[CITE-Forum] WFS 1.0.0 test failures do not provide descriptive error messages
Morris, Charles, E.
chuck.morris at ngc.com
Tue Dec 18 14:19:44 EST 2007
Better error messages would certainly be helpful, but it's unlikely that
OGC will put any more resources into the version 1.0.0 test suite.
From: Normand Savard [mailto:nsavard at mapgears.com]
Sent: Tuesday, December 18, 2007 1:09 PM
To: Morris, Charles, E.
Cc: Justin Deoliveira; cite-forum at opengeospatial.org
Subject: Re: [CITE-Forum] WFS 1.0.0 test failures do not provide
descriptive error messages
Morris, Charles, E. wrote:
>Thanks Justin That is exactly correct.
>From: Justin Deoliveira [mailto:jdeolive at openplans.org]
>Sent: Monday, December 17, 2007 3:19 PM
>To: Normand Savard
>Cc: Morris, Charles, E.; cite-forum at opengeospatial.org
>Subject: Re: [CITE-Forum] WFS 1.0.0 test failures do not provide
>descriptive error messages
>The problem is that (and someone correct me here if i am wrong) the
>validation engine is not actually using the "schemaLocation" of your
>returned document to validate. Its using a schema stored internally.
>In other words it is not using the reflective describeFeatureType call
>included in your output. And since in the copy of the schema it is
>using the geometry property is named "pointProperty" and not
>"msGeometry" you get an error.
As a suggestion I think it would be good to add to the error message
returned by the wfs:GMLValidatingParser that it does not validate
against the dataFeatures.xsd schema. This is not clear in the
documentation found at
this schema is used for validation. Also the GML validator
use the same schema as the one used by the TEAM Engine, not the one
provided by our WFS server response.
What do you think?
More information about the CITE-Forum