Re: [mapserver-users] GetStyles SLD does not include Opacity value
Now that I've advanced beyond the simple example included above, I see a possible shortcoming with the current implementation of the GetStyles request with respect to including rule's opacity and am wondering if this is something worth pursuing... While it is easy in my original example with a single CLASS in the layer, to move the opacity parameter within the CLASS' STYLE, what if my LAYER has a number of CLASSes? Given the current implementation, if I want to see the opacity in the GetStyles response, I will have to include the OPACITY setting in every CLASS STYLE, even though they are all the same and a major advantage of the LAYER's OPACITY value is avoiding that duplication (and potential for inconsistency). Since the GetMap and GetLegendGraphic renderers are capable of reviewing whether a CLASS STYLE's or the LAYER's OPACITY value is set and draw the image accordingly, it must be possible for the GetStyles response to do the same. My question is Should it? Or am I off the mark in my expectations here? Cheers jtm -- View this message in context: http://osgeo-org.1560.n6.nabble.com/GetStyles-SLD-does-not-include-Opacity-value-tp4979066p4979520.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] GetStyles SLD does not include Opacity value
Are we missing a few things like the response and the mapfile snippet? Also are you talking layer opacity or style opacity? I guess the email-er that sent around the post didn't include anything within the raw tags. Here's the full text again (without the tags): I have a Map file with a polygon layer that has an OPACITY value of 20 defined as: LAYER NAME Poly_wOpacity TYPE POLYGON OPACITY 20 STATUS ON PROJECTION init=epsg:3857 END INCLUDE ../../../mapserv/conf/sdeconnection.inc.map DATA ...,SHAPE,SDE.DEFAULT METADATA wms_title Polygon Layer w/Opacity of 20 wms_srs epsg:3857 END CLASS NAME Polygon Layer w/Opacity of 20 STYLE COLOR 255 0 0 END END END I see a correctly opaque display when I do a GetMap or a GetLegendGraphic, so I know the OPACITY value is set properly. However, when I perform a GetStyles request, the returned SLD doesn't include anything about the opacity. The response is: StyledLayerDescriptor version=1.0.0 ... NamedLayer NamePoly_wOpacity/Name UserStyle FeatureTypeStyle Rule NamePolygon Layer w/Opacity of 20/Name PolygonSymbolizer Fill CssParameter name=fill#ff/CssParameter /Fill /PolygonSymbolizer /Rule /FeatureTypeStyle /UserStyle /NamedLayer /StyledLayerDescriptor Am I wrong to expect to see a fill-opacity CssParameter for this layer? I am using Mapserver 6.0.1 and, since it might have some bearing of this question, the OUTPUTFORMAT for the map is defined as follows: OUTPUTFORMAT NAME png32 DRIVER AGG/PNG MIMETYPE image/png IMAGEMODE RGBA EXTENSION png FORMATOPTION INTERLACE=OFF TRANSPARENT ON END If there is some pointers someone can offer to get this information returned properly, I'd appreciate it. Cheers, jtm -- View this message in context: http://osgeo-org.1560.n6.nabble.com/GetStyles-SLD-does-not-include-Opacity-value-tp4979066p4979230.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] GetStyles SLD does not include Opacity value
That did the trick beautifully. Thank you, Steve!! In summary, the layer is now defined as: LAYER NAME Poly_wOpacity TYPE POLYGON STATUS ON PROJECTION init=epsg:3857 END INCLUDE ../../../mapserv/conf/sdeconnection.inc.map DATA ...,SHAPE,SDE.DEFAULT METADATA wms_title Polygon Layer w/Opacity of 20 wms_srs epsg:3857 END CLASS NAME Polygon Layer w/Opacity of 20 STYLE COLOR 255 0 0 OPACITY 20 END END END And, the GetStyles request returns the expected fill-opacity CssParameter: StyledLayerDescriptor version=1.0.0 ... NamedLayer NamenonContributingDrainage/Name UserStyle FeatureTypeStyle Rule NameNon-Contributing Drainage Area/Name PolygonSymbolizer Fill CssParameter name=fill#ff/CssParameter CssParameter name=fill-opacity0.20/CssParameter /Fill /PolygonSymbolizer /Rule /FeatureTypeStyle /UserStyle /NamedLayer /StyledLayerDescriptor -- View this message in context: http://osgeo-org.1560.n6.nabble.com/GetStyles-SLD-does-not-include-Opacity-value-tp4979066p4979243.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
[mapserver-users] GetStyles SLD does not include Opacity value
Hello everyone, I have a Map file with a polygon layer that has an OPACITY value of 20 defined as: I see a correctly opaque display when I do a GetMap or a GetLegendGraphic, so I know the OPACITY value is set properly. However, when I perform a GetStyles request, the returned SLD doesn't include anything about the opacity. The response is: Am I wrong to expect to see a fill-opacity CssParameter for this layer? I am using Mapserver 6.0.1 and, since it might have some bearing of this question, the OUTPUTFORMAT for the map is defined as follows: If there is some pointers someone can offer to get this information returned properly, I'd appreciate it. Cheers, jtm -- View this message in context: http://osgeo-org.1560.n6.nabble.com/GetStyles-SLD-does-not-include-Opacity-value-tp4979066.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
[mapserver-users] Re: Some PointSymbolizer values are being ignored. SLD Syntax problem?
Thomas, Thank you for your quick response and investigation into this issue. I introduced the code changes noted in the changeset you provided and the PointSymbolizer's fill-opacity, stroke-opacity and stroke-width are now properly applied. I tried to extrapolate what you did in msSLDParseGraphicFillOrStroke to msSLDGenerateLineSLD and msSLDGeneratePolygonSLD to ensure it also applied the stroke-opacity values for those symbolizers but the code for reading the CssParameters and applying them in those methods is different enough that I'm not clear on how to do so properly. If you have a moment to throw out a code snippet for that in those locations, I'll be happy to try it out and include it in the ticket I'll write up about all this stuff shortly. If you don't have time, no worries. Your help has already been tremendously useful. I'll keep trying... Cheers jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Some-PointSymbolizer-values-are-being-ignored-SLD-Syntax-problem-tp7028772p7115837.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
[mapserver-users] Re: Some PointSymbolizer values are being ignored. SLD Syntax problem?
I've created two separate tickets for the issues brought up in this thread: 1. http://trac.osgeo.org/mapserver/ticket/4131 - regarding the parameters ignored in the PointSymbolizer and highlighting Thomas Bonfort's proposed fix. 2. http://trac.osgeo.org/mapserver/ticket/4132 - regarding the stroke-opacity parameter being ignored in the PolygonSymbolizer and possibly identifying the area of code that is problematic. Cheers jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Some-PointSymbolizer-values-are-being-ignored-SLD-Syntax-problem-tp7028772p7116092.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
[mapserver-users] Are there Mapserver-specific Extensions to Layer in WMC?
Hello everyone, Hopefully, a quick question...a no or (better) a yes with an idea where to find some documentation would be splendid. Using mode=map and passing the URL to a WMC in the context parameter, I am receiving the image I expect...but not quite the one I want. I'd like to be able to limit the extents of a given WMS in the WMC. I don't see anything within the spec that would allow me to define that at the LAYER level, so I was wondering if there are any Mapserver-specific extensions to the Layer Extension tag that might allow this type of information to be included and used. Something like OpenLayers' ol:maxExtent...? Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Are-there-Mapserver-specific-Extension-s-to-Layer-in-WMC-tp7109187p7109187.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
[mapserver-users] Re: Some PointSymbolizer values are being ignored. SLD Syntax problem?
Hello everyone, I've come back to this problem I submitted to the forum last month re: my PointSymbolizer's fill-opacity, stroke-opacity and stroke-width being ignored. I'm no further ahead, so I thought I'd update what I've read and tried, just to see if that jogs anyone's thoughts in terms of some issue related to this that I'm not aware of. I am fairly certain this functionality is supported, given what I read in the SLD implementation - fill-opacity post thread (http://osgeo-org.1803224.n2.nabble.com/SLD-implementation-fill-opacity-td1971368.html) from 2008. I've worked through the StyledLayerDescriptor.xsd as best I can and believe my definition is correct syntactically. I also tried creating a Polygon layer and applying a similarly-defined PolygonSymbolizer SLD to the one I am using for the PointSymbolizer to see if my syntax was correct (as much as I could given the two different types of data). This, too, suggests it is fine. When using the PolygonSymbolizer definition below, the fill-opacity and stroke-width values are applied in the output. Granted, the the stroke-opacity is still not but at least something worked that doesn't with the PointSymbolizer. I'm including an attachment that shows the results of using the two SLDs included in this thread (my PointSymbolizer and the PolygonSymbolizer above). The definitions are as close to the same as I can make them but, as you can see, the output is not. In the Point symbology, the fill-opacity (0.4), stroke-opacity (0.6) and stroke-width (10.0) are being ignored; and, in the Polygon symbology, the stroke-opacity (0.6) is being ignored but the fill-opacity (0.4) and stoke-width (10.0) are being correctly applied. http://osgeo-org.1803224.n2.nabble.com/file/n7110031/Point_vs._Polygon.png Could there be an issue with my Mapfile definition? In case it matters, here is my OUTPUTFORMAT definition for the Point and Polygon layers used above and the Point layer definition: I really can't imagine what else to try...Is it possible this stuff is broken in 6.0.1? Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Some-PointSymbolizer-values-are-being-ignored-SLD-Syntax-problem-tp7028772p7110031.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
[mapserver-users] Re: Mapserver /tmp files appearing all of a sudden w/CONTEXT parameter used
With respect to this problem of writing to the /tmp, rather than our /ogc_tmp/ (IMAGEPATH) directory, one thing we have just tried though that is interesting... I mentioned earlier we were upgrading from 5.4 (I mis-spoke when I said 5.6) to 6.0.1. As we still have both executables available on the same machine, we tried switching them and making the same URL request. When the executable is 5.4, the *.context.xml is written to /ogc_tmp/ but when the executable is 6.0.1, it is /tmp. For this particular functionality (the copying of the WMC's contents), could there be an error reading the IMAGEPATH variable in the code? My debugging capabilities in this particular environment are visual only (just eyeballing the code and trying to find my way through it). I can't actually step through the code at run time to see the value being read when trying to put together the path for where to put the file but it's the only thing I can think of that might explain what we're seeing. Does anyone know if this is a valid possibility? Steve, Thank you for your response. To answer your questions specifically, we do not set IMAGEPATH and/or IMAGEURL in the Apache config files; only in the map files. To be certain, though, I verified this again this morning. With respect to the permissions, the Mapserver user doesn't have log on permissions in this environment, so I can't get a shell as that user. However, I checked the permissions for the folder and they include o+rwx. I would think that should allow Mapserver to write to it. Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Mapserver-tmp-files-appearing-all-of-a-sudden-w-CONTEXT-parameter-used-tp7009495p7028718.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
[mapserver-users] Some PointSymbolizer values are being ignored. SLD Syntax problem?
