From lists at dipole-consulting.com Tue Jul 1 09:04:21 2008 From: lists at dipole-consulting.com (Didrik Pinte) Date: Tue, 01 Jul 2008 09:04:21 +0200 Subject: Modifying layers / in-memory layers In-Reply-To: References: Message-ID: <1214895861.4352.24.camel@ddp.simpson> On Mon, 2008-06-30 at 13:21 -0300, Anthony Lenton wrote: > So, > - What do you think of the whole > allow-the-user-to-modify-shapes-and-info idea? Is it bound to fail? > Is somebody working on this already? I am really interested in this. Thuban is actually only a viewer but with such functionnalities, it could be much more than that. I never heard about someone working on that king of new feature for Thuban. I was planning to try to link Thuban with a mysql database to be able to edit some of it's data attributes, so this could be a really interesting thing. > - What do you think of the in-memory stores idea? Is there an > implementation I haven't seen? Is there a reason not to do this all > together? Is a postgis store a better idea? You can think about a "generic" in-memory store that you can feed with any type of data. Specialized stores like postgis and shapefile are there to increase performance or make use of specific feature of the backend. > - How much work does shptree do, is it a in-memory store that only > needs a shapefile to initialize from, or is it more of just a proxy to > the shapefile? I don't know. Isn't it related to the qix file ? > - Has anybody tried SpatiaLite? http://www.gaia-gis.it/spatialite/ I > haven't, but it might come in handy as a middle point between keeping > layers in memory and needing to set up a postgis database. It could be a way to persist the dataset on the filesystem. We could imagine an Extension that would allow a user to store/retrieve an in-memory layer from spatialite. Didrik -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080701/88ec17c5/attachment.bin From bernhard at intevation.de Wed Jul 2 12:59:34 2008 From: bernhard at intevation.de (Bernhard Reiter) Date: Wed, 2 Jul 2008 12:59:34 +0200 Subject: Modifying layers / in-memory layers In-Reply-To: References: Message-ID: <200807021259.37848.bernhard@intevation.de> Hi Anthony, On Monday 30 June 2008 18:21, Anthony Lenton wrote: > I have been thinking that a very neat feature for thuban would be to > allow you to modify a layer's shapes and info some how. For that a > previous step would be to have some sort of in-memory layers, to not > have to rewrite and reload a whole shapefile each time you modify > anything. I agree that this would be nice to have. It goes beyond the "interactive viewer" idea, but I consider this fine. While the original idea of Thuban was to first be a tool to very nicely do interactive data exploration and leave the "hard" GIS tasks to real GIS like GRASS, somehow this vision has become outdated. For once, developing GRASS frontends (for Thuban and similiar) did not really happen and there have been many other small goe-viewers that libraries that go into the direction of being a GIS (while mostly not fully there, but that would not be necessary in many cases anyway). > I looked in to this while coding the importMP extension and > I couldn't find in-memory shapestores. The SHPTree class needs to be > initted from a shapefile. I guess posgis shapestores could be used, > those are quite a bit faster than rewriting the whole shapefile, but > depend on you having postgis up and running. There is an > implementation of an in-memory table, MemoryTable, but it doesn't seem > to be used anywhere, is this so? I think the tests might use it. > And then there are also TransientDatabase tables that use SQLite. Yes, good for important query options. > So, > - What do you think of the whole > allow-the-user-to-modify-shapes-and-info idea? Is it bound to fail? It is a good idea that will work if done right. Having the possiblity of a diff to a current data-layer that is read-only would be cool. > Is somebody working on this already? There was a rough experimental extension to enable very simple line drawing, which probably is not a blueprint nor a good example to build upon. Somebody posted some changes for this a while ago: Barry. He stopped responding (maybe because we were slow to react as well) so that code was never taken up (nor copyright transfered). > - What do you think of the in-memory stores idea? Is there an > implementation I haven't seen? Is there a reason not to do this all > together? Is a postgis store a better idea? I believe in-memory is fine for a start. Most changes will be small. There should be a "commit" function to make the change permanent and possibly a "save-as" function to save to as many targets as we can which would include postfix/gis as well, if possible. > - How much work does shptree do, is it a in-memory store that only > needs a shapefile to initialize from, or is it more of just a proxy to > the shapefile? Don't know without inspection. > - Has anybody tried SpatiaLite? http://www.gaia-gis.it/spatialite/ I > haven't, but it might come in handy as a middle point between keeping > layers in memory and needing to set up a postgis database. I have not tried it, my strong guess is that it was not available when Thuban was designed. We were quite early in adoption sqlite. Could be a good idea for transient data, as it is used for tables already. > Comments? Ideas? Suggestions? Knights