Re: [Qgis-developer] Re: Raster providers
Hi, I have roughly implemented approximate reprojection. It should work for GDAL and GRASS, I have not tested WMS. Currently it is always using approximation (even if error is too big). Rendering with reprojection takes about 5 times more. You can get some info about times grepping debug output, e.g.: qgis rp.qgs 2>&1 | egrep "(time|CPMatrix size)" | sed 's/.*cpp: //' I have got for example for reprojected raster in raster providers branch: 89: (QgsRasterProjector) CPMatrix size: mCPRows = 5 mCPCols = 9 41: (readBlock) create projector time (ms): 10 53: (readBlock) read not projected block time (ms): 24 71: (readBlock) reproject block time (ms): 1022 1287: (draw) raster draw time (ms): 1446 1287: (draw) raster draw time (ms): 10 the same raster not reprojected in raster providers branch: 1287: (draw) raster draw time (ms): 323 in trunk: 1830: (draw) raster draw time (ms): 689 It is strange that it is slower in trunk. > On Thu, Jan 13, 2011 at 7:06 PM, John C. Tull wrote: >> Comparing the load time of this raster when the project CRS is the same as >> the raster (i.e., no transform should be occurring), the load time still >> requires several minutes. This is true whether or not I have enabled on the >> fly reprojection in the project options. In contrast, the same file takes >> several seconds to load in trunk. Are you talking about loading the raster first time or about redrawing? Maybe statistics are not necessarily computed. Radim ___ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
[Qgis-developer] Bug in QGIS 1.7 raster? Changing transparency values set the 3 RGB bands to band 1?
Hi Devs, With QGIS 1.7 release r15056 on WinXP Pro. I have a three-band raster (geotiff) loaded and it displays all-right. Then I open the raster properties window, go to the transparency tab, slide the transparency ruler and click Apply. The transparency is applied to the raster but the three bands (R,G,B) are all set to band 1 resulting in a black&white (grey) raster! The same behaviour is observed when I add a transparent pixel and click Apply... Can someone confirm these behaviours? Regards, Benoit___ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
Re: [Qgis-developer] Closest Feature Finder plugin error
Ups, sorry my fault On Thu, 20 Jan 2011 19:16:10 +0100 Giuseppe Sucameli wrote: > > seems the error is due to the fuzzytracer plugin. Look at the error ;) -- Alexander Bruy ___ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
Re: [Qgis-developer] Closest Feature Finder plugin error
Hi Alexander, On Thu, Jan 20, 2011 at 7:06 PM, Alexander Bruy wrote: > Traceback (most recent call last): > File "/home/alex/.qgis/python/plugins/fuzzytracer/fuzzytracer.py", line > 115, in mapToolChanged >self.iface.mapCanvas().setMapTool( self.mapTool ) > RuntimeError: underlying C/C++ object has been deleted > seems the error is due to the fuzzytracer plugin. Look at the error ;) Cheers. -- Giuseppe Sucameli ___ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
[Qgis-developer] Closest Feature Finder plugin error
Hi I've try to use Closest Feature Finder plugin and get next error Traceback (most recent call last): File "/home/alex/.qgis/python/plugins/fuzzytracer/fuzzytracer.py", line 115, in mapToolChanged self.iface.mapCanvas().setMapTool( self.mapTool ) RuntimeError: underlying C/C++ object has been deleted Python version: 2.5.2 (r252:60911, Sep 11 2008, 13:43:31) [GCC 4.2.4] QGIS version: 1.7.0-Trunk Trunk, 15053M Also I get same error with SelectPlus and Interactive Identify plugins. Seems that this related to the QgsMapTool usage. -- Alexander Bruy ___ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
Re: [Qgis-developer] Layer list widget for PyQGIS applications
Hi Martin, right, only old symbology is supported so far. Sounds great including the widget to PyQGIS, so I need to dive into the symbology-ng. Regards, Germán 2011/1/19 Martin Dobias > Hi German > > 2011/1/18 Germán Carrillo : > > Hi all, > > first of all, sorry for the delay in publishing this :-) > > At [1] you can find a basic layer list (legend) widget for PyQGIS > standalone > > applications. It's under GPL v.2. and it is based on the OpenOceanMap > legend > > [2] written by Aaron Racicot and, of course, on the QGIS layer list > widget. > > See [1] for further information and instructions to add it to your > > application. > > Thanks for providing us with the widget. In case you are going to > further polish it, we could eventually add it to the PyQGIS. > > I haven't tried it yet, but looking at the code I would suggest you to > add support also for symbology-ng - it seems that only old symbology > is supported as of now. > > Regards > Martin > -- --- |\__ (:>__)( |/ Soluciones Geoinformáticas Libres http://geotux.tuxfamily.org/ ___ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
Re: [Qgis-developer] Qgis usability tests