Hello all, Hopefully, this is a quickie and there's just something wrong with my SLD syntax that I can't see. I have a map file with a layer that has a single point in it. I am trying to apply an SLD to it using the following URL: http://mydomain/ogc/myWmsLAYERS=PointSERVICE=WMSVERSION=1.1.1REQUEST=GetMapEXCEPTIONS=application%2Fvnd.ogc.se_inimageFORMAT=image%2FjpegSRS=EPSG%3A42304BBOX=-2903297.000,-789911.,3431059.000,3959683.000WIDTH=256HEIGHT=256sld=http://mydomain/agmaf/PointSld.xml The content of the PointSld.xml is: In the resulting image, the point's fill-opacity, stroke-opacity and stroke-width values are being ignored, although its stroke and fill colours are correct. I've tried changing around the SLD but generally just seem to cause more problems. Is there something obviously wrong with the syntax that I'm just not seeing? It looks correct to me...but then it always does, doesn't it? I am using Mapserver 6.0.1. Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Some-PointSymbolizer-values-are-being-ignored-SLD-Syntax-problem-tp7028772p7028772.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
[mapserver-users] Re: Mapserver /tmp files appearing all of a sudden w/CONTEXT parameter used
Yes indeed, Daniel!! This is exactly what is going on. I introduced a definition for TEMPPATH in that imagepath.inc.map file (so it will get applied to all map files) and voila! The context.xml files were placed in the appointed location. Thank you very much!!! Cheers, jtm FYI, I had the impression from the ticket (http://trac.osgeo.org/mapserver/ticket/3354#comment:6) that the parameter should be TMPATH. Mapserver didn't recognize that one, but was happy with TEMPPATH. Perhaps I just don't know how to interpret the comment properly. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Mapserver-tmp-files-appearing-all-of-a-sudden-w-CONTEXT-parameter-used-tp7009495p7029107.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
[mapserver-users] Re: Mapserver /tmp files appearing all of a sudden w/CONTEXT parameter used
Well, I've tried debugging Mapserver to figure this out and have learned a couple of things...but I still haven't solved the problem re: why these tmp files are suddenly appearing in the /tmp directory. I thought perhaps that some debugging had gotten turned on somewhere to cause those files to appear now, but learned, as I stepped through mapcontext.c and maphttp.c that the context file is always written out in this process. It's simply the caching location that dictates where. So, I went looking around and determined that up until 2 weeks ago, these files were being written to a directory called ogc_tmp. That directory is the one we point to with the IMAGEPATH/IMAGEURL parameters included in the imagepath.inc.map file pointed to with this line in the wmsContextServer map file: INCLUDE ../../../mapserv/conf/imagepath.inc.map Its contents are: IMAGEPATH /data/nl18/httpd/htdocs/ogc_tmp/ IMAGEURL /ogc_tmp/ I thought, perhaps, that all of sudden maybe the directory had changed or something so now Mapserver couldn't find it from the location its defined at in the mapfile, but when I cd into the directory that contains the wmsContextServer.map file, I can do a directory listing of /data/nl18/httpd/htdocs/ogc_tmp/ and see the contents of the directory. I went to the location of the Mapserver executable in /cgi-bin and did the same and can see the directory from there also. But, am I on the right track here? All the other temporary files continue to be sent to the /ogc_tmp/ directory (image files, for example). It's only the contexts that are being placed in the /tmp directory now. So, besides this IMAGEPATH setting (which doesn't appear to have changed any, either in the wmsContextServer.map file or in the environment itself), is there some other place that that ogc_tmp directory might be getting specified? And when Mapserver is unable to find whatever place that other place is telling it to find /ogc_tmp it's defaulting to /tmp? Would Mapserver do that...default to /tmp if it can't find a specified location for something? I've been comparing source-controlled map files for days now trying to find what's changed and is causing these files to suddenly appear in this new location. If anyone can think of ANYTHING that might point me somewhere useful, I would certainly appreciate it. cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Mapserver-tmp-files-appearing-all-of-a-sudden-w-CONTEXT-parameter-used-tp7009495p7026365.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
[mapserver-users] Re: Mapserver /tmp files appearing all of a sudden w/CONTEXT parameter used
Oh, one major thing that has changed recently is that, because we are upgrading from Mapserver 5.6 to 6.0.1, we updated the OUTPUTFORMAT definition in the mapfile. It used to be: OUTPUTFORMAT NAME png DRIVER 'GD/PNG' MIMETYPE 'image/png' IMAGEMODE PC256 #RGB FORMATOPTION INTERLACE=OFF EXTENSION 'png' #TRANSPARENT ON END But, I wouldn't have thought that this would have any bearing on this particular problem. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Mapserver-tmp-files-appearing-all-of-a-sudden-w-CONTEXT-parameter-used-tp7009495p7026382.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
[mapserver-users] Mapserver /tmp files appearing all of a sudden w/CONTEXT parameter used
Hello all!! For years now, we have been using an empty mapfile to create an image from WMC Contexts. This continues to work fine but, all of a sudden within the last week or so, we've started seeing a buildup of temporary files (under the /tmp directory) that Mapserver is creating when we do this. The content of each temporary file is the content of the WMC whose URL is being passed in the request. The contents of our empty mapfile is included below and a typical request would look something like this: http://myserver/ogc/wmsContextServer?MODE=mapCONTEXT=http://anotherserver/contexts/somecontext.xmlLAYERS=* Any ideas where to look for what might have changed in the environment that these files are appearing here all of a sudden? I thought perhaps someone had changed the IMAGEPATH/IMAGEURL values located in the INCLUDEd imagepath.inc.map file but no, that's the same as it ever was. Does anyone know what the code is doing when a CONTEXT parameter is passed to know what might make it, all of a sudden, start writing to a given directory? I'm assuming these temporary files have always been written somewhere but don't know what configuration parameter directs them where. Or, if they weren't being written previously (just worked with in memory somehow) what configuration setting would all of sudden make it start actually saving the context contents to files? Any ideas or pointers of where to look would be gratefully accepted. I can't find what's changed... Cheers, jtm MAP NAME wmsContextServer FONTSET ../../../mapserv/etc/fonts.txt SYMBOLSET ../../../mapserv/etc/symbols.sym SIZE 400 300 STATUS ON IMAGETYPE PNG IMAGECOLOR 255 255 255 EXTENT -2338951 -80 3008692 390 UNITS METERS CONFIG CGI_CONTEXT_URL 1 PROJECTION init=epsg:42304 END RESOLUTION 96 REFERENCE # Parameter settings NOT to change STATUS ON COLOR -1 -1 -1 OUTLINECOLOR255 0 0 # Parameter settings that will change if you replace the default 'refmap.png' IMAGE ../htdocs/images/refmap.png SIZE100 74 EXTENT -2.2e+006 -712631 3.0728e+006 3.84e+006 END # REFERENCE SCALEBAR STATUS EMBED INTERVALS 1 POSITION LL TRANSPARENT false UNITS kilometers STYLE 1 SIZE 100 7 LABEL SIZE small END END WEB INCLUDE ../../../mapserv/conf/imagepath.inc.map METADATA # Used for App Page Title wms_title WMC Context Client wms_connectiontimeout 60 END END OUTPUTFORMAT NAME png8 DRIVER AGG/PNG8 MIMETYPE image/png; mode=8bit IMAGEMODE RGBA EXTENSION png FORMATOPTION INTERLACE=off FORMATOPTION QUANTIZE_FORCE=on FORMATOPTION QUANTIZE_COLORS=256 TRANSPARENT ON END END -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Mapserver-tmp-files-appearing-all-of-a-sudden-w-CONTEXT-parameter-used-tp7009495p7009495.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
[mapserver-users] Re: Can't WFS getFeature w/MapServer 6.0.1
While waiting for a patch which will correctly fix the problems described in this thread (OGC queries to SDE failing), we have introduced a change to our version of mapsde.c that is allowing us to move forward. In the msSDELayerFreeItemInfo function, we changed the boolean expressions at lines 2116 and 2121 to check that sde wasn't NULL before checking the basedefs and joindefs as shown below. For the time being, this is allowing us to continue our work with no serious problems encountered so far. If we do come across some, I'll be sure to let you know. Cheers, jtm void msSDELayerFreeItemInfo(layerObj *layer) { #ifdef USE_SDE msSDELayerInfo *sde = NULL; int i; if (!msSDELayerIsOpen(layer)) { msSetError( MS_SDEERR, SDE layer has not been opened., msSDELayerFreeItemInfo()); } sde = layer-layerinfo; if (sde != NULL sde-basedefs) { SE_table_free_descriptions(sde-basedefs); sde-basedefs = NULL; } if (sde != NULL sde-joindefs) { SE_table_free_descriptions(sde-joindefs); sde-joindefs = NULL; } if (layer-iteminfo) { : -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Can-t-WFS-getFeature-w-MapServer-6-0-1-tp6600286p6611100.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
