[Geoserver-users] Parameter Pass-Thru on KML_reflect

WMS might get super anal spec types a bit mad at us, I've had people tell me we shouldn't even offer the shortcuts that we do.  But I would see the value.  Could get around anal WMS types by making url like kml_reflect, wms/reflect?params.  Want to submit a feature request to JIRA?
  

sounds reasonable to me – don’t fix the /wms handler, but make a reflector. the things I think of as should-handle-defaults include the BBOX, the FORMAT, the STYLE, the WFS VERSION, TRANSPARENT, etc. etc.

should i do the feature request, or was that for b. owens?

thanks,
mike

Go ahead and make a feature request for it and set it to 1.6.

Brent Owens
(The Open Planning Project)

Michael Frumin wrote:

WMS might get super anal spec types a bit mad at us, I've had people tell me we shouldn't even offer the shortcuts that we do. But I would see the value. Could get around anal WMS types by making url like kml_reflect, wms/reflect?params. Want to submit a feature request to JIRA?
  

sounds reasonable to me -- don't fix the /wms handler, but make a reflector. the things I think of as should-handle-defaults include the BBOX, the FORMAT, the STYLE, the WFS VERSION, TRANSPARENT, etc. etc.

should i do the feature request, or was that for b. owens?

thanks,
mike

Michael Frumin wrote:
  

mutter mutter,
    

why cant this be the case for _all_ WMS and WFS requests in geoserver?
-mike
cholmes@anonymised.com wrote:
I wouldn't say it definitely makes things more complicated - the thing that's much nicer than a regular getmap request is it supplies defaults. But if you really want to change things, then let people over-ride.

Now, if it's a matter of having to over-ride each, then yeah, doing epsg over-rides is silly. But if it's easy to do it'd be better to just reflect everything by default, providing defaults if things aren't there.

C

Brent Owens wrote:
  >>> I agree that kmscore and kmattr should be passed through, but the rest > of the parameters shouldn't. It would make the kml reflector more > complicated than it needs to be. If you start passing in the other > parameters you essentially have the regular getMap request. So I would > like the reflector to remain simple, and a starting point for figuring > out how to get KML out of geoserver.
      

    >>>> I will take a look into the kmscore and kmattr values and see that they > get passed through, and set a jira for it.
        

Brent Owens
      >>> (The Open Planning Project)
          

    >>>>>> Michael Frumin wrote:
        

          >>>> I would think that these basic parameters that affect the way it works >> in GEarth, like KMScore and KMAttr, should be passed through. This >> still gets you the benefits of not having to worry about all the other >> crap (EPSG, VERSION, BBOX, etc. etc.) in a normal WMS request.
              

thanks,
mike

brentowens@anonymised.com wrote:
      >>>>> No, parameters get cut off and default parameters are set.
          

The only ones that are changeable are the bounding box, and layers, >>> and I think styles.

All the reflector does is return a regular WMS getMap request, with >>> KML output, so you can just use that once, get the proper URL, and >>> use that from then on with your specific parameters.
Its intent is just a quick and dirty way of getting data, with >>> default parameters, into Google Earth.

Brent Owens
(The Open Planning Project)

mfrumin wrote:
>>>> something I'm not completely clear on -- for the whole kml_reflect >>>> thing, are
        >>>>>> you supposed to be able to change the default parameters by tacking >>>> on more
            

GET params, eg:

http://localhost:8080/geoserver/wms/kml_reflect?layers=topp:states&KMScore=100&KMAttr=false >>>>

because when i add them onto the request, the WMS request inside the >>>> kml
file returned still has >
KMScore=30&KMAttr=true

thanks,
mike
  >>
          >>>> >>
              

-- Chris Holmes
The Open Planning Project
http://topp.openplans.org
  > !DSPAM:1003,45e5f931269171194215290!
      
-- Chris Holmes
The Open Planning Project
http://topp.openplans.org