I wrote a sample .gml file few months ago and I just find out that the
feature definition produced (.xsd) is odd!
The .xsd file used to have all input dataset feature types defined,
even if only one feature type was written. For instance, a source
dataset having Roads and Rivers would have produced a .xsd file
containing both even if only the Rivers feature was written. The .xsd
<element name="Roads" type="fme:RoadsType"
<element name="Rivers" type="fme:RiversType"
Now, only the feature type written in the sample is right (Roads), all
others have the following pattern in the .xsd file
<element name="XyZzY##" type="fme:XyZzYType"
where ## is sequential numbers corresponding to the number of feature
in the source dataset that were not written in the output .gml file.
Any ideas on what have caused that or what I can do to get the