[mapserver-users] OpenGIS CITE

[Brent Fraser]

Daniel M. (and others),

   First of all, thanks for implementing WMS in MapServer and continuing to
upgrade it.  I think giving the user a choice of client applications is a
great feature!

   I also believe conformance to the WMS spec (as much as possible) is
vital, especially since I've been struggling for the past few days to get a
couple of WMS client applications to speak to Mapserver: Iowa State U's
WMTOrthophoto extension to ArcView 3.x
(http://www.gis.iastate.edu/Links.html) (only supports WMS v1.0.0), and
ESRI's ArcExplorer 4.0.1 Java Edition (doesn't support WMS v1.1.0, although
v1.0.0 and v1.1.1 are ok).

    I followed the mapserver wiki on using ArcExplorer
(http://mapserver.gis.umn.edu/cgi-bin/wiki.pl?WMSMapserverArcExplorer) and
was eventually successful after using an ASP wrapper to spoof the WMS
version of Mapserver to be 1.1.1.  The problem with this method is that it
requires keeping the map file and script in sync.

  How about an optional "WMS_Version " keyword for the metadata section of
the WEB object in the map file?  If present, Mapserver would spit out a
<WMT_MS_Capabilities version='1.1.1'> tag (or whatever version was given in
the map file).  We could use it to get around client/server WMS version
incompatibilies.  I realize you're working to implement v1.1.1 in Mapserver
, but there could be other client applications coming out that support only
v1.1.2, etc.

Thanks again!

Brent Fraser
GeoAnalytic Inc.
Tel:(403) 213-2700
bfraser at geoanalytic.com
www.geoanalytic.com


----- Original Message -----
From: "Daniel Morissette" <morissette at dmsolutions.ca>
To: <bartvde at xs4all.nl>
Cc: <mapserver-users at lists.gis.umn.edu>
Sent: Friday, August 08, 2003 11:58 AM
Subject: Re: [mapserver-users] OpenGIS CITE


> Bart van den Eijnden wrote:
> >
> > has anyone from the mapserver community participated in the OpenGIS CITE
> > initiative, to see how conformant mapserver is to the OpenGIS
> > specifications? I know their test suite is only in beta, but perhaps
> > somebody from mapserver has done the beta test.
> >
>
> We are currently testing MapServer 4.0 against the CITE test suite for
> WMS, and it currently passes 56 out of 88 tests.
>
> Many of the tests that fail are due to the fact that it supports only
> WMS 1.1.0 and many tests require WMS 1.1.1.
>
> Also, as JF mentioned in his reply, in a few cases MapServer is more
> permissive than the spec allows and for this reason fails some tests.
> For instance, when passing invalid layer names MapServer silently
> ignores the invalid layer (or sends the whole map if LAYERS is omitted),
> but the spec requires that we send an exception in those cases.
>
> All the issues are being filed in bugzilla, if you do a query for
> "CITE", you'll currently see 12 bugs.  We will look into each issue and
> decide what we do about each one during the 4.1 development cycle.  Our
> goal would be to pass all WMS tests and get MapServer officially
certified.
>
> With respect to WFS, we are going to run the tests but don't expect to
> do very well since the MapServer WFS is very incomplete.  Also, based on
> the feedback I've seen on the CITE list, the WFS tests seem to push
> things quite a bit with respect to GML support, etc. and for this reason
> MapServer may never pass the whole WFS suite.
>
> Daniel
> --
> ------------------------------------------------------------
>   Daniel Morissette               morissette at dmsolutions.ca
>   DM Solutions Group              http://www.dmsolutions.ca/
> ------------------------------------------------------------
>
> _______________________________________________
> Mapserver-users mailing list
> Mapserver-users at lists.gis.umn.edu
> http://lists.gis.umn.edu/mailman/listinfo/mapserver-users