From jmckenna at gatewaygeomatics.com Mon May 4 14:24:48 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Mon, 4 May 2020 18:24:48 -0300 Subject: [mapserver-users] 7.6.0-rc4 released Message-ID: <23c55a34-15db-373f-9760-3c503870fe0b@gatewaygeomatics.com> The MapServer team is pleased to announce the forth release candidate of MapServer 7.6. Please help us test and let us know of any findings. (this RC fixes a critical regression for PostGIS layers) Details of the new features, migration instructions, and download links can be found on the dedicated announcement page at https://mapserver.org/development/announce/7-6.html . Thank you to everyone testing the beta/rc releases, and the developers providing fixes. If no other issues, 7.6.0 will be released at the end of this week. -- The MapServer Team From bob.basques at ci.stpaul.mn.us Wed May 6 10:25:00 2020 From: bob.basques at ci.stpaul.mn.us (Basques, Bob (CI-StPaul)) Date: Wed, 6 May 2020 17:25:00 +0000 Subject: [mapserver-users] REMINDER: May 13, 2020 - OSGeo Local Chapter Virtual Meeting (aka TCMUG) Message-ID: All, So, the last meeting was interesting in that we went virtual with it for the first time. All in all it worked out very well. We even had a few out of towners show up. We presented a couple of short presentations that seemed to be of interest to the group based on the questions raised. Operator error (on my part) prevented me from recording anything from the last meeting, I forgot to hit the record button, so no pictures , not even a screenshot. I'll attempt to do better this time around. We can still add a second presentation if anyone is interested, just let me know, or raise the topic at the meeting I'm liking the capabilities that Jitsi allows, but I also don't want the group to get bombed while a meeting is in progress. Still learning how to best secure the meeting. Until I figure out the basics on this, can I have everyone interested in attending send me their email address in order to send out the Jitsi meeting invite directly, vs announcing on the lists, which seems to be a no-no? Looking into some sort of Web based distro of the invite that folks would need to login to. But maybe someone out there has a better idea?? The May meeting will feature a guest speaker, Michael Battaglia (mjbattag at mtu.edu) - Bio. A blurb from Michael about the presentation: "Coastal wetlands are a critical component of the Great Lakes ecosystem. Fluctuating lake levels, anthropogenic activity, and invasion by non-native vegetation have led to rapid changes to these valuable, yet fragile, wetland areas. Frequent monitoring is necessary to understand where and when changes are occurring, however, field visits are costly and time consuming. Techniques leveraging spaceborne synthetic aperture radar data enable timely remote monitoring of a variety of important variables, including wetland type and inundation extent. This talk will show methods used to generate map products for those parameters and share results generated with RADARSAT-2 SAR data for several pilot sites around the Great Lakes. Duration: approx.. 30 min. Meeting details: Where: Jitsi (send email for invite) When: 4:30 pm, May 13th, 2020. bobb -------------- next part -------------- An HTML attachment was scrubbed... URL: From luis.de.sousa at protonmail.ch Thu May 7 01:03:43 2020 From: luis.de.sousa at protonmail.ch (=?UTF-8?Q?Lu=C3=AD=C2=ADs_Moreira_de_Sousa?=) Date: Thu, 07 May 2020 08:03:43 +0000 Subject: [mapserver-users] Using Raster Attribute Table (RAT) with WCS Message-ID: Dear all, I have a WCS serving a few thematic maps, where each cell represents a class, like in a land use or ecological map. The maps include a Raster Attribute Table (RAT) which essentially relates cell values to classes (e.g. 1 - Grassland). What would be the correct way of providing the RAT to users of these services? As far as I can tell, this is not supported in the WCS core. Is it somehow considered in an extension? Besides WCS, would it be possible to publish the RAT through MapServer in some other way? Thank you. -- Lu?s -------------- next part -------------- An HTML attachment was scrubbed... URL: From schroeter at netgis.de Thu May 7 02:36:06 2020 From: schroeter at netgis.de (Sven Schroeter) Date: Thu, 7 May 2020 11:36:06 +0200 Subject: [mapserver-users] HTML Legends with MapServer Message-ID: <99c089b2-e01c-7acf-c9b4-9a2c314ae363@netgis.de> Hi all, i have been using html legend templates for the mapserver for a very long time. With many layers such a template can become very long and confusing. Is there a possibility to split such a template into several files and e.g. include certain areas like in the mapfile? Maybe I could solve this with PHP Mapscript?? Up to now I have been using processLegendTemplate