Ticket #2512 (new defect)

Opened 7 months ago

Last modified 2 months ago

A GetCapabilities issue to a SOS 1.0.0 server with no service parameter must trigger an OWS 1.0.0 exception

Reported by: nsavard Assigned to: tomkralidis
Priority: normal Milestone: 5.4 release
Component: SOS Server Version: svn-trunk (development)
Severity: normal Keywords: OGC, CITE, TEAM, SOS 1.0.0
Cc: sdlime, warmerdam, assefa, dmorissette

Description

When a GetCapabilities? request is sent to a SOS 1.0.0 server with no service parameter, the response must be an OWS 1.0.0 exception.

Actually a WFS 1.0.0 exception. The test try to validate against the schema found here:

http://schemas.opengis.net/ows/1.0.0/owsExceptionReport.xsd

The concerned test is: owsTests:ows-OWS.GetCapabilities-Exceptions.1 (s0003)

and the request is: http://dev1.lan.mapgears.com/manwe/cgi-bin/mssos100_ogc_cite?version=1.0.0&request=GetCapabilities

Change History

02/14/08 17:37:02 changed by tomkralidis

  • cc changed from sdlime to sdlime, warmerdam, assefa, dmorissette.

This one is tough. If no SERVICE= is passed, that doesn't necessarily constitute an ExceptionReport, because MapServer offers up CGI mode as well.

CC'ing the other WxS folks on comments here.

(follow-up: ↓ 3 ) 02/14/08 17:47:40 changed by sdlime

I don't see a workaround for this one. Why does SOS have this requirement on not other OWS specs?

Steve

(in reply to: ↑ 2 ; follow-up: ↓ 4 ) 02/14/08 17:53:46 changed by tomkralidis

Replying to sdlime:

I don't see a workaround for this one. Why does SOS have this requirement on not other OWS specs? Steve

Steve: I agree.

Norm: does this type of assertion occur in CITE for WMS? I believe you have done tests which passed for WMS 1.1.1, so I'm wondering how this was dealt with there.

(in reply to: ↑ 3 ) 02/18/08 10:00:55 changed by nsavard

Replying to tomkralidis:

Replying to sdlime:

I don't see a workaround for this one. Why does SOS have this requirement on not other OWS specs? Steve

Steve: I agree. Norm: does this type of assertion occur in CITE for WMS? I believe you have done tests which passed for WMS 1.1.1, so I'm wondering how this was dealt with there.

Tom: I don't remember to have seen a test like that in any of the OWS before.

02/18/08 10:03:19 changed by tomkralidis

Can you ask CITE what we should do about this? Because we support more than OWS, we can't restrict in this manner.

02/19/08 15:23:36 changed by dmorissette

Perhaps we could add a "mode=ows" vendor-specific CGI parameter that would force treating a given request as a OWS request, this mode param could be included in the onlineresource in the capabilities and unless I'm mistaken that would allow us too bypass the CGI and pass all those tests.

We could also add support for a MS_MODE environment variable that could be set inside a wrapper script or in Apache config to any of the valid mode values and could be used as an alternative to the mode=ows CGI param to force MapServer to act as a OWS. This would be used the same way we use MS_MAPFILE already today in WMS wrapper scripts, see http://mapserver.gis.umn.edu/docs/howto/wms_server/#more-about-the-online-resource-url

(follow-up: ↓ 10 ) 02/21/08 18:22:08 changed by tomkralidis

Daniel: good idea. Would there be more cases where we would need this other than CITE? Norm: can you check w/ CITE w.r.t. what their position is on service URLs submitted to CITE which support over and above the OGC standards (like MapServer)?

(follow-up: ↓ 9 ) 02/22/08 08:50:51 changed by sdlime

The CITE tests allow you to *require* a vendor specific parameter? I'm ok with the idea it just seems like a workaround for a silly requirement for SOS.

Steve

(in reply to: ↑ 8 ) 02/22/08 10:24:54 changed by tomkralidis

Replying to sdlime:

The CITE tests allow you to *require* a vendor specific parameter? I'm ok with the idea it just seems like a workaround for a silly requirement for SOS. Steve

So long as it's part of the server base URL, it's fine. We'll see if we really need to do this before moving ahead.

(in reply to: ↑ 7 ) 02/25/08 21:48:06 changed by nsavard

Replying to tomkralidis:

Daniel: good idea. Would there be more cases where we would need this other than CITE? Norm: can you check w/ CITE w.r.t. what their position is on service URLs submitted to CITE which support over and above the OGC standards (like MapServer)?

I'll check.

02/28/08 08:48:39 changed by tomkralidis

Here is what everyone seemed to agree upon:

""" I agree with Daniel. Making 'mode=ows' a static part of the OnlineResource? URL is probably the best way to be OGC-compliant with the least impact on your existing code base. --- Raj

On Feb 27, 2008, at 4:39 PM, Daniel Morissette wrote:

Norm, I think for the mode=ows vendor-specific parameter to work it needs to be included in the onlineresource of all services (i.e. in the GetCapabilities? response), and then your question should really be: "Is there any problem with a server including a 'mode=ows' vendor-specific parameter in the onlineresource that it advertizes in GetCapabilities? for all its services?" Since the onlineresource is the base URL string which is used as-is by the client to connect to the service then I can't imagine that there would be any problem with that, a server should be allowed to put anything in the onlineresource URL as long as it's a valid URL and nothing in the base URL conflicts with documented parameters of the relevant OGC spec. Daniel

"""

Daniel: should we open another ticket just for the implementation of this addition?

02/28/08 09:38:02 changed by dmorissette

Yes, I think a separate ticket would be better, then the approach can be discussed in a single place and all tickets with related issues can refer to it in their 'fixed' comment.

05/20/08 12:31:23 changed by tomkralidis

FYI #2531 should fix this ticket.

07/04/08 16:22:55 changed by tomkralidis

  • milestone changed from 5.2 release to 5.4 release.