[gdal-dev] [mapserver-dev] New env. var. to make it easier to test/debug web services

[Even Rouault]

Le jeudi 02 octobre 2014 21:01:58, Even Rouault a écrit :
> Le jeudi 02 octobre 2014 20:57:46, Lime, Steve D (MNIT) a écrit :
> > How about making -nh set the environment variable programatically and
> > losing the sendheaders checks?
> 
> Yes, that's a possibility. Although I think -nh could just call a function
> to disable the boolean in msio instead of messing with environmenet
> variables.

I've just implemented that. So no more env. variable trick. Just the standard 
-nh mechanism.

Would be cool to have a "mapserv --help" mode to display available options, 
wouldn't it ?

> 
> But I don't think we can remove all the sendheaders logic completely.  As
> mentionned in my above email, there's seem to have places where you want to
> disable/enable headers temporarily. Seems related to template. But I'm not
> familiar with the part of the code.
> 
> > ________________________________________
> > From: Even Rouault [even.rouault at spatialys.com]
> > Sent: Thursday, October 02, 2014 1:55 PM
> > To: Lime, Steve D (MNIT)
> > Cc: gdal-dev at lists.osgeo.org
> > Subject: Re: [mapserver-dev] New env. var. to make it easier to
> > test/debug web services
> > 
> > Le jeudi 02 octobre 2014 20:45:26, Lime, Steve D (MNIT) a écrit :
> > > You were aware of the -nh option for mapserv, right?
> > 
> > Steve,
> > 
> > Ah, good point, actually, I was not.... I should have asked first.  But
> > -nh doesn't seem to work. At least not in the WMS or WFS cases. It seems
> > mapio.c doesn't honour mapserv->sendheaders, and the WMS or WFS code
> > doesn't test it either
> > 
> > By looking at the places where sendheaders is used, it seems that you
> > must test sendheaders at each place where msIO_setHeader() is called.
> > Whereas my change does put the test inside msIO_setHeader() itself.
> > Which is less error prone I think. But we cannot decently let both
> > mechanisms in place
> > 
> > I see also some places that directly mess with sendheaders. Not sure
> > why...
> > 
> > ../maptemplate.c:  mapserv->sendheaders = MS_TRUE;
> > ../maptemplate.c:    mapserv->sendheaders = MS_FALSE;
> > ../mapwfs.c:    mapserv->sendheaders = MS_TRUE;
> > 
> > Even
> > 
> > > ________________________________________
> > > From: mapserver-dev-bounces at lists.osgeo.org
> > > [mapserver-dev-bounces at lists.osgeo.org] on behalf of Even Rouault
> > > [even.rouault at spatialys.com] Sent: Thursday, October 02, 2014 1:44 PM
> > > To: mapserver-dev at lists.osgeo.org
> > > Subject: [mapserver-dev] New env. var. to make it easier to test/debug
> > > web
> > > 
> > >     services
> > > 
> > > Hi,
> > > 
> > > I always get annoyed when running mapserv from command line and wanting
> > > to test/debug WMS, WFS, etc... without launching mapserv as a CGI,
> > > because the HTTP headers get emitted and thus produce an invalid PNG or
> > > GML. So I've pushed in master a small enhancement. If MS_HTTP_HEADER=NO
> > > is set as environmenet variable, the HTTP headers are not emitted.
> > > 
> > > So now you can do things like :
> > > 
> > > MS_HTTP_HEADER=NO mapserv QUERY_STRING="map=test.map&SERVICE=WMS&..." >
> > > out.png && display out.png MS_HTTP_HEADER=NO mapserv
> > > QUERY_STRING="map=test.map&SERVICE=WFS&..."  | xmllint -noout -
> > > 
> > > Even
> > > 
> > > --
> > > Spatialys - Geospatial professional services
> > > http://www.spatialys.com
> > > _______________________________________________
> > > mapserver-dev mailing list
> > > mapserver-dev at lists.osgeo.org
> > > http://lists.osgeo.org/mailman/listinfo/mapserver-dev
> > 
> > --
> > Spatialys - Geospatial professional services
> > http://www.spatialys.com

-- 
Spatialys - Geospatial professional services
http://www.spatialys.com