Hi Jens,

Yes, wfs 1.1 maintains compatibility via gml2, however, the spec defines
gml3 as the default, which means that clients have to start asking for
gml2 explicitly, and I would guess that many probably don't.

Also, there is a minor difference in the gml3 GetFeature output. And it
is the use of gml:id as opposed to fid for feature ids.

-Justin

Jens Stutte wrote:
> On 2 Nov 2006 at 8:04, Paul Ramsey wrote:
> 
>> Jens,
>>
>> The WFS client code (most of it, anyways) resides in the Geotools 
>> code base.  The main issue with moving to WFS 1.1 is not the WFS 1.1 
>> specification, it's the GML 3.1.1 specification.  So if you're 
>> interested in helping, hop over the the geotools side of the fence 
>> and start talking there about what needs to be done for GML 3.1.1 
>> support.  Some of it is already being done by the geoserver team 
>> (working in the geotools code base (we really should have a simple 
>> architecture diagram at the top of our developers page)) for an 
>> opengis project they are on right now.  But it is a limited subset of 
>> GML3.
>>
>> Sorry to give you a research project when all you wanted was a simple 
>> answer!
>>
>> P
> 
> Hi Paul,
> 
> some research on the server side (deegree-devel list) led to the following:
> 
> <snip>
> /Hello Jens,/
> /
> /
> /Jens Stutte schrieb:/
> /> the WFS 1.1 spec defines for backwards compatibility as possible
> ouputFormat GML 2.1.2 /
> /> and 3.1.1 (see for example: page 48 of 04-094 Web Feature Service
> Implementation /
> /> Specification_V1.1.pdf). As far as I can see in current deegree
> examples, only GML 3.1.1 is /
> /> used. Is it possible to "downgrade" the WFS to use GML 2.1.2? Would
> it be very hard to /
> /> implement this?/
> />   /
> /Actually, the WFS GML output does not use anything that is special to /
> /GML 3. The GetFeature-reponses from the WFS should be valid GML 2 as
> well.../
> /> The use would be, to create a "simple" XML transformation between WFS
> 1.0 and WFS 1.1 /
> /> requests and responses in  order to be able to implement a fully
> backwards compatible WFS. /
> />   /
> /Yes, this could be realized as a proxy server for example./
> /
> /
> /Best regards,/
> /Markus S./
> </snip>
> 
> If this is true, why not realize (as a temporary hack) "proxied" WFS 1.1
> support on the client side inside uDig? With some XSLT you should be
> done - not very performant, but at least it would work. Just to get a
> step ahead without being blocked by others...
> 
> Best regards,
> 
> Jens
> 
> 
> 
>  
> !DSPAM:1004,454f4af2176681425493344!
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> User-friendly Desktop Internet GIS (uDig)
> http://udig.refractions.net
> http://lists.refractions.net/mailman/listinfo/udig-devel
> 
> 
> !DSPAM:1004,454f4af2176681425493344!


-- 
Justin Deoliveira
The Open Planning Project
[EMAIL PROTECTED]
_______________________________________________
User-friendly Desktop Internet GIS (uDig)
http://udig.refractions.net
http://lists.refractions.net/mailman/listinfo/udig-devel

Reply via email to