to output the HTML code. Thanks Sven From Anton.Bakker at kadaster.nl Thu May 7 05:21:16 2020 From: Anton.Bakker at kadaster.nl (Bakker, Anton) Date: Thu, 7 May 2020 12:21:16 +0000 Subject: [mapserver-users] WCS GetCoverage request one by one pixel Message-ID: Hi Jukka, I finally managed to check up on your suggestion. I set the RESAMPLE mode to NEAREST in the mapfile on a MapServer instance running locally on my machine. This results in the same response as setting the resampling method in the url (see my previous email) : http://localhost:32788/ahn3/wcs?service=WCS&Request=GetCoverage&version=2.0.1&CoverageId=ahn3_05m_dsm&format=image/tiff&subset=x(196396.75,196396.75)&subset=y(450031.75,450031.75) msImageCreate(): Image handling error. Attempt to allocate raw image failed, out of memory. Also all resampling methods result in the same response (tested by supplying through the query parameter in the url). When I look in the logs I see that MapServer determines the dimensions of the request to be 0x0 pixels, causing the creation of the image to fail: 2020-05-06 17:38:11: (mod_fastcgi.c.421) FastCGI-stderr: Subset for X-axis found: x 2020-05-06 17:38:11: (mod_fastcgi.c.421) FastCGI-stderr: Subset for Y-axis found: y 2020-05-06 17:38:11: (mod_fastcgi.c.421) FastCGI-stderr: msWCSGetCoverage20(): Set parameters from originaldata. Width: 0, height: 0, cellsize: -nan, extent: -nan,-nan,-nan,-nan 2020-05-06 17:38:11: (mod_fastcgi.c.421) FastCGI-stderr: msImageCreate(): Image handling error. Attempt to allocate raw image failed, out of memory. ?I also quickly checked whether requesting a 1x1 subset works, but this fails by returning geotiff image of 1x1 pixel with a pixel size of "-135000,162500": http://localhost:32788/ahn3/wcs?service=WCS&Request=GetCoverage&version=2.0.1&CoverageId=ahn3_05m_dsm&format=image/tiff&subset=x(196396.75,196397.25)&subset=y(450031.75,450032.25) However requesting an image of 2x2 pixels still works as expected: http://localhost:32788/ahn3/wcs?service=WCS&Request=GetCoverage&version=2.0.1&CoverageId=ahn3_05m_dsm&format=image/tiff&subset=x(196396.75,196397.75)&subset=y(450031.75,450032.75) Kind regards, Anton ________________________________ From: Rahkonen Jukka (MML) > Sent: Tuesday, April 7, 2020 5:49 PM To: Bakker, Anton >; mapserver-users at lists.osgeo.org > Subject: Re: WCS GetCoverage request one by one pixel Hi Anton, What I meant was to have a try by commenting out this line from your mapfile PROCESSING "RESAMPLE=BILINEAR" Actually it would be interesting to know what really happens if the interpolation/resampling method is configured in the mapfile and GetCoverage comes also with Interpolation parameter. -Jukka- -------------- next part -------------- An HTML attachment was scrubbed... URL: From sdlime at gmail.com Thu May 7 07:30:10 2020 From: sdlime at gmail.com (Steve Lime) Date: Thu, 7 May 2020 09:30:10 -0500 Subject: [mapserver-users] HTML Legends with MapServer In-Reply-To: <99c089b2-e01c-7acf-c9b4-9a2c314ae363@netgis.de> References: <99c089b2-e01c-7acf-c9b4-9a2c314ae363@netgis.de> Message-ID: Query templates already support the concept of includes: [include src="some other template"] so I wonder if it might be possible to extend that functionality to HTML legend processing. Do you have a particularly complex example you could share with me? --Steve On Thu, May 7, 2020 at 4:42 AM Sven Schroeter wrote: > Hi all, > > i have been using html legend templates for the mapserver for a very > long time. > With many layers such a template can become very long and confusing. > Is there a possibility to split such a template into several files and > e.g. include certain areas like in the mapfile? > Maybe I could solve this with PHP Mapscript? Up to now I have been > using processLegendTemplate to output the HTML code. > > Thanks > Sven > > > > _______________________________________________ > 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 atlantageek at gmail.com Thu May 7 11:31:34 2020 From: atlantageek at gmail.com (Atlanta Geek) Date: Thu, 7 May 2020 14:31:34 -0400 Subject: [mapserver-users] Data Stores for mapserver. Message-ID: How difficult is it to add a new datastore to mapserver. We currently use postgres but with a bigger need for deeper data analysis we are looking at other datastores for our report engine. Mapserver is one component of this engine. Does anyone have any experience using any other datastore with mapserver that is analytics specific? -- http://www.atlantageek.