| 05:26:40 | rduif: | building qgis on Jaunty here. Compiling OK but crash when running and trying to add classes in symbology |
| 05:27:50 | rduif: | building with my own gdal 1.5.4 after somebody told me that that in the past these crashes had something to do with gdal... |
| 05:28:13 | rduif: | did somebody build qgis succesfully on Jaunty? |
| 05:28:34 | rduif: | or other options? |
| 06:04:58 | rduif: | ok, now with ( homebrew ) gdal 1.6.0 no crashes with trunk on Jaunty |
| 06:06:41 | dassouki: | rduif: i didn't have troules installing on jaunty; however, when i did apt-get autoremove, it removed some libraries |
| 06:07:20 | dassouki: | troubles* |
| 07:08:01 | CIA-70: | aghisla * r10711 /docs/trunk/english_us/user_guide/ ( features_at_a_glance.tex forward.tex ): some updates: features and plugins list. |
| 07:08:03 | sigq: | tracfeed: Changeset [10711]: some updates: features and plugins list. <https://trac.osgeo.org/qgis/changeset/10711> |
| 08:32:26 | springmeyer: | wonder_sk: are you around? I get 'cannot import name uic' when trying to install symbology_ng plugin... |
| 08:33:55 | springmeyer: | I'm running the KyngChaos 1.01 build on mac os x and appears that that PyQt module is not being bundled... |
| 08:34:37 | wonder_sk: | springmeyer: i'm here from time to time... uic is part of pyqt |
| 08:35:12 | wonder_sk: | then you should probably ask kyngchaos to add it |
| 08:35:24 | springmeyer: | okay, sounds good |
| 08:36:00 | wonder_sk: | springmeyer: anyway i hope i'll finish soon the new symbology in c++ : ) |
| 08:36:11 | springmeyer: | oh, really? |
| 08:36:31 | springmeyer: | cool, well I'll try to get this plugin working so I can see what you have in store : ) |
| 08:37:18 | springmeyer: | trying to match/crosswalk qgis symbology <--> mapnik symbology has highlighted gaps in both |
| 08:39:18 | wonder_sk: | i would be interested in results of this research |
| 08:41:07 | springmeyer: | cool, I got the plugin working! |
| 08:41:08 | jef: | wonder_sk: btw does new symbology allow scaling of svg symbols? ie. fixed size in real world coordinates. |
| 08:41:28 | springmeyer: | just copied the uic folder into the QGIS bundle ( from my separate PyQT4 install ) |
| 08:42:02 | springmeyer: | yes, wonder_sk: I will gather my thoughts and send them to the list |
| 08:42:32 | springmeyer: | wonder_sk: for now I have implemented only partial support so I'll try to get a little farther first... |
| 08:45:00 | springmeyer: | does/will the new symbology support scale-based visibility? ( MAX/MIN SCALE denoms )? |
| 08:45:13 | springmeyer: | I notice there is support for only text now, but not for the other symbols... |
| 08:46:36 | springmeyer: | if anyone want to try the mapnik plugin I just set up a repo: http://qgis.dbsgeo.com/ |
| 08:47:19 | jef: | springmeyer: but for the layer as a whole. |
| 08:48:04 | springmeyer: | right, so the method now is to just make multiple copies of the layer with different scale visibility? |
| 08:50:39 | jef: | springmeyer: so you were asking about scale based symbology and not just scale-based visibility? |
| 08:51:15 | springmeyer: | right, ya sorry about mixing up my terms |
| 08:51:22 | springmeyer: | right, scale-based symbology |
| 08:51:41 | jef: | springmeyer: but that's not there for labels either |
| 08:51:47 | springmeyer: | so different symbology visible at different scales |
| 08:51:50 | springmeyer: | oh, okay |
| 08:52:20 | springmeyer: | ah, gocha |
| 08:52:46 | springmeyer: | right, you just have the ability to set min max scale they will be viewable, okay |
| 08:53:04 | jef: | springmeyer: yes |
| 08:53:50 | springmeyer: | hmm... |
