From bh at intevation.de Wed Dec 3 12:12:26 2003 From: bh at intevation.de (Bernhard Herzog) Date: Wed, 03 Dec 2003 12:12:26 +0100 Subject: [Thuban-list] Thuban 1.0rc1 released Message-ID: <6qoeuq17fp.fsf@salmakis.intevation.de> Hi all, I've just released the first release candidate for Thuban 1.0. Download from the usual place: http://thuban.intevation.org/download.html The changes since Thuban 0.9 include: - Faster rendering - Extensions to import APR files, to convert GNS files to shapefiles or to use Thuban as a WMS client, among others. Most of these extensions are somewhat experimental though. - Thuban files written with the Windows version should be readable on Unix systems now in most cases - Support for EPSG projections - A command line switch to enable attribute editing. This is a bit dangerous so it's off by default. What's missing for the final 1.0 release: - Updated translations. The strings to translate should not change anymore till 1.0 so now is the time to get the translations up to date. - There's a bug in the windows binary: The startup file thubanstart.py will not be executed. The reasons are detailed in the bug tracker (https://intevation.de/rt/webrt?serial_num=2241). I'm not sure what we can do about it until 1.0. Since the bug is more a Python 2.2 bug than a Thuban bug the easiest solution may be to build the windows binary with Python 2.3. Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From mlennert at club.worldonline.be Wed Dec 3 14:26:23 2003 From: mlennert at club.worldonline.be (Moritz Lennert) Date: Wed, 3 Dec 2003 14:26:23 +0100 (CET) Subject: [Thuban-list] Thuban 1.0rc1 released In-Reply-To: <6qoeuq17fp.fsf@salmakis.intevation.de> References: <6qoeuq17fp.fsf@salmakis.intevation.de> Message-ID: <37500.164.15.134.155.1070457983.squirrel@moritz.homelinux.org> Bernhard Herzog said: > > Hi all, > > I've just released the first release candidate for Thuban 1.0. Congratulations ! > Download from the usual place: > http://thuban.intevation.org/download.html I used alien to install it on Debian testing/unstable with python2.3. When I open it it gives quite a lot of RuntimeWarnings about API version mismatches: ******************* $ thuban /usr/lib/thuban/Thuban/../Lib/shapelib.py:2: RuntimeWarning: Python C API version mismatch for module shapelibc: This Python has API version 1012, module shapelibc has version 1011. import shapelibc /usr/lib/thuban/Thuban/Model/data.py:21: RuntimeWarning: Python C API version mismatch for module shptree: This Python has API version 1012, module shptree has version 1011. import shptree /usr/lib/thuban/Thuban/../Lib/dbflib.py:2: RuntimeWarning: Python C API version mismatch for module dbflibc: This Python has API version 1012, module dbflibc has version 1011. import dbflibc /usr/lib/thuban/Thuban/../Lib/Projection.py:2: RuntimeWarning: Python C API version mismatch for module Projectionc: This Python has API version 1012, module Projectionc has version 1011. import Projectionc /usr/lib/thuban/Thuban/UI/renderer.py:25: RuntimeWarning: Python C API version mismatch for module wxproj: This Python has API version 1012, module wxproj has version 1011. from wxproj import draw_polygon_shape, draw_polygon_init No thubanstart module available ****************** Then, when I try to open a sample session or a shapefile it segfaults immediately. I don't have this behaviour when I install the official thuban 0.9.0-1 Debian package. Moritz From bh at intevation.de Wed Dec 3 14:41:43 2003 From: bh at intevation.de (Bernhard Herzog) Date: Wed, 03 Dec 2003 14:41:43 +0100 Subject: [Thuban-list] Thuban 1.0rc1 released In-Reply-To: <37500.164.15.134.155.1070457983.squirrel@moritz.homelinux.org> (Moritz Lennert's message of "Wed, 3 Dec 2003 14:26:23 +0100 (CET)") References: <6qoeuq17fp.fsf@salmakis.intevation.de> <37500.164.15.134.155.1070457983.squirrel@moritz.homelinux.org> Message-ID: <6qfzg210iw.fsf@salmakis.intevation.de> "Moritz Lennert" writes: > When I open it it gives quite a lot of RuntimeWarnings about API version > mismatches: > > > ******************* > $ thuban > /usr/lib/thuban/Thuban/../Lib/shapelib.py:2: RuntimeWarning: Python C API > version mismatch for module shapelibc: This Python has API version 1012, > module shapelibc has version 1011. Debian testing uses Python 2.3. The RPM was built with 2.2. hence this version mismatch. Most of the time it's harmless in my experience, though. > Then, when I try to open a sample session or a shapefile it segfaults > immediately. You need to use exactly the same wxWindows and wxPython versions that Thuban was compiled with. It's unlikely that you will get the binary RPM to work on your debian system. The tarball contains the debian build files so you could build your own .deb file. I'm not sure how that's done, though :). However, Silke says she's going to build a new thuban debian package, so there might be an unofficial (i.e. not official from debian) package on our FTP server later today. Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From jan at intevation.de Wed Dec 3 14:42:29 2003 From: jan at intevation.de (Jan-Oliver Wagner) Date: Wed, 3 Dec 2003 14:42:29 +0100 Subject: [Thuban-list] Translations for Thuban. Was: Thuban 1.0rc1 released In-Reply-To: <6qoeuq17fp.fsf@salmakis.intevation.de> References: <6qoeuq17fp.fsf@salmakis.intevation.de> Message-ID: <20031203134229.GC29414@intevation.de> On Wed, Dec 03, 2003 at 12:12:26PM +0100, Bernhard Herzog wrote: > I've just released the first release candidate for Thuban 1.0. > Download from the usual place: > http://thuban.intevation.org/download.html I've updated all relevant web pages and submitted an update to freshmeat. So now it is upto the translators to updated the corresponding languages as there should be no string changes anymore for the final release 1.0. So far we had the following translators: French: Daniel Calvelo Aros German : Bj?rn Broscheit (already up-to-date) Italian: Maurizio Napolitano Russian: Alex Shevlakov Spanish: Daniel