From sethg at geographika.co.uk Thu Oct 1 05:46:38 2020 From: sethg at geographika.co.uk (Seth G) Date: Thu, 01 Oct 2020 14:46:38 +0200 Subject: [mapserver-users] Layer Metadata Questions Message-ID: <8aae88f0-241d-4a43-bf4f-ee46087f181f@www.fastmail.com> Hi all, I've been looking at using the Layer Metadata as described at https://mapserver.org/ogc/layer_metadata.html "MapServer?s layer API supports the ISO 19115:2003 geospatial standard" - although looking at https://www.iso.org/standard/26020.html this has now been withdrawn and replaced with ISO 19115-1:2014 - https://www.iso.org/standard/53798.html I presume there is not much difference between these 2 schemas? There is a ~180 EUR charge to get a copy of the PDF. The source code has a comment "Generate an ISO 19139:2007 representation of layer metadata " [1] - so maybe the docs need an update? The second question regards the mapping of the LAYER METADATA block to the ISO output. There is a good test and expected output at [2] and [3]. Looking at the code it appears the "contact" metadata can only be retrieved at the MAP rather than LAYER level [2]. Is this something that should be available at LAYER level also, perhaps using "wms_attribution_title"? Any comments appreciated, Seth [1] https://github.com/MapServer/MapServer/blob/e0f3ae291598abe8196d47a98723f702d9ee025f/mapmetadata.c#L694 [2] https://github.com/MapServer/MapServer/blob/master/msautotest/wxs/ows_metadata.map [3] https://github.com/MapServer/MapServer/blob/master/msautotest/wxs/expected/ows_metadata_layer_raster.xml -- web:http://geographika.co.uk twitter: @geographika From bslunab at correo.udistrital.edu.co Fri Oct 2 09:54:22 2020 From: bslunab at correo.udistrital.edu.co (BRIAN STIFFENN LUNA BOLIVAR) Date: Fri, 2 Oct 2020 11:54:22 -0500 Subject: [mapserver-users] Mapfile WMS configuration fail Message-ID: Hello everyone, good day. I have a problem when making the wms request through QGIS, where when loading the layer I get the following error: "connection rejected". 1. How can I solve it? 2. In Mapbender, I have an error when connecting the WMS, because the error that appears is the following: Error in the source mapping: Error: request failed with status 404; can it be because of the problem with the configuration of my mapfile? -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmckenna at gatewaygeomatics.com Fri Oct 2 10:05:52 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Fri, 2 Oct 2020 14:05:52 -0300 Subject: [mapserver-users] Mapfile WMS configuration fail In-Reply-To: References: Message-ID: You can use the "QGIS Network logger" plugin to open the GetMap request url, and examine its response in your browser: https://github.com/mapserver/mapserver/wiki/Get-the-Raw-WMS-Request-Generated-by-QGIS If the WMS server is configured by you, then I would also definitely execute a GetCapabilities request on it, and remove all warning/error messages from the GetCapabilities response. -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-02 1:54 p.m., BRIAN STIFFENN LUNA BOLIVAR wrote: > Hello everyone, good day. I have a problem when making the wms request > through QGIS, where when loading the layer I get the following error: > "connection rejected". > 1. How can I solve it? > 2. In Mapbender, I have an error when connecting the WMS, because the > error that appears is the following: Error in the source mapping: Error: > request failed with status 404; can it be because of the problem with > the configuration of my mapfile? > > From lars.schylberg at blixtmail.se Fri Oct 2 10:08:26 2020 From: lars.schylberg at blixtmail.se (Lars Schylberg) Date: Fri, 2 Oct 2020 19:08:26 +0200 Subject: [mapserver-users] Mapfile WMS configuration fail In-Reply-To: References: Message-ID: <733b8ac0-e489-d874-226a-3418e433f582@blixtmail.se> Hi, This usually the order I test mapserver configurations. 1) Test your mapfile with shp2img https://www.mapserver.org/utilities/shp2img.html 2) Test your WMS config with the built in open layers viewer: https://mapserver.org/cgi/openlayers.html 3) Test with QGIS If you really want test the mapserver syntax, install and use: https://mappyfile.readthedocs.io/en/latest/? and use the validate command on the command line. Good luck ! /Lars Schylberg Den 2020-10-02 kl. 18:54, skrev BRIAN STIFFENN LUNA BOLIVAR: > Hello everyone, good day. I have a problem when making the wms request > through QGIS, where when loading the layer I get the following error: > "connection rejected". > 1. How can I solve it? > 2. In Mapbender, I have an error when connecting the WMS, because the > error that appears is the following: Error in the source mapping: > Error: request failed with status 404; can it be because of the > problem with the configuration of my mapfile? > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From tom.kralidis at canada.ca Sun Oct 4 07:03:28 2020 From: tom.kralidis at canada.ca (Kralidis, Tom (EC)) Date: Sun, 4 Oct 2020 14:03:28 +0000 Subject: [mapserver-users] Layer Metadata Questions In-Reply-To: <8aae88f0-241d-4a43-bf4f-ee46087f181f@www.fastmail.com> References: <8aae88f0-241d-4a43-bf4f-ee46087f181f@www.fastmail.com> Message-ID: Hi Seth, comments interleaved: > -----Original Message----- > From: mapserver-users On > Behalf Of Seth G > Sent: October 1, 2020 08:47 > To: mapserver-users at lists.osgeo.org > Subject: [mapserver-users] Layer Metadata Questions > > Hi all, > > I've been looking at using the Layer Metadata as described at > https://mapserver.org/ogc/layer_metadata.html > > "MapServer?s layer API supports the ISO 19115:2003 geospatial standard" - > although looking at https://www.iso.org/standard/26020.html this has now > been withdrawn and replaced with ISO 19115-1:2014 - > https://www.iso.org/standard/53798.html > I presume there is not much difference between these 2 schemas? There is a > ~180 EUR charge to get a copy of the PDF. > The source code has a comment "Generate an ISO 19139:2007 > representation of layer metadata " [1] - so maybe the docs need an update? > ISO 19139:2007 is the XML representation of ISO 19115:2003. I'll update the docs articulating this relationship. > The second question regards the mapping of the LAYER METADATA block to > the ISO output. There is a good test and expected output at [2] and [3]. > Looking at the code it appears the "contact" metadata can only be retrieved > at the MAP rather than LAYER level [2]. > Is this something that should be available at LAYER level also, perhaps using > "wms_attribution_title"? > Good point. Perhaps if MAP.LAYER.METADATA.wms_attribution_title is available we can override accordingly? Feel free to open a ticket and assign to me. Thanks ..Tom > Any comments appreciated, > > Seth > > > [1] > https://github.com/MapServer/MapServer/blob/e0f3ae291598abe8196d47a > 98723f702d9ee025f/mapmetadata.c#L694 > [2] > https://github.com/MapServer/MapServer/blob/master/msautotest/wxs/o > ws_metadata.map > [3] > https://github.com/MapServer/MapServer/blob/master/msautotest/wxs/e > xpected/ows_metadata_layer_raster.xml > > > > -- > web:http://geographika.co.uk > twitter: @geographika > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users From rob.marjot at gmail.com Fri Oct 9 17:44:53 2020 From: rob.marjot at gmail.com (Rob Marjot) Date: Sat, 10 Oct 2020 02:44:53 +0200 Subject: [mapserver-users] PHP MapScript request handling gives seg fault error in libgdal Message-ID: Hi All, I am running latest MapServer 7.6.1 compiled from source on Ubuntu 20.04 with (also from source): - proj-7.1.0 - gdal-3.1.2 Using the ppa:ondrej/php repo, I installed the following PHP packages: php7.4 php7.4-fpm php7.4-dev Apache/2.4.41 proxies into the FPM over unix socket Compiling MapServer went smoothly and through the configuration of the extension in PHP, MapScript is available. When I call (HTTP GET) my PHP script to handle a WMS-formatted request [ OWSDispatch() ], the first 2 requests are handled fine (returns a rendered map image), but the next two make the PHP-FPM crash: /var/log/syslog: Oct 10 01:59:25 as02 kernel: [3663532.034251] php-fpm7.4[1260175]: segfault at 28 ip 00007fbce21fb3eb sp 00007ffe87f600e0 error 6 in libgdal.so.27.0.2 [7fbce152e000+d50000] Oct 10 01:59:25 as02 kernel: [3663532.034262] Code: 8b 15 81 de 56 00 e9 24 2d 38 ff 0f 1f 40 00 f3 0f 1e fa 41 54 49 89 f4 55 48 89 d5 53 89 fb 48 89 cf e8 e8 f0 ff ff 48 63 fb <4c> 89 24 f8 48 89 ac f8 00 01 00 00 5b 5d 41 5c c3 0f 1f 40 00 f3 What could be the problem? Any help is much appreciated. Best, Rob Marjot -------------- next part -------------- An HTML attachment was scrubbed... URL: From rob.marjot at gmail.com Sun Oct 11 13:42:13 2020 From: rob.marjot at gmail.com (Rob Marjot) Date: Sun, 11 Oct 2020 22:42:13 +0200 Subject: [mapserver-users] PHP MapScript request handling gives seg fault error in libgdal In-Reply-To: References: Message-ID: Dear list, See my previous email; I didn't mention that the map is rendered from ECW. Maybe that's causing the trouble? I made some progress in finding a workaround. I configured child processes in the fpm pool to handle only 1 request: pm.max_requests = 1 Basically this turns off any recycling. Any further suggestions what to look into? Thanks, Rob Op za 10 okt. 2020 om 02:44 schreef Rob Marjot : > Hi All, > > I am running latest MapServer 7.6.1 compiled from source on Ubuntu 20.04 > with (also from source): > - proj-7.1.0 > - gdal-3.1.2 > > Using the ppa:ondrej/php repo, I installed the following PHP packages: > php7.4 php7.4-fpm php7.4-dev > Apache/2.4.41 proxies into the FPM over unix socket > > Compiling MapServer went smoothly and through the configuration of the > extension in PHP, MapScript is available. When I call (HTTP GET) my PHP > script to handle a WMS-formatted request [ OWSDispatch() ], the first 2 > requests are handled fine (returns a rendered map image), but the next two > make the PHP-FPM crash: > /var/log/syslog: > Oct 10 01:59:25 as02 kernel: [3663532.034251] php-fpm7.4[1260175]: segfault > at 28 ip 00007fbce21fb3eb sp 00007ffe87f600e0 error 6 in libgdal.so.27.0.2 > [7fbce152e000+d50000] > Oct 10 01:59:25 as02 kernel: [3663532.034262] Code: 8b 15 81 de 56 00 e9 > 24 2d 38 ff 0f 1f 40 00 f3 0f 1e fa 41 54 49 89 f4 55 48 89 d5 53 89 fb 48 > 89 cf e8 e8 f0 ff ff 48 63 fb <4c> 89 24 f8 48 89 ac f8 00 01 00 00 5b 5d > 41 5c c3 0f 1f 40 00 f3 > > What could be the problem? Any help is much appreciated. > > Best, > Rob Marjot > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bslunab at correo.udistrital.edu.co Sun Oct 11 15:19:30 2020 From: bslunab at correo.udistrital.edu.co (BRIAN STIFFENN LUNA BOLIVAR) Date: Sun, 11 Oct 2020 17:19:30 -0500 Subject: [mapserver-users] Mapcache WMS layer group - WMTS Themes Message-ID: Hello everyone, good day.I'm trying hard to set the WMS layer group as in the mapfile, and WMTS theme section too, but I couldn't find anything in the mapcache documentation. Do you know the correct way to set this? Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmckenna at gatewaygeomatics.com Mon Oct 12 06:38:05 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Mon, 12 Oct 2020 10:38:05 -0300 Subject: [mapserver-users] PHP MapScript request handling gives seg fault error in libgdal In-Reply-To: References: Message-ID: Hi Rob, I'm not sure what is happening in your case, but make sure that you are using the recommended PHP SWIG MapScript in your build (-DWITH_PHPNG=1), as the old PHP MapScript is unmaintained. Some other thoughts: - can you trigger this with shp2img at the commandline? (this will remove MapScript/Apache/WMS from the equation) https://mapserver.org/utilities/shp2img.html - also, try your GetMap request at the commandline with: mapserv -nh "QUERY_STRING=map=/ms4w/apps/local-demo/local-demo.map&SERVICE=WMS&VERSION=1.1.1&REQUEST=GetMap&..." > ttt.png But if I can speak openly here, I believe most production instances are still running PROJ<6, so everyone is beginning to experiment with settings of FCGI/PHP-FPM for GDAL3+PROJ7 (as there is a noticeable performance hit with the new PROJ). In other words, keep us posted here on your testing and configuration; also, do consider documenting your configuration and workaround on the MapServer wiki, so others can follow: https://github.com/MapServer/MapServer/wiki Thanks Rob! -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-09 9:44 p.m., Rob Marjot wrote: > Hi All, > > I am running latest MapServer 7.6.1 compiled from source on Ubuntu 20.04 > with (also from source): > -?proj-7.1.0 > -?gdal-3.1.2 > > Using the ppa:ondrej/php repo, I installed the following PHP packages: > php7.4 php7.4-fpm php7.4-dev > Apache/2.4.41 proxies into the FPM over unix socket > > Compiling MapServer went smoothly and through the configuration of the > extension in PHP, MapScript is available. When I call (HTTP GET) my PHP > script to handle a WMS-formatted request [ OWSDispatch() ], the first 2 > requests are handled fine (returns a rendered map image), but the next > two make the PHP-FPM crash: > /var/log/syslog: > Oct 10 01:59:25 as02 kernel: [3663532.034251] php-fpm7.4[1260175]: > segfault at 28 ip 00007fbce21fb3eb sp 00007ffe87f600e0 error 6 in > libgdal.so.27.0.2[7fbce152e000+d50000] > Oct 10 01:59:25 as02 kernel: [3663532.034262] Code: 8b 15 81 de 56 00 e9 > 24 2d 38 ff 0f 1f 40 00 f3 0f 1e fa 41 54 49 89 f4 55 48 89 d5 53 89 fb > 48 89 cf e8 e8 f0 ff ff 48 63 fb <4c> 89 24 f8 48 89 ac f8 00 01 00 00 > 5b 5d 41 5c c3 0f 1f 40 00 f3 > > What could be the problem? Any help is much appreciated. > > Best, > Rob Marjot > > From rob.marjot at gmail.com Mon Oct 12 10:11:52 2020 From: rob.marjot at gmail.com (Rob Marjot) Date: Mon, 12 Oct 2020 19:11:52 +0200 Subject: [mapserver-users] PHP MapScript request handling gives seg fault error in libgdal In-Reply-To: References: Message-ID: Hi Jeff, all, Thank you for your thoughts on this. Can you elaborate on the PROJ<6 remark; serious performance hits? Indeed, I am on the NG (SWIG) MapScript. Trying with shp2img and mapserv -nh will require to have the equivalent .map file on disk; I will work on that soon. Taking a complete different approach, I have been able to reproduce the same problem by: 1. Installing *binaries* for MapServer 7.6.1, GDAL 3.1.3 and PROJ 7.1.1 from UbuntuGIS Unstable 2. Add ECW support through compiling the simple GDAL-plugin (tested: gdal_translate from ECW to GTiff works brilliantly) 3. Configured this to run in the context of 7.4 PHP-FPM Again the seg fault. Connecting to PHP-FPM from either Apache or NGINX does not make a difference, but that's no surprise. Rob Op ma 12 okt. 2020 om 15:38 schreef Jeff McKenna < jmckenna at gatewaygeomatics.com>: > Hi Rob, > > I'm not sure what is happening in your case, but make sure that you are > using the recommended PHP SWIG MapScript in your build (-DWITH_PHPNG=1), > as the old PHP MapScript is unmaintained. > > Some other thoughts: > > - can you trigger this with shp2img at the commandline? (this will > remove MapScript/Apache/WMS from the equation) > https://mapserver.org/utilities/shp2img.html > > - also, try your GetMap request at the commandline with: > > mapserv -nh > "QUERY_STRING=map=/ms4w/apps/local-demo/local-demo.map&SERVICE=WMS&VERSION=1.1.1&REQUEST=GetMap&..." > > > ttt.png > > But if I can speak openly here, I believe most production instances are > still running PROJ<6, so everyone is beginning to experiment with > settings of FCGI/PHP-FPM for GDAL3+PROJ7 (as there is a noticeable > performance hit with the new PROJ). > > In other words, keep us posted here on your testing and configuration; > also, do consider documenting your configuration and workaround on the > MapServer wiki, so others can follow: > https://github.com/MapServer/MapServer/wiki > > Thanks Rob! > > -jeff > > > > -- > Jeff McKenna > MapServer Consulting and Training Services > co-founder of FOSS4G > http://gatewaygeo.com/ > > > > On 2020-10-09 9:44 p.m., Rob Marjot wrote: > > Hi All, > > > > I am running latest MapServer 7.6.1 compiled from source on Ubuntu 20.04 > > with (also from source): > > - proj-7.1.0 > > - gdal-3.1.2 > > > > Using the ppa:ondrej/php repo, I installed the following PHP packages: > > php7.4 php7.4-fpm php7.4-dev > > Apache/2.4.41 proxies into the FPM over unix socket > > > > Compiling MapServer went smoothly and through the configuration of the > > extension in PHP, MapScript is available. When I call (HTTP GET) my PHP > > script to handle a WMS-formatted request [ OWSDispatch() ], the first 2 > > requests are handled fine (returns a rendered map image), but the next > > two make the PHP-FPM crash: > > /var/log/syslog: > > Oct 10 01:59:25 as02 kernel: [3663532.034251] php-fpm7.4[1260175]: > > segfault at 28 ip 00007fbce21fb3eb sp 00007ffe87f600e0 error 6 in > > libgdal.so.27.0.2[7fbce152e000+d50000] > > Oct 10 01:59:25 as02 kernel: [3663532.034262] Code: 8b 15 81 de 56 00 e9 > > 24 2d 38 ff 0f 1f 40 00 f3 0f 1e fa 41 54 49 89 f4 55 48 89 d5 53 89 fb > > 48 89 cf e8 e8 f0 ff ff 48 63 fb <4c> 89 24 f8 48 89 ac f8 00 01 00 00 > > 5b 5d 41 5c c3 0f 1f 40 00 f3 > > > > What could be the problem? Any help is much appreciated. > > > > Best, > > Rob Marjot > > > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmckenna at gatewaygeomatics.com Mon Oct 12 10:17:43 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Mon, 12 Oct 2020 14:17:43 -0300 Subject: [mapserver-users] PHP MapScript request handling gives seg fault error in libgdal In-Reply-To: References: Message-ID: <77626380-2a15-ba3f-d5d5-5d72e721d679@gatewaygeomatics.com> Hi Rob, Next I would remove the ECW support/reliance and try to trigger the error without ECW data (yes with a local mapfile called from shp2img). Maybe others have ideas. -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-12 2:11 p.m., Rob Marjot wrote: > Hi Jeff, all, > > Thank you for your thoughts on this. Can you elaborate on the PROJ<6 > remark; serious performance hits? > > Indeed, I am on the NG (SWIG) MapScript. Trying with shp2img and mapserv > -nh will require to have the equivalent .map file on disk; I will work > on that soon. > > Taking a complete different approach, I have been able to reproduce the > same problem by: > 1. Installing *binaries* for MapServer 7.6.1, GDAL 3.1.3 and PROJ 7.1.1 > from UbuntuGIS Unstable > 2. Add ECW support through compiling the simple GDAL-plugin (tested: > gdal_translate from ECW to GTiff works brilliantly) > 3. Configured this to run in the context of 7.4 PHP-FPM > > Again the seg fault. Connecting to PHP-FPM from either Apache or NGINX > does not make a difference, but that's no surprise. > > Rob > > Op ma 12 okt. 2020 om 15:38 schreef Jeff McKenna > >: > > Hi Rob, > > I'm not sure what is happening in your case, but make sure that you are > using the recommended PHP SWIG MapScript in your build > (-DWITH_PHPNG=1), > as the old PHP MapScript is unmaintained. > > Some other thoughts: > > - can you trigger this with shp2img at the commandline? (this will > remove MapScript/Apache/WMS from the equation) > https://mapserver.org/utilities/shp2img.html > > - also, try your GetMap request at the commandline with: > > ? ? ?mapserv -nh > "QUERY_STRING=map=/ms4w/apps/local-demo/local-demo.map&SERVICE=WMS&VERSION=1.1.1&REQUEST=GetMap&..." > > ?> ttt.png > > But if I can speak openly here, I believe most production instances are > still running PROJ<6, so everyone is beginning to experiment with > settings of FCGI/PHP-FPM for GDAL3+PROJ7 (as there is a noticeable > performance hit with the new PROJ). > > In other words, keep us posted here on your testing and configuration; > also, do consider documenting your configuration and workaround on the > MapServer wiki, so others can follow: > https://github.com/MapServer/MapServer/wiki > > Thanks Rob! > > -jeff > > > > -- > Jeff McKenna > MapServer Consulting and Training Services > co-founder of FOSS4G > http://gatewaygeo.com/ > > > > On 2020-10-09 9:44 p.m., Rob Marjot wrote: > > Hi All, > > > > I am running latest MapServer 7.6.1 compiled from source on > Ubuntu 20.04 > > with (also from source): > > -?proj-7.1.0 > > -?gdal-3.1.2 > > > > Using the ppa:ondrej/php repo, I installed the following PHP > packages: > > php7.4 php7.4-fpm php7.4-dev > > Apache/2.4.41 proxies into the FPM over unix socket > > > > Compiling MapServer went smoothly and through the configuration > of the > > extension in PHP, MapScript is available. When I call (HTTP GET) > my PHP > > script to handle a WMS-formatted request [ OWSDispatch() ], the > first 2 > > requests are handled fine (returns a rendered map image), but the > next > > two make the PHP-FPM crash: > > /var/log/syslog: > > Oct 10 01:59:25 as02 kernel: [3663532.034251] php-fpm7.4[1260175]: > > segfault at 28 ip 00007fbce21fb3eb sp 00007ffe87f600e0 error 6 in > > libgdal.so.27.0.2[7fbce152e000+d50000] > > Oct 10 01:59:25 as02 kernel: [3663532.034262] Code: 8b 15 81 de > 56 00 e9 > > 24 2d 38 ff 0f 1f 40 00 f3 0f 1e fa 41 54 49 89 f4 55 48 89 d5 53 > 89 fb > > 48 89 cf e8 e8 f0 ff ff 48 63 fb <4c> 89 24 f8 48 89 ac f8 00 01 > 00 00 > > 5b 5d 41 5c c3 0f 1f 40 00 f3 > > > > What could be the problem? Any help is much appreciated. > > > > Best, > > Rob Marjot > > > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users > From rob.marjot at gmail.com Mon Oct 12 10:24:54 2020 From: rob.marjot at gmail.com (Rob Marjot) Date: Mon, 12 Oct 2020 19:24:54 +0200 Subject: [mapserver-users] PHP MapScript request handling gives seg fault error in libgdal In-Reply-To: <77626380-2a15-ba3f-d5d5-5d72e721d679@gatewaygeomatics.com> References: <77626380-2a15-ba3f-d5d5-5d72e721d679@gatewaygeomatics.com> Message-ID: Yes, I will. Any additional background information you can share on the adviced PROJ<6 (or PROJ<=6, e.g. the current 6.3.2) Thanks, Rob Op ma 12 okt. 2020 om 19:17 schreef Jeff McKenna < jmckenna at gatewaygeomatics.com>: > Hi Rob, > > Next I would remove the ECW support/reliance and try to trigger the > error without ECW data (yes with a local mapfile called from shp2img). > > Maybe others have ideas. > > -jeff > > > > -- > Jeff McKenna > MapServer Consulting and Training Services > co-founder of FOSS4G > http://gatewaygeo.com/ > > > > On 2020-10-12 2:11 p.m., Rob Marjot wrote: > > Hi Jeff, all, > > > > Thank you for your thoughts on this. Can you elaborate on the PROJ<6 > > remark; serious performance hits? > > > > Indeed, I am on the NG (SWIG) MapScript. Trying with shp2img and mapserv > > -nh will require to have the equivalent .map file on disk; I will work > > on that soon. > > > > Taking a complete different approach, I have been able to reproduce the > > same problem by: > > 1. Installing *binaries* for MapServer 7.6.1, GDAL 3.1.3 and PROJ 7.1.1 > > from UbuntuGIS Unstable > > 2. Add ECW support through compiling the simple GDAL-plugin (tested: > > gdal_translate from ECW to GTiff works brilliantly) > > 3. Configured this to run in the context of 7.4 PHP-FPM > > > > Again the seg fault. Connecting to PHP-FPM from either Apache or NGINX > > does not make a difference, but that's no surprise. > > > > Rob > > > > Op ma 12 okt. 2020 om 15:38 schreef Jeff McKenna > > >: > > > > Hi Rob, > > > > I'm not sure what is happening in your case, but make sure that you > are > > using the recommended PHP SWIG MapScript in your build > > (-DWITH_PHPNG=1), > > as the old PHP MapScript is unmaintained. > > > > Some other thoughts: > > > > - can you trigger this with shp2img at the commandline? (this will > > remove MapScript/Apache/WMS from the equation) > > https://mapserver.org/utilities/shp2img.html > > > > - also, try your GetMap request at the commandline with: > > > > mapserv -nh > > > "QUERY_STRING=map=/ms4w/apps/local-demo/local-demo.map&SERVICE=WMS&VERSION=1.1.1&REQUEST=GetMap&..." > > > > > ttt.png > > > > But if I can speak openly here, I believe most production instances > are > > still running PROJ<6, so everyone is beginning to experiment with > > settings of FCGI/PHP-FPM for GDAL3+PROJ7 (as there is a noticeable > > performance hit with the new PROJ). > > > > In other words, keep us posted here on your testing and > configuration; > > also, do consider documenting your configuration and workaround on > the > > MapServer wiki, so others can follow: > > https://github.com/MapServer/MapServer/wiki > > > > Thanks Rob! > > > > -jeff > > > > > > > > -- > > Jeff McKenna > > MapServer Consulting and Training Services > > co-founder of FOSS4G > > http://gatewaygeo.com/ > > > > > > > > On 2020-10-09 9:44 p.m., Rob Marjot wrote: > > > Hi All, > > > > > > I am running latest MapServer 7.6.1 compiled from source on > > Ubuntu 20.04 > > > with (also from source): > > > - proj-7.1.0 > > > - gdal-3.1.2 > > > > > > Using the ppa:ondrej/php repo, I installed the following PHP > > packages: > > > php7.4 php7.4-fpm php7.4-dev > > > Apache/2.4.41 proxies into the FPM over unix socket > > > > > > Compiling MapServer went smoothly and through the configuration > > of the > > > extension in PHP, MapScript is available. When I call (HTTP GET) > > my PHP > > > script to handle a WMS-formatted request [ OWSDispatch() ], the > > first 2 > > > requests are handled fine (returns a rendered map image), but the > > next > > > two make the PHP-FPM crash: > > > /var/log/syslog: > > > Oct 10 01:59:25 as02 kernel: [3663532.034251] php-fpm7.4[1260175]: > > > segfault at 28 ip 00007fbce21fb3eb sp 00007ffe87f600e0 error 6 in > > > libgdal.so.27.0.2[7fbce152e000+d50000] > > > Oct 10 01:59:25 as02 kernel: [3663532.034262] Code: 8b 15 81 de > > 56 00 e9 > > > 24 2d 38 ff 0f 1f 40 00 f3 0f 1e fa 41 54 49 89 f4 55 48 89 d5 53 > > 89 fb > > > 48 89 cf e8 e8 f0 ff ff 48 63 fb <4c> 89 24 f8 48 89 ac f8 00 01 > > 00 00 > > > 5b 5d 41 5c c3 0f 1f 40 00 f3 > > > > > > What could be the problem? Any help is much appreciated. > > > > > > Best, > > > Rob Marjot > > > > > > > > _______________________________________________ > > mapserver-users mailing list > > mapserver-users at lists.osgeo.org mapserver-users at lists.osgeo.org> > > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > > > > _______________________________________________ > > mapserver-users mailing list > > mapserver-users at lists.osgeo.org > > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From sethg at geographika.co.uk Mon Oct 12 13:49:26 2020 From: sethg at geographika.co.uk (Seth G) Date: Mon, 12 Oct 2020 22:49:26 +0200 Subject: [mapserver-users] =?utf-8?q?PHP_MapScript_request_handling_gives_?= =?utf-8?q?seg_fault_error_in_libgdal?= In-Reply-To: References: <77626380-2a15-ba3f-d5d5-5d72e721d679@gatewaygeomatics.com> Message-ID: Hi Rob, More details on the PROJ6 performance implications at https://www.mapserver.org/development/rfc/ms-rfc-126.html#performance-implications As Jeff wrote if you can recreate with shp2img or mapserv then it can rule out MapScript. A minimal MAP file and dataset would then be required to debug further. Do you create a new mapscript.OWSRequest() object for each request? Seth -- web:http://geographika.co.uk twitter: @geographika On Mon, Oct 12, 2020, at 7:24 PM, Rob Marjot wrote: > Yes, I will. > > Any additional background information you can share on the adviced PROJ<6 (or PROJ<=6, e.g. the current 6.3.2) > > Thanks, > Rob > > Op ma 12 okt. 2020 om 19:17 schreef Jeff McKenna : >> Hi Rob, >> >> Next I would remove the ECW support/reliance and try to trigger the >> error without ECW data (yes with a local mapfile called from shp2img). >> >> Maybe others have ideas. >> >> -jeff >> >> >> >> -- >> Jeff McKenna >> MapServer Consulting and Training Services >> co-founder of FOSS4G >> http://gatewaygeo.com/ >> >> >> >> On 2020-10-12 2:11 p.m., Rob Marjot wrote: >> > Hi Jeff, all, >> > >> > Thank you for your thoughts on this. Can you elaborate on the PROJ<6 >> > remark; serious performance hits? >> > >> > Indeed, I am on the NG (SWIG) MapScript. Trying with shp2img and mapserv >> > -nh will require to have the equivalent .map file on disk; I will work >> > on that soon. >> > >> > Taking a complete different approach, I have been able to reproduce the >> > same problem by: >> > 1. Installing *binaries* for MapServer 7.6.1, GDAL 3.1.3 and PROJ 7.1.1 >> > from UbuntuGIS Unstable >> > 2. Add ECW support through compiling the simple GDAL-plugin (tested: >> > gdal_translate from ECW to GTiff works brilliantly) >> > 3. Configured this to run in the context of 7.4 PHP-FPM >> > >> > Again the seg fault. Connecting to PHP-FPM from either Apache or NGINX >> > does not make a difference, but that's no surprise. >> > >> > Rob >> > >> > Op ma 12 okt. 2020 om 15:38 schreef Jeff McKenna >> > >: >> > >> > Hi Rob, >> > >> > I'm not sure what is happening in your case, but make sure that you are >> > using the recommended PHP SWIG MapScript in your build >> > (-DWITH_PHPNG=1), >> > as the old PHP MapScript is unmaintained. >> > >> > Some other thoughts: >> > >> > - can you trigger this with shp2img at the commandline? (this will >> > remove MapScript/Apache/WMS from the equation) >> > https://mapserver.org/utilities/shp2img.html >> > >> > - also, try your GetMap request at the commandline with: >> > >> > mapserv -nh >> > "QUERY_STRING=map=/ms4w/apps/local-demo/local-demo.map&SERVICE=WMS&VERSION=1.1.1&REQUEST=GetMap&..." >> > >> > > ttt.png >> > >> > But if I can speak openly here, I believe most production instances are >> > still running PROJ<6, so everyone is beginning to experiment with >> > settings of FCGI/PHP-FPM for GDAL3+PROJ7 (as there is a noticeable >> > performance hit with the new PROJ). >> > >> > In other words, keep us posted here on your testing and configuration; >> > also, do consider documenting your configuration and workaround on the >> > MapServer wiki, so others can follow: >> > https://github.com/MapServer/MapServer/wiki >> > >> > Thanks Rob! >> > >> > -jeff >> > >> > >> > >> > -- >> > Jeff McKenna >> > MapServer Consulting and Training Services >> > co-founder of FOSS4G >> > http://gatewaygeo.com/ >> > >> > >> > >> > On 2020-10-09 9:44 p.m., Rob Marjot wrote: >> > > Hi All, >> > > >> > > I am running latest MapServer 7.6.1 compiled from source on >> > Ubuntu 20.04 >> > > with (also from source): >> > > - proj-7.1.0 >> > > - gdal-3.1.2 >> > > >> > > Using the ppa:ondrej/php repo, I installed the following PHP >> > packages: >> > > php7.4 php7.4-fpm php7.4-dev >> > > Apache/2.4.41 proxies into the FPM over unix socket >> > > >> > > Compiling MapServer went smoothly and through the configuration >> > of the >> > > extension in PHP, MapScript is available. When I call (HTTP GET) >> > my PHP >> > > script to handle a WMS-formatted request [ OWSDispatch() ], the >> > first 2 >> > > requests are handled fine (returns a rendered map image), but the >> > next >> > > two make the PHP-FPM crash: >> > > /var/log/syslog: >> > > Oct 10 01:59:25 as02 kernel: [3663532.034251] php-fpm7.4[1260175]: >> > > segfault at 28 ip 00007fbce21fb3eb sp 00007ffe87f600e0 error 6 in >> > > libgdal.so.27.0.2[7fbce152e000+d50000] >> > > Oct 10 01:59:25 as02 kernel: [3663532.034262] Code: 8b 15 81 de >> > 56 00 e9 >> > > 24 2d 38 ff 0f 1f 40 00 f3 0f 1e fa 41 54 49 89 f4 55 48 89 d5 53 >> > 89 fb >> > > 48 89 cf e8 e8 f0 ff ff 48 63 fb <4c> 89 24 f8 48 89 ac f8 00 01 >> > 00 00 >> > > 5b 5d 41 5c c3 0f 1f 40 00 f3 >> > > >> > > What could be the problem? Any help is much appreciated. >> > > >> > > Best, >> > > Rob Marjot >> > > >> > > >> > _______________________________________________ >> > mapserver-users mailing list >> > mapserver-users at lists.osgeo.org >> > https://lists.osgeo.org/mailman/listinfo/mapserver-users >> > >> > >> > _______________________________________________ >> > mapserver-users mailing list >> > mapserver-users at lists.osgeo.org >> > https://lists.osgeo.org/mailman/listinfo/mapserver-users >> > >> >> _______________________________________________ >> mapserver-users mailing list >> mapserver-users at lists.osgeo.org >> https://lists.osgeo.org/mailman/listinfo/mapserver-users > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From akrherz at iastate.edu Tue Oct 13 11:29:34 2020 From: akrherz at iastate.edu (Herzmann, Daryl E [AGRON]) Date: Tue, 13 Oct 2020 18:29:34 +0000 Subject: [mapserver-users] GDAL 3.1 linking issue with TIFFGetStrileByteCountWithErr Message-ID: Howdy, I've been playing around attempting to build mapserver (7.6 branch or master) against https://anaconda.org (conda-forge) provided GDAL 3.1. My build command is below. Whilst I am able to build mapserver, I hit runtime errors like so: PHP Warning: PHP Startup: Unable to load dynamic library 'php_mapscript' (tried: /usr/lib64/php/modules/php_mapscript (/usr/lib64/php/modules/php_mapscript: cannot open shared object file: No such file or directory), /usr/lib64/php/modules/php_mapscript.so (/opt/miniconda3/envs/prod/lib/libgdal.so.27: undefined symbol: TIFFGetStrileByteCountWithErr)) in Unknown on line 0 For right or wrong, this used to work with GDAL 3.0. A workaround is to tell PHP to LD_PRELOAD gdal.so like so: LD_PRELOAD=/opt/miniconda3/envs/prod/lib/libgdal.so.27 php ... Any comments on this situation. Is it just expected that the GDAL mapserver links against is compiled with --with-libtiff=internal ? thanks daryl export MYCONDA=/opt/miniconda3/envs/prod cmake .. -DWITH_CLIENT_WFS=1 \ -DWITH_CLIENT_WMS=1 \ -DWITH_GIF=1 \ -DWITH_PHP=1 \ -DWITH_PYTHON=1 \ -DWITH_PROTOBUFC=0 \ -DCMAKE_INSTALL_PREFIX:PATH=$MYCONDA \ -DPNG_LIBRARY=$MYCONDA/lib/libpng.so \ -DPNG_INCLUDE_DIR=$MYCONDA/include \ -DJPEG_LIBRARY=$MYCONDA/lib/libjpeg.so \ -DJPEG_INCLUDE_DIR=$MYCONDA/include \ -DFREETYPE_LIBRARY=$MYCONDA/lib/libfreetype.so \ -DFREETYPE_INCLUDE_DIR=$MYCONDA/include \ -DGIF_LIBRARY=$MYCONDA/lib/libgif.so \ -DGIF_INCLUDE_DIR=$MYCONDA/include \ -DCAIRO_LIBRARY=$MYCONDA/lib/libcairo.so \ -DCAIRO_INCLUDE_DIR=$MYCONDA/include/cairo/ \ -DCURL_LIBRARY=$MYCONDA/lib/libcurl.so \ -DCURL_INCLUDE_DIR=$MYCONDA/include \ -DPROJ_LIBRARY=$MYCONDA/lib/libproj.so \ -DPROJ_INCLUDE_DIR=$MYCONDA/include \ -DLIBXML2_LIBRARY=$MYCONDA/lib/libxml2.so \ -DLIBXML2_INCLUDE_DIR=$MYCONDA/include/libxml2/ \ -DHARFBUZZ_LIBRARY=$MYCONDA/lib/libharfbuzz.so \ -DHARFBUZZ_INCLUDE_DIR=$MYCONDA/include/harfbuzz/ \ -DFRIBIDI_LIBRARY=$MYCONDA/lib/libfribidi.so \ -DFRIBIDI_INCLUDE_DIR=$MYCONDA/include \ -DZLIB_LIBRARY=$MYCONDA/lib/libz.so \ -DZLIB_INCLUDE_DIR=$MYCONDA/include \ -DGEOS_LIBRARY=$MYCONDA/lib/libgeos_c.so \ -DGEOS_INCLUDE_DIR=$MYCONDA/include \ -DGDAL_LIBRARY=$MYCONDA/lib/libgdal.so \ -DGDAL_INCLUDE_DIR=$MYCONDA/include \ -DPOSTGRESQL_LIBRARY=$MYCONDA/lib/libpq.so \ -DPOSTGRESQL_INCLUDE_DIR=$MYCONDA/include -- /** * daryl herzmann * Systems Analyst III -- Iowa Environmental Mesonet * https://mesonet.agron.iastate.edu */ From akrherz at iastate.edu Tue Oct 13 11:46:12 2020 From: akrherz at iastate.edu (Herzmann, Daryl E [AGRON]) Date: Tue, 13 Oct 2020 18:46:12 +0000 Subject: [mapserver-users] GDAL 3.1 linking issue with TIFFGetStrileByteCountWithErr In-Reply-To: References: Message-ID: Howdy again, I figured I would figure it out after hitting send :/ The issue is that the RHEL8 provided libtiff is 4.0.9, so one needs to LD_PRELOAD anaconda's libtiff (4.1.0) to get that libtiff loaded first before the system provided one. LD_PRELOAD=/opt/miniconda3/envs/prod/lib/libtiff.so php .... Sorry for the noise and hopefully this helps somebody. daryl -- /** * daryl herzmann * Systems Analyst III -- Iowa Environmental Mesonet * https://mesonet.agron.iastate.edu */ ________________________________________ From: mapserver-users on behalf of Herzmann, Daryl E [AGRON] Sent: Tuesday, October 13, 2020 1:29 PM To: mapserver-users at lists.osgeo.org Subject: [mapserver-users] GDAL 3.1 linking issue with TIFFGetStrileByteCountWithErr Howdy, I've been playing around attempting to build mapserver (7.6 branch or master) against https://anaconda.org (conda-forge) provided GDAL 3.1. My build command is below. Whilst I am able to build mapserver, I hit runtime errors like so: PHP Warning: PHP Startup: Unable to load dynamic library 'php_mapscript' (tried: /usr/lib64/php/modules/php_mapscript (/usr/lib64/php/modules/php_mapscript: cannot open shared object file: No such file or directory), /usr/lib64/php/modules/php_mapscript.so (/opt/miniconda3/envs/prod/lib/libgdal.so.27: undefined symbol: TIFFGetStrileByteCountWithErr)) in Unknown on line 0 For right or wrong, this used to work with GDAL 3.0. A workaround is to tell PHP to LD_PRELOAD gdal.so like so: LD_PRELOAD=/opt/miniconda3/envs/prod/lib/libgdal.so.27 php ... Any comments on this situation. Is it just expected that the GDAL mapserver links against is compiled with --with-libtiff=internal ? thanks daryl export MYCONDA=/opt/miniconda3/envs/prod cmake .. -DWITH_CLIENT_WFS=1 \ -DWITH_CLIENT_WMS=1 \ -DWITH_GIF=1 \ -DWITH_PHP=1 \ -DWITH_PYTHON=1 \ -DWITH_PROTOBUFC=0 \ -DCMAKE_INSTALL_PREFIX:PATH=$MYCONDA \ -DPNG_LIBRARY=$MYCONDA/lib/libpng.so \ -DPNG_INCLUDE_DIR=$MYCONDA/include \ -DJPEG_LIBRARY=$MYCONDA/lib/libjpeg.so \ -DJPEG_INCLUDE_DIR=$MYCONDA/include \ -DFREETYPE_LIBRARY=$MYCONDA/lib/libfreetype.so \ -DFREETYPE_INCLUDE_DIR=$MYCONDA/include \ -DGIF_LIBRARY=$MYCONDA/lib/libgif.so \ -DGIF_INCLUDE_DIR=$MYCONDA/include \ -DCAIRO_LIBRARY=$MYCONDA/lib/libcairo.so \ -DCAIRO_INCLUDE_DIR=$MYCONDA/include/cairo/ \ -DCURL_LIBRARY=$MYCONDA/lib/libcurl.so \ -DCURL_INCLUDE_DIR=$MYCONDA/include \ -DPROJ_LIBRARY=$MYCONDA/lib/libproj.so \ -DPROJ_INCLUDE_DIR=$MYCONDA/include \ -DLIBXML2_LIBRARY=$MYCONDA/lib/libxml2.so \ -DLIBXML2_INCLUDE_DIR=$MYCONDA/include/libxml2/ \ -DHARFBUZZ_LIBRARY=$MYCONDA/lib/libharfbuzz.so \ -DHARFBUZZ_INCLUDE_DIR=$MYCONDA/include/harfbuzz/ \ -DFRIBIDI_LIBRARY=$MYCONDA/lib/libfribidi.so \ -DFRIBIDI_INCLUDE_DIR=$MYCONDA/include \ -DZLIB_LIBRARY=$MYCONDA/lib/libz.so \ -DZLIB_INCLUDE_DIR=$MYCONDA/include \ -DGEOS_LIBRARY=$MYCONDA/lib/libgeos_c.so \ -DGEOS_INCLUDE_DIR=$MYCONDA/include \ -DGDAL_LIBRARY=$MYCONDA/lib/libgdal.so \ -DGDAL_INCLUDE_DIR=$MYCONDA/include \ -DPOSTGRESQL_LIBRARY=$MYCONDA/lib/libpq.so \ -DPOSTGRESQL_INCLUDE_DIR=$MYCONDA/include -- /** * daryl herzmann * Systems Analyst III -- Iowa Environmental Mesonet * https://mesonet.agron.iastate.edu */ _______________________________________________ mapserver-users mailing list mapserver-users at lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/mapserver-users From bruno.domenget.roch at gmail.com Thu Oct 15 05:14:26 2020 From: bruno.domenget.roch at gmail.com (Bruno Domenget) Date: Thu, 15 Oct 2020 08:14:26 -0400 Subject: [mapserver-users] Mapserver 7.6.1 - PHP 7 Swig - Kml output issue Message-ID: Hi all ! On a brand new debian server with Mapserver 7.6.1 / SWIG PHP Mapscript (+ GDal 3.13 and Proj 7.1.1) I am encountering a strange issue with KML output format. The code and mapfiles I am using were working perfectly on an old Mapserver 6.4 / PHP mapscript installation, I just converted the PHP functions to the SWIG interface. On a simple map file and PHP script the map shows fine, but when I select KML as outputformat ( $map->selectOutputFormat('kml'); ) I get the following error : *Fatal error: Uncaught Exception: msDrawMap(): Image handling error. Failed to draw layer named 'Points'. in /home/mapserver/public_html/lib/mapscript.php:2637* *Stack trace: #0 /home/mapserver/public_html/lib/mapscript.php(2637): mapobj_draw() #1 /home/mapserver/public_html/test/test_points_export.php(15): mapObj->draw() #2 {main} thrown in /home/mapserver/public_html/lib/mapscript.php on line 2637* There is no detail in the logs concerning that error, and I haven?t found any suitable solution. *It works if I remove the LABEL entry from the layer class, but then there is no label?* I tried with several PHP versions (7.1, 7.3, 7.4) and it always gave me the same result. Everything else works perfectly, I have no issue with Shapefile or Mapinfo output. Does anybody know what is going on ? Here is an example of mapfile that causes this issue : https://pastebin.com/LFxZ1L9T And the PHP code : https://pastebin.com/Tp30bzf3 Thanks, Bruno -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmckenna at gatewaygeomatics.com Thu Oct 15 05:41:14 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Thu, 15 Oct 2020 09:41:14 -0300 Subject: [mapserver-users] Mapserver 7.6.1 - PHP 7 Swig - Kml output issue In-Reply-To: References: Message-ID: Hi Bruno, What happens if you call that mapfile at the commandline with shp2img (and pass the "i" switch with your format name)? such as: shp2img -m mymap.map -o ttt.png -i kml -all_debug 5 You can also then try png output: shp2img -m mymap.map -o ttt.png -i png -all_debug 5 Does it generate a valid map image? -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-15 9:14 a.m., Bruno Domenget wrote: > Hi all?! > > On a brand new debian server with Mapserver 7.6.1 / SWIG PHP Mapscript > (+ GDal 3.13 and Proj 7.1.1) I am encountering a strange issue with KML > output format. The code and mapfiles I am using were working perfectly > on an old Mapserver 6.4 / PHP mapscript installation, I just converted > the PHP functions to the SWIG interface. > > > On a simple map file and PHP script the map shows fine, but when I > select KML as outputformat ( $map->selectOutputFormat('kml'); ) I get > the following error : *Fatal error: Uncaught Exception: msDrawMap(): > Image handling error. Failed to draw layer named 'Points'. in > /home/mapserver/public_html/lib/mapscript.php:2637* > > /Stack trace: #0 /home/mapserver/public_html/lib/mapscript.php(2637): > mapobj_draw() #1 > /home/mapserver/public_html/test/test_points_export.php(15): > mapObj->draw() #2 {main} thrown in > /home/mapserver/public_html/lib/mapscript.php on line 2637/ > > There is no detail in the logs concerning that error, and I haven?t > found any suitable solution. *It works if I remove the LABEL entry from > the layer class, but then there is no label?* > > I tried with several PHP versions (7.1, 7.3, 7.4) and it always gave me > the same result. Everything else works perfectly, I have no issue with > Shapefile or Mapinfo output. > > Does anybody know what is going on ? > > Here is an example of mapfile that causes this issue : > > https://pastebin.com/LFxZ1L9T > > And the PHP code : > > https://pastebin.com/Tp30bzf3 > > Thanks, > > Bruno > From jmckenna at gatewaygeomatics.com Thu Oct 15 05:42:50 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Thu, 15 Oct 2020 09:42:50 -0300 Subject: [mapserver-users] Mapserver 7.6.1 - PHP 7 Swig - Kml output issue In-Reply-To: References: Message-ID: <093a26a4-c788-55d7-c933-8c3b8c9b8250@gatewaygeomatics.com> oops here is the correct command for kml output: shp2img -m mymap.map -o ttt.kml -i kml -all_debug 5 (does that generate a valid KML file?) -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-15 9:41 a.m., Jeff McKenna wrote: > Hi Bruno, > > What happens if you call that mapfile at the commandline with shp2img > (and pass the "i" switch with your format name)?? such as: > > ? shp2img -m mymap.map -o ttt.png -i kml -all_debug 5 > > You can also then try png output: > > ? shp2img -m mymap.map -o ttt.png -i png -all_debug 5 > > Does it generate a valid map image? > > > > > -jeff > > > From bruno.domenget.roch at gmail.com Thu Oct 15 06:48:10 2020 From: bruno.domenget.roch at gmail.com (Bruno Domenget) Date: Thu, 15 Oct 2020 09:48:10 -0400 Subject: [mapserver-users] Mapserver 7.6.1 - PHP 7 Swig - Kml output issue In-Reply-To: <093a26a4-c788-55d7-c933-8c3b8c9b8250@gatewaygeomatics.com> References: <093a26a4-c788-55d7-c933-8c3b8c9b8250@gatewaygeomatics.com> Message-ID: Hi Jeff, Thanks for your quick answer ! So I tried with shp2img but got the same error with KML output... *msDrawMap(): Image handling error. Failed to draw layer named 'points'.
GDAL: In GDALDestroy - unloading GDAL shared library.* Logfile : [Thu Oct 15 15:36:02 2020].174254 msLoadMap(): 0.005s [Thu Oct 15 15:36:02 2020].174344 msDrawMap(): rendering using outputformat named kml (KML). [Thu Oct 15 15:36:02 2020].174348 msDrawMap(): WMS/WFS set-up and query, 0.000s [Thu Oct 15 15:36:02 2020].174369 msINLINELayerOpen: Layer is already open! [Thu Oct 15 15:36:02 2020].174564 msSaveImage(/home/mapserver/public_html/test/././5f885042_26ab_0.png) total time: 0.000s [Thu Oct 15 15:36:02 2020].174629 msDrawMap(): Image handling error. Failed to draw layer named 'points'. [Thu Oct 15 15:36:02 2020].174649 msFreeMap(): freeing map at 0x559913f1af60. [Thu Oct 15 15:36:02 2020].174698 freeLayer(): freeing layer at 0x559914207ad0. It works perfectly with PNG output, and with KML output if I remove the LABEL lines from my mapfile. So it is not a Mapscript issue, there is definitely an issue with my Mapserver instance... Bruno Le jeu. 15 oct. 2020 ? 08:42, Jeff McKenna a ?crit : > oops here is the correct command for kml output: > > shp2img -m mymap.map -o ttt.kml -i kml -all_debug 5 > > (does that generate a valid KML file?) > > > -jeff > > > > -- > Jeff McKenna > MapServer Consulting and Training Services > co-founder of FOSS4G > http://gatewaygeo.com/ > > > On 2020-10-15 9:41 a.m., Jeff McKenna wrote: > > Hi Bruno, > > > > What happens if you call that mapfile at the commandline with shp2img > > (and pass the "i" switch with your format name)? such as: > > > > shp2img -m mymap.map -o ttt.png -i kml -all_debug 5 > > > > You can also then try png output: > > > > shp2img -m mymap.map -o ttt.png -i png -all_debug 5 > > > > Does it generate a valid map image? > > > > > > > > > > -jeff > > > > > > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmckenna at gatewaygeomatics.com Thu Oct 15 06:52:23 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Thu, 15 Oct 2020 10:52:23 -0300 Subject: [mapserver-users] Mapserver 7.6.1 - PHP 7 Swig - Kml output issue In-Reply-To: References: <093a26a4-c788-55d7-c933-8c3b8c9b8250@gatewaygeomatics.com> Message-ID: Hi Bruno, Next test is to test that your local GDAL was built with KML support: ogrinfo --formats (you should see a 'LIBKML' or 'KML' driver listed there, either one can be used for this) -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-15 10:48 a.m., Bruno Domenget wrote: > Hi Jeff, > > Thanks for your quick answer ! > > So I tried with shp2img but got the same error with KML output... > *msDrawMap(): Image handling error. Failed to draw layer named 'points'. >