with rubber chickens? :) -- Managing Director - Owner: www.intevation.net (Free Software Company) Germany Coordinator: fsfeurope.org. Coordinator: www.Kolab-Konsortium.com. Intevation GmbH, Osnabr?ck, DE; Amtsgericht Osnabr?ck, HRB 18998 Gesch?ftsf?hrer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080702/01d6f1d6/attachment.bin From bernhard at intevation.de Wed Jul 2 13:01:59 2008 From: bernhard at intevation.de (Bernhard Reiter) Date: Wed, 2 Jul 2008 13:01:59 +0200 Subject: Docs for the MapServer extension In-Reply-To: References: Message-ID: <200807021301.59404.bernhard@intevation.de> On Monday 30 June 2008 18:25, Anthony Lenton wrote: > Is there some sort of English (or Spanish!) translation of > http://ftp.intevation.de/thuban/doc/Diplomarbeit_JanSchuengel_v1.01_20040917.pdf ? Not that I am aware of. You've probably seen the English Abstract on one of the early pages. Most of it deals with the surrounding, and the examination of other solutions. The code should have most of the contents regarding what was implemented towards Thuban. Are you interested in something in particular? Bernhard -- Managing Director - Owner: www.intevation.net (Free Software Company) Germany Coordinator: fsfeurope.org. Coordinator: www.Kolab-Konsortium.com. Intevation GmbH, Osnabr?ck, DE; Amtsgericht Osnabr?ck, HRB 18998 Gesch?ftsf?hrer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080702/86eb2370/attachment.bin From bh at intevation.de Wed Jul 2 14:24:14 2008 From: bh at intevation.de (Bernhard Herzog) Date: Wed, 02 Jul 2008 14:24:14 +0200 Subject: Modifying layers / in-memory layers In-Reply-To: <200807021259.37848.bernhard@intevation.de> (Bernhard Reiter's message of "Wed, 2 Jul 2008 12:59:34 +0200") References: <200807021259.37848.bernhard@intevation.de> Message-ID: Bernhard Reiter writes: >> There is an >> implementation of an in-memory table, MemoryTable, but it doesn't seem >> to be used anywhere, is this so? > > I think the tests might use it. It's also used in GREAT-ER: http://great-er.intevation.org/ >> - What do you think of the in-memory stores idea? Is there an >> implementation I haven't seen? Is there a reason not to do this all >> together? Is a postgis store a better idea? > > I believe in-memory is fine for a start. Most changes will be small. > There should be a "commit" function to make the change permanent > and possibly a "save-as" function to save to as many targets as we can which > would include postfix/gis as well, if possible. [...] >> - Has anybody tried SpatiaLite? http://www.gaia-gis.it/spatialite/ I >> haven't, but it might come in handy as a middle point between keeping >> layers in memory and needing to set up a postgis database. > > I have not tried it, my strong guess is that it was not available when Thuban > was designed. We were quite early in adoption sqlite. > Could be a good idea for transient data, as it is used for tables > already. If we're talking about serious additions to Thuban's data model, it might be good to move to an integrated geometry + attribute data model first. That's basically what PostGISShapeStore already is. See also the comment in Thuban/Model/data.py. Bernhard -- Bernhard Herzog Intevation GmbH, Osnabr?ck Amtsgericht Osnabr?ck, HR B 18998 http://www.intevation.de/ Gesch?ftsf?hrer: Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner From antoniolenton at gmail.com Thu Jul 10 22:32:08 2008 From: antoniolenton at gmail.com (Anthony Lenton) Date: Thu, 10 Jul 2008 17:32:08 -0300 Subject: Clipping and zoom Message-ID: Hello list, I've added a patch on Wald here https://wald.intevation.org/tracker/index.php?func=detail&aid=670&group_id=6&atid=107 let me know what you think. -- Anthony Lenton Except - The good guys in software development http://except.com.ar From antoniolenton at gmail.com Thu Jul 10 22:37:17 2008 From: antoniolenton at gmail.com (Anthony Lenton) Date: Thu, 10 Jul 2008 17:37:17 -0300 Subject: Another couple of ideas Message-ID: Hello list, I started writing this within my previous mail, and then realized that it was about something completely different, so I'm sending another mail :-) I'd really like to see add a couple of characteristics from MapServer's data model be worked in to Thuban, namely: - The ability to set a max/min scale to a layer, so that it isn't shown outside of that range. - The notion of a style (with many styles to a class) to allow to draw lines with a border, dashed lines with a continuous background, and several other pretty things. Would you recommend / disencourage me to start looking in to coding one of these? Kind regards, -- Anthony Lenton Except - The good guys in software development http://except.com.ar From thuban-patches at wald.intevation.org Thu Jul 10 16:35:52 2008 From: thuban-patches at wald.intevation.org (thuban-patches@wald.intevation.org) Date: Thu, 10 Jul 2008 16:35:52 +0200 (CEST) Subject: =?UTF-8?B?W3RodWJhbi1QYXRjaGVzXVs2NzBdIENsaXBwaW5nIHRoZSBwb2ludHMgYXZvaWRzIHd4J3Mgc2NhbGUgbGltaXRhdGlvbnM=?