Re: [mapserver-users] WFS filter creates a query using a number instead of text
Mapserver 7 beta solved this issue.At least for my use cases.. with and without using gml_[item name]_type John Abraham j...@hbaspecto.com@lists.osgeo.org Envoyé par : mapserver-users-boun...@lists.osgeo.org 2015-02-27 09:05 A Rahkonen Jukka (MML) jukka.rahko...@maanmittauslaitos.fi cc mapserver-users@lists.osgeo.org mapserver-users@lists.osgeo.org, bte...@gmail.com bte...@gmail.com Objet Re: [mapserver-users] WFS filter creates a query using a number instead of text On Feb 26, 2015, at 11:55 PM, Rahkonen Jukka (MML) jukka.rahko...@maanmittauslaitos.fi wrote: You can also try to add gml_[item name]_type into the layer metadata I have to add gml_[item name]_type in all my map files, to make them work as wfs in my context. -- John Abraham Sent from my iPhone, please excuse any typos. ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
Re: [mapserver-users] How reproduce a false color NDVI.
Le dimanche 01 mars 2015 10:17:34, Andrea Peri a écrit : Hi, we have a set of tiffs georef with 4th bands of color RGB and Infrared. We need to activate a wms with the color from the NDVI index NDVI = Normalized Difference Vegetation Index NIR - RED NDVI = - NIR + RED Where NIR is the 4th band, RED s the 1th band. We prefer dont' do a physical float raster, but retrieve this realtime from the original tiffs. I try to reproduce this index NDVI using to a LUT table in MapServer, but without any success. There is some other solution to obtain this on mapserver. Andrea, It turns out that the same question was asked more than 10 years ago : http://lists.osgeo.org/pipermail/mapserver-users/2004-July/053754.html Since them, GDAL has the VRT format. With derived bands (described near the end of http://gdal.org/gdal_vrttut.html), you could do that but that requires writing a custom function in C, which isn't very convenient for integration with MapServer. It would be nice to add the capability of writing pixel functions in Python. Even -- Spatialys - Geospatial professional services http://www.spatialys.com ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
Re: [mapserver-users] WFS filter creates a query using a number instead of text
Thanks Jukka - I'll try your suggestion for OGR connection-type first, since that's easy for me to test on our existing platform. I have tried the PropertyIsLike and that definitely works. It'll take me a little bit longer to spin up a more recent version of the software packages... Regards, Bill. -- View this message in context: http://osgeo-org.1560.x6.nabble.com/WFS-filter-creates-a-query-using-a-number-instead-of-text-tp5130361p5190806.html Sent from the Mapserver - User mailing list archive at Nabble.com. ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
Re: [mapserver-users] Getcapabilities extremely slow in version 6.4.1
After profiling version 6.0.4 and 6.2.2 MapServer we were able to isolate the
issue. It seems that the issues is related to “OWS_LAYER_GROUP” under
LAYERMETADATA in our map files.
When “OWS_LAYER_GROUP” is not used in the map file, GetCapabilities request
on version 6.2.2 is significantly faster (was comparable with the timing on
version 6.0.2)
Profiling showed us that version 6.2.2 is taking significantly more time on
“msWMSPrepareNestedGroups” function in mapwms.c file. Especially
“strcasecmp” function in side “msWMSPrepareNestedGroups” is taking
significant time.
When we compare the source of mapwms.c, we noticed that there are some
additional processing happening in the “else” block “ /* split into
subgroups. Start at address + 1 because the first '/' would cause an extra
empty group */” .
“else” block in Version 6.0.4 looks like following ( line 1900 in mapwms.c)
//
else
{
/* split into subgroups. Start at adres + 1 because the first '/'
would cause an extra emtpy group */
nestedGroups[i] = msStringSplit(groups + 1, '/',
numNestedGroups[i]);
}
But the “else” block in version 6.2.2 looks like following with the
additional processing ( line 371 in mapwms.c)
/
else {
/* split into subgroups. Start at address + 1 because the first
'/' would cause an extra empty group */
nestedGroups[i] = msStringSplit(groups + 1, '/',
numNestedGroups[i]);
/* */
for (j = 0; j map-numlayers; j++) {
if (isUsedInNestedGroup[j])
continue;
for (k=0; knumNestedGroups[i]; k++) {
if ( GET_LAYER(map, j)-name strcasecmp(GET_LAYER(map,
j)-name, nestedGroups[i][k]) == 0 ) {
isUsedInNestedGroup[j] = 1;
break;
}
}
}
//
--
View this message in context:
http://osgeo-org.1560.x6.nabble.com/Getcapabilities-extremely-slow-in-version-6-4-1-tp5189325p5190810.html
Sent from the Mapserver - User mailing list archive at Nabble.com.
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users
Re: [mapserver-users] WFS filter creates a query using a number instead of text
Hi Jukka - I've tried testing your other suggestion: OK - I've tried using the OGR CONNECTIONTYPE - it no longer generates an error message, but comes back with a null result set. Checking the logs reveals that the error is now apparently being pushed further back up the stack - eg the Postgres logs are clear (strangely) but the MapServer debug log reports an error as follows: I guess effectively the same error as before - there's no quoting in the string literal (not sure why the Postgres log shouldn't still report this error...?) Testing the OGR config with WFS operator PropertyIsLike, works (as did the PostGIS driver.) I think I'll use this method for the moment as it will work with our production system. Thanks again! Steve - good to know that v7 works. -- View this message in context: http://osgeo-org.1560.x6.nabble.com/WFS-filter-creates-a-query-using-a-number-instead-of-text-tp5130361p5190809.html Sent from the Mapserver - User mailing list archive at Nabble.com. ___ mapserver-users mailing list mapserver-users@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/mapserver-users