> GDAL: In GDALDestroy - unloading GDAL shared library.* > > Logfile : > [Thu Oct 15 15:36:02 2020].174254 msLoadMap(): 0.005s > [Thu Oct 15 15:36:02 2020].174344 msDrawMap(): rendering using > outputformat named kml (KML). > [Thu Oct 15 15:36:02 2020].174348 msDrawMap(): WMS/WFS set-up and query, > 0.000s > [Thu Oct 15 15:36:02 2020].174369 msINLINELayerOpen: Layer is already open! > [Thu Oct 15 15:36:02 2020].174564 > msSaveImage(/home/mapserver/public_html/test/././5f885042_26ab_0.png) > total time: 0.000s > [Thu Oct 15 15:36:02 2020].174629 msDrawMap(): Image handling error. > Failed to draw layer named 'points'. > [Thu Oct 15 15:36:02 2020].174649 msFreeMap(): freeing map at > 0x559913f1af60. > [Thu Oct 15 15:36:02 2020].174698 freeLayer(): freeing layer at > 0x559914207ad0. > > It works perfectly with PNG output, and with KML output if I remove the > LABEL lines from my mapfile. > > So it is not a Mapscript issue, there is definitely an issue with my > Mapserver instance... > > Bruno > > > > Le?jeu. 15 oct. 2020 ??08:42, Jeff McKenna > > a > ?crit?: > > oops here is the correct command for kml output: > > ? ?shp2img -m mymap.map -o ttt.kml -i kml -all_debug 5 > > (does that generate a valid KML file?) > > > -jeff > > > > -- > Jeff McKenna > MapServer Consulting and Training Services > co-founder of FOSS4G > http://gatewaygeo.com/ > > > On 2020-10-15 9:41 a.m., Jeff McKenna wrote: > > Hi Bruno, > > > > What happens if you call that mapfile at the commandline with > shp2img > > (and pass the "i" switch with your format name)?? such as: > > > >? ? shp2img -m mymap.map -o ttt.png -i kml -all_debug 5 > > > > You can also then try png output: > > > >? ? shp2img -m mymap.map -o ttt.png -i png -all_debug 5 > > > > Does it generate a valid map image? > > > > > > > > > > -jeff > > > > > > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users > From bruno.domenget.roch at gmail.com Thu Oct 15 06:58:34 2020 From: bruno.domenget.roch at gmail.com (Bruno Domenget) Date: Thu, 15 Oct 2020 09:58:34 -0400 Subject: [mapserver-users] Mapserver 7.6.1 - PHP 7 Swig - Kml output issue In-Reply-To: References: <093a26a4-c788-55d7-c933-8c3b8c9b8250@gatewaygeomatics.com> Message-ID: Sorry I should have said at first that I already checked that. I have both KML and LIBKML drivers listed. And KML output works if I remove the LABEL entry from my mapfile... Le jeu. 15 oct. 2020 ? 09:52, Jeff McKenna a ?crit : > Hi Bruno, > > Next test is to test that your local GDAL was built with KML support: > > ogrinfo --formats > > (you should see a 'LIBKML' or 'KML' driver listed there, either one can > be used for this) > > -jeff > > > > -- > Jeff McKenna > MapServer Consulting and Training Services > co-founder of FOSS4G > http://gatewaygeo.com/ > > > On 2020-10-15 10:48 a.m., Bruno Domenget wrote: > > Hi Jeff, > > > > Thanks for your quick answer ! > > > > So I tried with shp2img but got the same error with KML output... > > *msDrawMap(): Image handling error. Failed to draw layer named 'points'. > >
> > GDAL: In GDALDestroy - unloading GDAL shared library.* > > > > Logfile : > > [Thu Oct 15 15:36:02 2020].174254 msLoadMap(): 0.005s > > [Thu Oct 15 15:36:02 2020].174344 msDrawMap(): rendering using > > outputformat named kml (KML). > > [Thu Oct 15 15:36:02 2020].174348 msDrawMap(): WMS/WFS set-up and query, > > 0.000s > > [Thu Oct 15 15:36:02 2020].174369 msINLINELayerOpen: Layer is already > open! > > [Thu Oct 15 15:36:02 2020].174564 > > msSaveImage(/home/mapserver/public_html/test/././5f885042_26ab_0.png) > > total time: 0.000s > > [Thu Oct 15 15:36:02 2020].174629 msDrawMap(): Image handling error. > > Failed to draw layer named 'points'. > > [Thu Oct 15 15:36:02 2020].174649 msFreeMap(): freeing map at > > 0x559913f1af60. > > [Thu Oct 15 15:36:02 2020].174698 freeLayer(): freeing layer at > > 0x559914207ad0. > > > > It works perfectly with PNG output, and with KML output if I remove the > > LABEL lines from my mapfile. > > > > So it is not a Mapscript issue, there is definitely an issue with my > > Mapserver instance... > > > > Bruno > > > > > > > > Le jeu. 15 oct. 2020 ? 08:42, Jeff McKenna > > > > a > > ?crit : > > > > oops here is the correct command for kml output: > > > > shp2img -m mymap.map -o ttt.kml -i kml -all_debug 5 > > > > (does that generate a valid KML file?) > > > > > > -jeff > > > > > > > > -- > > Jeff McKenna > > MapServer Consulting and Training Services > > co-founder of FOSS4G > > http://gatewaygeo.com/ > > > > > > On 2020-10-15 9:41 a.m., Jeff McKenna wrote: > > > Hi Bruno, > > > > > > What happens if you call that mapfile at the commandline with > > shp2img > > > (and pass the "i" switch with your format name)? such as: > > > > > > shp2img -m mymap.map -o ttt.png -i kml -all_debug 5 > > > > > > You can also then try png output: > > > > > > shp2img -m mymap.map -o ttt.png -i png -all_debug 5 > > > > > > Does it generate a valid map image? > > > > > > > > > > > > > > > -jeff > > > > > > > > > > > > > _______________________________________________ > > mapserver-users mailing list > > mapserver-users at lists.osgeo.org mapserver-users at lists.osgeo.org> > > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > > > > _______________________________________________ > > mapserver-users mailing list > > mapserver-users at lists.osgeo.org > > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From flavio at tydac.ch Thu Oct 15 08:20:54 2020 From: flavio at tydac.ch (Flavio Hendry) Date: Thu, 15 Oct 2020 17:20:54 +0200 Subject: [mapserver-users] GEOMTRANSFROM Units on STYLE Level - Illustrated In-Reply-To: <44910a18-2ec3-4cfd-ad84-f8ed8f92ee25@www.fastmail.com> References: <0c84b242-318a-47c9-a118-007b1286311b@www.fastmail.com> <23656816-9b5c-8eb9-09e7-3d4aa74685eb@mapgears.com> <44910a18-2ec3-4cfd-ad84-f8ed8f92ee25@www.fastmail.com> Message-ID: <1a49b3a8-6ceb-8e40-2442-5eec25c0f6aa@tydac.ch> Hi all ps, this to illustrate the desired result: https://www.mapplus.ch/buffer.png -- Mit freundlichem Gruss / Best regards Flavio Hendry ---------------------------------------------------------------- TYDAC Inc. - http://www.tydac.ch Swiss Maps - http://www.mapplus.ch ---------------------------------------------------------------- ############ Mit freundlichen Gruessen / Kind Regards ############ Flavio Hendry, CEO - mailto:flavio at tydac.ch ############ TYDAC AG - http://www.tydac.ch #### #### Geographic Information Solutions #### #### Optingenstrasse 27 -- CH-3013 Bern ############ Tel +41 (0)31 368 0180 ---------------------------------------------------------------- Location: http://www.mapplus.ch/adr/bern/optingenstrasse/27 ---------------------------------------------------------------- From flavio at tydac.ch Thu Oct 15 08:05:19 2020 From: flavio at tydac.ch (Flavio Hendry) Date: Thu, 15 Oct 2020 17:05:19 +0200 Subject: [mapserver-users] GEOMTRANSFROM Units on STYLE Level In-Reply-To: <44910a18-2ec3-4cfd-ad84-f8ed8f92ee25@www.fastmail.com> References: <0c84b242-318a-47c9-a118-007b1286311b@www.fastmail.com> <23656816-9b5c-8eb9-09e7-3d4aa74685eb@mapgears.com> <44910a18-2ec3-4cfd-ad84-f8ed8f92ee25@www.fastmail.com> Message-ID: <5630ee35-6c28-6a9d-d03c-851c55b25982@tydac.ch> Hi all I did something really fancy ... Multiple overlapping Distance Buffers on a MultiPoint Layer (distance from Hydrants, one layer for all of them) ... so great. Until I looked at the result and read that: "At the LAYER level (since 6.4), the original vector geometry (?real world? coordinates) is used. At the STYLE level, pixel coordinates are used." eh? why that? Things like Buffer one almost always wants ?real world?, no? Like a distance around hydrants, a bus stations? In pixels? Would be great if GEOMTRANSFROM would support SIZEUNITS. Or at least the opposite of this, such as have real world on the STYLE level ...: "It may be useful to apply pixel values also at the LAYER level, and that is possible. If UNITS is defined in the LAYER, the [map_cellsize] variable can be used to convert to pixel values at the LAYER level:" -- Mit freundlichem Gruss / Best regards Flavio Hendry ---------------------------------------------------------------- TYDAC Inc. - http://www.tydac.ch Swiss Maps - http://www.mapplus.ch ---------------------------------------------------------------- ############ Mit freundlichen Gruessen / Kind Regards ############ Flavio Hendry, CEO - mailto:flavio at tydac.ch ############ TYDAC AG - http://www.tydac.ch #### #### Geographic Information Solutions #### #### Optingenstrasse 27 -- CH-3013 Bern ############ Tel +41 (0)31 368 0180 ---------------------------------------------------------------- Location: http://www.mapplus.ch/adr/bern/optingenstrasse/27 ---------------------------------------------------------------- From jmckenna at gatewaygeomatics.com Thu Oct 15 08:52:29 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Thu, 15 Oct 2020 12:52:29 -0300 Subject: [mapserver-users] Mapserver 7.6.1 - PHP 7 Swig - Kml output issue In-Reply-To: References: <093a26a4-c788-55d7-c933-8c3b8c9b8250@gatewaygeomatics.com> Message-ID: Hi Bruno, I can trigger this problem with any mapfile that uses an inline FEATURE, a label, and KML output (only with those 3 combinations enabled, as it works with other outputformats as SVG). I don't think it has anything to do with your settings, as I can trigger that here on Windows/MS4W as well. Sorry for the bad news. -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-15 10:58 a.m., Bruno Domenget wrote: > Sorry I should have said at first that?I already checked that. I have > both KML and LIBKML drivers listed. > > And KML output works if I remove the LABEL entry from my mapfile... > > Le?jeu. 15 oct. 2020 ??09:52, Jeff McKenna > > a > ?crit?: > > Hi Bruno, > > Next test is to test that your local GDAL was built with KML support: > > ? ?ogrinfo --formats > > (you should see a 'LIBKML' or 'KML' driver listed there, either one can > be used for this) > > -jeff > > > > -- > Jeff McKenna > MapServer Consulting and Training Services > co-founder of FOSS4G > http://gatewaygeo.com/ > > > On 2020-10-15 10:48 a.m., Bruno Domenget wrote: > > Hi Jeff, > > > > Thanks for your quick answer ! > > > > So I tried with shp2img but got the same error with KML output... > > *msDrawMap(): Image handling error. Failed to draw layer named > 'points'. > >
