadata definitions, and will not contain "fake" data objects like the Link data object we saw above.
You can also create completely new data objects on this wizard page. For example, you can use this option for creating data objects that only live on the mobile device and are not populated through REST web services.