Calvelo Aros Please let us know if you don't think you can deliver an updated translation until mid of december. Are there any volunteers for further languages? All the best Jan -- Jan-Oliver Wagner http://intevation.de/~jan/ Intevation GmbH http://intevation.de/ FreeGIS http://freegis.org/ From mlennert at club.worldonline.be Wed Dec 3 14:59:01 2003 From: mlennert at club.worldonline.be (Moritz Lennert) Date: Wed, 3 Dec 2003 14:59:01 +0100 (CET) Subject: [Thuban-list] Thuban 1.0rc1 released In-Reply-To: <6qfzg210iw.fsf@salmakis.intevation.de> References: <6qoeuq17fp.fsf@salmakis.intevation.de><37500.164.15.134.155.1070457983.squirrel@moritz.homelinux.org> <6qfzg210iw.fsf@salmakis.intevation.de> Message-ID: <37801.164.15.134.155.1070459941.squirrel@moritz.homelinux.org> Bernhard Herzog said: > > The tarball contains the debian build files so you could build your own > .deb file. I'm not sure how that's done, though :). However, Silke says > she's going to build a new thuban debian package, so there might be an > unofficial (i.e. not official from debian) package on our FTP server > later today. > Ok, I'll wait for that. Thanks, Moritz From napo at itc.it Wed Dec 3 18:19:18 2003 From: napo at itc.it (Maurizio Napolitano) Date: Wed, 03 Dec 2003 18:19:18 +0100 Subject: [Thuban-list] Translations for Thuban. Was: Thuban 1.0rc1 released In-Reply-To: <20031203134229.GC29414@intevation.de> References: <6qoeuq17fp.fsf@salmakis.intevation.de> <20031203134229.GC29414@intevation.de> Message-ID: <1070471957.16300.76.camel@toriamos> > Italian: Maurizio Napolitano > Please let us know if you don't think you can deliver an updated > translation until mid of december. I have time after the 19 december. Probaly i can traslate the program before, but now i have to solve the problem to run the last version of Thuban. From Silke.Reimer at intevation.de Wed Dec 3 19:56:30 2003 From: Silke.Reimer at intevation.de (Silke Reimer) Date: Wed, 3 Dec 2003 19:56:30 +0100 Subject: [Thuban-list] Thuban 1.0rc1 released In-Reply-To: <37801.164.15.134.155.1070459941.squirrel@moritz.homelinux.org> References: <6qfzg210iw.fsf@salmakis.intevation.de> <37801.164.15.134.155.1070459941.squirrel@moritz.homelinux.org> Message-ID: <20031203185630.GE11542@intevation.de> Hallo, On Wed, Dec 03, 2003 at 02:59:01PM +0100, Moritz Lennert wrote: > Bernhard Herzog said: > > > > The tarball contains the debian build files so you could build your own > > .deb file. I'm not sure how that's done, though :). However, Silke says > > she's going to build a new thuban debian package, so there might be an > > unofficial (i.e. not official from debian) package on our FTP server > > later today. > > > > Ok, I'll wait for that. As promised by Bernhard earlier this day I built an debian package of thuban. You can download it from ftp://intevation.de/thuban/debian/thuban_1.0rc1-0intevation_i386.deb It has been built on Debian-Unstable (sid). Please note, that this is an unofficial debian package. The official one is in preparation though and will hopefully appear soon in the debian archives. Greetings, Silke -- Silke Reimer Intevation GmbH http://intevation.de/ FreeGIS http://freegis.org/ -------------- 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-list/attachments/20031203/6c48b5ad/attachment.bin From jan at intevation.de Wed Dec 3 20:28:55 2003 From: jan at intevation.de (Jan-Oliver Wagner) Date: Wed, 3 Dec 2003 20:28:55 +0100 Subject: [Thuban-list] Translations for Thuban. Was: Thuban 1.0rc1 released In-Reply-To: <1070471957.16300.76.camel@toriamos> References: <6qoeuq17fp.fsf@salmakis.intevation.de> <20031203134229.GC29414@intevation.de> <1070471957.16300.76.camel@toriamos> Message-ID: <20031203192855.GB4118@intevation.de> On Wed, Dec 03, 2003 at 06:19:18PM +0100, Maurizio Napolitano wrote: > > Italian: Maurizio Napolitano > > > Please let us know if you don't think you can deliver an updated > > translation until mid of december. > > I have time after the 19 december. > Probaly i can traslate the program before, but now i have to solve > the problem to run the last version of Thuban. please let us know about any problems with 1.0rc1. It is the idea of a release candidate to iron out any remaining problem. Best Jan -- Jan-Oliver Wagner http://intevation.de/~jan/ Intevation GmbH http://intevation.de/ FreeGIS http://freegis.org/ From matt.wilkie at gov.yk.ca Wed Dec 3 20:55:10 2003 From: matt.wilkie at gov.yk.ca (Matt Wilkie) Date: Wed, 03 Dec 2003 11:55:10 -0800 Subject: [Thuban-list] thuban, windows and python 2.3 Message-ID: <3FCE3F9E.5040104@gov.yk.ca> Hello, It's good to see that a full on 1.0 release is approaching. Congratulations to everybody for their hard work! Can the windows binary be used with python 2.3? If not is a py2.3 win-binary forthcoming? I'd rather not have 2.2 and 2.3 concurrent if I can avoid it (or maybe that is not even possible?) thanks in advance, -- matt wilkie -------------------------------------------- Geographic Information, Information Management and Technology, Yukon Department of Environment 10 Burns Road * Whitehorse, Yukon * Y1A 4Y9 867-667-8133 Tel * 867-393-7003 Fax -------------------------------------------- From napo at itc.it Thu Dec 4 09:13:36 2003 From: napo at itc.it (Maurizio Napolitano) Date: Thu, 04 Dec 2003 09:13:36 +0100 Subject: [Thuban-list] Translations for Thuban. Was: Thuban 1.0rc1 released In-Reply-To: <20031203192855.GB4118@intevation.de> References: <6qoeuq17fp.fsf@salmakis.intevation.de> <20031203134229.GC29414@intevation.de> <1070471957.16300.76.camel@toriamos> <20031203192855.GB4118@intevation.de> Message-ID: <1070525615.16300.87.camel@toriamos> > please let us know about any problems with 1.0rc1. > It is the idea of a release candidate to iron out any remaining Yesterday, before to send my email, i write a messagge about the problems that i have on the last Thuban version. Probaly the anti-spam filter of the sendmail server where i work has block the messagge (is a very bad antispam filter!!!) I