[mapserver-users] Re: WFS GetFeature w/Spatial Filter
This problem has turned out to be related to that described in thread http://osgeo-org.1803224.n2.nabble.com/Can-t-WFS-getFeature-w-MapServer-6-0-1-td6600286.html. It's working in our compiled version (with the fix described in that thread). -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WFS-GetFeature-w-Spatial-Filter-tp6604244p6611136.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
[mapserver-users] Re: Can't WFS getFeature w/MapServer 6.0.1
This is a correction and additional information to previous post. In my first post, I said However, the SDE log suggests the query took place and returned successfully. Upon closer inspection, this is not actually the case. Watching the SDE logs while trying to perform my query with 5.6.3, I see the following: Thu Jul 21 09:01:52 2011 - SDE Server Pid 11962 Registered, User: Username Schema: sde. Thu Jul 21 09:01:52 2011 - SDE Server 11962 exit'd with status 0 Thu Jul 21 09:01:52 2011 - SDE Server Pid 11962 Stopped, User: Username Schema: sde. Thu Jul 21 09:01:54 2011 - SDE Server Pid 11963 Registered, User: Username Schema: sde. Thu Jul 21 09:01:58 2011 - SDE Server 11963 exit'd with status 0 Thu Jul 21 09:01:59 2011 - SDE Server Pid 11963 Stopped, User: Username Schema: sde. When I perform the same query with 6.0.1, only one connection is made: Thu Jul 21 09:00:37 2011 - SDE Server Pid 11224 Registered, User: Username Schema: sde. Thu Jul 21 09:00:37 2011 - SDE Server 11224 exit'd with status 0 Thu Jul 21 09:00:37 2011 - SDE Server Pid 11224 Stopped, User: Username Schema: sde. So, while that one connection appears to perform successfully, there is no second connection at all (which according to what happens with 5.6.3, there should be). Somewhere between the handshaking (or whatever the heck it is) and the actual query being performed, something is going off the rails. Perhaps the connection value in layer-layerinfo being invalid in some way is stopping the second connection? It's also unable to free it. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Can-t-WFS-getFeature-w-MapServer-6-0-1-tp6600286p6606885.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
[mapserver-users] Re: Can't WFS getFeature w/MapServer 6.0.1
I've done some additional debugging and have learned that the mapwfs.c's msWFSGetFeature has changed significantly between 5.6.3 and 6.0.1. In particular, the problem seems to be starting at line 2457 in the 6.0.1 code (1248 in 5.6.3). At that point, a call to msOWSGetLayerExtent is made and part of that process creates and frees a connection to SDE. That is the first connection that is identified in the above posts (and in Michelle Douvil's post discussing her GDB findings). The second SDE connection is not made in 6.0.1 because, at line 2500 (still in mapwfs.c's), there is a call to msQueryByRect(map) which did not happen in 5.6.3. In that function (msQueryByRect), at line 1024, there is a msLayerClose(lp) call that ultimately takes us to the msSDELayerFreeItemInfo where the layer-layerinfo (and therefore the sde variable at line 2116) is already null because of the previous close (in the msOWSGetLayerExtent process) and so the if (sde-basedefs) condition gives a segmentation fault because it's trying to access a property in an object that doesn't exist. I really don't know anything about any of this code but know that what I describe is definitely what is happening as far as I can observe. I am hoping that someone who knows more about the code can identify whether there is a problem with the source code related to all this. And, if so, should I be putting something in the DEV forum about this? -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Can-t-WFS-getFeature-w-MapServer-6-0-1-tp6600286p6607659.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
[mapserver-users] Re: Can't WFS getFeature w/MapServer 6.0.1
Absolutely, we'd be happy to test a patch. I have some additional info (which you probably already know at this point but I'll pass it along just in case). I started looking into the WMS GetFeatureInfo request failure that Michelle was reporting in her post and found that ultimately it was the exact same problem. In that case, mapquery.c's msQueryByPoint makes an msLayerClose call on a layer that is not open (line 1535). Again, it ultimately gets down to the msSDELayerFreeItemInfo function where the layer-layerinfo is null, so sde-basedefs causes a segmentation fault. So, somewhere, there just needs to be a check that stops the attempt to free memory if the memory is not allocated at that time. Anyway, I'll send you an email directly to give you my coordinates and we'll arrange to try the patch. Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Can-t-WFS-getFeature-w-MapServer-6-0-1-tp6600286p6608020.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
[mapserver-users] Re: Upgrade MapServer from 5.4 to 6.0
Just in case someone is looking around for some answers and comes across this thread ... I'll reference another post where the originally posted problem (that is, the segmentation faults that occurs when doing a WMS getFeatureInfo request to SDE data, not the subsequent posts about the PNG background) is discussed in more detail. http://osgeo-org.1803224.n2.nabble.com/Can-t-WFS-getFeature-w-MapServer-6-0-1-td6600286.html#a6608020 -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Upgrade-MapServer-from-5-4-to-6-0-tp6563752p6608083.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
[mapserver-users] Re: Can't WFS getFeature w/MapServer 6.0.1
Is it possible that this post and Michelle Douvil's http://osgeo-org.1803224.n2.nabble.com/ArcSDE-9-3-1-QUERYMAP-QUERY-or-GETFEATUREINFO-not-workin-Mapserver-6-td6600091.html are having the same problem? I ask because a co-worker has tried using MS4W 3.0.3 to do a WMS GetFeatureInfo request and he gets an error (mapserv.exe has encountered a problem and needs to close.) when the Mapfile is using SDE to access the data, but it works fine when he uses a Shapefile...kind of like what's happening to me with my WFS GetFeature query on Linux. Further to my own problem, I now have a 5.6.3 compiled with debug also and, using Data Display Debugger, find that in 5.6.3 the connection and stream properties of layer-layerinfo are 0x0 just before the point of failure in 6.0.1 but when I break at the same location with 6.0.1, there are addresses still referenced there. Like I said in my previous post, SDE appears to believe the connection has been closed successfully and given the error gdb gave it's almost like the connection has already been closed but the address not cleared from the pointer. I don't know. I'm grasping at straws here and hoping someone who knows more about the SDE connection code can provide some guidance. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Can-t-WFS-getFeature-w-MapServer-6-0-1-tp6600286p6604123.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
[mapserver-users] WFS GetFeature w/Spatial Filter
Hello, I am trying to do a getFeature request with a Filter using the WITHIN spatial operator. I have a MapServer with the GEOS library compiled in and I am able to get an example working (or not, as the case may be) that exhibits my problem. I am getting an incorrect response to my getFeature request. The request is against a line layer (Shapefile, not SDE as noted in another of my active posts at the moment...they're all really just trying to get to this point) and includes a spatial filter delineating a polygon. I want to learn what features in the line layer (Roads25M) actually fall within the polygon submitted (I am limiting the response to 2 features just for testing purposes; ultimately I will want to know all the roads that fall within the polygon.) My request is: http://domain/ogc/joanne?typename=Roads25Mmaxfeatures=2filter=ogc:Filter xmlns:ogc=http://www.opengis.net/ogc;ogc:Orogc:Intersectsogc:PropertyNamemsGeometry/ogc:PropertyNamegml:Polygon xmlns:gml=http://www.opengis.net/gml; srsName=EPSG:42304gml:outerBoundaryIsgml:LinearRinggml:coordinates decimal=. cs=, ts= -1679350.2423216,1550622.4768356 -1816404.334979,992616.52815892 -1552085.7277111,718508.34284404 -1258398.3863023,845772.85745452 -1385662.9009128,1296093.4476147 -1679350.2423216,1550622.4768356/gml:coordinates/gml:LinearRing/gml:outerBoundaryIs/gml:Polygon/ogc:Intersects/ogc:Or/ogc:Filterservice=wfsversion=1.0.0request=getfeature Here's the filter string a little easier to read: filter=ogc:Filter xmlns:ogc=http://www.opengis.net/ogc; ogc:Or ogc:Intersects ogc:PropertyNamemsGeometry/ogc:PropertyName gml:Polygon xmlns:gml=http://www.opengis.net/gml; srsName=EPSG:42304 gml:outerBoundaryIs gml:LinearRing gml:coordinates decimal=. cs=, ts= -1679350.2423216,1550622.4768356 -1816404.334979,992616.52815892 -1552085.7277111,718508.34284404 -1258398.3863023,845772.85745452 -1385662.9009128,1296093.4476147 -1679350.2423216,1550622.4768356/gml:coordinates /gml:LinearRing /gml:outerBoundaryIs /gml:Polygon /ogc:Intersects /ogc:Or /ogc:Filter The response is: lt;wfs:FeatureCollection xsi:schemaLocation=quot;http://www.opengis.net/wfs http://schemas.opengis.net/wfs/1.0.0/WFS-basic.xsd http://mapserver.gis.umn.edu/mapserver http://lt;domaingt;/ogc/joanne?SERVICE=WFSVERSION=1.0.0REQUEST=DescribeFeatureTypeTYPENAME=Roads25MOUTPUTFORMAT=XMLSCHEMA; gml:boundedBy gml:nullmissing/gml:null /gml:boundedBy /wfs:FeatureCollection There are no errors in logs that I can find and as far as MapServer is concerned it appears to have successfully processed the request. I have created a Shapefile out of the Polygon coordinates I provide in the request and when I display it and the Roads Shapefile in ArcMap, there is obviously some overlap. In ArcMap, I can perform a Select by Location where I Select features from Roads that intersect Polygon. It highlights 9 records. I assume there is something wrong with the XML defining my spatial filter that is causing no results to be returned but can't determine what it might be. Does anything jump out at anyone? Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WFS-GetFeature-w-Spatial-Filter-tp6604244p6604244.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
[mapserver-users] Re: Mapfile layer not displaying in MapServer 6.0.1 (does in 5.4)
Steve, Thank you for your suggestions (including the one you sent me directly suggesting explicitly setting a width in the roads style). Today, when I went back to this, the request worked without my changing anything in the Mapfile. I have returned the data to be loaded through SDE (not the Shapefile) and it still seems to be working. I hate it when that happens!! The only thing I can think of is that something got cached somewhere along the way on Friday and the requests were never truly being processed as I was changing the Mapfile trying to get it to work. Obviously, somewhere along the way I got the Mapfile contents right, but wasn't seeing it in the browser. Okay, well. I guess I'll continue along the road to try and figure out my real problem now. Sorry for wasting everyone's time but I'll probably be back. Cheers jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Mapfile-layer-not-displaying-in-MapServer-6-0-1-does-in-5-4-tp6587562p6596058.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
[mapserver-users] Mapfile layer not displaying in MapServer 6.0.1 (does in 5.4)
I'm having a problem upgrading from MapServer 5.4 to 6.0.1, so thought I'd start from scratch trying to isolate my ultimate problem. To start, all I am trying to do is display a simple Mapfile line layer in my browser. I have (what I believe to be) the simplest Mapfile I can make that allows me to do a mode=map display in 5.4. But, when I try to use the same Mapfile with the 6.0.1 installation, the layer does not display. I have reviewed the MapServer Migration Guide and the online 6.0 documentation but can't see anything in this simple Mapfile that would cause the lack of display...although it's probably staring me in the face, I know. Anyway, I don't believe there's an actual problem per se (debugging results attached below), I assume there's just a difference between how I'm supposed to define something for 5.4 and how it should be defined for 6.0. The only thing I can think of that might be affecting this is the OUTPUTFORMAT changes but I certainly can't see what the problem might be. I also looked into some of the line styling changes, but don't think any of them apply to this example. I'm hoping another set of eyes (that have more experience with these changes than I) can identify what I am defining incorrectly for 6.0. Cheers, jtm I provide the URL http://domain/ogc/joanne?mode=maplayers=Roads25 and when I point to the MapServer 5.4's domain I see the roads layer, but if I point to the MapServer 6.0.1 domain (which is basically the same as the 5.4 environment - essentially a fully complete mirror), the result is an empty image. No errors, no problem; but no roads either. The wrapper file contains: MS_MAPFILE=${MS_APPSPATH}joanne/map/joanne_sub.map export MS_MAPFILE ${MS_CGIBIN}mapserv And the Mapfile joanne_sub.map contents are below. It's as simple as I could make it. MAP NAME mySubMap FONTSET ../../../mapserv/etc/fonts.txt SYMBOLSET ../../../mapserv/etc/symbols.sym SIZE 800 600 IMAGETYPE PNG IMAGECOLOR 255 255 255 EXTENT -2903297.697389 -789911.662978 3431059.382789 3959683.443689 UNITS METERS PROJECTION init=epsg:42304 END WEB IMAGEPATH /path/httpd/htdocs/ogc_tmp/ IMAGEURL /ogc_tmp/ END #end WEB OUTPUTFORMAT NAME 'AGG' DRIVER AGG/PNG MIMETYPE image/png IMAGEMODE RGB EXTENSION png END #end OUTPUTFORMAT LAYER NAME 'Roads25M' TYPE LINE STATUS ON CONNECTION host,port,sde,username,password CONNECTIONTYPE SDE DATA 'XDM_XDATA.TRN_NSF_ROAD_V6_CT25M_R2,SHAPE,SDE.DEFAULT' CLASS STYLE COLOR 0 0 0 END #end STYLE END #end CLASS END #end LAYER END #end MAP 5.4 debug results for http://5.4Domain/ogc/joanne?mode=maplayers=Roads25 request: [Fri Jul 15 12:48:39 2011].722684 msDrawMap(): Layer 0 (Roads25M), 0.742s [Fri Jul 15 12:48:39 2011].722837 msDrawMap(): Drawing Label Cache, 0.000s [Fri Jul 15 12:48:39 2011].722846 msDrawMap() total time: 0.742s [Fri Jul 15 12:48:39 2011].735882 msSaveImage() total time: 0.013s [Fri Jul 15 12:48:39 2011].735953 mapserv request processing time (loadmap not incl.): 0.756s [Fri Jul 15 12:48:39 2011].735960 msFreeMap(): freeing map at 0x8b5f1a0. 6.0.1 debug results for http://6.0.1Domain/ogc/joanne?mode=maplayers=Roads25 request: [Fri Jul 15 12:48:41 2011].658817 msDrawMap(): rendering using outputformat named png (AGG/PNG). [Fri Jul 15 12:48:41 2011].658911 msDrawMap(): WMS/WFS set-up and query, 0.000s [Fri Jul 15 12:48:41 2011].658996 msDrawMap(): Drawing Label Cache, 0.000s [Fri Jul 15 12:48:41 2011].659004 msDrawMap() total time: 0.002s [Fri Jul 15 12:48:41 2011].743425 msSaveImage(stdout) total time: 0.084s [Fri Jul 15 12:48:41 2011].743545 mapserv request processing time (loadmap not incl.): 0.087s [Fri Jul 15 12:48:41 2011].743557 msFreeMap(): freeing map at 0x8261138. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Mapfile-layer-not-displaying-in-MapServer-6-0-1-does-in-5-4-tp6587562p6587562.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
[mapserver-users] Re: Mapfile layer not displaying in MapServer 6.0.1 (does in 5.4)
I tried making the Mapfile even simpler and removing SDE from the mix. I created a Shapefile of the layer and now the Mapfile is: MAP NAME mySubMap FONTSET ../../../mapserv/etc/fonts.txt SYMBOLSET ../../../mapserv/etc/symbols.sym SHAPEPATH /path/httpd/apps/joanne/data SIZE 800 600 IMAGETYPE PNG IMAGECOLOR 255 255 255 EXTENT -2903297.697389 -789911.662978 3431059.382789 3959683.443689 UNITS METERS PROJECTION init=epsg:42304 END WEB IMAGEPATH /path/httpd/htdocs/ogc_tmp/ IMAGEURL /ogc_tmp/ END #end WEB OUTPUTFORMAT NAME 'AGG' DRIVER AGG/PNG MIMETYPE image/png IMAGEMODE RGB EXTENSION png END #end OUTPUTFORMAT LAYER NAME 'Roads25M' TYPE LINE STATUS ON DATA roads CLASS STYLE COLOR 0 0 0 END #end STYLE END #end CLASS END #end LAYER END #end MAP Still no luck. Works in 5.4; not in 6.0.1. The only other thing I can think of to try next is to actually compile 5.4 and 6.0.1 on the same host and just call them appropriately in the exact same environment. I'll try that next, but if someone sees something glaringly erroneous for a MapServer 6.0 Mapfile I'd appreciate a heads-up. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Mapfile-layer-not-displaying-in-MapServer-6-0-1-does-in-5-4-tp6587562p6588088.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
[mapserver-users] Re: EPSG:42304 - EPSG:3978? or 3979?
I'll ignore the questions and replies that have hijacked the thread and say a big thank you to both Frank and Jeff. I'll start looking into specifically what's required to switch our apps from EPSG:42304 to EPSG:3978 (now that I know which one it should be). Thanks to you both! And, Frank...nice to hear from you. Hope all is well! -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/EPSG-42304-EPSG-3978-or-3979-tp6530222p6533451.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
[mapserver-users] EPSG:42304 - EPSG:3978? or 3979?
