David,
I notice you made the following change for GEOS-3192:
--- trunk/src/wms/src/main/java/org/vfny/geoserver/wms/responses/helpers/WMSCapsTransformer.java 2009-06-24 20:56:00 UTC (rev 12700)
+++ trunk/src/wms/src/main/java/org/vfny/geoserver/wms/responses/helpers/WMSCapsTransformer.java 2009-06-24 21:50:46 UTC (rev 12701)
@@ -682,7 +682,7 @@
AttributesImpl qatts = new AttributesImpl();
qatts.addAttribute("", "queryable", "queryable", "", "1");
start("Layer", qatts);
- element("Name", layer.getName());
+ element("Name", layer.getResource().getNamespace().getPrefix() + ":" + layer.getName());
//REVISIT: this is bad, layer should have title and anbstract by itself
element("Title", layer.getResource().getTitle());
element("Abstract", layer.getResource().getAbstract());
While I realize this is a single line change, and that the change is probably worth while as it fixes an issue, changing anything (no matter how trivial) in the capabilities document can seriously affect existing clients, so such changes should not be made lightly.
I am not against the change, but it went forward with absolutely no discussion on the developers list. Was there discussion on IRC that I missed? Did you verify this change with existing clients? I am guessing openlayers but what about udig, qgis and other clients?
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
Justin Deoliveira ha scritto:
David,
I notice you made the following change for GEOS-3192:
--- trunk/src/wms/src/main/java/org/vfny/geoserver/wms/responses/helpers/WMSCapsTransformer.java 2009-06-24 20:56:00 UTC (rev 12700)
+++ trunk/src/wms/src/main/java/org/vfny/geoserver/wms/responses/helpers/WMSCapsTransformer.java 2009-06-24 21:50:46 UTC (rev 12701)
@@ -682,7 +682,7 @@
AttributesImpl qatts = new AttributesImpl();
qatts.addAttribute("", "queryable", "queryable", "", "1");
start("Layer", qatts);
- element("Name", layer.getName());
+ element("Name", layer.getResource().getNamespace().getPrefix() + ":" + layer.getName());
//REVISIT: this is bad, layer should have title and anbstract by itself
element("Title", layer.getResource().getTitle());
element("Abstract", layer.getResource().getAbstract());
While I realize this is a single line change, and that the change is probably worth while as it fixes an issue, changing anything (no matter how trivial) in the capabilities document can seriously affect existing clients, so such changes should not be made lightly.
I am not against the change, but it went forward with absolutely no discussion on the developers list. Was there discussion on IRC that I missed? Did you verify this change with existing clients? I am guessing openlayers but what about udig, qgis and other clients?
I saw no discussion. Looking at the diff it would seem this change
fixes a regression introduced during the NY sprint? (using
layer and their names instead of resources and their qualified names?)
Cheers
Andrea
I saw no discussion. Looking at the diff it would seem this change
fixes a regression introduced during the NY sprint? (using
layer and their names instead of resources and their qualified names?)
Right, I guess a chance to chime in would have been nice. Especially since a better fix might just be to have LayerInfoImpl.getName() call through to ResourceInfo.getPrefixedName() since we don't really allow the two to differ, and this probably will probably come up again in other places.
Cheers
Andrea
------------------------------------------------------------------------------
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
Justin Deoliveira ha scritto:
I saw no discussion. Looking at the diff it would seem this change
fixes a regression introduced during the NY sprint? (using
layer and their names instead of resources and their qualified names?)
Right, I guess a chance to chime in would have been nice. Especially since a better fix might just be to have LayerInfoImpl.getName() call through to ResourceInfo.getPrefixedName() since we don't really allow the two to differ, and this probably will probably come up again in other places.
Yup, agreed it's a better way and one we discussed and agreed upon
already. Soo... let's do it?
Cheers
Andrea
Andrea Aime wrote:
Justin Deoliveira ha scritto:
I saw no discussion. Looking at the diff it would seem this change
fixes a regression introduced during the NY sprint? (using
layer and their names instead of resources and their qualified names?)
Right, I guess a chance to chime in would have been nice. Especially since a better fix might just be to have LayerInfoImpl.getName() call through to ResourceInfo.getPrefixedName() since we don't really allow the two to differ, and this probably will probably come up again in other places.
Yup, agreed it's a better way and one we discussed and agreed upon
already. Soo... let's do it?
Agreed. Note it also forces layer name to be read only for the time being and to make the UI layer name validation check weaker (for instance, ArcSDE type names have been non wfs conformant since 2003, they contain periods, something that can be fixed as per the resource/publish split when all the services use layer name, but right now it simply means we couldn't add any ArcSDE feature type).
Gabriel
PS: whether to make ArcSDE type names WFS friendly as default may be a separate discussion. GeoServer may also choose to escape names somehow which in the end seems cleverer to me, as there's no real mandate on geotools type names to be GML valid?
Cheers
Andrea
------------------------------------------------------------------------------
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel
--
Gabriel Roldan
OpenGeo - http://opengeo.org
Expert service straight from the developers.
Gabriel Roldan ha scritto:
Agreed. Note it also forces layer name to be read only for the time being and to make the UI layer name validation check weaker (for instance, ArcSDE type names have been non wfs conformant since 2003, they contain periods, something that can be fixed as per the resource/publish split when all the services use layer name, but right now it simply means we couldn't add any ArcSDE feature type).
Imho we can have weaker name validation for the moment
Gabriel
PS: whether to make ArcSDE type names WFS friendly as default may be a separate discussion. GeoServer may also choose to escape names somehow which in the end seems cleverer to me, as there's no real mandate on geotools type names to be GML valid?
Indeed
Cheers
Andrea
--
Andrea Aime
OpenGeo - http://opengeo.org
Expert service straight from the developers.