= Message-ID: <20080710143552.0DC5D406CC@pyrosoma.intevation.org> Patches item #670, was opened at 2008-07-10 11:35 Status: Open Priority: 3 Submitted By: Anthony Lenton (elachuni) Assigned to: Nobody (None) Summary: Clipping the points avoids wx's scale limitations Version: None Category: None Initial Comment: Attached is a patch that allows you to lift the max_scale limit, by clipping points within wx's limits before drawing. * Optimized renderers are disabled in the patch as clipping is only implemented in the Python default renderers. As clipping is done after projecting the points, if the idea seems sound we'd have to implement clipping in the C wxproj renderers too, or pass the points back to Python to be clipped after projecting. * in Thuban.UI.viewport.ViewPort's calc_min_max_scales, max_scale is now set arbitrarily to 500000. We can instead choose a different maximum scale policy, as now we're not restricted by wx's limitations. * I've added quite a bit of testing but quite likely a couple of cases are still not handled correctly. Failing test cases are welcome. * I haven't tried clipping with ExportRenderer or PrinterRenderer, I don't know if those will be trouble. ---------------------------------------------------------------------- You can respond by visiting: http://wald.intevation.org/tracker/?func=detail&atid=107&aid=670&group_id=6 From bernhard at intevation.de Fri Jul 11 10:58:16 2008 From: bernhard at intevation.de (Bernhard Reiter) Date: Fri, 11 Jul 2008 10:58:16 +0200 Subject: Clipping and zoom In-Reply-To: References: Message-ID: <200807111058.21079.bernhard@intevation.de> Anthony, On Thursday 10 July 2008 22:32, Anthony Lenton wrote: > I've added a patch on Wald here the idea is good, so I am positive about it in general. As the change has a few lines, I would need a bit of time to read and try it first. :) What is your failcount on all tests with your change? You could also run the tests for svgexport for instance. Bernhard -- Managing Director - Owner: www.intevation.net (Free Software Company) Germany Coordinator: fsfeurope.org. Coordinator: www.Kolab-Konsortium.com. Intevation GmbH, Osnabr?ck, DE; Amtsgericht Osnabr?ck, HRB 18998 Gesch?ftsf?hrer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080711/dcf1f2ab/attachment.bin From bernhard at intevation.de Fri Jul 11 11:04:39 2008 From: bernhard at intevation.de (Bernhard Reiter) Date: Fri, 11 Jul 2008 11:04:39 +0200 Subject: Another couple of ideas In-Reply-To: References: Message-ID: <200807111104.40256.bernhard@intevation.de> Hi Anthony, On Thursday 10 July 2008 22:37, Anthony Lenton wrote: > I'd really like to see add a couple of characteristics from > MapServer's data model be worked in to Thuban, namely: > > ?- The ability to set a max/min scale to a layer, so that it isn't > shown outside of that range. For some layers this certainly makes sense, though maybe we should have some function that calculates this level of detail without fix settings, because the settings will depend a lot on the output device. (E.g. a layer that cannot be shown on screen with 72 dpi, might easily be seen in a few dots on a 600 lpi printer.) > ?- The notion of a style (with many styles to a class) to allow to > draw lines with a border, dashed lines with a continuous background, > and several other pretty things. We did not add this before out of two ideas: a) we wanted to avoid making Thuban a layout application as vector drawing applications usually do a better job at this. This is why I came up with the printing pipeline Thuban -> MapSVG -> Vector drawing application b) For trying to explore data, some simpler drawings might be better (we do not know for sure). Technically I believe that the old wxWidgets drawing engine might also pose some limits on what could be done easily. I read somewhere that I would have a new system, but did not followup up when this was included. Also there is much more computing power today so doing better drawings should be fine. If you need those features, by all means, give them a coding attempt. :) Best Regards, BErnhard -- Managing Director - Owner: www.intevation.net (Free Software Company) Germany Coordinator: fsfeurope.org. Coordinator: www.Kolab-Konsortium.com. Intevation GmbH, Osnabr?ck, DE; Amtsgericht Osnabr?ck, HRB 18998 Gesch?ftsf?hrer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080711/41124951/attachment.bin From antoniolenton at gmail.com Fri Jul 11 15:18:19 2008 From: antoniolenton at gmail.com (Anthony Lenton) Date: Fri, 11 Jul 2008 10:18:19 -0300 Subject: Clipping and zoom In-Reply-To: <200807111058.21079.bernhard@intevation.de> References: <200807111058.21079.bernhard@intevation.de> Message-ID: Hello Berhnard, On Fri, Jul 11, 2008 at 5:58 AM, Bernhard Reiter wrote: > the idea is good, so I am positive about it in general. > As the change has a few lines, I would need a bit of time > to read and try it first. :) > > What is your failcount on all tests with your change? > You could also run the tests for svgexport for instance. There are two tests that fail, and they also fail in a clean repo here: test_Columns and test_Column from Extensions/ogr/test/testOGRShapestore.py The rest of the suite runs clean. -- Anthony Lenton Except - The good guys in software development http://except.com.ar From antoniolenton at gmail.com Fri Jul 11 16:05:06 2008 From: antoniolenton at gmail.com (Anthony Lenton) Date: Fri, 11 Jul 2008 11:05:06 -0300 Subject: Another couple of ideas In-Reply-To: <200807111104.40256.bernhard@intevation.de> References: <200807111104.40256.bernhard@intevation.de> Message-ID: Hello Bernhard, On Fri, Jul 11, 2008 at 6:04 AM, Bernhard Reiter wrote: > Hi Anthony, > > On Thursday 10 July 2008 