Hello all, I know this isn't really a MapServer question, but I know that some of the people involved in this stuff read this forum and don't know where else to post it. I'm trying to understand if the old, unofficial, NRCan EPSG:42304 code (often used in Canadian MapServer installations) translates to the new official EPSG:3978 or EPSG:3979 code. But, I can't figure out the answer and am hoping that someone who reads this can straighten me out. I know little about projections at this level and apologize if I'm using the wrong terminology or am passing along mis-information in asking for clarification. There was an custom EPSG code that many, in Canada, used in MapServer installations, EPSG:42304. From what I understood, it was a projection defined by Natural Resources Canada and others adopted it as a kind of unofficial standard. The only definition of that projection I've been able to find, in terms of how to define it in a PROJ4 file (which is why I'm looking for it) is at: http://spatialreference.org/ref/sr-org/6754/, where its parameters are specified as: +proj=lcc +lat_1=49 +lat_2=77 +lat_0=49 +lon_0=-95 +x_0=0 +y_0=0 +ellps=GRS80 +datum=NAD83 +units=m +no_defs I was led to believe this projection was submitted for inclusion as an official EPSG code and that it had been approved. Now, there's two official EPSG codes that I've found that seem to be similar. They are EPSG:3978 and EPSG:3979. Using the EPSG Registry's (http://www.epsg-registry.org/) retrieve by code tool, I've looked at the definitions of 3978 and 3979 and can find the only difference between the two to be their Geodetic Datum (one being Canada and one being North America) which ultimately only seems to be different with respect to their Area of Use. When I look at the PROJ.4 site's EPSG list (http://svn.osgeo.org/metacrs/proj/trunk/proj/nad/epsg), the entries for the two are the exact same, which suggests my interpretation of the Area of Use being the only difference might be correct. # NAD83 / Canada Atlas Lambert 3978 +proj=lcc +lat_1=49 +lat_2=77 +lat_0=49 +lon_0=-95 +x_0=0 +y_0=0 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs # NAD83(CSRS) / Canada Atlas Lambert 3979 +proj=lcc +lat_1=49 +lat_2=77 +lat_0=49 +lon_0=-95 +x_0=0 +y_0=0 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs Unfortunately, neither of these parameter lists are exactly the same as the 42304 definition from above (these have +towgs84=0,0,.. and the 42304 one has +datum=NAD83...for all I know, those two parameters are defining the same thing). So, my question is: Are either of these official EPSG codes actually the same as that unofficial 42304 code? Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/EPSG-42304-EPSG-3978-or-3979-tp6530222p6530222.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
[mapserver-users] Re: EPSG:42304 - EPSG:3978? or 3979?
Oh Jeff, It seems I'm always saying no to that question from you and then finding out later that I should have said yes but didn't realize it. So, first of all, thank you for continuing to respond to my questions even when I'm repeatedly too slow to follow something that's readily apparent to others. Unfortunately, in answer to your question this time, I'm going to go with sort of. You have confirmed my understanding of what happened, but I'm still not sure which of these two new EPSG codes I should be picking to update my own application and get away from using 42304. Can I just change the EPSG code to one or the other of the new ones (or either, since they seem to be the same as far as I can tell)? Or, do I have to reproject the data from the old 42304 to one of them before I can do this? When I compare the PROJ4 definitions of the new and old, they aren't the same and that suggests I need to reproject but someone told me that I should not have to do that... that I should be able to just change the code in my Mapfiles and OpenLayers WMS definitions. How do I proceed? Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/EPSG-42304-EPSG-3978-or-3979-tp6530222p6530859.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
[mapserver-users] WMS GetFeatureInfo - is it possible to support multiple INFO_FORMATs?
Hello all, I've been trying to figure out how to do this for a while this afternoon and thought perhaps I'd better step back and find out if it's even possible before I keep struggling with it. Is it possible to support multiple INFO_FORMAT properties to a WMS GetFeatureInfo request? That is, is there a way to configure a MAP file that is acting as a WMS Server to return HTML or XML based on what is asked for in the request itself? Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WMS-GetFeatureInfo-is-it-possible-to-support-multiple-INFO-FORMATs-tp6349109p6349109.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
[mapserver-users] Re: WMS GetFeatureInfo - is it possible to support multiple INFO_FORMATs?
Thanks for your response, Jeff; and, to be honest, I didn't think it could. However, it's quite interesting that you would reference that particular documentation because it was exactly that section that started me looking into how it might be possible. There are three links to what I was guessing was ultimately the same MAP file at http://demo.mapserver.org/cgi-bin/wms: Here are working examples of GetFeatureInfo requests: text/plain, text/html, gml ... However, given your confirmation and my own previous understanding, I imagine it's likely a script that is looking at different MAP files based on the value passed for the INFO_FORMAT parameter. Again, thanks. It's always nice to know when I can stop beating my head against a wall. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WMS-GetFeatureInfo-is-it-possible-to-support-multiple-INFO-FORMATs-tp6349109p6349243.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
[mapserver-users] Re: WMS GetFeatureInfo - is it possible to support multiple INFO_FORMATs?
Ah, you're two steps ahead of me again! Thanks, -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WMS-GetFeatureInfo-is-it-possible-to-support-multiple-INFO-FORMATs-tp6349109p6349248.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
[mapserver-users] Re: WMS GetFeatureInfo - is it possible to support multiple INFO_FORMATs?
To clarify, I want to be able to make two completely separate GetFeatureInfo requests to a WMS Server; one that will return HTML and the other will return XML. As far as I'm understand from the MapServer documentation, I need to set wms_feature_info_mime_type parameter to one of three values: - text/xml - text/html - application/vnd.ogc.gml How can I set it to two of those values? Or, am I supposed to be able to do something like the following in a LAYER's WEB object? wms_feature_info_mime_type text/html wms_feature_info_mime_type text/xml And, if I can do that, how can I point to two different TEMPLATES? An HTML one and an XSL for the XML? Or, again, am I supposed to be able to have two TEMPLATE keywords in a single LAYER? Or, am I completely missing some very important item in all these discussions? BTW, Rahkonen Jukka, I particularly appreciated your description of these items at: http://osgeo-org.1803224.n2.nabble.com/GetFeatureInfo-request-doubts-td5083816.html http://osgeo-org.1803224.n2.nabble.com/GetFeatureInfo-request-doubts-td5083816.html . -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WMS-GetFeatureInfo-is-it-possible-to-support-multiple-INFO-FORMATs-tp6349109p6349329.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
[mapserver-users] Re: WMS GetFeatureInfo - is it possible to support multiple INFO_FORMATs?
I'm starting to think that what I might do is support HTML and GML and simply not return any geometry for the GML. Does that sound like a way to do what I need? -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WMS-GetFeatureInfo-is-it-possible-to-support-multiple-INFO-FORMATs-tp6349109p6349334.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
[mapserver-users] Re: WMS GetFeatureInfo - is it possible to support multiple INFO_FORMATs?
There is no such thing as text/xml. Where the heck did I come up with that? Okay, I know how to support HTML and GML at the same time. Sorry about all that... -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WMS-GetFeatureInfo-is-it-possible-to-support-multiple-INFO-FORMATs-tp6349109p6349382.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
[mapserver-users] Re: Font/Symbol size change with different MapServer
Since you folks are usually so good on this forum, the lack of response to my previous reply suggests there is something wrong with my question. I'm guessing there's something about this that I just don't get which is glaringly obvious to anyone who knows anything about it. I'm hoping someone can help me understand what might be causing the discrepancy I am seeing between our two different environments and further, where to look to verify those differences exist. In the map file in question no DEFRESOLUTION is set; so, as I understand the documentation I have found, it defaults to 72. The RESOLUTION is set in the map file to 96. Since the DEFRESOLUTION and RESOLUTION are different, the symbology in the layer definitions will be adjusted by a factor corresponding to RESOLUTION/DEFRESOLUTION. More specifically, this is done by multiplying the layer-scalefactor value by the resolution/defresolution ratio in msPrepareImage(). Finally, the documentation suggests that these parameters will affect size (for symbols and labels)...sure sounds like it's related to my problem. However, as the map file is the same on both machines (extracted directly from our source control), there has to be something different about the MapServer configuration itself that is causing the factor computed by multiplying the layer-scalefactor value by the resolution/defresolution ratio to come up with different values on the different machines and drawing the symbols and labels different sizes. Somewhere along the way, some value is being used that is different and I'm trying to identify what and where that might be so I can make sure they are the same. If any of this suggests that I have a completely wrong understanding of this stuff, please let me know where I can find a more detailed explanation. Or, if this is the wrong track, I'd appreciate being pointed in another direction. As it is at the moment, I'm completely stuck. Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Font-Symbol-size-change-with-different-MapServer-tp6310467p6313449.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
[mapserver-users] Re: Font/Symbol size change with different MapServer
Good lord! I've discovered the problem. They have upgraded the version of MapServer on one of the machines and not the other. One is 5.4 (before all this DEFRESOLUTION stuff was introduced) and the other is 5.6.4 (after). So, the calculations are completely different. Okay, got it. Sorry for wasting your time with the longer explanation above. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Font-Symbol-size-change-with-different-MapServer-tp6310467p6313512.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
[mapserver-users] Font/Symbol size change with different MapServer
I have created a WMS containing a layer of points and labels that looks fine in our development environment. The font and symbol sizes are as I expect them to be. However, when I move this same WMS to our production environment, they display bigger...markedly so. I'll try to attach the pictures (with fingers crossed that it works). Is there some default value that is part of the MapServer configuration somewhere that could be affecting the size of the fonts and symbols? Is there a resolution thing somewhere I need to be concerned about? So far I haven't read anything to this effect anywhere. Thanks, jtm http://osgeo-org.1803224.n2.nabble.com/file/n6310467/Development.png http://osgeo-org.1803224.n2.nabble.com/file/n6310467/Production.png -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Font-Symbol-size-change-with-different-MapServer-tp6310467p6310467.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
[mapserver-users] Re: Font/Symbol size change with different MapServer
I came across that in the documentation, but couldn't find where I would check that. I understand setting it in the map files; but where is the default set? Where do I look to see if they are different on the two systems? -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Font-Symbol-size-change-with-different-MapServer-tp6310467p6310801.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
[mapserver-users] Yet another Dynamic WMS server question