> > GDAL: In GDALDestroy - unloading GDAL shared library.* > > > > Logfile : > > [Thu Oct 15 15:36:02 2020].174254 msLoadMap(): 0.005s > > [Thu Oct 15 15:36:02 2020].174344 msDrawMap(): rendering using > > outputformat named kml (KML). > > [Thu Oct 15 15:36:02 2020].174348 msDrawMap(): WMS/WFS set-up and > query, > > 0.000s > > [Thu Oct 15 15:36:02 2020].174369 msINLINELayerOpen: Layer is > already open! > > [Thu Oct 15 15:36:02 2020].174564 > > > msSaveImage(/home/mapserver/public_html/test/././5f885042_26ab_0.png) > > total time: 0.000s > > [Thu Oct 15 15:36:02 2020].174629 msDrawMap(): Image handling error. > > Failed to draw layer named 'points'. > > [Thu Oct 15 15:36:02 2020].174649 msFreeMap(): freeing map at > > 0x559913f1af60. > > [Thu Oct 15 15:36:02 2020].174698 freeLayer(): freeing layer at > > 0x559914207ad0. > > > > It works perfectly with PNG output, and with KML output if I > remove the > > LABEL lines from my mapfile. > > > > So it is not a Mapscript issue, there is definitely an issue with my > > Mapserver instance... > > > > Bruno > > > > > > > > Le?jeu. 15 oct. 2020 ??08:42, Jeff McKenna > > > >> a > > ?crit?: > > > >? ? ?oops here is the correct command for kml output: > > > >? ? ? ? ?shp2img -m mymap.map -o ttt.kml -i kml -all_debug 5 > > > >? ? ?(does that generate a valid KML file?) > > > > > >? ? ?-jeff > > > > > > > >? ? ?-- > >? ? ?Jeff McKenna > >? ? ?MapServer Consulting and Training Services > >? ? ?co-founder of FOSS4G > > http://gatewaygeo.com/ > > > > > >? ? ?On 2020-10-15 9:41 a.m., Jeff McKenna wrote: > >? ? ? > Hi Bruno, > >? ? ? > > >? ? ? > What happens if you call that mapfile at the commandline with > >? ? ?shp2img > >? ? ? > (and pass the "i" switch with your format name)?? such as: > >? ? ? > > >? ? ? >? ? shp2img -m mymap.map -o ttt.png -i kml -all_debug 5 > >? ? ? > > >? ? ? > You can also then try png output: > >? ? ? > > >? ? ? >? ? shp2img -m mymap.map -o ttt.png -i png -all_debug 5 > >? ? ? > > >? ? ? > Does it generate a valid map image? > >? ? ? > > >? ? ? > > >? ? ? > > >? ? ? > > >? ? ? > -jeff > >? ? ? > > >? ? ? > > >? ? ? > From jmckenna at gatewaygeomatics.com Thu Oct 15 09:00:53 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Thu, 15 Oct 2020 13:00:53 -0300 Subject: [mapserver-users] Mapserver 7.6.1 - PHP 7 Swig - Kml output issue In-Reply-To: References: <093a26a4-c788-55d7-c933-8c3b8c9b8250@gatewaygeomatics.com> Message-ID: Bruno: since you get all the credit for this research, can you please file this as a ticket at : https://github.com/MapServer/MapServer/issues (include your mapfile and commands, nothing about mapscript). I believe this is something to do with how inline features are handled, for labels, in KML output. (maybe related to the message "msINLINELayerOpen: Layer is already open!" that MapServer is yelling at us, ha) Thanks and hello from the east coast of Canada, -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-15 10:58 a.m., Bruno Domenget wrote: > Sorry I should have said at first that?I already checked that. I have > both KML and LIBKML drivers listed. > > And KML output works if I remove the LABEL entry from my mapfile... > > Le?jeu. 15 oct. 2020 ??09:52, Jeff McKenna > > a > ?crit?: > > Hi Bruno, > > Next test is to test that your local GDAL was built with KML support: > > ? ?ogrinfo --formats > > (you should see a 'LIBKML' or 'KML' driver listed there, either one can > be used for this) > > -jeff > > > > -- > Jeff McKenna > MapServer Consulting and Training Services > co-founder of FOSS4G > http://gatewaygeo.com/ > > > On 2020-10-15 10:48 a.m., Bruno Domenget wrote: > > Hi Jeff, > > > > Thanks for your quick answer ! > > > > So I tried with shp2img but got the same error with KML output... > > *msDrawMap(): Image handling error. Failed to draw layer named > 'points'. > >
> > GDAL: In GDALDestroy - unloading GDAL shared library.* > > > > Logfile : > > [Thu Oct 15 15:36:02 2020].174254 msLoadMap(): 0.005s > > [Thu Oct 15 15:36:02 2020].174344 msDrawMap(): rendering using > > outputformat named kml (KML). > > [Thu Oct 15 15:36:02 2020].174348 msDrawMap(): WMS/WFS set-up and > query, > > 0.000s > > [Thu Oct 15 15:36:02 2020].174369 msINLINELayerOpen: Layer is > already open! > > [Thu Oct 15 15:36:02 2020].174564 > > > msSaveImage(/home/mapserver/public_html/test/././5f885042_26ab_0.png) > > total time: 0.000s > > [Thu Oct 15 15:36:02 2020].174629 msDrawMap(): Image handling error. > > Failed to draw layer named 'points'. > > [Thu Oct 15 15:36:02 2020].174649 msFreeMap(): freeing map at > > 0x559913f1af60. > > [Thu Oct 15 15:36:02 2020].174698 freeLayer(): freeing layer at > > 0x559914207ad0. > > > > It works perfectly with PNG output, and with KML output if I > remove the > > LABEL lines from my mapfile. > > > > So it is not a Mapscript issue, there is definitely an issue with my > > Mapserver instance... > > > > Bruno > > > > > > > > Le?jeu. 15 oct. 2020 ??08:42, Jeff McKenna > > > >> a > > ?crit?: > > > >? ? ?oops here is the correct command for kml output: > > > >? ? ? ? ?shp2img -m mymap.map -o ttt.kml -i kml -all_debug 5 > > > >? ? ?(does that generate a valid KML file?) > > > > > >? ? ?-jeff > > > > > > > >? ? ?-- > >? ? ?Jeff McKenna > >? ? ?MapServer Consulting and Training Services > >? ? ?co-founder of FOSS4G > > http://gatewaygeo.com/ > > > > > >? ? ?On 2020-10-15 9:41 a.m., Jeff McKenna wrote: > >? ? ? > Hi Bruno, > >? ? ? > > >? ? ? > What happens if you call that mapfile at the commandline with > >? ? ?shp2img > >? ? ? > (and pass the "i" switch with your format name)?? such as: > >? ? ? > > >? ? ? >? ? shp2img -m mymap.map -o ttt.png -i kml -all_debug 5 > >? ? ? > > >? ? ? > You can also then try png output: > >? ? ? > > >? ? ? >? ? shp2img -m mymap.map -o ttt.png -i png -all_debug 5 > >? ? ? > > >? ? ? > Does it generate a valid map image? > >? ? ? > > >? ? ? > > >? ? ? > > >? ? ? > > >? ? ? > -jeff > >? ? ? > > >? ? ? > > >? ? ? > > > > >? ? ?_______________________________________________ > >? ? ?mapserver-users mailing list > > mapserver-users at lists.osgeo.org > > > > > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > > > > _______________________________________________ > > mapserver-users mailing list > > mapserver-users at lists.osgeo.org > > > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users > From bruno.domenget.roch at gmail.com Thu Oct 15 09:23:40 2020 From: bruno.domenget.roch at gmail.com (Bruno Domenget) Date: Thu, 15 Oct 2020 12:23:40 -0400 Subject: [mapserver-users] Mapserver 7.6.1 - PHP 7 Swig - Kml output issue In-Reply-To: References: <093a26a4-c788-55d7-c933-8c3b8c9b8250@gatewaygeomatics.com> Message-ID: Hi Jeff, Thanks for taking the time to investigate that ! That is what I feared, but it is not such a big deal as we can still use csv export and write our own KML creation function. It would have been much worse if the issue had been about Shapefile export. I will create the ticket as soon as possible. Thanks and hello from Quebec ! Bruno Le jeu. 15 oct. 2020 ? 12:01, Jeff McKenna a ?crit : > Bruno: since you get all the credit for this research, can you please > file this as a ticket at : https://github.com/MapServer/MapServer/issues > (include your mapfile and commands, nothing about mapscript). I > believe this is something to do with how inline features are handled, > for labels, in KML output. (maybe related to the message > "msINLINELayerOpen: Layer is already open!" that MapServer is yelling at > us, ha) > > Thanks and hello from the east coast of Canada, > > -jeff > > > -- > Jeff McKenna > MapServer Consulting and Training Services > co-founder of FOSS4G > http://gatewaygeo.com/ > > > > > > On 2020-10-15 10:58 a.m., Bruno Domenget wrote: > > Sorry I should have said at first that I already checked that. I have > > both KML and LIBKML drivers listed. > > > > And KML output works if I remove the LABEL entry from my mapfile... > > > > Le jeu. 15 oct. 2020 ? 09:52, Jeff McKenna > > > > a > > ?crit : > > > > Hi Bruno, > > > > Next test is to test that your local GDAL was built with KML support: > > > > ogrinfo --formats > > > > (you should see a 'LIBKML' or 'KML' driver listed there, either one > can > > be used for this) > > > > -jeff > > > > > > > > -- > > Jeff McKenna > > MapServer Consulting and Training Services > > co-founder of FOSS4G > > http://gatewaygeo.com/ > > > > > > On 2020-10-15 10:48 a.m., Bruno Domenget wrote: > > > Hi Jeff, > > > > > > Thanks for your quick answer ! > > > > > > So I tried with shp2img but got the same error with KML output... > > > *msDrawMap(): Image handling error. Failed to draw layer named > > 'points'. > > >
> > > GDAL: In GDALDestroy - unloading GDAL shared library.* > > > > > > Logfile : > > > [Thu Oct 15 15:36:02 2020].174254 msLoadMap(): 0.005s > > > [Thu Oct 15 15:36:02 2020].174344 msDrawMap(): rendering using > > > outputformat named kml (KML). > > > [Thu Oct 15 15:36:02 2020].174348 msDrawMap(): WMS/WFS set-up and > > query, > > > 0.000s > > > [Thu Oct 15 15:36:02 2020].174369 msINLINELayerOpen: Layer is > > already open! > > > [Thu Oct 15 15:36:02 2020].174564 > > > > > msSaveImage(/home/mapserver/public_html/test/././5f885042_26ab_0.png) > > > total time: 0.000s > > > [Thu Oct 15 15:36:02 2020].174629 msDrawMap(): Image handling > error. > > > Failed to draw layer named 'points'. > > > [Thu Oct 15 15:36:02 2020].174649 msFreeMap(): freeing map at > > > 0x559913f1af60. > > > [Thu Oct 15 15:36:02 2020].174698 freeLayer(): freeing layer at > > > 0x559914207ad0. > > > > > > It works perfectly with PNG output, and with KML output if I > > remove the > > > LABEL lines from my mapfile. > > > > > > So it is not a Mapscript issue, there is definitely an issue with > my > > > Mapserver instance... > > > > > > Bruno > > > > > > > > > > > > Le jeu. 15 oct. 2020 ? 08:42, Jeff McKenna > > > > > > > >> a > > > ?crit : > > > > > > oops here is the correct command for kml output: > > > > > > shp2img -m mymap.map -o ttt.kml -i kml -all_debug 5 > > > > > > (does that generate a valid KML file?) > > > > > > > > > -jeff > > > > > > > > > > > > -- > > > Jeff McKenna > > > MapServer Consulting and Training Services > > > co-founder of FOSS4G > > > http://gatewaygeo.com/ > > > > > > > > > On 2020-10-15 9:41 a.m., Jeff McKenna wrote: > > > > Hi Bruno, > > > > > > > > What happens if you call that mapfile at the commandline > with > > > shp2img > > > > (and pass the "i" switch with your format name)? such as: > > > > > > > > shp2img -m mymap.map -o ttt.png -i kml -all_debug 5 > > > > > > > > You can also then try png output: > > > > > > > > shp2img -m mymap.map -o ttt.png -i png -all_debug 5 > > > > > > > > Does it generate a valid map image? > > > > > > > > > > > > > > > > > > > > -jeff > > > > > > > > > > > > > > > > > > _______________________________________________ > > > mapserver-users mailing list > > > mapserver-users at lists.osgeo.org > > > > > > > > > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > > > > > > > _______________________________________________ > > > mapserver-users mailing list > > > mapserver-users at lists.osgeo.org > > > > > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > > > > > _______________________________________________ > > mapserver-users mailing list > > mapserver-users at lists.osgeo.org mapserver-users at lists.osgeo.org> > > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > > > > _______________________________________________ > > mapserver-users mailing list > > mapserver-users at lists.osgeo.org > > https://lists.osgeo.org/mailman/listinfo/mapserver-users > > > > _______________________________________________ > mapserver-users mailing list > mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From bruno.domenget.roch at gmail.com Mon Oct 19 17:22:42 2020 From: bruno.domenget.roch at gmail.com (Bruno Domenget) Date: Mon, 19 Oct 2020 20:22:42 -0400 Subject: [mapserver-users] Mapserver 7.6.1 - PHP 7 Swig - Kml output issue In-Reply-To: References: <093a26a4-c788-55d7-c933-8c3b8c9b8250@gatewaygeomatics.com> Message-ID: Hi Jeff, Thanks for taking the time to investigate that ! That is what I feared, but it is not such a big deal as we can still use csv export and write our own KML creation function. It would have been much worse if the issue had been about Shapefile export. I will create the ticket as soon as possible. Thanks and hello from Quebec ! Bruno -------------- next part -------------- An HTML attachment was scrubbed... URL: From zmitserk at mail.ru Wed Oct 21 03:14:58 2020 From: zmitserk at mail.ru (=?UTF-8?B?Wm1pdHNlciBLb3podWto?