#QGIS IRC Log - 2008-09-12

For logs after Feb 3, 2007, all times are GMT-8. Prior logs are GMT-9.
Back to Logs
02:53:39 spacedman: hmmm i think qgis.org needs some new screenshots - current ones are 0.7.4!
02:53:50 : * spacedman gets busy with scrot
03:03:16 ndlovu: does anyone know of a linux app / webpage that will convert shapefiles to kml ( attributes not necessary )?
03:05:40 jef: hm, did you try the ogr converter plugin?
03:06:03 ndlovu: jef, nope, what does that do?
03:06:42 jef: convert a ogr datasource to an other format. like ogr2ogr just with a gui.
03:07:37 ndlovu: do you know offhand what repository it's in?
03:07:44 jef: trunk
03:12:51 ndlovu: jef, so I would use svn to install from trunk?
03:15:48 mapmagic: wildintellect: I installed that python plugin. Now how do i import the osm file ? Thanks.
09:01:43 CIA-9: jef * r9305 /trunk/qgis/i18n/qgis_de.ts: update german translation
09:02:03 mapmagic: how do i open an osm map in qgis after installing the osm plugin?
09:03:14 sigq: tracfeed: Changeset [9305]: update german translation <https://trac.osgeo.org/qgis/changeset/9305>
09:12:36 ADKPete: mapmagic: Sorry can't help too much never use the OSM plugin, but there have been a lot of changes recently to the core qgis API
09:12:58 ADKPete: mapmagic: it is possible it is not compatible with the current version you are using
09:13:39 ADKPete: you building your own binaries using the trunk?
10:31:22 ChrisGo: Does Quantum GIS serve GIS or output to web at all?
10:33:43 ChrisGo: Seems like a nice alternative to Map3D or ArcMap.
10:35:40 ADKPete: ChrisGo: http://karlinapp.ethz.ch/qgis_wms/index.html
10:35:41 sigq: Title: Welcome to QGIS mapserver homepage ( at karlinapp.ethz.ch )
10:39:43 CIA-9: ersts * r9306 /trunk/qgis/ ( 5 files in 4 dirs ):
10:39:43 CIA-9: -Patch for building pyramids
10:39:43 CIA-9: -Closes ticket #1264
10:40:21 wildintellect: ChrisGo, not exactly
10:40:39 wildintellect: QGIS is a desktop application as is ArcMap
10:40:40 sigq: tracfeed: Ticket #1264 ( bug closed ): pyramids marked as built while they are not <https://trac.osgeo.org/qgis/ticket/1264#comment:1> || Changeset [9306]: -Patch for building pyramids -Closes ticket #1264 <https://trac.osgeo.org/qgis/changeset/9306>
10:41:13 wildintellect: What you want is something like Mapserver or Geoserver which is more like ArcIMS or ArcGIS Server
10:41:36 wildintellect: QGIS does have a tool to help build a mapserver configuration file
10:43:56 wildintellect: ADKPete, have you ever tried that app?
10:51:19 ADKPete: wildintellect: I did try one one line example but not sure if it was that exact one....is was interesting
11:00:53 wildintellect: I agree, just not sure it still works, it's a year old, the QGIS api has changed a lot since then
11:04:01 ChrisGo: So QGIS will allow creation of SHP files for example or is it more of a ArcMap Viewer type application?
11:04:32 ChrisGo: Also the QGIS NSIS installer is awesome where can I get a copy of the script : )
11:06:50 wildintellect: QGIS is a fully functional GIS application
11:09:39 wildintellect: ChrisGo, you'll probably need to send an email to the windows packagers, I'm not sure it's in the main code but you can go look trac.osgeo.org/qgis
11:10:07 ChrisGo: wildintellect Thank you!
