So my reading of the spec actually implies that it is required. The
section, in both 1.1.1 and 1.3 says:
The mandatory "map request part" entry in Table 9 represents a sequence
of parameters from the GetMap
request that generated the original map. Two of the GetMap parameters
are omitted because GetFeatureInfo
provides its own values: VERSION and REQUEST. The remainder of the
GetMap request shall be embedded
contiguously in the GetFeatureInfo request.
So the FORMAT part is part of the original map request, and thus it's
one of the parameters that needs to be embedded. CITE obviously sends
requests this way, or we wouldn't be passing.
That said, I agree it's silly, and thought it was silly the first time I
read it, since there's no way it'd ever be used.
I'd prefer this wasn't a config option, as I don't believe it will
actually mess up cite tests, since they'll just send the param and
nothing will happen.
It shouldn't be that hard for us to fix, and if every one else is
letting it slide than I'm fine to. We actually just use the same
parser for GetFeatureInfo and GetMap, we just pass that portion along,
and it is mandatory for a GetMap request. So we'd have to pass a dummy
value in.
Note that we do implement INFO_FORMAT, it's not confusion on our part,
it's the spec.
Go ahead and re-open the bug and attach my comments. We just have to
make sure to run through cite after the fix, but I don't believe it
will affect the tests.
Chris
Quoting Justin Deoliveira <jdeolive@anonymised.com>:
This might be a good thing to apply at the start of 1.4.x. If need be
we
could use the scrict cite conformance hack to enable / disable the
check
for the format paramter.
Gabriel Roldán wrote:
> Guys, Michael is requiring FORMAT parameter in WMS GetFeatureInfo
being not
> mandatory. I certainly understand it.
> My only concern about it is if WMS CITE tests requires it? I guess
some of you
> used to run WMS CITE?
>
>
>
> Michael Adair commented on GEOS-532:
> ------------------------------------
>
> I'll let you guys decide if you shoudl re-open, but I think this
should be
> fixed in Geoserver. FORMAT should not be a required parameter for
the WMS
> GetFeatureInfo request.
>
> OGC has buried their older spec documents so I can't check v1.1.1,
but in WMS
> 1.3 it is definitely not required, and any other WMS 1.1.1 we have
tested
> against didn't require FORMAT as a param.
>
> We implemented a workaround in MapBuilder that adds on an empty
FORMAT param
> to the request, but that doesn't seem like the correct way to do
that.
>
>
>>WMS GetFeatureInfo FORMAT parameter
>>-----------------------------------
>>
>> Key: GEOS-532
>> URL: http://jira.codehaus.org/browse/GEOS-532
>> Project: GeoServer
>> Type: Bug
>>
>> Components: WMS
>> Reporter: Michael Adair
>> Assignee: Gabriel Roldán
>>
>>
>>
>>The GetFeatureInfo WMS request takes an INFO_FORMAT parameter, not
FORMAT.
>>for reference: http://jira.codehaus.org/browse/MAP-8
>
>
> --
> This message is automatically generated by JIRA.
> -
> If you think it was sent incorrectly contact one of the
administrators:
> http://jira.codehaus.org/secure/Administrators.jspa
> -
> For more information on JIRA, see:
> http://www.atlassian.com/software/jira
>
> -------------------------------------------------------
>
--
Justin Deoliveira
The Open Planning Project
http://topp.openplans.org
***
Chris Holmes
The Open Planning Project
thoughts at: http://cholmes.wordpress.com
----------------------------------------------------------
This mail sent through IMP: https://webmail.limegroup.com/