22:37, Anthony Lenton wrote: >> I'd really like to see add a couple of characteristics from >> MapServer's data model be worked in to Thuban, namely: >> >> - The ability to set a max/min scale to a layer, so that it isn't >> shown outside of that range. > > For some layers this certainly makes sense, though maybe we should have > some function that calculates this level of detail without fix settings, > because the settings will depend a lot on the output device. > (E.g. a layer that cannot be shown on screen with 72 dpi, might easily be seen > in a few dots on a 600 lpi printer.) Hm, true... though those changes in ouput-resolution should already bring along other problems, for example a line that looks neat at 2px-wide on the screen will look real scrawny and thin on a high-res printer. How are you working around those problems currently? >> - The notion of a style (with many styles to a class) to allow to >> draw lines with a border, dashed lines with a continuous background, >> and several other pretty things. > > We did not add this before out of two ideas: > a) we wanted to avoid making Thuban a layout application as vector > drawing applications usually do a better job at this. > This is why I came up with the printing pipeline > Thuban -> MapSVG -> Vector drawing application > b) For trying to explore data, some simpler drawings might > be better (we do not know for sure). > > Technically I believe that the old wxWidgets drawing engine might also pose > some limits on what could be done easily. I read somewhere that I would have > a new system, but did not followup up when this was included. I think the wx map renderer is an *excellent* piece of software, I'm impressed at just how fast it works. But yes, maybe at some point it would be good to start using a more featureful backend. While talking about features that probably need a major rewrite, let me add these to the list :-) - Antialias - Line width zooming (showing lines wider when you get closer) - Rotated labels > Also there is much more computing power today so doing better drawings should > be fine. > > If you need those features, by all means, give them a coding attempt. :) Wee :-D -- Anthony Lenton Except - The good guys in software development http://except.com.ar From bernhard at intevation.de Wed Jul 16 23:19:48 2008 From: bernhard at intevation.de (Bernhard Reiter) Date: Wed, 16 Jul 2008 23:19:48 +0200 Subject: Clipping and zoom In-Reply-To: References: <200807111058.21079.bernhard@intevation.de> Message-ID: <200807162319.49581.bernhard@intevation.de> Hi Anthony, On Friday 11 July 2008 15:18, Anthony Lenton wrote: > On Fri, Jul 11, 2008 at 5:58 AM, Bernhard Reiter wrote: > > the idea is good, so I am positive about it in general. > > As the change has a few lines, I would need a bit of time > > to read and try it first. :) > > > > What is your failcount on all tests with your change? > > You could also run the tests for svgexport for instance. > > There are two tests that fail, and they also fail in a clean repo > here: test_Columns and test_Column from > Extensions/ogr/test/testOGRShapestore.py ah, those are in the beta code and mentioned in the releasenotes, so they are expected to fail. > The rest of the suite runs clean. That is very promissing. :) Do you expect any other problems? Currently speed of course will be a problem. Bernhard -- Managing Director - Owner: www.intevation.net (Free Software Company) Germany Coordinator: fsfeurope.org. Coordinator: www.Kolab-Konsortium.com. Intevation GmbH, Osnabr?ck, DE; Amtsgericht Osnabr?ck, HRB 18998 Gesch?ftsf?hrer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080716/53589b47/attachment.bin From bernhard at intevation.de Wed Jul 16 23:25:00 2008 From: bernhard at intevation.de (Bernhard Reiter) Date: Wed, 16 Jul 2008 23:25:00 +0200 Subject: Another couple of ideas In-Reply-To: References: <200807111104.40256.bernhard@intevation.de> Message-ID: <200807162325.01045.bernhard@intevation.de> [I am on vacation that is why my reaction time is longer than usual.] On Friday 11 July 2008 16:05, Anthony Lenton wrote: > On Fri, Jul 11, 2008 at 6:04 AM, Bernhard Reiter wrote: > > On Thursday 10 July 2008 22:37, Anthony Lenton wrote: > >> I'd really like to see add a couple of characteristics from > >> MapServer's data model be worked in to Thuban, namely: > >> > >> ?- The ability to set a max/min scale to a layer, so that it isn't > >> shown outside of that range. > > > > For some layers this certainly makes sense, though maybe we should have > > some function that calculates this level of detail without fix settings, > > because the settings will depend a lot on the output device. > > (E.g. a layer that cannot be shown on screen with 72 dpi, might easily be > > seen in a few dots on a 600 lpi printer.) > > Hm, true... though those changes in ouput-resolution should already > bring along other problems, for example a line that looks neat at > 2px-wide on the screen will look real scrawny and thin on a high-res > printer. ?How are you working around those problems currently? Good question, I believe that there is a minium line width for vector output, and again the printing pipeline via map-svg should take care of this on the vector drawing application side of things. Probably it is an unresolved problem. Again, if you need this in Thuban and it is reasonably coded, we'll put it in, I just wanted to respond with my thoughts on the matter. > >> ?