Similar to the type of thing I'm doing in: http://osgeo-org.1803224.n2.nabble.com/Changing-map-file-parameter-Label-Color-via-URL-td5771138.html We have a dynamic map file that we use to display a number of dynamically generated pieces of information. Currently, the Map file has a projection defined and, as long as we're working with data that is in that projection, all is well. However, sometimes we're working with data in a different projection, so my question is whether it is possible to tell MapServer what the projection of the data is in within the WMS request also? So, given a URL like this: http://domain/dynamicWms_en?map.layer[mylayer]=DATA+'path2/shapefile'+ENDLAYERS=mylayerFORMAT=image/pngSRS=EPSG:2957... I'd like to introduce something like: map.projection='init%22epsg:2957' to indicate what projection the data the map file is pointing to is in that projection. Is that even possible? And, if yes, can anyone give me an idea what the syntax is because I've been trying and can't get it to work. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Yet-another-Dynamic-WMS-server-question-tp6288703p6288703.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
[mapserver-users] Re: msProcessProjection(): Projection library error. no options in 'init' file
I can confirm that the PROJ_LIB environment variable is not set in the httpd configuration file. Adding it has solved the problem. Yay!! In addition, I got some answers to my questions regarding where MapServer was getting this information if that environment variable wasn't in the Apache configuration. By default, it was using a directory provided at compile time, as suggested in a previous post. Obviously, that path wasn't the same as the one we thought it was using. Now we know and our choice is either to leave the PROJ_LIB environment variable set or to just overwrite the epsg file in the proper location. Special thanks to Jeff for his input here. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/msProcessProjection-Projection-library-error-no-options-in-init-file-tp6273726p6277431.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
[mapserver-users] msProcessProjection(): Projection library error. no options in 'init' file
We are trying to deploy a WMS to production that we have working in a development environment. When I try to do a getCapabilities request in the prod environment, I get the message provided in the subject above. I looked at the documentation for this error message at http://mapserver.org/errors.html. It tells me: The message tells you that the projection you’re trying to use isn’t defined in the epsg file. Open your epsg file in a text editor and search for your projection to make sure that it exists. I have checked the epsg file in the prodution environment and the projection I am asking for is included. Further, I have run a file comparison on the epsg file in dev and the one in prod and they are the same. Does anyone know of other reasons MapServer could be unable to find the projection definition? Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/msProcessProjection-Projection-library-error-no-options-in-init-file-tp6273726p6273726.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
[mapserver-users] Re: msProcessProjection(): Projection library error. no options in 'init' file
I tried putting: CONFIG PROJ_LIB /usr/share/proj/ into the map file on production and it is now working. So, the problem is that the MapServer in production is not looking where I expect it to for the epsg file. Is there a configuration thing I need to do? What dictates where MapServer looks for that file by default? -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/msProcessProjection-Projection-library-error-no-options-in-init-file-tp6273726p6273806.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
[mapserver-users] Re: msProcessProjection(): Projection library error. no options in 'init' file
Thanks for your response Jeff. I'm guessing you responded before you received my followup post regarding the location of the epsg file. However, to address the concern, I did check and the MAP's projection is defined with lowercase epsg and the wms_srs list is uppercase EPSG following the example found at: http://mapserver.org/ogc/wms_server.html discussing Map PROJECTION and wms_srs metadata. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/msProcessProjection-Projection-library-error-no-options-in-init-file-tp6273726p6273896.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
[mapserver-users] Re: msProcessProjection(): Projection library error. no options in 'init' file
No, no, no! I didn't solve it. I found out more specifically what the problem is and have got a work around but I have not solved the problem as far as I am concerned. I have an epsg file in a proj directory on the MapServer host. As far as I was concerned, MapServer is using that file to indicate which epsg projections are supported. However, MapServer is obviously not looking at that file because the projection in question definitely exists in it and when I specifically point to the file's location within the Map file itself (using the CONFIG PROJ_LIB directive), and thereby overriding where ever it is MapServer is looking by default, everything works. I need to know how to ensure the configuration of MapServer is properly looking at the location I expect it to. At http://mapserver.org/installation/unix.html, I see a --with-proj=/usr/local compile option. What happens if someone didn't include that when they compiled? Or, if that's not even possible, is there any way to find out from the MapServer executable itself where it is looking for the epsg file? And, further, if it is looking somewhere and doesn't find one at all, does it use some kind of default list? -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/msProcessProjection-Projection-library-error-no-options-in-init-file-tp6273726p6274097.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
[mapserver-users] Re: msProcessProjection(): Projection library error. no options in 'init' file
Ooops! Sorry! I just reacted to the Correct, glad you solved it! line. I'll look into the Apache configuration you suggested. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/msProcessProjection-Projection-library-error-no-options-in-init-file-tp6273726p6274126.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
[mapserver-users] Re: SLD parameter in GetMap and GetLegendGraphic requests
Okay, well, it was worth a shot... We are moving the location of dynamically-generated SLDs to a publicly-accessible area of our website. They will be remote from the Shapefiles they are used for, but we'll survive. I thought I should add a final comment to provide (what appears to be) the only resolution in case someone else is trying to find a way around the same thing in future. Cheers all, -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/SLD-parameter-in-GetMap-and-GetLegendGraphic-requests-tp6155608p6161896.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
[mapserver-users] Re: WMS w/SLD - Can I get rid of the labels returned with a GetLegendGraphic request?
Another one bites the dustI guess this one isn't possible as asked either. I did want to let any future readers know what we have done to address this is create a legend generator (for lack of a better term) web service that uses a getStyles request for a given layer, parses the SLD and generates image files representing each rule. Then, it builds a JSON object with all the details to build the legend presentation in whatever client calls it (including the paths to the legend icons). The client can decide if it wants to display the text that goes with a given legend icon itself. Cheers all, -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WMS-w-SLD-Can-I-get-rid-of-the-labels-returned-with-a-GetLegendGraphic-request-tp6146719p6161941.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
[mapserver-users] Another dynamic WMS server and WMC question - dynamic Name anyone?
Just looking for ideas regarding another dynamic WMS and WMC question ... As described in other posts about this project I am working on, we have a process that dynamically generates Shapefiles and then, using a blank Map file acting as a WMS server, we do GetMap requests that include the DATA path to the new file to add the individual vector layers to a single image. We have another blank Map file that we use to print these layers. For this we pass a WMC context that defines all the layers that should be included. While I hope that background explanation is good enough for the purposes of this post, if it is not, a more detailed description of this process (including examples) can be found at:http://osgeo-org.1803224.n2.nabble.com/mode-map-with-WMC-of-dynamic-lt-Layer-gt-failing-with-getString-error-td5824774.html#a5831237 Now, my problem... The blank WMS server Map file has 3 layers in it: a point, a line and a polygon. Each have their own unique name: Point, Line, Polygon. When we add a dynamic vector/WMS layer individually, the WMS GetMap request uses that name to indicate which layer provides the basic configuration. When that layer is added to the WMC for printing, that name is used as the value of the Layer's Name entity in the LayerList to indicate which Map file layer to use. All this is fine unless there are more than one of the same type of layer added to the WMC. At that time, there are multiple LayerList Layers with the same Name value (say, Line) in the WMC. So, my question is this... Is there any way of dynamically assigning a name to a layer in the mapfile. I really can't imagine that there is but am wondering if there's some way of using something like the DATAPATTERN for a layer where, say, as long as the format of characters match it will use that layer. Something like: LAYER NAME line NAMEFORMAT Line* STATUS ON TYPE line PROJECTION init=epsg:42304 END CLASS COLOR 0 0 0 NAME Some Line Layer END METADATA wms_title Some Line Layer END END So that if I made a request as below it would resolve to figuring out what the correct layer to use is but still allow that unique name to be used to reference it: http://localhost/cgi-bin/mapserv.exe?map=/ms4w/apps/testing/dynamic.maprequest=getMapservice=wmsversion=1.1.1map.layer[LineOfRoads]=DATA ‘id3/vector’ ENDlayers=LineOfRoadssrs=EPSG:42304bbox=-2338951,-80,3008692,390format=image/pngheight=300width=300 I don't really expect that it would actually work like this...the amount of time necessary for processing and resolving requests would be prohibitive. I'm just trying to suggest something that might convey what we're trying to accomplish in the hopes of causing some creative juices to spill out there and asking if anyone knows of any way (even an obscure one) that will allow us to get unique Name values into the printing WMC. Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Another-dynamic-WMS-server-and-WMC-question-dynamic-Name-anyone-tp6162084p6162084.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
[mapserver-users] SLD parameter in GetMap and GetLegendGraphic requests
The MapServer documentation says the value of an SLD parameter must point to a remote SLD (SLD=http://URL_TO_SLD). Okay, I understand, but I still have to ask this question... Is there any way the path can be provided to a directory that exists on the MapServer host itself? One of our processes generates a shapefile and the SLD to symbolize it dynamically. We then use a blank MAP file (acting as a WMS Server) to load up the data with the SLD for display (using a GetMap request). All the generated files are stored in the same directory on the same system as MapServer. MapServer can get at the shapefile because the blank MAP file has a SHAPEPATH in it and we just pass along the rest of the path to the files through the GetMap request's DATA parameter. For example: http:///ogc/printVecWms_en?map.layer[LineString]=DATA '6543c79b-8c56-4dd1-8a89-ef8dde819e02/LineString' ENDVERSION=1.1.1SLD=http:///printVecData/SLD/6543c79b-8c56-4dd1-8a89-ef8dde819e02/LineString.sldTRANSPARENT=trueLAYER=LineStringSERVICE=WMSREQUEST=GetLegendGraphicSTYLES=EXCEPTIONS=application/vnd.ogc.se_inimageFORMAT=image/jpegSRS=EPSG:42304BBOX=-397165.01997824,-789911,2108966.9600435,1716220.9800218WIDTH=256HEIGHT=256 Unfortunately, the area where these files are created is not accessible externally, nor do we want them to be (particularly the shapefiles). This means we have to move the location of the SLDs to a place where we do not forbid access in order to use the syntax above. And that works fine. However, we'd really, really like to keep the SLDs closely associated (physically) with the Shapefiles they were built for. So, if anyone knows some syntax or a way that we can pass a value through the SLD parameter that is not publically accessible via http, I'd love to here it!! Is there some way of specifying the SLD to use within the MAP file itself and consequently a way that I can set it in the map.layer[LineString]= value instead? I've never heard of such a thing nor have I found anything that suggests I can, I'm just throwing it out there to see if anything jumps in to anyone's mind. Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/SLD-parameter-in-GetMap-and-GetLegendGraphic-requests-tp6155608p6155608.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
[mapserver-users] WMS w/SLD - Can I get rid of the labels returned with a GetLegendGraphic request?
