Hi all,

I recently created https://osgeo-org.atlassian.net/browse/GEOT-6704

>From the ticket:

------ ticket start ------
The WFS 2.0 request objects all have an "extendedProperties" attribute, 
inherited from net.opengis.wfs20.impl.BaseRequestTypeImpl.
Expected: Every instance of BaseRequestTypeImpl has an own properties instance.
Actual: All instances of BaseRequestTypeImpl share the same properties instance.
Reason: Since the upgrade of the EMF version the "extendedProperties" are 
initialized with a static default map.
This is only the case for WFS 2.0. On WFS 1.x the default map is null.
// The assert below currently fails


@Test

public void testOwnExtendedProperties() {

       TransactionType lTransaction1 = 
Wfs20Factory.eINSTANCE.createTransactionType();

       lTransaction1.getExtendedProperties().put("foo", "bar");

       TransactionType lTransaction2 = 
Wfs20Factory.eINSTANCE.createTransactionType();

       assertFalse(lTransaction2.getExtendedProperties().containsKey("foo"));

}
I suppose the generator config needs to be tweaked.
------ ticket end  ------

In the meantime I have done some further investigations. I now I have some 
questions:


  1.  The class net.opengis.wfs20.impl.BaseRequestTypeImpl is apparently not 
generated by the GT XSD generator framework, but is generated "manually" using 
Eclipse. Right?
  2.  I deleted BaseRequestTypeImpl.java file and regenerated it (Opened 
"wfs20.genmodel" with Eclipse, selected BaseRequestTypeImpl and called 
"Generate Model Code".). Is this a valid approach?
  3.  The result class differs from the original: The default value for the 
extendedProperties is now null. As expected. So this would fix this issue. Any 
idea why GeoTools contains a different result?
  4.  Is the problem fixed for the future when I create a PR with the correctly 
generated class or is there the chance that the next generation results in the 
same mistake?

I use Eclipse 2020-09, with contains EMF 2.21.0.v20200708-0547.
Maybe it was a problem of an older EMF version...

Best regards,
Andreas

This e-mail may contain confidential information. If you are not the intended 
recipient (or have received this e-mail in error) please notify the sender 
immediately and destroy this e-mail. Any unauthorised copying, disclosure or 
distribution of the material in this e-mail is strictly forbidden.

_______________________________________________
GeoTools-Devel mailing list
GeoTools-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to