| 08:54:31 | springmeyer: | ya, thats kinda a big reason folks use Mapnik, so i'll need to figure out a way to map the use of many layers with different scale dependencies in QGIS.... |
| 08:54:59 | springmeyer: | maybe to one layer in Mapnik with many scale-based symbologies |
| 08:58:35 | wonder_sk: | jef: how do you mean the scaling of svg symbols? to set that symbol is e.g. 2 map units wide? |
| 08:59:46 | wonder_sk: | springmeyer: scale-based visibility is available currently only for the whole layer - either to show it or not |
| 09:00:16 | springmeyer: | okay, thx wonder_sk |
| 09:00:58 | wonder_sk: | springmeyer: later i'd like to develop an advanced renderer that would be rule-based - someting like mapnik rules |
| 09:01:27 | springmeyer: | great, yes that would be very useful |
| 09:06:28 | springmeyer: | hey guys, I'm running into odd unicode errors when trying to work with the upper/lower values of a graduated symbol in the python bindings |
| 09:06:43 | springmeyer: | anyone have ideas of what might be going wrong? http://dpaste.de/YK89/ |
| 09:06:44 | sigq: | Title: dpaste.de: Snippet #1077 ( at dpaste.de ) |
| 09:10:34 | wonder_sk: | springmeyer: what is that graduated_filter? |
| 09:11:01 | wonder_sk: | unicode in python is always a bit pain : ) |
| 09:11:25 | springmeyer: | ya, essentially I have to turn all QStrings into standard str to be able to send to Mapnik |
| 09:11:36 | wonder_sk: | the usual problem of this error is when trying to convert a python unicode string or QString to str |
| 09:11:46 | springmeyer: | and depending on where I call str( QString ) qgis either crashes or fails to decode |
| 09:11:49 | wonder_sk: | does mapnik accept unicode? |
| 09:11:50 | springmeyer: | okay |
| 09:11:56 | springmeyer: | no, not yet |
| 09:12:04 | wonder_sk: | it would be wiser to use unicode( QString ) |
| 09:12:27 | springmeyer: | wonder_sk: essentially I am taking the sym.upperValue/Lower and pushing through this: |
| 09:12:28 | springmeyer: | http://bitbucket.org/springmeyer/quantumnik/src/tip/sync.py#cl-30 |
| 09:12:29 | sigq: | Title: springmeyer / Quantumnik / source bitbucket.org ( at bitbucket.org ) |
| 09:14:49 | epifanio: | hi All, hi springmeyer |
| 09:15:01 | wonder_sk: | so it looks like that the attribute name 'val' contains some special chars, right? |
| 09:15:01 | springmeyer: | hey epifanio : ) |
| 09:15:17 | epifanio: | i know the uic problem, i can rty to help you |
| 09:15:44 | epifanio: | you need to add few lines inside the xcode project |
| 09:16:43 | epifanio: | to have uic and other "pyqt-internal" files, like qwt too |
| 09:16:43 | springmeyer: | wonder_sk: 'val' in this case should just be the field name string |
| 09:17:40 | wonder_sk: | springmeyer: yes, but field name could also contain some special chars |
| 09:18:33 | springmeyer: | hmm, the field name is 'SpeciesRic' |
| 09:18:43 | springmeyer: | if I wrap that in unicode( ) I get http://dpaste.de/oHpG/ |
| 09:18:46 | sigq: | Title: dpaste.de: Snippet #1078 ( at dpaste.de ) |
| 09:19:27 | springmeyer: | hmm, looks like I am calling something on an integer 'QString::at( int )' ? |
| 09:20:10 | wonder_sk: | qstring::at is a function for accessing characters from a qstring |
| 09:20:35 | springmeyer: | okay |
| 09:21:04 | wonder_sk: | weird stuff |
| 09:21:43 | wonder_sk: | conversion of qstring to unicode fails? hm... |
| 09:22:47 | springmeyer: | cool! got it working |
| 09:23:04 | springmeyer: | this works..... |
| 09:23:07 | springmeyer: | idx = self.layer.renderer( ).classificationAttributes( )[0] |
| 09:23:07 | springmeyer: | return unicode( QString( self.fields[idx].name( ) ) ) |
| 09:23:11 | springmeyer: | this crashes.... |