- The notion of a style (with many styles to a class) to allow to > >> draw lines with a border, dashed lines with a continuous background, > >> and several other pretty things. > > > > We did not add this before out of two ideas: > > a) we wanted to avoid making Thuban a layout application as vector > > ? drawing applications usually do a better job at this. > > ? This is why I came up with the printing pipeline > > ? ? ? ?Thuban -> MapSVG -> Vector drawing application > > b) For trying to explore data, some simpler drawings might > > ? be better (we do not know for sure). > > > > Technically I believe that the old wxWidgets drawing engine might also > > pose some limits on what could be done easily. I read somewhere that I > > would have a new system, but did not followup up when this was included. > > I think the wx map renderer is an *excellent* piece of software, I'm > impressed at just how fast it works. > But yes, maybe at some point it would be good to start using a more > featureful backend. Are there other backend available with wx? Okay there is OpenGL, but I always thought that drawing maps should work well without resorting to something like it. > While talking about features that probably need a major rewrite, let > me add these to the list :-) > - Antialias Would be a wx internal feature, right? > - Line width zooming (showing lines wider when you get closer) > - Rotated labels Those would be within Thuban, both look reasonable to me. > > Also there is much more computing power today so doing better drawings > > should be fine. > > > > If you need those features, by all means, give them a coding attempt. :) > > Wee :-D >;-) -- Managing Director - Owner: www.intevation.net (Free Software Company) Germany Coordinator: fsfeurope.org. Coordinator: www.Kolab-Konsortium.com. Intevation GmbH, Osnabr?ck, DE; Amtsgericht Osnabr?ck, HRB 18998 Gesch?ftsf?hrer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080716/3a35d82f/attachment.bin From thuban-patches at wald.intevation.org Sat Jul 19 23:43:10 2008 From: thuban-patches at wald.intevation.org (thuban-patches@wald.intevation.org) Date: Sat, 19 Jul 2008 23:43:10 +0200 (CEST) Subject: =?UTF-8?B?W3RodWJhbi1QYXRjaGVzXVs2ODhdIEJ1Z2ZpeGVzIGZvciB1bW5fbWFwc2VydmVyIGV4dGVuc2lvbg==?= Message-ID: <20080719214310.06E8A406CC@pyrosoma.intevation.org> Patches item #688, was opened at 2008-07-19 18:43 Status: Open Priority: 3 Submitted By: Anthony Lenton (elachuni) Assigned to: Nobody (None) Summary: Bugfixes for umn_mapserver extension Version: None Category: None Initial Comment: Hello, I'm using mapscript 5.0.0 here and I get the following failures when I run the umn_mapserver extension tests: anthony at daphne:~/svn/thuban/trunk/thuban/Extensions/umn_mapserver/test$ python test_mapserver.py ..F...FE........ ====================================================================== ERROR: test_MF_Scalebar (__main__.mapserver_Classes) ---------------------------------------------------------------------- Traceback (most recent call last): File "test_mapserver.py", line 234, in test_MF_Scalebar self.eq(testscalebar.get_position(mode="string"),"ll") File "/home/anthony/svn/thuban/trunk/thuban/Extensions/umn_mapserver/mapfile.py", line 786, in get_position return scalebar_position_type[self._scalebar.position] KeyError: 104 ====================================================================== FAIL: test_MF_Layer (__main__.mapserver_Classes) ---------------------------------------------------------------------- Traceback (most recent call last): File "test_mapserver.py", line 198, in test_MF_Layer self.eq(testlayer.get_projection().get_projection(), "+proj=latlong+ellps=clrk66") AssertionError: '+proj=latlong +ellps=clrk66' != '+proj=latlong+ellps=clrk66' ====================================================================== FAIL: test_MF_Projection (__main__.mapserver_Classes) ---------------------------------------------------------------------- Traceback (most recent call last): File "test_mapserver.py", line 278, in test_MF_Projection self.eq(testproj.get_params(), ['proj=utm','ellps=WGS84','zone=26','north']) AssertionError: ['proj=utm ', 'ellps=WGS84 ', 'zone=26 ', 'north'] != ['proj=utm', 'ellps=WGS84', 'zone=26', 'north'] ---------------------------------------------------------------------- Ran 16 tests in 0.023s FAILED (failures=2, errors=1) anthony at daphne:~/svn/thuban/trunk/thuban/Extensions/umn_mapserver/test$ The attached patch fixes the failing tests. I have the feeling that several of the enumerations at the beginning of mapfile.py should be changed to use the MS_* constants instead of assuming that enumerations start from 0, label_position_type and legend_position_type at least; at the moment there isn't a test that fails for those, but I think that's because of missing tests, not because of working code. ---------------------------------------------------------------------- You can respond by visiting: http://wald.intevation.org/tracker/?func=detail&atid=107&aid=688&group_id=6 From thuban-patches at wald.intevation.org Sun Jul 20 07:09:16 2008 From: thuban-patches at wald.intevation.org (thuban-patches@wald.intevation.org) Date: Sun, 20 Jul 2008 07:09:16 +0200 (CEST) Subject: =?UTF-8?B?W3RodWJhbi1QYXRjaGVzXVs2ODldIFN1cHBvcnQgZm9yIFBvc3RHSVMgbGF5ZXJzIGluIHVtbl9tYXBzZXJ2ZXIgZXhwb3J0?