Is it possible to get the GetLegendGraphic response from MapServer to not include the labels for the classes (I'd just like a column of icons)? Further info that may, may not, be relevant is that I am using an SLD to display the WMS. I can get the legend graphic fine, but the problem is that is says default next to the colour swatch because default is the value provided in the of the in the SLD. The data and the SLD are generated dynamically as part of a larger process and there really is no useful name that can be associated with the colour swatch. However, Name is a required entity of NamedLayer in the SLD, so I have to put something in there. But, I don't want to see it in the legend... Is there any way I can stop it? -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WMS-w-SLD-Can-I-get-rid-of-the-labels-returned-with-a-GetLegendGraphic-request-tp6146719p6146719.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
[mapserver-users] Re: WMS w/SLD - Can I get rid of the labels returned with a GetLegendGraphic request?
Oh, sorry, I thought I'd add that I looked around in the docs for the MapFile to see if I could set something there to turn off labels in the legend but didn't have any luck. If someone knows of something through that route, I'd be happy to hear about it. Really, anything would be fine. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/WMS-w-SLD-Can-I-get-rid-of-the-labels-returned-with-a-GetLegendGraphic-request-tp6146719p6146892.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
[mapserver-users] Dynamic WMS Service - Projections problem?
First, the background... I have a WMS Service that has an empty point layer in it. I use it to display labels at dynamically generated coordinates by passing x,y within the request. For example: http://domain/ogc/mapserv?map=/path/fenceLabelWms_en.mapmap.layer[fencelabel]=FEATURE POINTS -105.5620661354989 51.14394472618872 END TEXT Fence 1 ENDmap.layer[fencelabel].class[0]=COLOR 255 0 0 ENDmap.layer[fencelabel].class[0]=LABEL COLOR 255 0 0 SIZE 10 POSITION UR ENDLAYERS=fencelabelmode=map The entire map file is as follows: MAP NAME fenceLabelWms_en FONTSET ../../../mapserv/etc/fonts.txt SYMBOLSET ../../../mapserv/etc/symbols.sym SIZE 400 300 IMAGETYPE PNG IMAGECOLOR 255 255 255 RESOLUTION 96 EXTENT -141.5 41.5 -51.5 83.5 # all of Canada UNITS METERS PROJECTION init=epsg:4269 # lat/long END WEB METADATA wms_srs EPSG:42304 END END OUTPUTFORMAT NAME png DRIVER 'GD/PNG' MIMETYPE 'image/png' IMAGEMODE PC256 #RGB FORMATOPTION INTERLACE=OFF EXTENSION 'png' TRANSPARENT ON END LAYER NAME fencelabel STATUS ON TYPE point CLASS SIZE 0 COLOR 0 0 0 LABEL TYPE truetype FONT verdana ANTIALIAS false END END END END A number of applications use this WMS and each has its own projection but, as in the request above, all pass the x,y as lat/long (EPSG:4269). The MAP METADATA's wms_srs lists the supported projections (for the purposes of this example, I'm simplifying by only including EPSG:42304). So, this allows me to provide the point coordinates in lat/long, but display them using the projection from the application itself. For example: http://domain/ogc/mapserv?map=/path/fenceLabelWms_en.mapmap.layer[fencelabel]=FEATURE POINTS -105.5620661354989 51.14394472618872 END TEXT 'Fence 1' ENDmap.layer[fencelabel].class[0]=COLOR 255 0 0 ENDmap.layer[fencelabel].class[0]=LABEL COLOR 255 0 0 SIZE 10 POSITION UR ENDLAYERS=fencelabelFORMAT=image/pngSRS=EPSG:42304TRANSPARENT=trueSERVICE=WMSVERSION=1.1.1REQUEST=GetMapSTYLES=EXCEPTIONS=application/vnd.ogc.se_inimageBBOX=-2903297.000,-789911.,3431059.000,3959683.000WIDTH=1219HEIGHT=601 Knowing this, I can define this dynamic layer within a WMC Layer tag and load it into an OpenLayers Map (whose projection is set to EPSG:42304 and extents encompass all of Canada). The WMC definition for the above layer is as follows (we use a batch script to set the path to the MapServer executable and MAP file, fenceLabelWms_en): Layer queryable=0 hidden=0 Server service=OGC:WMS version=1.1.1 OnlineResource xlink:type=simple xlink:href=http://domain/ogc/fenceLabelWms_en?map.layer[fencelabel]=FEATURE%20POINTS%20-105.5620661354989%2051.14394472618872%20END%20TEXT%20'Fence%201'%20ENDamp;map.layer[fencelabel].class[0]=COLOR%20255%20105%200%20ENDamp;map.layer[fencelabel].class[0]=LABEL%20COLOR%20255%20105%200%20SIZE%2010%20POSITION%20LR%20END/ /Server Namefencelabel/Name TitleFence 1/Title SRSEPSG:42304/SRS FormatList Format current=1image/png/Format /FormatList Extension ol:maxExtent minx=-2903297.000 miny=-789911. maxx=3431059.000 maxy=3959683.000/ ol:transparenttrue/ol:transparent ol:numZoomLevels5/ol:numZoomLevels ol:unitsm/ol:units ol:isBaseLayerfalse/ol:isBaseLayer ol:displayInLayerSwitcherfalse/ol:displayInLayerSwitcher ol:singleTiletrue/ol:singleTile /Extension /Layer Okay, everything works fine up to this point. The fence label WMS is displayed in the OpenLayers Map correctly in relation to the other layers it contains, all of which have projections of EPSG:42304. Now, the problem ... I’d like to print the image displayed to the user. To do this, I have another empty MAP file that I use to print the WMC; the same WMC I loaded into OpenLayers. The entire contents of this MAP file are: MAP NAME wmsContextServer FONTSET ../../../mapserv/etc/fonts.txt SYMBOLSET ../../../mapserv/etc/symbols.sym SIZE 400 300 STATUS ON IMAGETYPE PNG IMAGECOLOR 255 255 255 EXTENT -2338951 -80 3008692 390 UNITS METERS CONFIG CGI_CONTEXT_URL 1 PROJECTION init=epsg:42304 END RESOLUTION 96 REFERENCE # Parameter settings NOT to change STATUS ON COLOR -1 -1 -1 OUTLINECOLOR255 0 0 # Parameter settings that will change if you replace the default 'refmap.png' IMAGE ../htdocs/images/refmap.png SIZE100 74 EXTENT -2.2e+006 -712631 3.0728e+006 3.84e+006 END # REFERENCE SCALEBAR STATUS EMBED INTERVALS 1 POSITION LL TRANSPARENT true UNITS kilometers STYLE 1 SIZE 100 7 LABEL SIZE small END END WEB INCLUDE ../../../mapserv/conf/imagepath.inc.map METADATA # Used for App Page Title wms_title WMC Context Client wms_connectiontimeout 60 END END
[mapserver-users] Re: ‘mode=map’ with WMC of dynamic Layer failing with getString() error
That error usually occurs when there is a problem with the syntax of your CGI query string. That is consistent with the post I found at: http://osgeo-org.1803224.n2.nabble.com/Clarification-on-RFC-44-changes-td2585073.html which suggests that there may be a problem with the syntax of “map.layer[line]=DATA ‘id3/vector’ END” in my URL but I don’t know what an alternative correct way to send this information would be that would work for both the WMS GetMap requests and the “mode=map” method. I haven’t been able to get either the WMS GetMap request or the “mode=map” request to work trying to guess. Is there anyone out there who can provide any help with this? I have reviewed the URL Jeff provided and tried using an underscore rather than the dot notation but continue to receive the getString() error. Is there some alternative syntax to try for specifying the DATA parameter that would work for both a getMap request and using “mode=map”? I don't see anything else in my URL that could be wrong...but maybe I've been looking at it too long. Also, to respond to Ophidian, I have tried inserting %20 and '+' in the blanks and tried replacing the single quotes as shown below. All to no effect. Thanks for the suggestion though. http://localhost/cgi-bin/mapserv.exe?map=/ms4w/apps/testing/dynamic.mapmap.layer[line]=DATA%20%91id3/vector%92%20ENDlayers=linemode=map -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/mode-map-with-WMC-of-dynamic-Layer-failing-with-getString-error-tp5824774p5831237.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
[mapserver-users] Re: ‘mode=map’ with WMC of dynamic Layer - Solved!