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jukka.rahkonen at maanmittauslaitos.fi Thu May 7 13:51:09 2020 From: jukka.rahkonen at maanmittauslaitos.fi (Rahkonen Jukka (MML)) Date: Thu, 7 May 2020 20:51:09 +0000 Subject: [mapserver-users] Data Stores for mapserver. Message-ID: <04dbe612c92d49e5a333a7a4a81e8412@C119S212VM042.msvyvi.vaha.local> Hi, Do you find anything suitable for you from this list https://gdal.org/drivers/vector/index.html? They are all available for you through OGR connection but it can well be that some formats do not work totally out-of-the-box. -Jukka Rahkonen- L?hett?j?: mapserver-users Puolesta Atlanta Geek L?hetetty: torstai 7. toukokuuta 2020 21.32 Vastaanottaja: mapserver-users at lists.osgeo.org Aihe: [mapserver-users] Data Stores for mapserver. How difficult is it to add a new datastore to mapserver. We currently use postgres but with a bigger need for deeper data analysis we are looking at other datastores for our report engine. Mapserver is one component of this engine. Does anyone have any experience using any other datastore with mapserver that is analytics specific? -- http://www.atlantageek.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From schroeter at netgis.de Thu May 7 23:06:33 2020 From: schroeter at netgis.de (Sven Schroeter) Date: Fri, 8 May 2020 08:06:33 +0200 Subject: [mapserver-users] HTML Legends with MapServer In-Reply-To: References: <99c089b2-e01c-7acf-c9b4-9a2c314ae363@netgis.de> Message-ID: the templates are usually not very complex, but rather long due to the large number of layers, here is a shortened example: https://mapcache.de/ex/legend_template.html Sven Am 07.05.2020 um 16:30 schrieb Steve Lime: > Query templates already support the concept of includes: > > ? [include src="some other template"] > > so I wonder if it might be possible to extend that functionality to > HTML legend processing. Do you have a particularly complex example you > could share with me? > > --Steve > > On Thu, May 7, 2020 at 4:42 AM Sven Schroeter > wrote: > > Hi all, > > i have been using html legend templates for the mapserver for a very > long time. > With many layers such a template can become very long and confusing. > Is there a possibility to split such a template into several files > and > e.g. include certain areas like in the mapfile? > Maybe I could solve this with PHP Mapscript?? Up to now I have been > using processLegendTemplate to output the HTML code. > > Thanks > Sven > > > > _______________________________________________ > 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 atlantageek at gmail.com Fri May 8 05:27:53 2020 From: atlantageek at gmail.com (Atlanta Geek) Date: Fri, 8 May 2020 08:27:53 -0400 Subject: [mapserver-users] Data Stores for mapserver. In-Reply-To: <04dbe612c92d49e5a333a7a4a81e8412@C119S212VM042.msvyvi.vaha.local> References: <04dbe612c92d49e5a333a7a4a81e8412@C119S212VM042.msvyvi.vaha.local> Message-ID: ElasticSearch is one of our options. Are these available in mapserver. Though we've been using mapserver for a year now its mostly been crafting the postgis queries and not much else On Thu, May 7, 2020 at 4:51 PM Rahkonen Jukka (MML) < jukka.rahkonen at maanmittauslaitos.fi> wrote: > Hi, > > > > Do you find anything suitable for you from this list > https://gdal.org/drivers/vector/index.html? > > They are all available for you through OGR connection but it can well be > that some formats do not work totally out-of-the-box. > > > > -Jukka Rahkonen- > > > > *L?hett?j?:* mapserver-users *Puolesta > *Atlanta Geek > *L?hetetty:* torstai 7. toukokuuta 2020 21.32 > *Vastaanottaja:* mapserver-users at lists.osgeo.org > *Aihe:* [mapserver-users] Data Stores for mapserver. > > > > How difficult is it to add a new datastore to mapserver. We currently use > postgres but with a bigger need for deeper data analysis we are looking at > other datastores for our report engine. Mapserver is one component of this > engine. Does anyone have any experience using any other datastore with > mapserver that is analytics specific? > > > > -- > > http://www.atlantageek.com > -- http://www.atlantageek.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From even.rouault at spatialys.com Fri May 8 05:32:09 2020 From: even.rouault at spatialys.com (Even Rouault) Date: Fri, 08 May 2020 14:32:09 +0200 Subject: [mapserver-users] Data Stores for mapserver. In-Reply-To: References: <04dbe612c92d49e5a333a7a4a81e8412@C119S212VM042.msvyvi.vaha.local> Message-ID: <2434726.qBjym3WITG@even-i700> On vendredi 8 mai 2020 08:27:53 CEST Atlanta Geek wrote: > ElasticSearch is one of our options. Are these available in mapserver. > Though we've been using mapserver for a year now its mostly been crafting > the postgis queries and not much else MapServer is definitely used with Elasticsearch through the OGR ElasticSearch driver: https://gdal.org/drivers/vector/elasticsearch.html I did fixes recently in it for better translation of SQL filters (coming from MapServer) to Elasticsearch query language. Even -- Spatialys - Geospatial professional services http://www.spatialys.