= Message-ID: <20080720050916.2A246406C2@pyrosoma.intevation.org> Patches item #689, was opened at 2008-07-20 02:09 Status: Open Priority: 3 Submitted By: Anthony Lenton (elachuni) Assigned to: Nobody (None) Summary: Support for PostGIS layers in umn_mapserver export Version: None Category: None Initial Comment: The attached patch allows the umn_mapserver extension to correctly handle PostGIS layers, setting the data, connection and connectiontype attributes in the created mapfile. Mapfiles with PostGIS layers can't be imported yet, so it's a bit of a lopsided support, but export-only is probably better than no support at all. ---------------------------------------------------------------------- You can respond by visiting: http://wald.intevation.org/tracker/?func=detail&atid=107&aid=689&group_id=6 From thuban-bugs at wald.intevation.org Mon Jul 21 05:12:52 2008 From: thuban-bugs at wald.intevation.org (thuban-bugs@wald.intevation.org) Date: Mon, 21 Jul 2008 05:12:52 +0200 (CEST) Subject: =?UTF-8?B?W3RodWJhbi1CdWdzXVs2OTBdIEVycm9yIGluIHByb2plY3Rpb24gZGlhbG9n?= Message-ID: <20080721031252.B826E406E5@pyrosoma.intevation.org> Bugs item #690, was opened at 2008-07-21 00:12 Status: Open Priority: 3 Submitted By: Anthony Lenton (elachuni) Assigned to: Nobody (None) Summary: Error in projection dialog Resolution: None Version: None Category: None Initial Comment: I'm not sure I'm not doing something wrong, but I'm pretty sure that the app shouldn't produce a traceback: * Open the app * Go to Map->Projection... * Select a projection, for example "UTM Zone 27" (any other can do). Here on my machine the specific UI for the particular projection (the ellipsoid combo, the zone spinbox and the southern hemisphere checkbox for this example) appears misplaced in the top left corner of the window instead of appearing in the right panel, but that's not what this bug report is about... * Click the "New" button at the right What I would expect to happen: The right panel is cleared and I can create a new projection for my map. What happens instead: An exception occurs, with the following traceback: anthony at daphne:~/svn/thuban/trunk/thuban$ ./thuban.py An unhandled exception occurred: type object 'ProjPanel' has no attribute 'Clear' (please report to http://thuban.intevation.org/bugtracker.html) Traceback (most recent call last): File "/home/anthony/svn/thuban/trunk/thuban/Thuban/UI/projdialog.py", line 263, in _OnNew self.curProjPanel.Clear() File "/home/anthony/svn/thuban/trunk/thuban/Thuban/UI/projdialog.py", line 953, in Clear ProjPanel.Clear(self) AttributeError: type object 'ProjPanel' has no attribute 'Clear' anthony at daphne:~/svn/thuban/trunk/thuban$ ---------------------------------------------------------------------- You can respond by visiting: http://wald.intevation.org/tracker/?func=detail&atid=105&aid=690&group_id=6 From antoniolenton at gmail.com Tue Jul 22 01:10:01 2008 From: antoniolenton at gmail.com (Anthony Lenton) Date: Mon, 21 Jul 2008 20:10:01 -0300 Subject: Thuban at JRSL Message-ID: Hello list! I'm planning on presenting a talk for the upcoming regional Free Software jornadas, JRSL http://jornadas.cafelug.org.ar/8/en/home.php The talk is going to be about GIS in general with a bit of web mapping probably, I haven't decided well, but I wanted to give Thuban a central position in the talk, hopefully to rally more developers in to the project. I wanted to ask your thoughts, if there was anything in particular that you think that's a strong point for Thuban, at the moment or in the not-so-near future, what you've used it successfully for, what other apps you've made it play with. Kind regards, -- Anthony Lenton Except - The good guys in software development http://except.com.ar From thuban-bugs at wald.intevation.org Fri Jul 25 01:42:58 2008 From: thuban-bugs at wald.intevation.org (thuban-bugs@wald.intevation.org) Date: Fri, 25 Jul 2008 01:42:58 +0200 (CEST) Subject: =?UTF-8?B?W3RodWJhbi1CdWdzXVs2OTRdIEEgY291cGxlIG9mIHVuY2F1Z2h0IGV4Y2VwdGlvbnMgd2l0aCB1bW5fbWFwc2VydmVy?= Message-ID: <20080724234258.CC742406E9@pyrosoma.intevation.org> Bugs item #694, was opened at 2008-07-24 20:42 Status: Open Priority: 3 Submitted By: Anthony Lenton (elachuni) Assigned to: Nobody (None) Summary: A couple of uncaught exceptions with umn_mapserver Resolution: None Version: None Category: None Initial Comment: Hello, Playing around with the umn_mapserver extension throws a couple exceptions that probably should be handled with warning dialogs: 1) If you attempt to export the map before setting its size, you get an uncaught exception. To reproduce this traceback: * Open Thuban * Click on Experimental->MapServer->Create new mapfile * Click on Experimental->MapServer->Export mapfile * Enter any filename, and click Ok. An unhandled exception occurred: msCalculateScale(): General error message. Invalid image extent, minx=nan, miny=nan, maxx=nan, maxy=nan. (please report to http://thuban.intevation.org/bugtracker.html) Traceback (most recent call last): File "/home/anthony/svn/thuban/trunk/thuban/Thuban/UI/mainwindow.py", line 300, in invoke_command command.Execute(self.Context()) File "/home/anthony/svn/thuban/trunk/thuban/Thuban/UI/command.py", line 121, in Execute apply(self.function, (context,) + self.args + args, kw) File "/home/anthony/svn/thuban/trunk/thuban/Extensions/umn_mapserver/mf_export.py", line 163, in export_mapfile thuban_to_map(context, theMap) File "/home/anthony/svn/thuban/trunk/thuban/Extensions/umn_mapserver/mf_export.py", line 104, in thuban_to_map map.set_extent(tb_bbox) File "/home/anthony/svn/thuban/trunk/thuban/Extensions/umn_mapserver/mapfile.py", line 1027, in set_extent newextent[2],newextent[3]) File "/usr/lib/python2.5/site-packages/mapscript.py", line 1275, in setExtent def setExtent(*args): return _mapscript.mapObj_setExtent(*args) MapServerError: msCalculateScale(): General error message. Invalid image extent, minx=nan, miny=nan, maxx=nan, maxy=nan. Checking the code in mapfile.py there seems to be an attempt to fix this by doing: if self.get_size()[0] == - 1: print "define the size first to set extent" print "size is now set to (1,1)" self.set_size(1,1) just before setting the map extent. It seems that this isn't enough on all platforms to avoid set_extent to fail. 2) If you attempt to enter non-numeric characters in the size field of the map configuration dialog you get an uncaught exception. To reproduce this: * Open Thuban * Click on Experimental->MapServer->Create new mapfile * Click on Experimental->MapServer->Edit mapfile->Map * Enter anything that isn't a number in width or height field. An unhandled exception occurred: invalid literal for int() with base 10: '-1g' (please report to http://thuban.intevation.org/bugtracker.html) Traceback (most recent call last): File "/home/anthony/svn/thuban/trunk/thuban/Extensions/umn_mapserver/mf_handle.py", line 217, in OnOK self.tb_map_umn.set_size(int(self.text_width.GetValue()), ValueError: invalid literal for int() with base 10: '-1blah' ---------------------------------------------------------------------- You can respond by visiting: http://wald.intevation.org/tracker/?func=detail&atid=105&aid=694&group_id=6 From thuban-bugs at wald.intevation.org Sun Jul 27 07:57:14 2008 From: thuban-bugs at wald.intevation.org (thuban-bugs@wald.intevation.org) Date: Sun, 27 Jul 2008 07:57:14 +0200 (CEST) Subject: =?UTF-8?B?W3RodWJhbi1CdWdzXVs2OTZdIFVuY2F1Z2h0IGV4Y2VwdGlvbiB3aGVuIHRyeWluZyB0byBleHBvcnQgYSBtYXBmaWxlIHR3aWNl?= Message-ID: <20080727055714.EAECE406FB@pyrosoma.intevation.org> Bugs item #696, was opened at 2008-07-27 02:57 Status: Open Priority: 3 Submitted By: Anthony Lenton (elachuni) Assigned to: Nobody (None) Summary: Uncaught exception when trying to export a mapfile twice Resolution: None Version: None Category: None Initial Comment: If you try to export a map using the umn_mapserver extension twice, an uncaught exception occurs. The map needs to have at least one non-default classification for error to occur. To reproduce this bug: * Open thuban * Add a layer to the map, say Iceland's political.shp * Click twice on the only layer in the left bar to bring up the Layer Properties dialog. * Select any Classification field, for example Area. * Add a singleton classification, and set it up any how. Extra points for pretty colors. * Click Ok to close the Layer Properties dialog. * Click on Experimental->MapServer->Create mapfile * Click on Experimental->MapServer->Edit mapfile->Map * Set some width and height (say 50x50) * Click on Experimental->MapServer->Export mapfile * Select any filename and click Ok * Click on Experimental->MapServer->Export mapfile, again * Select any filename (can be the same as before or not) and click Ok. An unhandled exception occurred: removeClass(): Child array error. Cannot remove class, invalid index 1 (please report to http://thuban.intevation.org/bugtracker.html) Traceback (most recent call last): File "/home/anthony/svn/thuban/trunk/thuban/Thuban/UI/mainwindow.py", line 300, in invoke_command command.Execute(self.Context()) File "/home/anthony/svn/thuban/trunk/thuban/Thuban/UI/command.py", line 121, in Execute apply(self.function, (context,) + self.args + args, kw) File "/home/anthony/svn/thuban/trunk/thuban/Extensions/umn_mapserver/mf_export.py", line 163, in export_mapfile thuban_to_map(context, theMap) File "/home/anthony/svn/thuban/trunk/thuban/Extensions/umn_mapserver/mf_export.py", line 105, in thuban_to_map tblayer_to_map(tb_map, map) File "/home/anthony/svn/thuban/trunk/thuban/Extensions/umn_mapserver/mf_export.py", line 73, in tblayer_to_map map.add_thubanlayer(layer) File "/home/anthony/svn/thuban/trunk/thuban/Extensions/umn_mapserver/mapfile.py", line 1071, in add_thubanlayer new_layer.remove_allclasses() File "/home/anthony/svn/thuban/trunk/thuban/Extensions/umn_mapserver/mapfile.py", line 735, in remove_allclasses self._mf_layer.removeClass(i) File "/usr/lib/python2.5/site-packages/mapscript.py", line 1114, in removeClass def removeClass(*args): return _mapscript.layerObj_removeClass(*args) MapServerChildError: removeClass(): Child array error. Cannot remove class, invalid index 1 I believe that this is a problem with the umn_mapserver that keeps track of the classes in the map and doesn't realise when the mapObj is created anew underneath, so it tries to clear the classes on an already clear layer when it comes round to exporting the second time. ---------------------------------------------------------------------- You can respond by visiting: http://wald.intevation.org/tracker/?func=detail&atid=105&aid=696&group_id=6 From bernhard at intevation.de Mon Jul 28 16:50:42 2008 From: