#QGIS IRC Log - 2009-06-08

For logs after Feb 3, 2007, all times are GMT-8. Prior logs are GMT-9.
Back to Logs
04:07:07 : * lasma is away: Gone away for now
06:55:33 FloH: hi, any windows user there who can help me with my compiling problems?
06:59:36 jef: FloH: on mingw?
07:00:29 FloH: yes, msys and mingw
07:03:11 jctull: has anyone else tested the "Fetch Python plugins..." menu item in trunk? It is non-responsive on my OS X build, and provides no debug feedback either.
07:34:52 jctull: strangely, the menu item for the python plugin is not working, but the installer worked with the plugin notification link on the main window. Fresh build, rebuilt sip and pyqt beforehand.
11:56:14 CGI593: hello all. when I install qgis1.1 in osgeo4w, do I also have to install qgis-grass-plugin too?
11:57:06 CGI593: I notice that qgis-grass-plugin is not selected when I select qgis1.1
11:58:13 CGI593: brb
11:59:03 FrankW: CGI593: I do believe you need to select it if you want grass support.
12:02:23 jef: CGI593: no, 1.1 has python+grass support builtin. same for qgis-dev
12:02:50 FrankW: I stand corrected!
12:07:52 CGI593: jef: thanks for the info. I also notice that gdal-grass, psycopg2, pyqwt5-qt4, python-qgis, and python-rpy2 are not selected too.
12:10:25 CGI593: I would imagine that I'll have to select them
12:11:22 jef: CGI593: python-qgis is builtin too. psycopg2, pyqwt5-qt4 and python-rpy2 are not used by qgis ( just some optional plugins ). but gdal-grass is a valid point, I think.
12:15:58 CGI593: jef: I'm trying to gather up info for our IT dept. I want the make sure I give them simple instructions on how to install the osgeo4w stack. that's why I'm asking these questions
12:18:11 wildintellect: sampson, we should probably chat at some point later this week - you're bringing up a lot of ideas that I've already started on
12:22:55 jef: CGI593: fixed.
12:25:35 CGI593: jef: I see that gdal-grass is included now...thanks!
12:33:41 timlinux: hi all
12:38:38 epifanio: hi timlinux ;- )
12:38:59 epifanio: jctull: .. a question, have you PyQwt working ?
12:39:22 jctull: not sure. I have the source. What binaries does it install?
12:39:56 epifanio: it create a direcxtory qwt inside the pyqt folder in site packages
12:41:30 epifanio: on osx i means, here with qt4.5 , pyqt4.5 , sip4.8 .. it fails to build :-( qwt5.1 install fine and works too ( i tested the example file ) while pyqwt fails with an error about qwt-slider
12:41:37 jctull: I do not have it installed. I think I tried briefly some time ago and gave up
12:41:52 jctull: sorry
12:42:34 epifanio: ok .. np :- ) i reinstalled osx on my previouse system i had it working .. so maybi is a "version ( pyqt-sip )" problem
12:43:10 epifanio: i mailed pyqwt list too .. so i hope it will be fixed soon
12:43:30 epifanio: thanks to check :- )
12:50:32 timlinux: springmeyer, I see on twitter your quantumnik was released....
12:51:14 springmeyer: yeppers
12:51:47 springmeyer: tires have been kicked enough now
12:52:31 springmeyer: timlinux: hopefully we can get martin set up with it, so he can flip between qgs and mapnik as he perfects labels
12:53:03 springmeyer: and we'll know when the plugin is obsolete : )
12:54:24 springmeyer: timlinux: btw has anyone ever tried to bundle a dynamic library with a python plugin?
12:54:32 timlinux: springmeyer, cool.....need to try to install it here
12:54:55 timlinux: have 40mb left on my internet cap so will need to wait post top up
12:55:15 springmeyer: : )
12:55:38 timlinux: springmeyer, I havent tried much at all with python to be honest - the extent of my python knowledge is what I hack about with in django
12:55:49 timlinux: ( still lovin django btw )
12:55:52 ADKPete: springmeyer: you may be opening a whole can of worms trying to do that
12:56:01 timlinux: hi ADKPete
12:56:02 springmeyer: I've been getting mapnik built using static linking lately and I bet I could eventually just package the mapnik.so that are the c++ python bindings to mapnik alongside the plugin
12:56:06 timlinux: long time no see
12:56:08 ADKPete: howdy timlinux
12:56:10 ADKPete: aye
12:56:15 springmeyer: ADKPete: I am sure I am : )
12:56:31 springmeyer: and a can o whoopass against myself too probably : )
12:57:08 springmeyer: django can be frustrating like anything but hard not to love
12:57:08 ADKPete: springmeyer: of course it depends on how many dependencies it has
12:57:32 springmeyer: well, 'none' if its statically linked, eh?
