Justin Deoliveira ha scritto:
> Hi Andrea,
> 
> Turning off validation at construction time is something that I 
> generally think everyone agrees with. And is something that is there 
> simply as a sanity check while switching to the new model. So it should 
> not be an issue to turn it off.
> 
> As for the default value... if we do in fact *not* set a value when the 
> attribute is non-nillable, i believe that breaks the rules of the 
> feature model. I would suggest we proceed in one of two ways:
> 
> 1) we open a debate about whether default value is useful in the model, 
> and take it out if need be. Gabriel has some good insight on this one.
> 
> 2) we add a flag to SimpleFeaureBuilder to allow the versioning 
> datastore to change the behavior, and ignore the default even for 
> non-nillable attributes.
> 
> If we do that will that fix your issue?

Hum, not exactly because versioning is a wrapper around the postgis
datastore. The feature are really created in some pieces of code
shared with all jdbc datastores.
It should be possible to set that as a query hint... but it does
not seem like a super clear option to me...

For the record, the actual feature construction that breaks the
code is (apparently) in JDBCFeatureWriter.next(), line 94:
SimpleFeature temp = SimpleFeatureBuilder.template(featureType, null);

So one would have to instruct the datastore to tell the jdbc feature
write to tell the SimpleFeatureBuilder not to force default values in...

Cheers
Andrea

Cheers
Andrea

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to