| 09:23:19 | springmeyer: | idx = self.layer.renderer( ).classificationAttributes( )[0] |
| 09:23:19 | springmeyer: | return unicode( self.fields[idx].name( ) ) |
| 09:23:45 | springmeyer: | ( where self.fields is 'self.layer.dataProvider( ).fields( )' ) |
| 09:26:12 | wonder_sk: | i can imagine why this could crash |
| 09:27:26 | wonder_sk: | name( ) returns reference to a QString |
| 09:28:25 | springmeyer: | oh, so should I do something different in python to get the field name? |
| 09:28:57 | wonder_sk: | and given that self.fields[idx] is just a temporary object ( is it?? ) it gets deleted by garbage collector before the returned reference is used, so the conversion uses invalid object |
| 09:29:53 | wonder_sk: | springmeyer: hard to say, this is just my hypothesis : ) |
| 09:29:53 | springmeyer: | yes, surely temporary |
| 09:30:16 | wonder_sk: | but if you save the reference, e.g. fld = self.fields[idx] |
| 09:30:25 | springmeyer: | seems like a smart hypothesis, thanks |
| 09:30:29 | wonder_sk: | and then do return unicode( fld.name( ) ) |
| 09:30:35 | wonder_sk: | it should work |
| 09:30:37 | springmeyer: | right, headed to do that now... : ) |
| 09:30:38 | springmeyer: | great |
| 09:31:59 | springmeyer: | nice : ) |
| 09:35:08 | wonder_sk: | springmeyer: btw. what are your plans with the quantumnik plugin? : ) i mean who is the target audience etc |
| 09:36:08 | springmeyer: | hey, good question |
| 09:36:37 | springmeyer: | the only plan that got me started was to have the ability to style simple maps in QGIS and export to Mapnik XML |
| 09:36:48 | springmeyer: | so to match the ability to export to Mapserver Mapfile |
| 09:37:17 | springmeyer: | with the audience being Mapnik users solely to help give a kick start to authoring XML |
| 09:37:26 | springmeyer: | but... |
| 09:38:05 | springmeyer: | this my first python plugin so I've been seeing the potential for more |
| 09:38:59 | wonder_sk: | ah right, so no more XML handwriting ;- ) |
| 09:39:18 | springmeyer: | well, ideally : ) |
| 09:39:48 | springmeyer: | but nearly all that use Mapnik will end up hand authoring to add the scale-based symbology |
| 09:40:20 | springmeyer: | since most current Mapnik users tend toward producing large tilesets |
| 09:40:38 | springmeyer: | but Mapnik also has potental for thematic cartography |
| 09:41:06 | springmeyer: | so, the ability of QGIS to do unique values symbolization/graduated symbolizer is excellent |
| 09:42:04 | springmeyer: | that works so well ( esp with the TopoColour plugin ) that I am seeing less need for adding stuff like that into Mapnik itself |
| 09:42:33 | epifanio: | springmeyer: have you solved the "import uic" problem ? |
| 09:42:57 | springmeyer: | epifanio: yes, by hand installing the uic folder inside QGIS bundle |
| 09:43:29 | epifanio: | ok, to have uic and pyqwt working, i added it in the "bundle-Qt" script ( this : http://img206.imageshack.us/img206/5544/immagine1h.png ) |
| 09:43:42 | epifanio: | double click on it : http://rafb.net/p/vaLSUq53.html i added the lines : "49,50,98,99" |
| 09:43:43 | sigq: | Title: Nopaste - No description ( at rafb.net ) |
| 09:44:14 | springmeyer: | cool, is that going to be added to trunk? official builds? |
| 09:44:42 | epifanio: | for uic maybe yes ... but pyqwt no, beacouse it is optional |
| 09:45:14 | springmeyer: | wonder_sk: last main short term goal for me ( other than learning more qgis ) is to provide the ability to export high resolution maps directly from QGIS via Mapnik |
| 09:45:19 | epifanio: | william has done a readme about custom build, he suggest me that way |
| 09:45:31 | springmeyer: | ah, okay |