=) Date: Wed, 21 Oct 2020 13:14:58 +0300 Subject: [mapserver-users] =?utf-8?q?problem_with_getcapabilities_xml_gene?= =?utf-8?q?ration_when_layer_is_using_data_from_postgresql_inherited_table?= Message-ID: <1603275298.195396547@f418.i.mail.ru> Good time of the day, may be it is my misconception, but I used to think that generation of?GetCapabilities xml?doesn?t require checking that the data source of the layer is valid and it really has data?etc. I was thinking that this is just a matter of converting metadata from mapfile to xml format?by certain rules. But it appears, that it is different from what I have been thinking. I am having a wms service that reads data from inherited postgresql table (table inherits data from many thousands of childs table) and it takes infinity to generate getcapabilities xml document: https://gis.lesprojekt.cz/cgi-bin/mapserv?map=/home/dima/maps/olu/european_openlandusemap.map&service=WMS&request=GetCapabilities With other wms services that use just ordinary tables as data source it (this instance of mapserver)?works just fine. So my questions is : is it possible to change the code of mapserver not to check all child tables and instead just generate getcapabilities xml from the metadata in mapfile? Or may be someone had this issue and found other solution? I think in older versions of mapserver (this concrete instance is?version 7.2.0 )?inherited tables didn?t cause any troubles. Sincerely, Dimitri ? -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmckenna at gatewaygeomatics.com Wed Oct 21 04:31:39 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Wed, 21 Oct 2020 08:31:39 -0300 Subject: [mapserver-users] problem with getcapabilities xml generation when layer is using data from postgresql inherited table In-Reply-To: <1603275298.195396547@f418.i.mail.ru> References: <1603275298.195396547@f418.i.mail.ru> Message-ID: Hi Dimitri, For database connections such as PostgreSQL I always manually set the LAYER metadata "wms_extent" for each layer (the actual bounding box of that specific table in PostGIS), so that MapServer does not have to calculate the extents for the GetCapabilities for each layer. (I believe this is also mentioned as a "WARNING" in the GetCapabilities response). -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-21 7:14 a.m., Zmitser Kozhukh wrote: > Good time of the day, may be it is my misconception, but I used to think > that generation of?GetCapabilities xml?doesn?t require checking that the > data source of the layer is valid and it really has data?etc. I was > thinking that this is just a matter of converting metadata from mapfile > to xml format?by certain rules. > But it appears, that it is different from what I have been thinking. I > am having a wms service that reads data from inherited postgresql table > (table inherits data from many thousands of childs table) and it takes > infinity to generate getcapabilities xml document: > https://gis.lesprojekt.cz/cgi-bin/mapserv?map=/home/dima/maps/olu/european_openlandusemap.map&service=WMS&request=GetCapabilities > With other wms services that use just ordinary tables as data source it > (this instance of mapserver)?works just fine. > So my questions is : is it possible to change the code of mapserver not > to check all child tables and instead just generate getcapabilities xml > from the metadata in mapfile? > Or may be someone had this issue and found other solution? > I think in older versions of mapserver (this concrete instance > is?version 7.2.0 )?inherited tables didn?t cause any troubles. > Sincerely, Dimitri > From jmckenna at gatewaygeomatics.com Wed Oct 21 04:35:56 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Wed, 21 Oct 2020 08:35:56 -0300 Subject: [mapserver-users] problem with getcapabilities xml generation when layer is using data from postgresql inherited table In-Reply-To: <1603275298.195396547@f418.i.mail.ru> References: <1603275298.195396547@f418.i.mail.ru> Message-ID: I should also mention that for each of my PostgreSQL layers I always enable connection pooling for each LAYER: PROCESSING "CLOSE_CONNECTION=DEFER" Although my notes don't answer your specific question, these are the steps that I always follow. -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-21 7:14 a.m., Zmitser Kozhukh wrote: > Good time of the day, may be it is my misconception, but I used to think > that generation of?GetCapabilities xml?doesn?t require checking that the > data source of the layer is valid and it really has data?etc. I was > thinking that this is just a matter of converting metadata from mapfile > to xml format?by certain rules. > But it appears, that it is different from what I have been thinking. I > am having a wms service that reads data from inherited postgresql table > (table inherits data from many thousands of childs table) and it takes > infinity to generate getcapabilities xml document: > https://gis.lesprojekt.cz/cgi-bin/mapserv?map=/home/dima/maps/olu/european_openlandusemap.map&service=WMS&request=GetCapabilities > With other wms services that use just ordinary tables as data source it > (this instance of mapserver)?works just fine. > So my questions is : is it possible to change the code of mapserver not > to check all child tables and instead just generate getcapabilities xml > from the metadata in mapfile? > Or may be someone had this issue and found other solution? > I think in older versions of mapserver (this concrete instance > is?version 7.2.0 )?inherited tables didn?t cause any troubles. > Sincerely, Dimitri > From zmitserk at mail.ru Wed Oct 21 23:28:39 2020 From: zmitserk at mail.ru (=?UTF-8?B?Wm1pdHNlciBLb3podWto?=) Date: Thu, 22 Oct 2020 09:28:39 +0300 Subject: [mapserver-users] =?utf-8?q?mapserver-users_Digest=2C_Vol_153=2C_?= =?utf-8?q?Issue_13?= In-Reply-To: References: Message-ID: <1603348119.554298250@f450.i.mail.ru> Thank you, Jeff. Specifying wms_extent in metadata has helped. I didn?t think that all these efforts (reading tens of thousands child tables) was just to figure out layer extent. Somehow was thinking that it is taken automatically from mapfile header. I need to check just for an interest to read such layer through OGR library and see how much it will take to complete layer.GetExtent() :D? Regarding close_connection=defer directive. Does it mean ?that if I use it the connection will be opened for the whole?time wms is being?viewed by user and all the select queries will use this exact connection. While if I wouldn?t use this directive:??different select queries would have different connections opened? Or do I misunderstand something? Sincerely, Dimitri ? ? >?????, 21 ??????? 2020, 22:01 +03:00 ?? mapserver-users-request at lists.osgeo.org: >? >Send mapserver-users mailing list submissions to >mapserver-users at lists.osgeo.org > >To subscribe or unsubscribe via the World Wide Web, visit >https://lists.osgeo.org/mailman/listinfo/mapserver-users >or, via email, send a message with subject or body 'help' to >mapserver-users-request at lists.osgeo.org > >You can reach the person managing the list at >mapserver-users-owner at lists.osgeo.org > >When replying, please edit your Subject line so it is more specific >than "Re: Contents of mapserver-users digest..." > > >Today's Topics: > >???1. problem with getcapabilities xml generation when layer is >??????using data from postgresql inherited table (Zmitser Kozhukh) >???2. Re: problem with getcapabilities xml generation when layer is >??????using data from postgresql inherited table (Jeff McKenna) >???3. Re: problem with getcapabilities xml generation when layer is >??????using data from postgresql inherited table (Jeff McKenna) > > >---------------------------------------------------------------------- > >Message: 1 >Date: Wed, 21 Oct 2020 13:14:58 +0300 >From: Zmitser Kozhukh < zmitserk at mail.ru > >To: mapserver-users < mapserver-users at lists.osgeo.org > >Subject: [mapserver-users] problem with getcapabilities xml generation >when layer is using data from postgresql inherited table >Message-ID: < 1603275298.195396547 at f418.i.mail.ru > >Content-Type: text/plain; charset="utf-8" > > >Good time of the day, may be it is my misconception, but I used to think that generation of?GetCapabilities xml?doesn?t require checking that the data source of the layer is valid and it really has data?etc. I was thinking that this is just a matter of converting metadata from mapfile to xml format?by certain rules. >But it appears, that it is different from what I have been thinking. I am having a wms service that reads data from inherited postgresql table (table inherits data from many thousands of childs table) and it takes infinity to generate getcapabilities xml document: >https://gis.lesprojekt.cz/cgi-bin/mapserv?map=/home/dima/maps/olu/european_openlandusemap.map&service=WMS&request=GetCapabilities >With other wms services that use just ordinary tables as data source it (this instance of mapserver)?works just fine. >So my questions is : is it possible to change the code of mapserver not to check all child tables and instead just generate getcapabilities xml from the metadata in mapfile? >Or may be someone had this issue and found other solution? >I think in older versions of mapserver (this concrete instance is?version 7.2.0 )?inherited tables didn?t cause any troubles. >Sincerely, Dimitri >? >-------------- next part -------------- >An HTML attachment was scrubbed... >URL: < http://lists.osgeo.org/pipermail/mapserver-users/attachments/20201021/397456cd/attachment-0001.html > > >------------------------------ > >Message: 2 >Date: Wed, 21 Oct 2020 08:31:39 -0300 >From: Jeff McKenna < jmckenna at gatewaygeomatics.com > >To: mapserver-users at lists.osgeo.org >Subject: Re: [mapserver-users] problem with getcapabilities xml >generation when layer is using data from postgresql inherited table >Message-ID: >< aec3cd68-66bb-237e-5028-cad64d11e575 at gatewaygeomatics.com > >Content-Type: text/plain; charset=utf-8; format=flowed > >Hi Dimitri, > >For database connections such as PostgreSQL I always manually set the >LAYER metadata "wms_extent" for each layer (the actual bounding box of >that specific table in PostGIS), so that MapServer does not have to >calculate the extents for the GetCapabilities for each layer. (I >believe this is also mentioned as a "WARNING" in the GetCapabilities >response). > >-jeff > > > >-- >Jeff McKenna >MapServer Consulting and Training Services >co-founder of FOSS4G >http://gatewaygeo.com/ > > > >On 2020-10-21 7:14 a.m., Zmitser Kozhukh wrote: >> Good time of the day, may be it is my misconception, but I used to think >> that