copy and paste my words here. ---- Problems with Thuban I have tested it on two computer. A computer is based on a RedHat 9.0 and is a Intel P4 2.4 Ghz with 1Ghz of RAM. I have "compiled" Thuban from the source. The program work but the command to change the color of a shape by "class" (range or class) is disable. If i have load a "Thuban session" with a layer with this features, the display is right ... but the command is always disabled. On another computer (my laptop) with a Mandrake 9.2 (Intel P4 Mobile 2.2Ghz and 512Mb ram) if have compiled Thuban from the source (i have'nt used the RPM for RH) and the result is a "core dump" if i try to load some ESRI Shape file. In this case i have the python binding for GDAL and the psycopg but the command for the "Database connection" is broken. Any hint? PS: sorry for my bad english!!! --- From mlennert at club.worldonline.be Thu Dec 4 10:22:07 2003 From: mlennert at club.worldonline.be (Moritz Lennert) Date: Thu, 4 Dec 2003 10:22:07 +0100 (CET) Subject: [Thuban-list] visual classifier Message-ID: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org> On July 11, 2003, Daniel Calvelo wrote: >In the attachment you will find a set of files and a patch against what was >the June 20th CVS. They provide the beginnings of a custom classifier. Let me >elaborate. > >The patch adds a "Custom breaks" option to the class generator. When >selected, >it opens a graphical display of the selected field that includes: a histogram >(a slider controls the number of bins used), a "pointgram" (a point, or >rather >a vertical bar for each data value) and a set of triangular markers >indicating >breaks. These markers can be deleted (double-click) and added (single-click >on > new location). They represent cut points from which a set of ranges will be >generated. Has this been taken any further ? Moritz From jan at intevation.de Thu Dec 4 10:26:41 2003 From: jan at intevation.de (Jan-Oliver Wagner) Date: Thu, 4 Dec 2003 10:26:41 +0100 Subject: [Thuban-list] visual classifier In-Reply-To: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org> References: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org> Message-ID: <20031204092641.GB30424@intevation.de> On Thu, Dec 04, 2003 at 10:22:07AM +0100, Moritz Lennert wrote: > Has this been taken any further ? it looked promising but was a bit buggy. Don't know whether Daniel worked further on it. -- Jan-Oliver Wagner http://intevation.de/~jan/ Intevation GmbH http://intevation.de/ FreeGIS http://freegis.org/ From dcalvelo at minag.gob.pe Thu Dec 4 14:41:44 2003 From: dcalvelo at minag.gob.pe (Daniel Calvelo) Date: Thu, 4 Dec 2003 08:41:44 -0500 Subject: [Thuban-list] visual classifier In-Reply-To: <20031204092641.GB30424@intevation.de> References: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org> <20031204092641.GB30424@intevation.de> Message-ID: <20031204133653.M91206@minag.gob.pe> Hi all. Sorry, I haven't had time to work on the thing. Furthermore, the current infrastructure should be overhauled to accomodate this kind of "extension", and the whole will have to be rewritten when transparency is considered. If there is interest, I could update the patch to work with recent CVS, and maybe add the most urgent features. (Yes, I'm very happy somebody's actually interested!) Cheers, Daniel. -- Daniel Calvelo Aros -- Direcci?n General de Informaci?n Agraria -- Ministerio de Agricultura del Per? -- (51-1) 424-9001 ---------- Original Message ----------- From: Jan-Oliver Wagner To: Thuban Mailing List Sent: Thu, 4 Dec 2003 10:26:41 +0100 Subject: Re: [Thuban-list] visual classifier > On Thu, Dec 04, 2003 at 10:22:07AM +0100, Moritz Lennert wrote: > > Has this been taken any further ? > > it looked promising but was a bit buggy. Don't know whether Daniel > worked further on it. > > -- > Jan-Oliver Wagner http://intevation.de/~jan/ > > Intevation GmbH http://intevation.de/ > FreeGIS http://freegis.org/ > > _______________________________________________ > Thuban-list mailing list > Thuban-list at intevation.de > https://intevation.de/mailman/listinfo/thuban-list ------- End of Original Message ------- From jan at intevation.de Thu Dec 4 14:49:02 2003 From: jan at intevation.de (Jan-Oliver Wagner) Date: Thu, 4 Dec 2003 14:49:02 +0100 Subject: [Thuban-list] visual classifier In-Reply-To: <20031204133653.M91206@minag.gob.pe> References: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org> <20031204092641.GB30424@intevation.de> <20031204133653.M91206@minag.gob.pe> Message-ID: <20031204134902.GC30714@intevation.de> Hi Daniel, On Thu, Dec 04, 2003 at 08:41:44AM -0500, Daniel Calvelo wrote: > Sorry, I haven't had time to work on the thing. > Furthermore, the current infrastructure should be overhauled to accomodate > this kind of "extension", and the whole will have to be rewritten when > transparency is considered. > > If there is interest, I could update the patch to work with recent CVS, and > maybe add the most urgent features. (Yes, I'm very happy somebody's actually > interested!) if it is not too much work for you (or you just love this work :-) it would be nice to have an update for current CVS. If possible, an extension would be nice, but I am not 100% sure right now whether this is achievable. An extension could go immediately to CVS (marked as experimental). Best Jan -- Jan-Oliver Wagner http://intevation.de/~jan/ Intevation GmbH http://intevation.de/ FreeGIS http://freegis.org/ From bh at intevation.de Thu Dec 4 16:07:27 2003 From: bh at intevation.de (Bernhard Herzog) Date: Thu, 04 Dec 2003 16:07:27 +0100 Subject: [Thuban-list] Translations for Thuban. Was: Thuban 1.0rc1 released In-Reply-To: <1070525615.16300.87.camel@toriamos> (Maurizio Napolitano's message of "Thu, 04 Dec 2003 09:13:36 +0100") References: <6qoeuq17fp.fsf@salmakis.intevation.de> <20031203134229.GC29414@intevation.de> <1070471957.16300.76.camel@toriamos> <20031203192855.GB4118@intevation.de> <1070525615.16300.87.camel@toriamos> Message-ID: <6qptf4vcy8.fsf@salmakis.intevation.de> Maurizio Napolitano writes: > A computer is based on a RedHat 9.0 and is a Intel P4 2.4 Ghz with 1Ghz > of RAM. > I have "compiled" Thuban from the source. The program work but the > command to change the color of a shape by "class" (range or class) is > disable. Which command do you mean exactly? The "Properties..." command in the layer menu? That one is only available when a layer is selected. Click in a layer in the legend window to select it. > On another computer (my laptop) with a Mandrake 9.2 (Intel P4 Mobile > 2.2Ghz and 512Mb ram) if have compiled Thuban from the source (i have'nt > used the RPM for RH) and the result is a "core dump" if i try to load > some ESRI Shape file. This kind of segfault usually comes from a binary incompatibility between Thuban, wxWindows and wxPython. One possible cause is that the wx-config and hence the headerfiles etc. used during Thuban compilation is from a different wxWindows version than the one used at runtime. > In this case i have the python binding for GDAL and the psycopg but the > command for the "Database connection" is broken. Does the About-box show a version for psycopg? Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From mlennert at club.worldonline.be Thu Dec 4 17:37:23 2003 From: mlennert at club.worldonline.be (Moritz Lennert) Date: Thu, 4 Dec 2003 17:37:23 +0100 (CET) Subject: [Thuban-list] visual classifier In-Reply-To: <20031204133653.M91206@minag.gob.pe> References: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org><20031204092641.GB30424@intevation.de> <20031204133653.M91206@minag.gob.pe> Message-ID: <35404.164.15.134.155.1070555843.squirrel@moritz.homelinux.org> Daniel Calvelo said: > Hi all. > > Sorry, I haven't had time to work on the thing. > > Furthermore, the current infrastructure should be overhauled to accomodate > this kind of "extension", and the whole will have to be rewritten when > transparency is considered. > > If there is interest, I could update the patch to work with recent CVS, > and > maybe add the most urgent features. (Yes, I'm very happy somebody's > actually > interested!) I am very interested. In the aim of getting my department to slowly move towards free software solutions, a cartography solution is one of the main issues. And data classification is an important element in that. It would actually be ideal to have an automatic "natural breaks" classification available in addition to the current uniform distribution and quantiles. When you said: > and a set of triangular markers indicating > breaks. These markers can be deleted (double-click) and added > (single-click on new location). They represent cut points from which a set > of ranges will be generated. does that mean that these markers are set automatically at the beginning ? How do you define these breaks ? Moritz From dcalvelo at minag.gob.pe Thu Dec 4 18:47:00 2003 From: dcalvelo at minag.gob.pe (Daniel Calvelo) Date: Thu, 4 Dec 2003 12:47:00 -0500 Subject: [Thuban-list] visual classifier References: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org><20031204092641.GB30424@intevation.de> <20031204133653.M91206@minag.gob.pe> <35404.164.15.134.155.1070555843.squirrel@moritz.homelinux.org> Message-ID: <20031204174649.M72607@minag.gob.pe> ["Moritz Lennert"] > When you said: > > > and a set of triangular markers indicating > > breaks. These markers can be deleted (double-click) and added > > (single-click on new location). They represent cut points from which a set > > of ranges will be generated. > > does that mean that these markers are set automatically at the > beginning ? How do you define these breaks ? Yes, they are set initially as evenly spaced. I looked hard at how you could initialize them using the other classifiers, but the infrastructure ties very closely the classifiers with the class definitions, and I couldn't find an elegant, unobtrusive solution. My idea was to have a permanent View of the classifier (in the cute world, it would also have visual cues for the corresponding drawing style), in which markers could be gridded, reset to automatic settings (quantiles, even breaks, "natural" breaks,...), hand-tuned either visually or by input, and so on. A kind of classification studio if you wish. Generating the theme classes would be either a side-effect or an extra action for this subsystem. I got stuck by wxWindows too. I couldn't produce something correct that gave me sliders that I could drag and drop. Could wx-savvy people give any hint on that? I also explored natural breaks. I found very scarce literature on the subject. >From what I gathered, natural breaks are based on a kernel approximation to the data distribution. From previous experience, programming these things needs tuning and lots of special casing. If you have ideas on the subject, they are most welcome. Daniel. From warmerdam at pobox.com Thu Dec 4 15:09:47 2003 From: warmerdam at pobox.com (Frank Warmerdam) Date: Thu, 04 Dec 2003 09:09:47 -0500 Subject: [Thuban-list] Re: Shapelib: dbfopen.c addition In-Reply-To: <20031204140323.A32479@itc.it> References: <20031204140323.A32479@itc.it> Message-ID: <3FCF402B.1050900@pobox.com> Markus Neteler wrote: > Frank, > > while trying to get Thuban (from Intevation) running, > I found that they have added a function to dbfopen.c in shapelib. > Maybe interesting for you to add into CVS? > Author should be Bernhard Herzog. > > Function attached. > > Best regards > > Markus > > > > ------------------------------------------------------------------------ > > /************************************************************************/ > /* DBFCommit() */ > /* */ > /* Write any changes made into the file. */ > /* */ > /************************************************************************/ > int SHPAPI_CALL > DBFCommit( DBFHandle psDBF ) > > { > DBFFlushRecord( psDBF ); > if (fflush( psDBF->fp ) == EOF) > return FALSE; > > return TRUE; > } Markus, I see there is already a DBFUpdateHeader() function