[Mapserver-users] WFS client howto

[Bart van den Eijnden]

Hi Gregor,

the best thing is to start with a very small BBOX.

GML is a slow medium, and the generating and parsing will take quite some 
time, depending on the size of your dataset, possibly longer than 60 
seconds. Are you requesting the whole map (i.e. the whole dataset)?

Try the most simple GET request in a browser using a very small BBOX on 
your data and get that working first.

BTW: Did you set your MAP file with env variables or something similar so 
that you don't need to specify ?map= in the CONNECTION? Just checking.

Best regards,
Bart

On Tue, 1 Jun 2004 14:16:35 -0700 (PDT), Gregor Mosheh 
<stigmata at blackangel.net> wrote:

>
> Thanks for your help, Bart. It still doesn't work, but at least I now 
> know
> why.
>
>> But you have probably run into something like the following:
>> http://mapserver.gis.umn.edu/bugs/show_bug.cgi?id=597
>
> Yes, that's the same result I was getting with my manually-entered URLs.
> So that "nothing there" response is the "junk" that makes the layer (and
> thus the whole map) fail?
>
>
>
>> Set your wfs_request_method metadata to GET then you can see in the 
>> Apache
>> log exactly the full request which gets sent.
>
> Yes, and that gave me the lead to figure out what was being sent and thus
> what was wrong. Now my WFS client looks like this, and is at least 
> sending
> requests to the httpd. This paragraph is nothing at all like the one in
> the howto, though. Any comments? (the CONNECTION is right. it's a
> dedicated wms/wfs-only mapserver)
>
>   CONNECTIONTYPE wfs
>   CONNECTION "http://maps.hostgis.com:800/mapserv?"
>   METADATA
>     "wfs_service"           "WFS"
>     "wfs_typename"          "streets"
>     "wfs_srs"               "EPSG:4269"
>     "wfs_name"              "streets"
>     "wfs_title"             "streets"
>     "wfs_version"           "1.0.0"
>     "wfs_request_method"    "GET"
>     "wfs_connectiontimeout" "60"
>   END
>
>
>
> I still have the problem that the connections are timing out. But I'll
> start a new thread on this... Thanks a lot, Bart.
>
> _______________________________________________
> Mapserver-users mailing list
> Mapserver-users at lists.gis.umn.edu
> http://lists.gis.umn.edu/mailman/listinfo/mapserver-users
>



--