Maestro layer validation incorrectly report missing primary key

[gabrimonfa]

Maestro 4.0.0.6282

I'm using the OGRFdoProvider with gdal/ogr recompiled with PostgreSQL support

Layer validation reports that primary key is missing, while it is not the case. I don't know if the problem is due to my particular setup or is a general one

[jng]

If you preview the Feature Source (the web-based one, not the Maestro one) does it also fail to indicate identity properties (with a *) when you look at the class definition?

If so, it's a problem with the FDO provider

If not, then yes this is a genuine defect

[gabrimonfa]

@jng

If you preview the Feature Source (the web-based one, not the Maestro one)

I'm not entirely sure of what you mean. I suppose I have to look to the page

​http://myserver:8008/mapguide/schemareport/describeschema.php

In that page I found that there is an int32 attribute marked with * in the tables that maestro reports to be without primary key

Property Name	Property Type
foglio	string(4)
gid	int32
particella	string(5)
...	...
* ogc_fid	int32

FDO chooses the wrong attribute (the real pk in the database is gid) but ogc_fid field is also a non-null unique int32 field (that moreover has exactly the same content as gid)

The same happens in the other tables, FDO founds valid identity attributes.

So probably it is a genuine defect

[jng]

Can you attach the FDO schema XML for this feature source? Just do a DESCRIBEFEATURE on that feature source in the mapagent test pages.

[gabrimonfa]

Here the results of DescribeFeatureSchema on the FeatureSource. It seems that the keys are present.

​http://pastebin.com/1maZmpg6

[jng]

Do you have a fully qualified version of this? (with xs: fdo:, etc prefixes)

MaestroAPI cannot parse unqualified XML schemas. The mapagent does not (should not) be returning unqualified XML content

[gabrimonfa]

I hope this is what you need. Otherwise let me know how to obtain it.

​http://pastebin.com/dfbEuKz3

[jng]

Fixed r6420