I know that it works the way you have specified it. But according to the wfs 1.0.0 specification section 13.7.3.1 the typename parameter is optional if a featureid parameter is specified. When there is no typename encoded in the feature id and the typename is missing, how should geoserver know to which datasource the fid applies? That’s why I think it’s reasonable to have a fid format of . (at least when typename is missing). On the other hand, I don’t know how wfs clients would handle this…
-----Ursprüngliche Nachricht-----
Von: Héctor Hugo Barriuso [mailto:hector@anonymised.com]
Gesendet: Dienstag, 25. November 2003 10:45
An: Simon Räss
Betreff: Re: [Geoserver-devel] Feature Id Format
I think the correct query is:
http://localhost:8080/geoserver/GetFeature?TYPENAME=RQS_POINT&FEATUREID=9
So, you are asking for the feature of type RQS_POINT with the id = 9.
It works for me in postGIS when I ask for a feature in a table with its id.
I don’t know if the correct sintax is typename or type_name, so try for the two.
Hope it helps.
----- Original Message -----
From: Simon Räss
To: geoserver-devel@lists.sourceforge.net
Sent: Tuesday, November 25, 2003 10:29 AM
Subject: [Geoserver-devel] Feature Id Format
I send the following request to geoserver:
http://localhost:8080/geoserver/GetFeature?FEATUREID=RQS_POINT.9
I see the point, that the format must be .. But the generated sql statement contains:
WHERE FID=‘RQS_POINT.9’
The FID column is a number, that’s why the above request throws an exception. It should be something like:
select … from RQS_POINT where FID=‘9’ …
Is this a problem of the OracleDataSource or is it the way it is intended to work?
Simon Räss