| 09:46:12 | wonder_sk: | springmeyer: nice! |
| 09:46:14 | springmeyer: | thx |
| 09:46:56 | wonder_sk: | still i hope that with the new symbology and better labeling as my summer of code project there will be no need for mapnik in qgis ;- ) |
| 09:47:23 | : | * wonder_sk is going to have a dinner |
| 09:47:24 | springmeyer: | wonder_sk: yes, I agree too : ) |
| 09:47:43 | springmeyer: | ideally mapnik integration would only be needed on mapnik side |
| 09:48:18 | springmeyer: | to take advantage of all the amazing symbology in qgis |
| 09:48:35 | springmeyer: | congrats on the GSOC ( I can't wait for updates ) : ) |
| 09:48:46 | springmeyer: | have a good din din |
| 10:00:21 | lynx_r: | hi all |
| 10:01:01 | lynx_r: | wonder_sk: what can you say about my patch with tabs? |
| 10:02:49 | Tim_B: | Anyone home that is familiar with the debian packaging of QGIS, specifically of the python modules? |
| 10:03:16 | lynx_r: | wonder_sk: had it been included? |
| 10:03:16 | Tim_B: | I am trying to rebuild the Jaunty QGIS packages from the launchpad ppa. |
| 10:04:26 | Tim_B: | During the creation of the deb package, dh_install complains of <debian/tmp/>usr/lib/python*/site-packages/qgis/__init__.py missing. |
| 10:06:34 | Tim_B: | Tracing this back it looks like my build is placing the python files into <debian/tmp/>usr/lib/python2.6/dist-packages/qgis/ |
| 10:08:12 | Tim_B: | The subdir site-packages *seems* like the right place, and dist-packages is the *wrong* subdir. Am I correct? |
| 10:09:13 | Tim_B: | Also, the python script from python/CMakeLists.txt that runs during configuration returns "/usr/lib/python2.6/dist-packages". |
| 10:09:45 | Tim_B: | I'm not sure what direction to go into next... |
| 10:11:01 | springmeyer: | wonder_sk: I clarified the goals: http://bitbucket.org/springmeyer/quantumnik/wiki/Home |
| 10:11:02 | sigq: | Title: springmeyer / Quantumnik / wiki / Home bitbucket.org ( at bitbucket.org ) |
| 10:14:06 | lynx_r: | wonder_sk: it is https://trac.osgeo.org/qgis/attachment/ticket/1622/qgis_tabs.patch jef said that you have done some modification to apply it to latest revision |
| 10:14:08 | sigq: | Title: #1622: qgis_tabs.patch - Quantum GIS - Trac ( at trac.osgeo.org ) |
| 10:40:14 | wonder_sk: | springmeyer: thanks |
| 10:40:43 | wonder_sk: | Tim_B: jef is the guy to ask about debian packaging |
| 10:41:39 | wonder_sk: | lynx_r: i haven't been doing modifications to your patch. jef was probably saying that due some recent commits by me your patch doesn't apply cleanly |
| 10:42:12 | wonder_sk: | lynx_r: but i haven't had time to look at it now, sorry |
| 10:42:45 | wonder_sk: | lynx_r: try updating the patch to latest svn trunk |
| 11:01:42 | Tim_B: | Thanks, winder_sk. I have to take ( another ) break from this to go clean the garage, so hopefully jef will see above ;- ) |
| 11:02:24 | lynx_r: | wonder_sk: humm, sorry |
| 11:04:13 | lynx_r: | i wrong understanded |
| 13:09:58 | springmeyer: | so, in python how can you loop through all layers in the legend and find out which ones are checked? |
| 13:10:14 | springmeyer: | and then programmaticaly uncheck or check them? |
| 13:18:28 | wonder_sk: | springmeyer: it's not possible to access legend directly. but for your purpose you can access mapRenderer( ) method of map canvas. In map renderer, access layerSet( ) method which gives you only the layers which will be rendered |
| 13:19:21 | springmeyer: | oh, okay thats helpful |
| 13:19:35 | wonder_sk: | i'm affraid there's no simple way to check and uncheck the layers in legend programmatically ... why do you need that? |