11:35:53 timlinux: hi all
11:35:58 timlinux: jef: hi
11:36:06 timlinux: ADKPete: hi
11:36:11 jef: timlinux: hi
11:36:15 ADKPete: timlinux: howdy
11:36:25 timlinux: ADKPete: I see you are getting into the swing of things with unit testing :- )
11:36:42 ADKPete: I am? : )
11:37:00 timlinux: ADKPete: I see the overlay function I wrote for raster is on by default now - at least on my work pc anyway
11:37:01 ADKPete: I compile them just don't always run them ; )
11:37:30 ADKPete: Yes, I was wonder where that came from...
11:37:32 jef: compiling counts, too?
11:37:49 timlinux: jef: you get one karma point every time you compile qgis :- )
11:38:05 ADKPete: wow, I should be somewhere else by now.....
11:38:18 timlinux: ADKPete: I removed the checkbox for overlay from the raster gui when I tidied it up a bit the other day
11:38:36 timlinux: and raster class has show debug overlay set to false
11:38:37 ADKPete: ah yes, I did see that but did not make the connection
11:38:44 timlinux: by default
11:39:01 timlinux: I was trying to think of a better way to make it available
11:39:08 timlinux: dunno if you ever used the overlay
11:39:15 ADKPete: nope, actually
11:39:21 timlinux: we could scrap it if you dont see any need for it
11:39:34 ADKPete: with all the debug messages, I think it is taken care of
11:39:41 timlinux: it was handy when I first started with raster stuff
11:39:47 timlinux: ok lets ditch it then
11:40:26 timlinux: its probably not something we want on the raster public api anyway
11:40:28 jef: speaking of which
11:40:30 CIA-9: jef * r9307 /trunk/qgis/src/app/qgsdbsourceselect.cpp: remove spurious connect
11:40:40 ADKPete: but there is still a big nasty in the draw( ) :( when project on the fly is on...
11:41:03 sigq: tracfeed: Changeset [9307]: remove spurious connect <https://trac.osgeo.org/qgis/changeset/9307>
11:41:06 ADKPete: I think I will try to attack that this weekend
11:42:01 : * timlinux is ditching overlay stuff quick
11:42:35 timlinux: ADKPete: I saw you discussing it with FrankW the other night
11:42:39 ADKPete: aye
11:42:40 timlinux: didnt get all the details
11:42:55 timlinux: basically the aoi is being projected on the fly when it shouldnt be?
11:43:12 ADKPete: I have to track back out of the raster layer a bit to understand a couple of things
11:43:43 ADKPete: now it is doing what it is supposed to be, just that rasters can't deal with the request correctly
11:44:15 ADKPete: I am just not understanding the units of the extent
11:44:18 timlinux: right
11:44:33 timlinux: we need to reproject rasters on the fly to sort it out :- )
11:45:52 ADKPete: all of the extends come back in the original coordinate system system ( including the viewport ) but MapToPixel or what every is giving the reprojected info
11:46:41 ADKPete: then you end up with requests that are trying to create an 8mill x 8 mill pixel image
11:46:50 ADKPete: anyway....
11:46:51 timlinux: ouch
11:47:15 : * timlinux hasnt really poked int he maptopoxel code too much
11:47:50 ADKPete: The easy thing is to just short circuit the draw routine for when the projections don't match, just have to figure out all of that logic
11:48:35 ADKPete: What was why I was looking at QgsRect the other day....
11:48:54 ADKPete: seems like the intersection between the view port and an unprojected raster should not overlap
11:49:06 ADKPete: if the viewport is a different projection
11:49:21 ADKPete: well their extents anyway
11:52:48 ADKPete: timlinux: speak of unit tests, did you update the landsat_875 expected image recently?
11:54:54 ChrisGo: wildintellect or anyone how do I get a hold of the windows packagers so I can get the nsis for review?