generation of?GetCapabilities xml?doesn?t require checking that the >> data source of the layer is valid and it really has data?etc. I was >> thinking that this is just a matter of converting metadata from mapfile >> to xml format?by certain rules. >> But it appears, that it is different from what I have been thinking. I >> am having a wms service that reads data from inherited postgresql table >> (table inherits data from many thousands of childs table) and it takes >> infinity to generate getcapabilities xml document: >> https://gis.lesprojekt.cz/cgi-bin/mapserv?map=/home/dima/maps/olu/european_openlandusemap.map&service=WMS&request=GetCapabilities >> With other wms services that use just ordinary tables as data source it >> (this instance of mapserver)?works just fine. >> So my questions is : is it possible to change the code of mapserver not >> to check all child tables and instead just generate getcapabilities xml >> from the metadata in mapfile? >> Or may be someone had this issue and found other solution? >> I think in older versions of mapserver (this concrete instance >> is?version 7.2.0 )?inherited tables didn?t cause any troubles. >> Sincerely, Dimitri >> > > > > > >------------------------------ > >Message: 3 >Date: Wed, 21 Oct 2020 08:35:56 -0300 >From: Jeff McKenna < jmckenna at gatewaygeomatics.com > >To: mapserver-users at lists.osgeo.org >Subject: Re: [mapserver-users] problem with getcapabilities xml >generation when layer is using data from postgresql inherited table >Message-ID: >< add5b6e5-479e-4f4e-61ef-5570c1e94ab1 at gatewaygeomatics.com > >Content-Type: text/plain; charset=utf-8; format=flowed > >I should also mention that for each of my PostgreSQL layers I always >enable connection pooling for each LAYER: > >???PROCESSING "CLOSE_CONNECTION=DEFER" > >Although my notes don't answer your specific question, these are the >steps that I always follow. > >-jeff > > > >-- >Jeff McKenna >MapServer Consulting and Training Services >co-founder of FOSS4G >http://gatewaygeo.com/ > > >On 2020-10-21 7:14 a.m., Zmitser Kozhukh wrote: >> Good time of the day, may be it is my misconception, but I used to think >> that generation of?GetCapabilities xml?doesn?t require checking that the >> data source of the layer is valid and it really has data?etc. I was >> thinking that this is just a matter of converting metadata from mapfile >> to xml format?by certain rules. >> But it appears, that it is different from what I have been thinking. I >> am having a wms service that reads data from inherited postgresql table >> (table inherits data from many thousands of childs table) and it takes >> infinity to generate getcapabilities xml document: >> https://gis.lesprojekt.cz/cgi-bin/mapserv?map=/home/dima/maps/olu/european_openlandusemap.map&service=WMS&request=GetCapabilities >> With other wms services that use just ordinary tables as data source it >> (this instance of mapserver)?works just fine. >> So my questions is : is it possible to change the code of mapserver not >> to check all child tables and instead just generate getcapabilities xml >> from the metadata in mapfile? >> Or may be someone had this issue and found other solution? >> I think in older versions of mapserver (this concrete instance >> is?version 7.2.0 )?inherited tables didn?t cause any troubles. >> Sincerely, Dimitri >> > > >------------------------------ > >Subject: Digest Footer > >_______________________________________________ >mapserver-users mailing list >mapserver-users at lists.osgeo.org >https://lists.osgeo.org/mailman/listinfo/mapserver-users > >------------------------------ > >End of mapserver-users Digest, Vol 153, Issue 13 >************************************************ -- Zmitser Kozhukh -------------- next part -------------- An HTML attachment was scrubbed... URL: From richard.greenwood at gmail.com Sat Oct 24 10:21:55 2020 From: richard.greenwood at gmail.com (Richard Greenwood) Date: Sat, 24 Oct 2020 11:21:55 -0600 Subject: [mapserver-users] Is mode=tile always epsg:3857? Message-ID: I'm trying to generate vector tiles in a non epsg:3857 (web mercator) projection but it looks to me like mapserver is treating it as 3857 despite passing a srs. My request generated by OpenLayers at the full map extent is: /cgi-bin/mapserv?map=mvt.map&SRS=EPSG:3738&mode=tile&tile=0+0+0&layers=ownership&map.imagetype=mvt My extents are: 1622000 639000 2308000 1284700 My mapserver log looks like: [Sat Oct 24 11:10:58 2020].947687 msTileSetParams(): tile_metatile_level = 0 [Sat Oct 24 11:10:58 2020].947690 msTileSetExtent(): gmaps coords (x: 0, y: 0, z: 0) [Sat Oct 24 11:10:58 2020].947692 msTileSetExtent(): gmaps metacoords (x: 0, y: 0, z: 0) [Sat Oct 24 11:10:58 2020].947695 msTileSetExtent(): base image size (256 x 256) [Sat Oct 24 11:10:58 2020].947697 msTileSetExtent(): buffered image size (256 x 256) [Sat Oct 24 11:10:58 2020].947700 msTileSetExtent (-19959236.823047, -19959236.823047) (19959236.823047, 19959236.823047) There's obviously something that I'm not getting... -- Richard W. Greenwood, PLS www.greenwoodmap.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From michael.smith.erdc at gmail.com Sat Oct 24 10:35:38 2020 From: michael.smith.erdc at gmail.com (Michael Smith) Date: Sat, 24 Oct 2020 13:35:38 -0400 Subject: [mapserver-users] Is mode=tile always epsg:3857? In-Reply-To: References: Message-ID: Rich, Looking at mode=tile, yes, it sets it to spherical Mercator https://github.com/MapServer/MapServer/blob/master/maptile.c#L259 From: mapserver-users on behalf of Richard Greenwood Date: Saturday, October 24, 2020 at 1:22 PM To: mapserver Subject: [mapserver-users] Is mode=tile always epsg:3857? I'm trying to generate vector tiles in a non epsg:3857 (web mercator) projection but it looks to me like mapserver is treating it as 3857 despite passing a srs. My request generated by OpenLayers at the full map extent is: /cgi-bin/mapserv?map=mvt.map&SRS=EPSG:3738&mode=tile&tile=0+0+0&layers=ownership&map.imagetype=mvt My extents are: 1622000 639000 2308000 1284700 My mapserver log looks like: [Sat Oct 24 11:10:58 2020].947687 msTileSetParams(): tile_metatile_level = 0 [Sat Oct 24 11:10:58 2020].947690 msTileSetExtent(): gmaps coords (x: 0, y: 0, z: 0) [Sat Oct 24 11:10:58 2020].947692 msTileSetExtent(): gmaps metacoords (x: 0, y: 0, z: 0) [Sat Oct 24 11:10:58 2020].947695 msTileSetExtent(): base image size (256 x 256) [Sat Oct 24 11:10:58 2020].947697 msTileSetExtent(): buffered image size (256 x 256) [Sat Oct 24 11:10:58 2020].947700 msTileSetExtent (-19959236.823047, -19959236.823047) (19959236.823047, 19959236.823047) There's obviously something that I'm not getting... -- Richard W. Greenwood, PLS www.greenwoodmap.com _______________________________________________ mapserver-users mailing list mapserver-users at lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/mapserver-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From richard.greenwood at gmail.com Sat Oct 24 11:43:44 2020 From: richard.greenwood at gmail.com (Richard Greenwood) Date: Sat, 24 Oct 2020 12:43:44 -0600 Subject: [mapserver-users] Is mode=tile always epsg:3857? In-Reply-To: References: Message-ID: Thanks Mike! On Sat, Oct 24, 2020 at 11:35 AM Michael Smith wrote: > Rich, > > > > Looking at mode=tile, yes, it sets it to spherical Mercator > > > > https://github.com/MapServer/MapServer/blob/master/maptile.c#L259 > > > > > > > > *From: *mapserver-users on > behalf of Richard Greenwood > *Date: *Saturday, October 24, 2020 at 1:22 PM > *To: *mapserver > *Subject: *[mapserver-users] Is mode=tile always epsg:3857? > > > > I'm trying to generate vector tiles in a non epsg:3857 (web mercator) > projection but it looks to me like mapserver is treating it as 3857 despite > passing a srs. > > > > My request generated by OpenLayers at the full map extent is: > > > /cgi-bin/mapserv?map=mvt.map&SRS=EPSG:3738&mode=tile&tile=0+0+0&layers=ownership&map.imagetype=mvt > > > > My extents are: 1622000 639000 2308000 1284700 > > > > My mapserver log looks like: > > [Sat Oct 24 11:10:58 2020].947687 msTileSetParams(): tile_metatile_level = > 0 > [Sat Oct 24 11:10:58 2020].947690 msTileSetExtent(): gmaps coords (x: 0, > y: 0, z: 0) > [Sat Oct 24 11:10:58 2020].947692 msTileSetExtent(): gmaps metacoords (x: > 0, y: 0, z: 0) > [Sat Oct 24 11:10:58 2020].947695 msTileSetExtent(): base image size (256 > x 256) > [Sat Oct 24 11:10:58 2020].947697 msTileSetExtent(): buffered image size > (256 x 256) > [Sat Oct 24 11:10:58 2020].947700 msTileSetExtent (-19959236.823047, > -19959236.823047) (19959236.823047, 19959236.823047) > > > > There's obviously something that I'm not getting... > > > > -- > > Richard W. Greenwood, PLS > www.greenwoodmap.com > > _______________________________________________ mapserver-users mailing > list mapserver-users at lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/mapserver-users > -- Richard W. Greenwood, PLS www.greenwoodmap.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmckenna at gatewaygeomatics.com Tue Oct 27 05:47:07 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Tue, 27 Oct 2020 09:47:07 -0300 Subject: [mapserver-users] Is mode=tile always epsg:3857? In-Reply-To: References: Message-ID: <7ead2a24-297c-bd7f-f6a5-8ece4ffde59f@gatewaygeomatics.com> Thanks Rich and Mike, I've done some more local tests and given some much needed love to the mode=tile documentation, so hopefully these questions are prevented in the future. Let me know if you notice anything else. https://mapserver.org/output/tile_mode.html -jeff -- Jeff McKenna MapServer Consulting and Training Services co-founder of FOSS4G http://gatewaygeo.com/ On 2020-10-24 3:43 p.m., Richard Greenwood wrote: > Thanks Mike! > > On Sat, Oct 24, 2020 at 11:35 AM Michael Smith > > wrote: > > Rich,____ > > __ __ > > Looking at mode=tile, yes, it sets it to spherical Mercator____ > > __ __ > > https://github.com/MapServer/MapServer/blob/master/maptile.c#L259____ > > __ __ > > __ __ > > __ __ > > *From: *mapserver-users > on behalf of > Richard Greenwood > > *Date: *Saturday, October 24, 2020 at 1:22 PM > *To: *mapserver > > *Subject: *[mapserver-users] Is mode=tile always epsg:3857?____ > > __ __ > > I'm trying to generate vector tiles in a non epsg:3857 (web > mercator) projection but it looks to me like mapserver is treating > it as 3857 despite passing a srs. ____ > > __ __ > > My request generated by OpenLayers at the full map extent is:____ > > > /cgi-bin/mapserv?map=mvt.map&SRS=EPSG:3738&mode=tile&tile=0+0+0&layers=ownership&map.imagetype=mvt____ > > __ __ > > My extents are:?1622000 639000 2308000 1284700____ > > __ __ > > My mapserver log looks like:____ > > [Sat Oct 24 11:10:58 2020].947687 msTileSetParams(): > tile_metatile_level = 0 > [Sat Oct 24 11:10:58 2020].947690 msTileSetExtent(): gmaps coords > (x: 0, y: 0, z: 0) > [Sat Oct 24 11:10:58 2020].947692 msTileSetExtent(): gmaps > metacoords (x: 0, y: 0, z: 0) > [Sat Oct 24 11:10:58 2020].947695 msTileSetExtent(): base image size > (256 x 256) > [Sat Oct 24 11:10:58 2020].947697 msTileSetExtent(): buffered image > size (256 x 256) > [Sat Oct 24 11:10:58 2020].947700 msTileSetExtent (-19959236.823047, > -19959236.823047) (19959236.823047, 19959236.823047) > ____ > > __ __ > > There's obviously something that I'm not getting...____ > > __ __ > > -- ____ > > Richard W. Greenwood, PLS > www.greenwoodmap.com ____ >