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmckenna at gatewaygeomatics.com Fri May 8 16:20:03 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Fri, 8 May 2020 20:20:03 -0300 Subject: [mapserver-users] 7.6.0 released Message-ID: <3d135b5c-832d-3ab0-18f8-e01f94ee2068@gatewaygeomatics.com> The MapServer team is very pleased to announce the major release of MapServer 7.6.0 Details of the new features, migration instructions + examples, and download links (and available binaries for Debian, Windows etc.) can be found on the dedicated announcement page at https://mapserver.org/development/announce/7-6.html A sincere thank you to everyone testing the beta/rc releases, the developers providing fixes, and the documentation/packaging teams. To celebrate, please share a little video of thanks for the wonderful 20 years (of logs) of commits to the MapServer project (!!) : https://vimeo.com/416494241 Happy (fast) mapserving... -- The MapServer Team From jmckenna at gatewaygeomatics.com Sat May 9 06:50:01 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Sat, 9 May 2020 10:50:01 -0300 Subject: [mapserver-users] 7.6.0 released In-Reply-To: <3d135b5c-832d-3ab0-18f8-e01f94ee2068@gatewaygeomatics.com> References: <3d135b5c-832d-3ab0-18f8-e01f94ee2068@gatewaygeomatics.com> Message-ID: <03d5a150-bb50-f05c-28fc-8be82a623291@gatewaygeomatics.com> The entire 7.6.0 documentation is also available as a PDF (984 pages, wow!?) to download locally, with a nice clickable table of contents in left panel etc. https://download.osgeo.org/mapserver/docs/MapServer.pdf 20 years, ~1000 pages. Thanks to the entire MapServer community. -jeff -- Jeff McKenna MapServer Consulting and Training Services http://gatewaygeo.com/ On 2020-05-08 8:20 p.m., Jeff McKenna wrote: > The MapServer team is very pleased to announce the major release of > MapServer 7.6.0 > > Details of the new features, migration instructions + examples, and > download links (and available binaries for Debian, Windows etc.) can be > found on the dedicated announcement page at > https://mapserver.org/development/announce/7-6.html > > A sincere thank you to everyone testing the beta/rc releases, the > developers providing fixes, and the documentation/packaging teams. > > To celebrate, please share a little video of thanks for the wonderful 20 > years (of logs) of commits to the MapServer project (!!) : > https://vimeo.com/416494241 > > Happy (fast) mapserving... > From even.rouault at spatialys.com Sat May 9 11:33:12 2020 From: even.rouault at spatialys.com (Even Rouault) Date: Sat, 09 May 2020 20:33:12 +0200 Subject: [mapserver-users] Using Raster Attribute Table (RAT) with WCS In-Reply-To: References: Message-ID: <4253434.HPj8BnJhPe@even-i700> Lu??s, > I have a WCS serving a few thematic maps, where each cell represents a > class, like in a land use or ecological map. The maps include a Raster > Attribute Table (RAT) which essentially relates cell values to classes > (e.g. 1 - Grassland). > > What would be the correct way of providing the RAT to users of these > services? As far as I can tell, this is not supported in the WCS core. Is > it somehow considered in an extension? > > Besides WCS, would it be possible to publish the RAT through MapServer in > some other way? One possibility would be for the WCS code in MapServer to return a .zip with let's say a .tif file and a GDAL .aux.xml file with a GDAL RAT object serialized in it. The input RAT could be in a .aux.xml sidecar of the input file too. MapServer would have to set it from the source to the target Even -- Spatialys - Geospatial professional services http://www.spatialys.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From even.rouault at spatialys.com Sat May 9 11:40:59 2020 From: even.rouault at spatialys.com (Even Rouault) Date: Sat, 09 May 2020 20:40:59 +0200 Subject: [mapserver-users] WCS GetCoverage request one by one pixel In-Reply-To: References: Message-ID: <3542836.WVHXShWDJB@even-i700> Anton, > I quickly skimmed the issue tracker on Github but could not find any issue > causing this problem. Any thoughts on the cause of the problem are highly > appreciated. I believe this is a bug in MapServer due to a difference between pixel-center vs pixel-corner convention between WCS conventions and MapServer conventions, which must lead to an empty bounding box when the request is 1x1, and thus causing invalid resolution to be computed at some point (division by zero). This could presumably be fixed Even -- Spatialys - Geospatial professional services http://www.spatialys.