It is Qgis 1.6. In the end of the report, there are suggestions that refers to the cases in the study and for better improvements of them. During the study, Assoc. Prof. Dr.Kursat Cağıltay was my advisor. It was just final homework of a Human Computer Interaction (HCI) course, but turned out to be a nice report I suppose. It is a user-based one. Actually, I do interface design because of my profession. I work in Modelling and Simulation Research and Development Center in METU. We do usability tests on simulator interfaces. But I am doing Phd at Geodetic and Geographic Information Technologies Division at METU.This time, I wanted study Gis area. How do you welcome these kind of research? I know that you have meetings that developers do presentations and share their report. I can send it here before or the committee? Here are some links: http://www.ggit.metu.edu.tr/ (GGIT department at METU) http://www.metu.edu.tr/ (METU web site) http://www.modsimmer.metu.edu.tr/ ( Modelling and Simulation Center at METU) best regards -- Aslı Yılmaz Industrial Designer METU'05 On Thu, Jan 20, 2011 at 3:12 PM, Werner Macho wrote: > Hi! > > > finallizing the report now. I would like to share it with you if >> possible when I finished. >> > > That sounds really great .. If there are some hints for improvement in the > report I'm sure we'll try to bring it to the current version. > > On what version of QGIS is your report based? > > kind regards > Werner > > -- Aslı Yılmaz Industrial Designer METU'05 ___ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
Re: [Qgis-developer] Qgis usability tests
Hi! finallizing the report now. I would like to share it with you if possible when I finished. That sounds really great .. If there are some hints for improvement in the report I'm sure we'll try to bring it to the current version. On what version of QGIS is your report based? kind regards Werner ___ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
Re: [Qgis-developer] Qgis usability tests
Hi, Ok thanks alot. Actually, I aimed to make a usability test for a Gis software because their interface is generally complex and few usabiliy studies are made on them. I chose Qgis because it is opensource and expandable. Also, as far as I understood from user and developer meetings and these kind of forums, the committee of Qgis is putting emphasize on developing the software. I made my study and I am finallizing the report now. I would like to share it with you if possible when I finished. Aslı On Wed, Jan 19, 2011 at 9:57 PM, Andreas Neumann wrote: > Hi Asli, > > I don't about usability studies around QGIS. So, if you want to start > something, you are welcome. > > I am sure that such a study would contribute some interesting aspects about > shortcomings of the existing GUI, or ideas to improve it. > > Andreas > > > On 1/19/11 6:58 PM, yilmaz asli wrote: > > Hie, > > I just subscribed to this group. I do not know whether my mail is sent or > may be I am sending it twice, if it is sorry. > > I am a Phd student at Geodetic and Geographic Information Technologies > Division at the Middle East Technical University in Turkey. My major is > industrial design. I am interested in usability topics. For my research, I > wonder if there is user based usability tests of Qgis and where can I find > their reports if there are? Could you help me about it? > > Thank you very much, > > -- > Aslı Yılmaz > Industrial Designer METU'05 > > > ___ > Qgis-developer mailing > listQgis-developer@lists.osgeo.orghttp://lists.osgeo.org/mailman/listinfo/qgis-developer > > > > ___ > Qgis-developer mailing list > Qgis-developer@lists.osgeo.org > http://lists.osgeo.org/mailman/listinfo/qgis-developer > > -- Aslı Yılmaz Industrial Designer METU'05 ___ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
Re: [Qgis-developer] Highlighting feature with 'Identify' tool
On 01/19/2011 09:44 PM, Alex Mandel wrote: On 01/19/2011 02:10 AM, Ivan Mincik wrote: Hi, is there any special reason that after using 'Identify' tool on polygon feature, it is highlighted by its outer shape (whole polygon extent area) rather than by exact geometry like it is done by 'Select' tool ? Example: 1. Draw multi-polygon feature containing few holes. 2. Click on polygon using 'Identify' tool. Whole polygon including holes is highlighted. 3. Click on polygon using 'Select' tool. Exact polygon shape is highlighted (without holes) I think, current behavior is quite confusing, when You have big polygons, containing holes where another polygons exists. I seams, there is related ticket on this [1]. [1] http://trac.osgeo.org/qgis/ticket/2190 Thanks It might have something to do with the fact that you can Identify multiple layers at the same time, so in that sense the Identify would include anything in the hole underneath your layer too. Looks like the ticket you found in trac is the same thing. I suggest you add your description to provide more details to it. Thanks, it makes some sense, but it is not very user friendly behavior. I have added my notes to that ticket. -- Ivan Mincik, Gista s.r.o. ___ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