| 13:19:50 | springmeyer: | essentially I'm trying to figure out a way to turn off all layers and then turn just those previously selected back on |
| 13:20:36 | springmeyer: | currently I'm calling 'self.iface.actionHideAllLayers( ).trigger( )' |
| 13:20:40 | wonder_sk: | you can probably use QgsMapCanvas::setLayerSet function... though i don't know how much does it fit into your needs |
| 13:20:42 | springmeyer: | and show... |
| 13:21:28 | springmeyer: | yes, I'm trying that now but getting SIGBUS crashes |
| 13:24:42 | springmeyer: | hmmm, probably not worth the mucking... |
| 13:24:53 | springmeyer: | thanks for the tip thought wonder_sk |
| 13:25:43 | wonder_sk: | springmeyer: you're welcome, hope you'll have luck |
| 13:47:22 | jef: | wonder_sk: kind of. but just keeping the initial size is enough - or is it that easy to scale a svg to a certain size? |
| 13:49:23 | wonder_sk: | jef: i'm thinking of creating a QgsUnit class that would work as an universal way how to express units ( not only for symbology ) |
| 13:50:24 | wonder_sk: | so user should be able to specify either milimeters, pixel, miles, map units or whatever he likes and qgis should calculate the correct size for the current view and output type |
| 13:53:59 | jef: | wonder_sk: sounds good. |
| 13:56:39 | wonder_sk: | jef: btw. do you have any idea of the state of 1.1 binary for win? |
| 13:57:45 | jef: | wonder_sk: I didn't do one yet. I'm not sure if we want 1.1, 1.0.2 or both |
| 14:00:18 | wonder_sk: | jef: no idea. but 1.1 is probably a better choice IMHO |
| 14:00:51 | wonder_sk: | i've just finished compiling a mingw/msys based build for my friend... |
| 14:02:17 | : | * jef wondering why we are actually maintain 1.0 - who's going to use that? |
| 14:03:06 | wonder_sk: | i think it's part of Tim's plan to keep a stable branch |
| 14:03:28 | jef: | sure. still... |
| 14:03:46 | wonder_sk: | anyway i'm still of the opinion that qgis is too young for to try to have a rockstable branch |
| 14:04:24 | wonder_sk: | so currently i feel the backporting and api stability more a burden than an advantage |
| 14:04:53 | wonder_sk: | though users and plugin devs are probably of the opposite opinion : ) |
| 14:05:17 | jef: | macho: can we have a poll on the website? |
| 14:06:35 | jef: | wonder_sk: and we don't have abi stability anyway. |
| 14:07:40 | wonder_sk: | right |
| 14:08:34 | jef: | wonder_sk: who's going to persuade tim? ; ) |
| 14:10:08 | wonder_sk: | jef: hehe :- ) i've been trying for some time, saying that qgis api is not really something we can consider stable |
| 14:12:33 | jef: | wonder_sk: ah, that's why he's on the run again ; ) |
| 14:39:05 | : | * wonder_sk waving... good night! |
| 14:40:39 | dassouki: | good night |
| 16:28:48 | jctull: | epifanio: ping |
| 19:20:08 | CIA-70: | cengel * r10712 /docs/trunk/english_us/user_guide/ ( 5 files in 3 dirs ): |
| 19:20:08 | CIA-70: | Added content to Spatiallite Layers chapter |
| 19:20:08 | CIA-70: | Addeed images/mActionAddSpatiaLiteLayer.png |
| 19:21:30 | CIA-70: | cengel * r10713 /docs/trunk/english_us/user_guide/working_with_vector.tex: Commented out Disclaimer note for Chap 5 |
| 19:21:31 | sigq: | tracfeed: Changeset [10713]: Commented out Disclaimer note for Chap 5 <https://trac.osgeo.org/qgis/changeset/10713> |
| 21:04:01 | gsherman: | @seen timlinux |
| 21:04:01 | sigq: | gsherman: timlinux was last seen in #qgis 1 week, 3 days, 6 hours, 15 minutes, and 24 seconds ago: <timlinux> cya springmeyer, jef |
| 22:52:58 | CIA-70: | mhugent * r10714 /branches/Version-1_0/src/ ( 3 files in 2 dirs ): backport wfs fix to 1.0 |
| 22:52:58 | sigq: | tracfeed: Changeset [10714]: backport wfs fix to 1.0 <https://trac.osgeo.org/qgis/changeset/10714> |