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jukka.rahkonen at maanmittauslaitos.fi Mon May 11 04:02:50 2020 From: jukka.rahkonen at maanmittauslaitos.fi (Rahkonen Jukka (MML)) Date: Mon, 11 May 2020 11:02:50 +0000 Subject: [mapserver-users] WCS GetCoverage request one by one pixel Message-ID: Hi, I verified the error and I am pretty sure that subsetting with min=max has been working for me couple of years ago with some Mapserver version. This sends a GeoTIFF: http://localhost:8060/cgi-bin/mapserv.exe?map=c:\ms4w_data\raster_test.map&service=WCS&version=2.0.1&request=GetCoverage&CoverageID=natural_earth&SUBSET=long(27,28)&SUBSET=lat(62,63) This sends an error: http://localhost:8060/cgi-bin/mapserv.exe?map=c:\ms4w_data\raster_test.map&service=WCS&version=2.0.1&request=GetCoverage&CoverageID=natural_earth&SUBSET=long(27,27)&SUBSET=lat(62,62) msImageCreate(): Image handling error. Attempt to allocate raw image failed, out of memory. I believe that 1x1 pixel image is expected because WCS 2.0.1 standard defines that the request is OK. "Then, for the trim bounds trimLow and trimHigh the following shall hold: L <= trimLow <= trimHigh <= H" Here is my MS4W mapfile for the Natural Earth raster https://www.naturalearthdata.com/http//www.naturalearthdata.com/download/50m/raster/NE1_50M_SR_W.zip MAP NAME "" STATUS ON MAXSIZE 5000 CONFIG "MS_ERRORFILE" "c:\ms4w_data\test_error.txt" #DEBUG 3 EXTENT -180 -90 180 90 SIZE 600 600 #SHAPEPATH "c:\ms4w\tmp\ms_tmp" IMAGECOLOR 255 255 255 UNITS METERS WEB #IMAGEPATH "c:\ms4w\tmp\ms_tmp" #IMAGEURL "c:\ms4w\tmp\ms_tmp" METADATA "ows_enable_request" "*" "wms_srs" "EPSG:4326" "ows_onlineresource" "http://localhost:8060/cgi-bin/mapserv.exe?map=c:\ms4w_data\test.map" "wms_title" "wms-test" "wcs_label" "wcs-test" ### required "wcs_description" "wcs-test" "wcs_enable_request" "*" END END PROJECTION "init=epsg:4326" END OUTPUTFORMAT NAME 'AGG_Q' MIMETYPE "image/png" DRIVER "AGG/PNG" EXTENSION "png" IMAGEMODE RGB FORMATOPTION "QUANTIZE_FORCE=ON" FORMATOPTION "QUANTIZE_DITHER=OFF" FORMATOPTION "QUANTIZE_COLORS=256" END OUTPUTFORMAT NAME GEOTIFF DRIVER "GDAL/GTiff" MIMETYPE "image/tiff" IMAGEMODE BYTE EXTENSION "tif" END LAYER NAME "natural_earth" TYPE RASTER STATUS ON DATA "c:\ms4w_data\ne_raster\NE1_50M_SR_W.tif" PROJECTION "init=epsg:4326" END METADATA "wms_title" "ne_raster" "wcs_title" "ne_raster" END END #layer END #map L?hett?j?: mapserver-users Puolesta Even Rouault L?hetetty: lauantai 9. toukokuuta 2020 21.41 Vastaanottaja: mapserver-users at lists.osgeo.org Aihe: Re: [mapserver-users] WCS GetCoverage request one by one pixel Anton, > I quickly skimmed the issue tracker on Github but could not find any issue > causing this problem. Any thoughts on the cause of the problem are highly > appreciated. I believe this is a bug in MapServer due to a difference between pixel-center vs pixel-corner convention between WCS conventions and MapServer conventions, which must lead to an empty bounding box when the request is 1x1, and thus causing invalid resolution to be computed at some point (division by zero). This could presumably be fixed Even -- Spatialys - Geospatial professional services http://www.spatialys.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From fl_v at inbox.ru Mon May 11 07:06:09 2020 From: fl_v at inbox.ru (=?UTF-8?B?VmxhZGltaXI=?=) Date: Mon, 11 May 2020 17:06:09 +0300 Subject: [mapserver-users] =?utf-8?q?Getting_cellsize_with_Mapscript?= Message-ID: <1589205969.299974072@f413.i.mail.ru> Hi All! ? As I noticed, the 'cellsize' is calculated during 'OWSDispatch' performing. Is it possible to get it right after setting of request parameters? ? Code sample: ? ? ? map_obj = mapscript.mapObj('mapfile.map') ? ? request = mapscript.OWSRequest() ? ? for key, value in params: ??? ? ? request.setParameter(key, value)#setting BBOX, HEIGHT, WIDTH etc ? ? ? ? cellsize = map_obj.cellsize ??? ? ? ? ? ? ? ? mapscript.msIO_installStdoutToBuffer() ? ? map_obj.OWSDispatch(request) ? ? mapscript.msIO_getStdoutBufferBytes() ? -- Regards, Vladimir. ? -------------- next part -------------- An HTML attachment was scrubbed... URL: From bob.basques at ci.stpaul.mn.us Mon May 11 13:38:08 2020 From: bob.basques at ci.stpaul.mn.us (Basques, Bob (CI-StPaul)) Date: Mon, 11 May 2020 20:38:08 +0000 Subject: [mapserver-users] REMINDER: May 13, 2020 - OSGeo Local Chapter Virtual Meeting (aka TCMUG) Message-ID: LAST REMINDER!! Remember to send your email for the invite/password. All, So, the last meeting was interesting in that we went virtual with it for the first time. All in all it worked out very well. We even had a few out of towners show up. We presented a couple of short presentations that seemed to be of interest to the group based on the questions raised. Operator error (on my part) prevented me from recording anything from the last meeting, I forgot to hit the record button, so no pictures , not even a screenshot. I'll attempt to do better this time around. We can