bernhard at intevation.de (Bernhard Reiter) Date: Mon, 28 Jul 2008 16:50:42 +0200 Subject: Commits unm and importpmf Message-ID: <200807281650.46191.bernhard@intevation.de> Anthony, just saw a couple of your commits! That is the right way, especially for the exensions you are working on, just fix the stuff or commit if no other developers comments (Especially I am sometimes slow to react, so do not let yourself be stopped by this.) Best, Bernhard -- Managing Director - Owner: www.intevation.net (Free Software Company) Germany Coordinator: fsfeurope.org. Coordinator: www.Kolab-Konsortium.com. Intevation GmbH, Osnabr?ck, DE; Amtsgericht Osnabr?ck, HRB 18998 Gesch?ftsf?hrer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080728/4c9464da/attachment.bin From bernhard at intevation.de Mon Jul 28 16:55:34 2008 From: bernhard at intevation.de (Bernhard Reiter) Date: Mon, 28 Jul 2008 16:55:34 +0200 Subject: Thuban at JRSL In-Reply-To: References: Message-ID: <200807281655.35689.bernhard@intevation.de> Hi Anthony, On Tuesday 22 July 2008 01:10, Anthony Lenton wrote: > ? I'm planning on presenting a talk for the upcoming regional Free > Software jornadas, JRSL http://jornadas.cafelug.org.ar/8/en/home.php > > The talk is going to be about GIS in general with a bit of web mapping > probably, I haven't decided well, but I wanted to give Thuban a > central position in the talk, hopefully to rally more developers in to > the project. wonderful! It is good to see Thuban getting more attention! > I wanted to ask your thoughts, if there was anything in particular > that you think that's a strong point for Thuban, at the moment or in > the not-so-near future, what you've used it successfully for, what > other apps you've made it play with. We have done a nice decision support system called Great-er with it: http://thuban.intevation.org/screenshot-details.html?sc=greater http://great-er.intevation.org/ Thuban is crossplattform and with a clear Free Software Strategy even for windows. It uses best of breed libraries. The table join support is something users like also the speed and extensibility with a strong programming language. We have also ussed Thuban to preprocess a complete map of Osnabrueck. Best, Bernhard -- Managing Director - Owner: www.intevation.net (Free Software Company) Germany Coordinator: fsfeurope.org. Coordinator: www.Kolab-Konsortium.com. Intevation GmbH, Osnabr?ck, DE; Amtsgericht Osnabr?ck, HRB 18998 Gesch?ftsf?hrer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080728/f7811898/attachment.bin From bernhard at intevation.de Mon Jul 28 16:56:22 2008 From: bernhard at intevation.de (Bernhard Reiter) Date: Mon, 28 Jul 2008 16:56:22 +0200 Subject: Thuban at JRSL In-Reply-To: <200807281655.35689.bernhard@intevation.de> References: <200807281655.35689.bernhard@intevation.de> Message-ID: <200807281656.23890.bernhard@intevation.de> On Monday 28 July 2008 16:55, Bernhard Reiter wrote: > > I wanted to ask your thoughts, if there was anything in particular > > that you think that's a strong point for Thuban, at the moment or in > > the not-so-near future, what you've used it successfully for, what > > other apps you've made it play with. You should also use at the thuban-list (not just the development list) or maybe even on the freegis.org list. -- Managing Director - Owner: www.intevation.net (Free Software Company) Germany Coordinator: fsfeurope.org. Coordinator: www.Kolab-Konsortium.com. Intevation GmbH, Osnabr?ck, DE; Amtsgericht Osnabr?ck, HRB 18998 Gesch?ftsf?hrer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080728/84e6ac76/attachment.bin From antoniolenton at gmail.com Wed Jul 30 08:00:24 2008 From: antoniolenton at gmail.com (Anthony Lenton) Date: Wed, 30 Jul 2008 03:00:24 -0300 Subject: New extension: gMapTiles Message-ID: Hello list, I've just committed the first version of a Google Map Tiles exporter extension. For now it allows you to generate tiles for using with the Google Maps API out of you thuban map. I'd like to make it also know how to import data from Google Maps, but that is a bit further down the road. Kind regards, -- Anthony Lenton Except - The good guys in software development http://except.com.ar From bernhard at intevation.de Thu Jul 31 19:31:22 2008 From: bernhard at intevation.de (Bernhard Reiter) Date: Thu, 31 Jul 2008 19:31:22 +0200 Subject: New extension: gMapTiles In-Reply-To: References: Message-ID: <200807311931.26322.bernhard@intevation.de> Anthony, On Wednesday 30 July 2008 08:00, Anthony Lenton wrote: > I've just committed the first version of a Google Map Tiles exporter > extension. ?For now it allows you to generate tiles for using with the > Google Maps API out of you thuban map. ?I'd like to make it also know > how to import data from Google Maps, but that is a bit further down > the road. sound wonderful! :) I haven't used the google maps API, would you care doing a cool screenshot if this is possible and an overview paragraph for people like me? I could include the text in the next Thuban release NEWS File. Best, Bernhard -- Managing Director - Owner: www.intevation.net (Free Software Company) Germany Coordinator: fsfeurope.org. Coordinator: www.Kolab-Konsortium.com. Intevation GmbH, Osnabr?ck, DE; Amtsgericht Osnabr?ck, HRB 18998 Gesch?ftsf?hrer Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20080731/0eb2c79d/attachment.bin