After preparing my last post, I noticed the strange numbers (really the fact that I got two different ones) for the single quotes when I encoded the URL (%91 and %92) that was failing: http://localhost/cgi-bin/mapserv.exe?map=/ms4w/apps/testing/dynamic.mapmap.layer[line]=DATA%20%91id3/vector%92%20ENDlayers=linemode=map And, after looking closer, I realized there were smart rather than regular (as in %27) single quotes. Once I changed them the following URL worked fine in the browser after all: http://localhost/cgi-bin/mapserv.exe?map=/ms4w/apps/testing/dynamic.mapmap.layer[line]=DATA 'id3/vector' ENDlayers=linemode=map So, I thought, my attempts at trying to figure out what the bigger problem I was having had failed to this point. I did not have smart quotes in the WMC for that Layer definition (as shown in the original post) but I was unable to create an output image with it. Then, I thought of Obsidian's suggestion and changed the LayerListLayer definition as follows (see the blanks replaced with %20 in the OnlineResource values): Layer queryable=0 hidden=0 clid=id3vector Server service=OGC:WMS version=1.1.1 title=Id3 Vector OnlineResource xlink:type=simple xlink:href=http://domain/ogc/printVecWms_en?map.layer[line]=DATA%20'id3/vector'%20END/ /Server Nameline/Name Titleid3 vector/Title SRSEPSG:42304/SRS FormatList Format current=1image/png/Format /FormatList StyleList Style current=1 Namedefault/Name Titledefault/Title LegendURL format=image/png width=47 height=294 OnlineResource xlink:type=simple xlink:href=http://domain/ogc/printVecWms_en?version=1.1.1service=WMSrequest=GetLegendGraphicmap.layer[line]=DATA%20'id3/vector'%20ENDlayer=lineformat=image/pngSTYLE=default/ /LegendURL /Style /StyleList Extension ol:maxExtent minx=-2903297 miny=-789911 maxx=3431059 maxy=3959683/ ol:transparenttrue/ol:transparent ol:isBaseLayerfalse/ol:isBaseLayer ol:displayInLayerSwitchertrue/ol:displayInLayerSwitcher /Extension /Layer And, low and behold, I have an image with the dynamic shapefile's contents included. Yay!!! Thank you, Jeff and Obsidian, for your pointers towards things to look more closely at. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/mode-map-with-WMC-of-dynamic-Layer-failing-with-getString-error-tp5824774p5831449.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
[mapserver-users] ‘mode=map’ with WMC of dynamic Layer failing with getString() error
I apologize profusely for how brutally long this post is, but I’m trying to avoid a bunch of questions by providing as much background to the full process I am trying to develop before I actually get to specific problem I am stuck at now. I think it will save time, but I could be wrong…in which case, I’m sorry. I have a MAP file with an empty line LAYER defined as follows: LAYER NAME line STATUS ON TYPE line PROJECTION init=epsg:42304 END CLASS COLOR 0 0 0 NAME Some Line Layer END METADATA wms_title Some Line Layer END END My process dynamically generates a shapefile. Then, within a WMS getMap request, I provide the DATA value to create an image of its contents as follows: http://localhost/cgi-bin/mapserv.exe?map=/ms4w/apps/testing/dynamic.maprequest=getMapservice=wmsversion=1.1.1map.layer[line]=DATA ‘id3/vector’ ENDlayers=linesrs=EPSG:42304bbox=-2338951,-80,3008692,390format=image/pngheight=300width=300 Knowing this, I can define this dynamic layer in a WMC Layer tag and load it into OpenLayers (or probably anything else that supports loading and displaying a WMC, OpenLayers is just what I am using). The WMC definition for the above layer is as follows: Layer queryable=0 hidden=0 clid=id3vector Server service=OGC:WMS version=1.1.1 title=Id3 Vector OnlineResource xlink:type=simple xlink:href=http://domain/ogc/printVecWms_en?map.layer[line]=DATA 'id3/vector' END/ /Server Nameline/Name Titleid3 vector/Title SRSEPSG:42304/SRS FormatList Format current=1image/png/Format /FormatList StyleList Style current=1 Namedefault/Name Titledefault/Title LegendURL format=image/png width=47 height=294 OnlineResource xlink:type=simple xlink:href=http://domain/ogc/printVecWms_en?version=1.1.1amp;service=WMSamp;request=GetLegendGraphicamp;map.layer[line]=DATA 'id3/vector' ENDamp;layer=lineamp;format=image/pngamp;STYLE=default/ /LegendURL /Style /StyleList Extension ol:maxExtent minx=-2903297 miny=-789911 maxx=3431059 maxy=3959683/ ol:transparenttrue/ol:transparent ol:isBaseLayerfalse/ol:isBaseLayer ol:displayInLayerSwitchertrue/ol:displayInLayerSwitcher /Extension /Layer Okay, that’s the background and everything works fine up to this point. Now, I’d like to print the image displayed to the user. To do this, I have another empty MAP file that I use to print a WMC. The entire contents of the MAP file are: MAP NAME wmsContextServer FONTSET ../../../mapserv/etc/fonts.txt SYMBOLSET ../../../mapserv/etc/symbols.sym SIZE 400 300 STATUS ON IMAGETYPE PNG IMAGECOLOR 255 255 255 EXTENT -2338951 -80 3008692 390 UNITS METERS CONFIG CGI_CONTEXT_URL 1 PROJECTION init=epsg:42304 END RESOLUTION 96 REFERENCE # Parameter settings NOT to change STATUS ON COLOR -1 -1 -1 OUTLINECOLOR255 0 0 # Parameter settings that will change if you replace the default 'refmap.png' IMAGE ../htdocs/images/refmap.png SIZE100 74 EXTENT -2.2e+006 -712631 3.0728e+006 3.84e+006 END # REFERENCE SCALEBAR STATUS EMBED INTERVALS 1 POSITION LL TRANSPARENT true UNITS kilometers STYLE 1 SIZE 100 7 LABEL SIZE small END END WEB INCLUDE ../../../mapserv/conf/imagepath.inc.map METADATA # Used for App Page Title wms_title WMC Context Client wms_connectiontimeout 60 END END OUTPUTFORMAT NAME png DRIVER 'GD/PNG' MIMETYPE 'image/png' IMAGEMODE RGB FORMATOPTION INTERLACE=OFF EXTENSION 'png' #TRANSPARENT ON END OUTPUTFORMAT NAME gif DRIVER 'GD/GIF' MIMETYPE 'image/gif' IMAGEMODE RGB FORMATOPTION INTERLACE=OFF EXTENSION 'gif' END END For any WMC I have tried so far that doesn’t contain one of these “dynamic” layers in it, I can use the following URL to get an image of all of the layers in the WMC successfully: http://localhost/cgi-bin/mapserv.exe?map=/ms4w/apps/testing/wmsContextServer.mapmode=mapcontext=http://localhost/contexts/non-dynamiclayer.xmllayers=all And, without using any dynamic layers, we have been following this approach in a production environment for a couple of years to provide our print functionality. The wmsContextServer.map file has not changed since February 2, 2009. However, now we are trying to introduce these dynamic layers into the mix and I am running into trouble. If I use the same URL to the wmsContextServer.map but point to a context with a dynamic layer added in to it (I start with one that works and just insert the WMC LayerListLayer contents I included above), I get the following error message and the *.img.tmp file mentioned is not created in the IMAGEURL directory (which is pointed to, BTW in
[mapserver-users] Changing map file parameter (Label Color) via URL
Hello, Can someone help me out with the proper syntax for the following? Or, tell me outright that it is not possible? I've been trying to apply the documentation found on the MapServer CGI Controls page (http://www.mapserver.org/cgi/controls.html#changing-map-file-parameters-via-a-form-or-a-url) under the Changing map file parameters via a form or a URL section but have not had any luck applying it to change the colour of the label text. I have a basic URL in which I pass the coordinates for a point and provide the label text as follows: http://localhost/cgi-bin/mapserv.exe?map=/ms4w/apps/wms/dynamicWms/map/dynamicWms.mapmap.layer[resultlabel]=FEATURE POINTS -749340.32786036 180081.55026124 END TEXT 'Fence 1' ENDLAYERS=resultlabelFORMAT=image/pngSRS=EPSG:42304TRANSPARENT=trueSERVICE=WMSVERSION=1.1.1REQUEST=GetMapSTYLES=EXCEPTIONS=application/vnd.ogc.se_inimageBBOX=-910537.6366469,102726.53880899,-587878.43588336,261873.32786947WIDTH=1219HEIGHT=601 I would further like to set the colour of the point symbol and its label text within the URL. At this point, I am able to set the point symbol's colour by adding map.layer[resultlabel].class[0]=COLOR 255 0 0; but I can't seem to change the label text's colour. It is always black. I've tried adding map.layer[resultlabel].class[0].label[0]=COLOR 255 0 0 and map.layer[resultlabel].class[0].label=COLOR 255 0 0 to no avail. If it matters, my map layer is defined as follows: LAYER NAME resultlabel STATUS ON TYPE point PROJECTION init=epsg:42304 END METADATA wms_title Result Label END CLASS SYMBOL thickcrosshair SIZE 9 COLOR 0 0 0 LABEL TYPE truetype FONT arial-bold SIZE 10 COLOR 0 0 0 OUTLINECOLOR 255 255 255 POSITION ur FORCE ON END END END Does anyone know if this is this just a syntax problem that I am having or if changing the label text's colour is just not supported? Cheers, jtm -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/Changing-map-file-parameter-Label-Color-via-URL-tp5771138p5771138.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