still add a second presentation if anyone is interested, just let me know, or raise the topic at the meeting I'm liking the capabilities that Jitsi allows, but I also don't want the group to get bombed while a meeting is in progress. Still learning how to best secure the meeting. Until I figure out the basics on this, can I have everyone interested in attending send me their email address in order to send out the Jitsi meeting invite directly, vs announcing on the lists, which seems to be a no-no? Looking into some sort of Web based distro of the invite that folks would need to login to. But maybe someone out there has a better idea?? The May meeting will feature a guest speaker, Michael Battaglia (mjbattag at mtu.edu) - Bio. A blurb from Michael about the presentation: "Coastal wetlands are a critical component of the Great Lakes ecosystem. Fluctuating lake levels, anthropogenic activity, and invasion by non-native vegetation have led to rapid changes to these valuable, yet fragile, wetland areas. Frequent monitoring is necessary to understand where and when changes are occurring, however, field visits are costly and time consuming. Techniques leveraging spaceborne synthetic aperture radar data enable timely remote monitoring of a variety of important variables, including wetland type and inundation extent. This talk will show methods used to generate map products for those parameters and share results generated with RADARSAT-2 SAR data for several pilot sites around the Great Lakes. Duration: approx.. 30 min. Meeting details: Where: Jitsi (send email for invite) When: 4:30 pm, May 13th, 2020. bobb -------------- next part -------------- An HTML attachment was scrubbed... URL: From paiu.georgiana at gmail.com Mon May 18 02:13:48 2020 From: paiu.georgiana at gmail.com (Georgiana Paiu) Date: Mon, 18 May 2020 11:13:48 +0200 Subject: [mapserver-users] List properties of mapserver objects Message-ID: I am using mapserver 7 with PHP and I was trying to list the properties and the values of some objects (example styleObj, labelObj) by using get_object_vars function from PHP. It seems the result is always an empty array. Can you help me with a suggestion, please? I mention that before I was using mapserver 5 and it was working fine. Thank you, *Georgiana Paiu* Web Developer (+40) 75 443 4340 <(+40)+75+443+4340> | paiu.georgiana at gmail.com | Skype: georgiana.paiu -------------- next part -------------- An HTML attachment was scrubbed... URL: From jukka.rahkonen at maanmittauslaitos.fi Mon May 18 06:16:32 2020 From: jukka.rahkonen at maanmittauslaitos.fi (Rahkonen Jukka (MML)) Date: Mon, 18 May 2020 13:16:32 +0000 Subject: [mapserver-users] KKJ images misaligned after update In-Reply-To: References: Message-ID: <08b7485843604ce0a5da1c319b2fd164@C119S212VM042.msvyvi.vaha.local> Hi, I got a report about a service that has been running fine for years but got broken now after following updates: sqlite 3.29 -> 3.31 proj 6.2.0 -> 6.3.0 geos 3.7.2 -> 3.8.0 gdal 2.4.2 -> 3.0.4 mapserver 7.4.2 -> 7.4.3 What breaks is an orthophoto layer that has source files in many CRSs. The source CRS may be EPSG:2391, 2392, 2393, 2395, and 3067, and the output is EPSG:3067. What happens is that the images in 239x series using the Finnish KKJ datum are not reprojected correctly. It seems that the affine transformation parameters (TOWGS84[-96.062,-82.428,-121.753,4.801,0.345,-1.376,1.496]) of the Finnish KKJ system http://epsg.io/2393 are not used. I made a small test-zip into http://latuviitta.org/downloads/kkjtest.zip. It works fine for me with MS4W (Mapserver 7.6.0dev, GDAL 2.4.4, Proj 5.2.0) but with the updated system the middle tile takes quite an offset (kkj_wrong.jpeg). The middle image is in EPSG:2393, the outmost two are in EPSG:3067, and the WMS GetMap was made with &CRS=EPSG:3067. The error is specific for the KKJ->ETRS-TM35FIN (EPSG:3067) transformation because if user reads the layer from WMS in EPSG:3857 then the middle tile stays where it should. Unfortunately I do not have myself Mapserver with GDAL 3 and Proj 6 available for re-producing the error. If somebody wants to have a try it should be relatively easy. Edit the kkjtest.map mapfile and the "location" field of the kkjtest.shp to match with your file paths. -Jukka Rahkonen- -------------- next part -------------- A non-text attachment was scrubbed... Name: kkj_wrong.jpg Type: image/jpeg Size: 23927 bytes Desc: kkj_wrong.jpg URL: From even.rouault at spatialys.com Mon May 18 06:44:21 2020 From: even.rouault at spatialys.com (Even Rouault) Date: Mon, 18 May 2020 15:44:21 +0200 Subject: [mapserver-users] KKJ images misaligned after update In-Reply-To: <08b7485843604ce0a5da1c319b2fd164@C119S212VM042.msvyvi.vaha.local> References: <08b7485843604ce0a5da1c319b2fd164@C119S212VM042.msvyvi.vaha.local> Message-ID: <1750948.fl4FrXcyKO@even-i700> On lundi 18 mai 2020 13:16:32 CEST Rahkonen Jukka (MML) wrote: > Hi, > > I got a report about a service that has been running fine for years but got > broken now after following updates: > > sqlite 3.29 -> 3.31 > proj 6.2.0 -> 6.3.0 > geos 3.7.2 -> 3.8.0 > gdal 2.4.2 -> 3.0.4 > mapserver 7.4.2 -> 7.4.3 I didn't try, but I'd strongly suspect the combination MapServer 7.4 + GDAL 3 doesn't work so well here. The issue you hit should be hopefully fixed with the following added code path in MapServer 7.6: https://github.com/mapserver/mapserver/blob/ 13607c79d8254326c83b1b8b2f81d2560169106c/mapogr.cpp#L959 Basically on reading the SRS from the TIFF file with GDAL, we used to export the WKT definition to a PROJ.4 string, and then built the Mapserver CRS object from it. Before GDAL 3.X, this export to PROJ.4 string included TOWGS84 parameters. Since GDAL 3.X, it no longer does. Hence the new code path, that detects that the CRS read from GDAL has an EPSG code attached, and when that's the case, instanciate a +init=epsg:XXXX string as the Mapserver CRS object to avoid going to potentially lossy PROJ.4 strings. So "good" version combinations are probably: - Mapserver 7.4 (or 7.6) + GDAL 2.4 + PROJ 5 - or MapServer 7.6 + GDAL 3.X + PROJ >= 6 Even -- Spatialys - Geospatial professional services http://www.spatialys.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmckenna at gatewaygeomatics.com Tue May 19 06:36:12 2020 From: jmckenna at gatewaygeomatics.com (Jeff McKenna) Date: Tue, 19 May 2020 10:36:12 -0300 Subject: [mapserver-users] KKJ images misaligned after update In-Reply-To: <1750948.fl4FrXcyKO@even-i700> References: <08b7485843604ce0a5da1c319b2fd164@C119S212VM042.msvyvi.vaha.local> <1750948.fl4FrXcyKO@even-i700> Message-ID: On 2020-05-18 10:44 a.m., Even Rouault wrote: > > So "good" version combinations are probably: > > - Mapserver 7.4 (or 7.6) + GDAL 2.4 + PROJ 5 > > - or MapServer 7.6 + GDAL 3.X + PROJ >= 6 > I have added this note to the 7.6 Migration Guide: https://mapserver.org/MIGRATION_GUIDE.html#migration-7-6 -jeff -- Jeff McKenna MapServer Consulting and Training Services http://gatewaygeo.com/ From sdlime at gmail.com Tue May 19 15:55:13 2020 From: sdlime at gmail.com (Steve Lime) Date: Tue, 19 May 2020 17:55:13 -0500 Subject: [mapserver-users] List properties of mapserver objects In-Reply-To: References: Message-ID: Are you using PHP/MapScript or the Swig-based version? On Mon, May 18, 2020 at 4:14 AM Georgiana Paiu wrote: > I am using mapserver 7 with PHP and I was trying to list the properties > and the values of some objects (example styleObj, labelObj) by > using get_object_vars function from PHP. It seems the result is always an > empty array. Can you help me with a suggestion, please? > > I mention that before I was using mapserver 5 and it was working fine. > > Thank you, > > *Georgiana Paiu* > Web Developer > (+40) 75 443 4340 <(+40)+75+443+4340> | paiu.georgiana at gmail.com | Skype: > georgiana.paiu > > _______________________________________________ > 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 paiu.georgiana at gmail.com Wed May 20 00:10:16 2020 From: paiu.georgiana at gmail.com (Georgiana Paiu) Date: Wed, 20 May 2020 09:10:16 +0200 Subject: [mapserver-users] List properties of mapserver objects In-Reply-To: <9EC3CD09-3CFB-4E13-92AA-7690B67C7DC5@gmail.com> References: <9EC3CD09-3CFB-4E13-92AA-7690B67C7DC5@gmail.com> Message-ID: I am using PHP/Mapscript *Georgiana Paiu* Web Developer (+40) 75 443 4340 <(+40)+75+443+4340> | paiu.georgiana at gmail.com | Skype: georgiana.paiu On Wed, 20 May 2020 at 08:30, Georgiana Paiu wrote: > I am using PHP/Mapscript > > > Sent from my iPhone > > On 20 May 2020, at 00:55, Steve Lime wrote: > > ? > Are you using PHP/MapScript or the Swig-based version? > > On Mon, May 18, 2020 at 4:14 AM Georgiana Paiu > wrote: > >> I am using mapserver 7 with PHP and I was trying to list the properties >> and the values of some objects (example styleObj, labelObj) by >> using get_object_vars function from PHP. It seems the result is always an >> empty array. Can you help me with a suggestion, please? >> >> I mention that before I was using mapserver 5 and it was working fine. >> >> Thank you, >> >> *Georgiana Paiu* >> Web Developer >> (+40) 75 443 4340 <(+40)+75+443+4340> | paiu.georgiana at gmail.com | >> Skype: georgiana.paiu >> >> _______________________________________________ >> 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 beu at live.jp Wed May 20 03:32:34 2020 From: beu at live.jp (beu) Date: Wed, 20 May 2020 03:32:34 -0700 (MST) Subject: [mapserver-users] Java MapScript on Tomcat servlet does not free GeoTIFF file handles Message-ID: <1589970754290-0.post@n6.nabble.com> Dear