11:56:40 CIA-9: marcopx * r9308 /trunk/qgis/AUTHORS: Replaced spaces with tabs in last two lines
11:56:59 sigq: tracfeed: Changeset [9308]: Replaced spaces with tabs in last two lines <https://trac.osgeo.org/qgis/changeset/9308>
11:58:31 ADKPete: ChrisGo: Best way would be to send a message to the dev listserv
11:58:51 ChrisGo: thanks on it
12:00:04 timlinux: ADKPete: I dont think I did ( update 875 )
12:00:50 timlinux: ChrisGo: cc Marco Pasetti
12:00:57 timlinux: hes the win guy
12:01:38 timlinux: ADKPete: I think in some cases similar but different crs'es could have intersecting extents still e.g. utm33s and 34s
12:01:38 ChrisGo: Was going to send to qgis-release-team@lists.osgeo.org not sure what Marco Pasetti email is.
12:01:54 timlinux: ChrisGo: qgis developer mailing list is better
12:02:03 ADKPete: yeah, but not utm and LongLat
12:02:27 timlinux: ADKPete: yup
12:03:07 timlinux: ChrisGo: jef also occassionally whips up msvc built win binaries
12:03:07 ADKPete: but currently it say they intersect and I need to figure out why
12:03:44 ADKPete: but I partially know why, all the extends, layer, viewport, etc, are all in the same coor system...but they should not be
12:03:51 ChrisGo: It's more about the NSIS than the files included in the installer. It's real sweet.
12:04:06 timlinux: ADKPete: well I guess if you have projected coords like -767676876876 76767676 312312 2312312 and then a lat long box its prolly got a pretty good chance of intersecting
12:04:14 timlinux: depends where you are I guess
12:04:45 timlinux: ChrisGo: I did encourage him to put it all into svn - not sure if he has done so yet
12:05:14 timlinux: ChrisGo: I do know he said he handcarved it all rather than using one of the traditional templates
12:05:21 ChrisGo: I checked and did not see but I could have missed.
12:05:35 ChrisGo: timlinux rub it in more : )
12:05:45 timlinux: if its there it will be under win_build/
12:05:46 ChrisGo: I sent to the email groups so ill wait thanks for the direction
12:06:18 timlinux: ChrisGo: cool - Ill drop some more hints to him about checking it into svn too
12:10:18 : * ADKPete has to run for a bit
12:10:27 ADKPete: later all
12:13:38 nhv: is QGis using bigtiff ?
12:19:50 timlinux: nhv: qgis is using gdal
12:20:14 timlinux: dont we get bigtiff automagically via gdal?
12:21:03 nhv: hmm not sure probably depends on the version
12:21:59 nhv: I was trying to get more inf to feed back to
12:22:01 FrankW: Right, it depends on whether you use the system libtiff, the internal gdal libtiff or a custom build of libtiff4.
12:28:57 CIA-9: timlinux * r9309 /trunk/qgis/src/ ( 3 files in 2 dirs ): Remove debug overlay from api and code base
12:29:39 sigq: tracfeed: Changeset [9309]: Remove debug overlay from api and code base <https://trac.osgeo.org/qgis/changeset/9309>
14:18:57 CIA-9: timlinux * r9310 /trunk/qgis/python/plugins/plugin_installer/ ( 14 files ): Updates for Borys and added some missing files
14:19:54 sigq: tracfeed: Changeset [9310]: Updates for Borys and added some missing files <https://trac.osgeo.org/qgis/changeset/9310>
15:12:37 sigq: tracfeed: Ticket #1312 ( bug created ): r9310 fails to build on Debian testing amd64, gcc 4.3.1 <https://trac.osgeo.org/qgis/ticket/1312>
15:47:57 CIA-9: jef * r9311 /trunk/qgis/python/core/ ( qgsrasterlayer.sip qgsrasterpyramid.sip ): fix #1312
15:49:03 sigq: tracfeed: Ticket #1312 ( bug closed ): r9310 fails to build on Debian testing amd64, gcc 4.3.1 <https://trac.osgeo.org/qgis/ticket/1312#comment:1> || Changeset [9311]: fix #1312 <https://trac.osgeo.org/qgis/changeset/9311>