12:57:54 ADKPete: then you only have to deal with platform differences
12:58:11 springmeyer: I can hear the headlines now... first python plugin to hit 80MB download ; )
12:58:46 springmeyer: yes, I would have no idea how to do that on windows, but mac would be very feasible and I figure linux could be no harder
12:59:15 springmeyer: but, honestly I'm unlikely to even try it : ) just curious if anyone had considered it
12:59:36 : * timlinux blows off some of his last remaining internet cap looking at wwdc liveblog: http://www.engadget.com/2009/06/08/phil-schiller-keynote-live-from-wwdc-2009/
12:59:37 sigq: Title: Engadget ( at www.engadget.com )
13:43:43 epifanio: a problem compiling last trunk : http://paste.debian.net/38437/ now i'm doing mak distclean in grass and try to rebuild it .. but i'm not sgure where the problem is
13:43:44 sigq: Title: #debian.de Pastezone ( at paste.debian.net )
13:55:31 jef: epifanio: does qgsgeometry.cpp compile? also depends on geos_c.h
13:58:25 epifanio: how to check it ? .. i'm looking inside the generated files
13:59:35 epifanio: this the cmake log tring to re-run make after the failure : http://paste.debian.net/38438/
13:59:36 sigq: Title: #debian.de Pastezone ( at paste.debian.net )
13:59:55 epifanio: at the top i can see that it find geos
14:00:33 epifanio: i dome make clean & make distclean in grass and recompiled it without errors, now i'll try again with qgis
14:07:28 epifanio: jef: qgsgeometry compile
14:08:47 epifanio: i can see in the cmake log : [ 3%] Building CXX object src/core/CMakeFiles/qgis_core.dir/qgsgeometry.o
14:42:49 CGI539: hello all. I just installed qgis from osgeo4w. I only see an OSGeo4W icon on the screen. Starting the shell and typing qgis, I get the error message "The system cannot execute the specified program."
14:46:49 markusN: hi - I get sip: /home/neteler/software/qgis_1_1_0/python/core/qgssymbol.sip:75: syntax error
14:47:03 markusN: using qgis_1_1_0 SVN
14:47:55 markusN: in that line is 'virtual QImage getPointSymbolAsImage( ...
14:48:58 markusN: http://pastebin.com/d27546e8d
14:48:59 sigq: Title: C++ pastebin - collaborative debugging tool ( at pastebin.com )
14:55:38 markusN: HA
14:55:41 markusN: found it
14:55:45 markusN: 1. -> 1.0
14:55:50 markusN: can't believe it
14:58:56 sigq: tracfeed: Ticket #1723 ( bug created ): qgis_1_1_0/python/core/qgssymbol.sip:75: syntax error + fix <https://trac.osgeo.org/qgis/ticket/1723>
14:59:05 markusN: voilĂ 
14:59:13 markusN: patch attached
14:59:20 markusN: notte
16:06:03 springmeyer: epifanio: ping
16:06:16 epifanio: hi springmeyer
16:06:47 springmeyer: hey there. I keep on forgetting - did you get mapnik to build against the cairo framework from gtk-osx?
16:07:01 springmeyer: or just cairo/cairomm from macports?
16:07:13 epifanio: yes it compile with framwork
16:07:26 springmeyer: ah, interesting.
16:07:38 springmeyer: and did you test the pdf/svg output at all?
16:09:21 epifanio: but the same framework is not good to compile grass with cairo support so i'm switching to this note : http://www.kyngchaos.com/macosx:build:cairo to build cairo from source
16:09:22 sigq: Title: Cairo Build [KyngChaos Wiki] ( at www.kyngchaos.com )
16:10:29 springmeyer: okay. sure
16:10:33 epifanio: springmeyer: no i've no tested it
16:10:36 springmeyer: so did you test ?
16:10:42 springmeyer: ah, okay
16:11:03 springmeyer: ya, the gtk framework does not provide libcairomm-1.0.dylib so I doubt it will work with mapnik
16:11:20 springmeyer: cairomm is the C++ bindings to the C cairo lib
16:11:32 springmeyer: maybe that is the grass problem too? is grass C++?
16:11:47 epifanio: i have formatted my mac ... some ( too many ) problems with duplicate grass version ... so now i'm starting from scratch to install all the GIS sw
16:11:51 epifanio: grass is C
16:12:02 springmeyer: hmm, okay
16:12:44 epifanio: so mapnic is on my road ... and having a fresh installation i'll finally try to annote the installation steps in a clean way
16:13:53 epifanio: the william notes works fine and the build is fast
16:14:44 epifanio: i hope it will work with pycairo too .. the bad is that pycaro ( >1.8 ) needs python 2.6 .. and osx come with 2.5
16:15:21 springmeyer: I know, that is a major annoyance. I've been thinking hacking that setyp.py
16:15:30 epifanio: i don't know if pycairo <= 1.8 can be a limitation for mapnick
16:15:47 springmeyer: nope, works fine if you can build it
16:16:08 springmeyer: but mapnik also works fine with py26
16:16:19 springmeyer: are you running QGIS against py26 at all yet?
16:16:29 epifanio: great, on my previouse hd i had quantumnick working :- ) nice!!!
16:16:51 epifanio: no for pytghon i'm using the system 2.5
16:16:57 springmeyer: ah, okay
16:26:07 springmeyer: facinating epifanio: cairo is hanging around everywhere : )
16:26:11 springmeyer: '/Developer/SDKs/MacOSX10.5.sdk/usr/X11/lib/libcairo.dylib'
16:28:26 epifanio: :- ) it is nice!
16:29:36 springmeyer: just no cairomm to be found ;(
16:33:12 epifanio: we will install it too .. maybe the William's note about cairo can help
16:42:22 springmeyer: yes, cairomm is not bad at all, really pixman and friends are the tricky ones
22:53:32 CIA-65: jef * r10893 /trunk/qgis/ ( 4 files in 4 dirs ): fix build problems with GRASS 6.5 GEOS support and #1723
22:54:37 sigq: tracfeed: Changeset [10893]: fix build problems with GRASS 6.5 GEOS support and #1723 <https://trac.osgeo.org/qgis/changeset/10893>
22:58:04 sigq: tracfeed: Ticket #1723 ( bug closed ): qgis_1_1_0/python/core/qgssymbol.sip:75: syntax error + fix <https://trac.osgeo.org/qgis/ticket/1723#comment:1>