all, I'm coding Java MapScript program as servlet webapp of Tomcat (on Windows), using "javamapscript.dll" and "mapscript.jar", to process "mapObj" object dynamically. But this program has some problems on file-handles. Could someone please tell me how to fix them? This program accepts WMS request (GetMap) and returns "image/png" data. rough flow of this program: get HTTP request body create OWSRequest instance create mapObj instance with some MapFile call msIO_installStdoutToBuffer() call mapObj#OWSDispatch() call mapObj#draw() and create imageObj instance call imageObj#save() and create temporary image file set image file data to HTTP response body delete the image file call imageObj#delete() call mapObj#delete() call OWSRequest#delete() call msIO_resetHandlers() Problem 1: Tomcat-process does not free GeoTIFF file-handles per (pooling) thread. With "Resource Monitor" of Windows, it is confirmed that Tomcat8.exe is holding file-handles of "foobar.tif". MapFile: MAP ... LAYER TYPE RASTER STATUS ON DATA "foobar.tif" END END Problem 2: Tomcat-process does not free *.shp, *.shx file-handles sometimes, when it is running multi-thread'ly (more than about 150 threads). Problem 3: When the total number of file-handles held reaches some value, Tomcat-process can not open more files, and MapScript always raise UnknownError. With best regards, MapServer (MapScript) 7.4.3 (on Windows) ... release-1911-x64-gdal-3-0-0-mapserver-7-4-3 at GISInternals Tomcat 8.5.51 (on Windows) JavaSE 8u251 (on Windows) Windows Server 2016 Beu -- Sent from: http://osgeo-org.1560.x6.nabble.com/Mapserver-User-f4226646.html From beu at live.jp Fri May 22 05:01:53 2020 From: beu at live.jp (beu) Date: Fri, 22 May 2020 05:01:53 -0700 (MST) Subject: [mapserver-users] Java MapScript on Tomcat servlet does not free GeoTIFF file handles In-Reply-To: <1589970754290-0.post@n6.nabble.com> References: <1589970754290-0.post@n6.nabble.com> Message-ID: <1590148913176-0.post@n6.nabble.com> Dear, That "rough flow" is running, but that is dirty and illegal. I modified that: get HTTP request body create OWSRequest instance create mapObj instance with some MapFile call msIO_installStdoutToBuffer() call mapObj#OWSDispatch() call msIO_stripStdoutBufferContentType() call msIO_getStdoutBufferBytes() and get binary data call msIO_resetHandlers() call HttpServletResponse#getOutputStream() call ServletOutputStream#write() and write binary data call ServletoutputStream#close() call mapObj#delete() call OWSRequest#delete() But, the problems are intact. I think that mapObj#OWSDispatch() does not free file handles ... what do you think? Could someone please tell me how to fix them? Regards, Beu -- Sent from: http://osgeo-org.1560.x6.nabble.com/Mapserver-User-f4226646.html From even.rouault at spatialys.com Fri May 22 06:27:17 2020 From: even.rouault at spatialys.com (Even Rouault) Date: Fri, 22 May 2020 15:27:17 +0200 Subject: [mapserver-users] Java MapScript on Tomcat servlet does not free GeoTIFF file handles In-Reply-To: <1589970754290-0.post@n6.nabble.com> References: <1589970754290-0.post@n6.nabble.com> Message-ID: <2833055.MPbBA8MxH1@even-i700> > Problem 1: > Tomcat-process does not free GeoTIFF file-handles per (pooling) thread. > With "Resource Monitor" of Windows, it is confirmed that Tomcat8.exe is > holding file-handles of "foobar.tif". > > MapFile: > MAP > ... > LAYER > TYPE RASTER > STATUS ON > DATA "foobar.tif" > END > END > You probably need to add PROCESSING "CLOSE_CONNECTION=ALWAYS" in your LAYER definition By default, datasets of raster layers are kept open until MapServer is terminated. CLOSE_CONNECTION is described in https://mapserver.org/mapfile/layer.html, but as far as I can see, not explicit about this particular behaviour of raster layers where the default is DEFER. The "documentation" is at: https://github.com/mapserver/mapserver/blob/ c4006da79bc59e30478d9e2354070ffb9dcbc614/mapraster.c#L675 Even -- Spatialys - Geospatial professional services http://www.spatialys.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From beu at live.jp Mon May 25 18:43:27 2020 From: beu at live.jp (beu) Date: Mon, 25 May 2020 18:43:27 -0700 (MST) Subject: [mapserver-users] Java MapScript on Tomcat servlet does not free GeoTIFF file handles In-Reply-To: <2833055.MPbBA8MxH1@even-i700> References: <1589970754290-0.post@n6.nabble.com> <2833055.MPbBA8MxH1@even-i700> Message-ID: <1590457407129-0.post@n6.nabble.com> Hi Even, Thanks for your response. Oh, I see. it's right! I append the PROCESSING statement to raster LAYER, and then any raster file handles were not held. But maybe performance will be decreased. Is there an API likes msCleanup for a thread? Regards, Beu -- Sent from: http://osgeo-org.1560.x6.nabble.com/Mapserver-User-f4226646.html