which accomplishing pretty much the same as DBFCommit(), but also ensures that the header is in a consistent state (the record count appears there). I think it would be best if Thuban used that instead if possible. I have cc:ed the Thuban list in case they want to change their code. DBFUpdateHeader() looks like: /************************************************************************/ /* DBFUpdateHeader() */ /************************************************************************/ void SHPAPI_CALL DBFUpdateHeader( DBFHandle psDBF ) { unsigned char abyFileHeader[32]; if( psDBF->bNoHeader ) DBFWriteHeader( psDBF ); DBFFlushRecord( psDBF ); fseek( psDBF->fp, 0, 0 ); fread( abyFileHeader, 32, 1, psDBF->fp ); abyFileHeader[4] = (unsigned char) (psDBF->nRecords % 256); abyFileHeader[5] = (unsigned char) ((psDBF->nRecords/256) % 256); abyFileHeader[6] = (unsigned char) ((psDBF->nRecords/(256*256)) % 256); abyFileHeader[7] = (unsigned char) ((psDBF->nRecords/(256*256*256)) % 256); fseek( psDBF->fp, 0, 0 ); fwrite( abyFileHeader, 32, 1, psDBF->fp ); fflush( psDBF->fp ); } Best regards, -- ---------------------------------------+-------------------------------------- I set the clouds in motion - turn up | Frank Warmerdam, warmerdam at pobox.com light and sound - activate the windows | http://pobox.com/~warmerdam and watch the world go round - Rush | Geospatial Programmer for Rent From mlennert at club.worldonline.be Thu Dec 4 19:41:34 2003 From: mlennert at club.worldonline.be (Moritz Lennert) Date: Thu, 4 Dec 2003 19:41:34 +0100 (CET) Subject: [Thuban-list] visual classifier In-Reply-To: <20031204174649.M72607@minag.gob.pe> References: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org><20031204092641.GB30424@intevation.de><20031204133653.M91206@minag.gob.pe><35404.164.15.134.155.1070555843.squirrel@moritz.homelinux.org> <20031204174649.M72607@minag.gob.pe> Message-ID: <36217.164.15.134.155.1070563294.squirrel@moritz.homelinux.org> Daniel Calvelo said: > > I also explored natural breaks. I found very scarce literature on the > subject. >>From what I gathered, natural breaks are based on a kernel approximation >> to > the data distribution. From previous experience, programming these things > needs tuning and lots of special casing. If you have ideas on the subject, > they are most welcome. > We have a fortran program here which was written ages ago and which uses a geometrical approach, exploring the slope between data points. If you are interested I can send it to you (I will have to check with the authors, but I would be very surprised if they weren't willing to release the code freely). Moritz From dcalvelo at minag.gob.pe Thu Dec 4 19:59:18 2003 From: dcalvelo at minag.gob.pe (Daniel Calvelo) Date: Thu, 4 Dec 2003 13:59:18 -0500 Subject: [Thuban-list] visual classifier In-Reply-To: <36217.164.15.134.155.1070563294.squirrel@moritz.homelinux.org> References: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org><20031204092641.GB30424@intevation.de><20031204133653.M91206@minag.gob.pe><35404.164.15.134.155.1070555843.squirrel@moritz.homelinux.org> <20031204174649.M72607@minag.gob.pe> <36217.164.15.134.155.1070563294.squirrel@moritz.homelinux.org> Message-ID: <20031204185536.M88053@minag.gob.pe> If you could get the code, I will try to include the algorithm. It might be better to first implement it as another classifier and then work on the visual part. Do you reckon any licence issues? Daniel. -- Daniel Calvelo Aros -- Direcci?n General de Informaci?n Agraria -- Ministerio de Agricultura del Per? -- (51-1) 424-9001 ---------- Original Message ----------- From: "Moritz Lennert" To: "Daniel Calvelo" Cc: thuban-list at intevation.de Sent: Thu, 4 Dec 2003 19:41:34 +0100 (CET) Subject: Re: [Thuban-list] visual classifier > Daniel Calvelo said: > > > > > I also explored natural breaks. I found very scarce literature on the > > subject. > >>From what I gathered, natural breaks are based on a kernel approximation > >> to > > the data distribution. From previous experience, programming these things > > needs tuning and lots of special casing. If you have ideas on the subject, > > they are most welcome. > > > > We have a fortran program here which was written ages ago and which > uses a geometrical approach, exploring the slope between data > points. If you are interested I can send it to you (I will have to > check with the authors, but I would be very surprised if they > weren't willing to release the code freely). > > Moritz ------- End of Original Message ------- From bh at intevation.de Thu Dec 4 20:16:23 2003 From: bh at intevation.de (Bernhard Herzog) Date: Thu, 04 Dec 2003 20:16:23 +0100 Subject: [Thuban-list] Re: Shapelib: dbfopen.c addition In-Reply-To: <3FCF402B.1050900@pobox.com> (Frank Warmerdam's message of "Thu, 04 Dec 2003 09:09:47 -0500") References: <20031204140323.A32479@itc.it> <3FCF402B.1050900@pobox.com> Message-ID: <6qd6b4v1fc.fsf@salmakis.intevation.de> Frank Warmerdam writes: > I see there is already a DBFUpdateHeader() function which accomplishing > pretty much the same as DBFCommit(), but also ensures that the header is > in a consistent state (the record count appears there). I ran into that (wrong record count when appending records) when I hacked on the drawshape extension :) > I think it would be best if Thuban used that instead if possible. Indeed. This is only available in CVS so far AFAICT. Is it safe to simply copy that function into our code or would it be better to use dbfopen.c from CVS? Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From warmerdam at pobox.com Thu Dec 4 20:31:56 2003 From: warmerdam at pobox.com (Frank Warmerdam) Date: Thu, 04 Dec 2003 14:31:56 -0500 Subject: [Thuban-list] Re: Shapelib: dbfopen.c addition In-Reply-To: <6qd6b4v1fc.fsf@salmakis.intevation.de> References: <20031204140323.A32479@itc.it> <3FCF402B.1050900@pobox.com> <6qd6b4v1fc.fsf@salmakis.intevation.de> Message-ID: <3FCF8BAC.4080006@pobox.com> Bernhard Herzog wrote: > Frank Warmerdam writes: > > >>I see there is already a DBFUpdateHeader() function which accomplishing >>pretty much the same as DBFCommit(), but also ensures that the header is >>in a consistent state (the record count appears there). > > > I ran into that (wrong record count when appending records) when I > hacked on the drawshape extension :) > > >>I think it would be best if Thuban used that instead if possible. > > > Indeed. This is only available in CVS so far AFAICT. Is it safe to > simply copy that function into our code or would it be better to use > dbfopen.c from CVS? Bernhard, The best would likely to be grab dbfopen.c, shpopen.c, and shapefil.h from CVS though it should be fine to just copy in that one function and the corresponding prototype. I do intend to do a new shapelib release sometime soon but haven't really gotten around to it. Best regards, -- ---------------------------------------+-------------------------------------- I set the clouds in motion - turn up | Frank Warmerdam, warmerdam at pobox.com light and sound - activate the windows | http://pobox.com/~warmerdam and watch the world go round - Rush | Geospatial Programmer for Rent From frank.koormann at intevation.de Fri Dec 5 09:50:42 2003 From: frank.koormann at intevation.de (Frank Koormann) Date: Fri, 5 Dec 2003 09:50:42 +0100 Subject: [Thuban-list] visual classifier In-Reply-To: <36217.164.15.134.155.1070563294.squirrel@moritz.homelinux.org> References: <20031204174649.M72607@minag.gob.pe> <36217.164.15.134.155.1070563294.squirrel@moritz.homelinux.org> Message-ID: <20031205085042.GA22952@intevation.de> * Moritz Lennert [031204 19:41]: > Daniel Calvelo said: > > > > > I also explored natural breaks. I found very scarce literature on the > > subject. > >>From what I gathered, natural breaks are based on a kernel approximation > >> to > > the data distribution. From previous experience, programming these things > > needs tuning and lots of special casing. If you have ideas on the subject, > > they are most welcome. > > > > We have a fortran program here which was written ages ago and which uses a > geometrical approach, exploring the slope between data points. If you are > interested I can send it to you (I will have to check with the authors, > but I would be very surprised if they weren't willing to release the code > freely). Just for completeness or if some problems arise from using the fortran code: The original paper on natural breaks ("Jenks optimization") is "Error on Choroplethic Maps: Definitions, Measurement, Reduction", G. F. Jenks & F. C. Caspall, Annals of the Association of American Geographers 61 (2), pp. 217-244 (1971) Should be available via JSTOR (http://www.jstor.org/journals/00045608.html) if somebody has an account there. Best regards, Frank -- Frank Koormann Professional Service around Free Software (http://intevation.net/) FreeGIS Project (http://freegis.org/) From mlennert at club.worldonline.be Fri Dec 5 10:00:27 2003 From: mlennert at club.worldonline.be (Moritz Lennert) Date: Fri, 5 Dec 2003 10:00:27 +0100 (CET) Subject: [Thuban-list] visual classifier In-Reply-To: <20031205085042.GA22952@intevation.de> References: <20031204174649.M72607@minag.gob.pe><36217.164.15.134.155.1070563294.squirrel@moritz.homelinux.org> <20031205085042.GA22952@intevation.de> Message-ID: <37311.164.15.134.155.1070614827.squirrel@moritz.homelinux.org> Frank Koormann said: > * Moritz Lennert [031204 19:41]: >> Daniel Calvelo said: >> >> > >> > I also explored natural breaks. I found very scarce literature on the >> > subject. >> >>From what I gathered, natural breaks are based on a kernel >> approximation >> >> to >> > the data distribution. From previous experience, programming these >> things >> > needs tuning and lots of special casing. If you have ideas on the >> subject, >> > they are most welcome. >> > >> >> We have a fortran program here which was written ages ago and which uses >> a >> geometrical approach, exploring the slope between data points. If you >> are >> interested I can send it to you (I will have to check with the authors, >> but I would be very surprised if they weren't willing to release the >> code >> freely). > > Just for completeness or if some problems arise from using the fortran > code: The original paper on natural breaks ("Jenks > optimization") is "Error on Choroplethic Maps: Definitions, > Measurement, Reduction", G. F. Jenks & F. C. Caspall, Annals of the > Association of American Geographers 61 (2), pp. 217-244 (1971) > > Should be available via JSTOR > (http://www.jstor.org/journals/00045608.html) > if somebody has an account there. > I do and I downloaded the article if anyone is interested. Moritz From dcalvelo at minag.gob.pe Sat Dec 6 02:25:08 2003 From: dcalvelo at minag.gob.pe (Daniel Calvelo) Date: Fri, 5 Dec 2003 20:25:08 -0500 Subject: [Thuban-list] visual classifier In-Reply-To: <37311.164.15.134.155.1070614827.squirrel@moritz.homelinux.org> References: <20031204174649.M72607@minag.gob.pe><36217.164.15.134.155.1070563294.squirrel@moritz.homelinux.org> <20031205085042.GA22952@intevation.de> <37311.164.15.134.155.1070614827.squirrel@moritz.homelinux.org> Message-ID: <20031206012209.M23935@minag.gob.pe> I am interested! BTW I looked into the fortran code you sent me. AFAICT, it's based on recursive partitionning of the dataset according to a measure of the goodness-of-fit with straight lines. I'm translating it into Python to test further; I'll keep the list informed. I'll try it also with K-means: a long time ago I did that myself. Thanks for the feedback!!! Daniel. -- Daniel Calvelo Aros -- Direcci?n General de Informaci?n Agraria -- Ministerio de Agricultura del Per? -- (51-1) 424-9001 ---------- Original Message ----------- From: "Moritz Lennert" To: "Frank Koormann" Cc: thuban-list at intevation.de, "Daniel Calvelo" Sent: Fri, 5 Dec 2003 10:00:27 +0100 (CET) Subject: Re: [Thuban-list] visual classifier > Frank Koormann said: > > * Moritz Lennert [031204 19:41]: > >> Daniel Calvelo said: > >> > >> > > >> > I also explored natural breaks. I found very scarce literature on the > >> > subject. > >> >>From what I gathered, natural breaks are based on a kernel > >> approximation > >> >> to > >> > the data distribution. From previous experience, programming these > >> things > >> > needs tuning and lots of special casing. If you have ideas on the > >> subject, > >> > they are most welcome. > >> > > >> > >> We have a fortran program here which was written ages ago and which uses > >> a > >> geometrical approach, exploring the slope between data points. If you > >> are > >> interested I can send it to you (I will have to check with the authors, > >> but I would be very surprised if they weren't willing to release the > >> code > >> freely). > > > > Just for completeness or if some problems arise from using the fortran > > code: The original paper on natural breaks ("Jenks > > optimization") is "Error on Choroplethic Maps: Definitions, > > Measurement, Reduction", G. F. Jenks & F. C. Caspall, Annals of the > > Association of American Geographers 61 (2), pp. 217-244 (1971) > > > > Should be available via JSTOR > > (http://www.jstor.org/journals/00045608.html) > > if somebody has an account there. > > > > I do and I downloaded the article if anyone is interested. > > Moritz ------- End of Original Message ------- From putler at sauder.ubc.ca Thu Dec 4 23:38:51 2003 From: putler at sauder.ubc.ca (Dan Putler) Date: Thu, 04 Dec 2003 14:38:51 -0800 Subject: [Thuban-list] visual classifier In-Reply-To: <20031204174649.M72607@minag.gob.pe> References: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org> <20031204092641.GB30424@intevation.de> <20031204133653.M91206@minag.gob.pe> <35404.164.15.134.155.1070555843.squirrel@moritz.homelinux.org> <20031204174649.M72607@minag.gob.pe> Message-ID: <200312041438510980.01145E7F@exchange.commerce.ubc.ca> Hi Daniel, The "natural breaks" method as done in ArcView is also known as Jenks Method. One citation that might help in your literature search (given the references it contains) is Jenks, George F. and Fred C. Caspall (1971), "Error in Choroplethic Maps: Definition, Measurement, Reduction," Annals of the Association of American Geographers 61 (2, June), 217-244. Jenks methods is actually based on a method proposed by Walter Fisher. The cite to Fisher's work is Fisher, Walter D. (1958), "On Grouping for Maximum Homogeneity," American Statistical Assoication Journal, December. Fisher's method turns out to be pretty computationally intensive, so my guess is that ESRI actually doesn't use it in ArcView. Ultimately, "natural breaks" amounts to single variable cluster analysis. Given this, a K-Means cluster analysis algorithm could be used. Another approach (that is used as an alternative approach to Fisher's method in xlstat, a set of add-in multivariate statistics macros of Excel) is an algorithm proposed by Anderberg (Anderberg M.R. (1973). Cluster analysis for applications. Academic Press, New York). A benefit to using K-Means to create the "natural breaks" is that the SciPy package already contains a K-Means method function written in Python. Although, it may require bringing in a lot of bagage to Thuban. Some thoughts. Given the level of detail, I've been thinking about this. I wrote two classifiers for OpenEV, and wanted to include a natural breaks classifier. However, I haven't had time to work on it, and don't think I will for awhile. Dan *********** REPLY SEPARATOR *********** On 12/4/2003 at 12:47 PM Daniel Calvelo wrote: >["Moritz Lennert"] >> When you said: >> >> > and a set of triangular markers indicating >> > breaks. These markers can be deleted (double-click) and added >> > (single-click on new location). They represent cut points from which a >set >> > of ranges will be generated. >> >> does that mean that these markers are set automatically at the >> beginning ? How do you define these breaks ? > >Yes, they are set initially as evenly spaced. I looked hard at how you >could >initialize them using the other classifiers, but the infrastructure ties >very >closely the classifiers with the class definitions, and I couldn't find an >elegant, unobtrusive solution. > >My idea was to have a permanent View of the classifier (in the cute world, >it >would also have visual cues for the corresponding drawing style), in which >markers could be gridded, reset to automatic settings (quantiles, even >breaks, >"natural" breaks,...), hand-tuned either visually or by input, and so on. A >kind of classification studio if you wish. Generating the theme classes >would >be either a side-effect or an extra action for this subsystem. > >I got stuck by wxWindows too. I couldn't produce something correct that >gave >me sliders that I could drag and drop. Could wx-savvy people give any hint >on >that? > >I also explored natural breaks. I found very scarce literature on the >subject. >>From what I gathered, natural breaks are based on a kernel approximation >to >the data distribution. From previous experience, programming these things >needs tuning and lots of special casing. If you have ideas on the subject, >they are most welcome. > >Daniel. > >_______________________________________________ >Thuban-list mailing list >Thuban-list at intevation.de >https://intevation.de/mailman/listinfo/thuban-list _________________________________________ Dan Putler Sauder School of Business The University of British Columbia Email: putler at sauder dot ubc dot ca Phone: 604-822-8329 From jan at intevation.de Sat Dec 6 11:28:05 2003 From: jan at intevation.de (Jan-Oliver Wagner) Date: Sat, 6 Dec 2003 11:28:05 +0100 Subject: [Thuban-list] visual classifier In-Reply-To: <200312041438510980.01145E7F@exchange.commerce.ubc.ca> References: <40211.164.15.134.155.1070529727.squirrel@moritz.homelinux.org> <20031204092641.GB30424@intevation.de> <20031204133653.M91206@minag.gob.pe> <35404.164.15.134.155.1070555843.squirrel@moritz.homelinux.org> <20031204174649.M72607@minag.gob.pe> <200312041438510980.01145E7F@exchange.commerce.ubc.ca> Message-ID: <20031206102805.GD7026@intevation.de> On Thu, Dec 04, 2003 at 02:38:51PM -0800, Dan Putler wrote: > A benefit to using K-Means to create the "natural breaks" is that the SciPy package already contains a K-Means method function written in Python. Although, it may require bringing in a lot of bagage to Thuban. Thuban will treat such bagage in the same it handles the other ones: If the packages are found, the functionality is available, else disabled. So actually not a too serious problem :-) Best Jan -- Jan-Oliver Wagner http://intevation.de/~jan/ Intevation GmbH http://intevation.de/ FreeGIS http://freegis.org/ From matt.wilkie at gov.yk.ca Tue Dec 9 19:18:14 2003 From: matt.wilkie at gov.yk.ca (Matt Wilkie) Date: Tue, 09 Dec 2003 10:18:14 -0800 Subject: [Thuban-list] thuban, windows and python 2.3 Message-ID: <3FD611E6.2000803@gov.yk.ca> Hello, It's good to see that a full on 1.0 release is approaching. Congratulations to everybody for their hard work! Can the windows binary be used with python 2.3? If not is a py2.3 win-binary forthcoming? I'd rather not have 2.2 and 2.3 concurrent if I can avoid it (or maybe that is not even possible?) thanks in advance, -- matt wilkie -------------------------------------------- Geographic Information, Information Management and Technology, Yukon Department of Environment 10 Burns Road * Whitehorse, Yukon * Y1A 4Y9 867-667-8133 Tel * 867-393-7003 Fax -------------------------------------------- From bh at intevation.de Wed Dec 10 14:20:34 2003 From: bh at intevation.de (Bernhard Herzog) Date: Wed, 10 Dec 2003 14:20:34 +0100 Subject: [Thuban-list] thuban, windows and python 2.3 In-Reply-To: <3FD611E6.2000803@gov.yk.ca> (Matt Wilkie's message of "Tue, 09 Dec 2003 10:18:14 -0800") References: <3FD611E6.2000803@gov.yk.ca> Message-ID: <6qzne0byhp.fsf@salmakis.intevation.de> Matt Wilkie writes: > Hello, > > It's good to see that a full on 1.0 release is approaching. > Congratulations to everybody for their hard work! Thanks. > Can the windows binary be used with python 2.3? No. You'd have to recompile some modules. > If not is a py2.3 win-binary forthcoming? Good question. Given that the startup file doesn't work on windows with python 2.2 but will probably work with 2.3 (see https://intevation.de/rt/webrt?serial_num=2241) I think that we should build a Python 2.3 binary at least for the final release or the next release candidate should there be one. > I'd rather not have 2.2 and > 2.3 concurrent if I can avoid it (or maybe that is not even possible?) Having two python versions installed simultaneously should be possible. See e.g. this posting by Tim Peters: http://mail.python.org/pipermail/tutor/2003-August/024465.html -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From bh at intevation.de Tue Dec 16 11:27:43 2003 From: bh at intevation.de (Bernhard Herzog) Date: Tue, 16 Dec 2003 11:27:43 +0100 Subject: [Thuban-list] Thuban 1.0 Translation Status Message-ID: <6qy8tdys4g.fsf@salmakis.intevation.de> The main thing missing for the final Thuban 1.0 release are the translations. Here's the current Thuban translation status: de.po: 362 translated es.po: 303 translated, 22 fuzzy, 37 untranslated fr.po: 303 translated, 22 fuzzy, 37 untranslated it.po: 287 translated, 30 fuzzy, 45 untranslated ru.po: 274 translated, 12 fuzzy, 76 untranslated Comments: Only de is completely up to date (I admit I have an unfair advantage :) ). Maurizio Napolitano says he has time for the Italian translation after 19th December. Alex Shevlakov has updated the russian translation but one remaining problem there is that the tooltips for some reason show latin1 text IIRC. It would be nice if we could make the final release including updated translations before Christmas, which means next Monday. Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From dcalvelo at agrario.minag.gob.pe Tue Dec 16 20:19:09 2003 From: dcalvelo at agrario.minag.gob.pe (Daniel) Date: Tue, 16 Dec 2003 14:19:09 -0500 Subject: [Thuban-list] es and fr translations for 1.0 Message-ID: <20031216191601.M88061@minag.gob.pe> Thubanions, Attached are .po files for ES and FR as of 1.0rc1. I updated the version number to 1.0; my setup produced UTF-8 po files, which shouldn't be a problem according to BH. Happy thubaning. Daniel. -- Daniel Calvelo Aros -- Direcci?n General de Informaci?n Agraria -- Ministerio de Agricultura del Per? -- (51-1) 424-9001 -------------- next part -------------- A non-text attachment was scrubbed... Name: es_fr.po.zip Type: application/zip Size: 18766 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-list/attachments/20031216/a733a20f/es_fr.po.zip From dcalvelo at minag.gob.pe Thu Dec 18 22:34:14 2003 From: dcalvelo at minag.gob.pe (Daniel Calvelo Aros) Date: Thu, 18 Dec 2003 16:34:14 -0500 Subject: [Thuban-list] Classifications Message-ID: <20031218210216.M64523@minag.gob.pe> Hi all. I've read the extraordinary article by Jenks and Caspall on classification for choroplethic maps. A must read, BTW. I'm willing to implement their technique and maybe use it as a framework for all the classification options in Thuban. I will need some help, not in the algorithms themselves but rather on how to hook them into Thuban. Let me give an overview of the classification problem as presented by Jenks and Caspall, then I will present the problems I'm facing. Jenks and Caspall do three important things in the article. First they state the problem as an optimization one: the minimization of an "interpretation error" induced by the given (classified) representation of choroplethic maps. Second, they divide the problem into three different perspectives and finally they give algorithms for partial and overall solutions of the problem. Let me elaborate on the second point: the authors state that reading a choroplethic map may be performed for three reasons (or phrased differently, choropletic maps have three different uses). First, (the "O" use) they are read as an overview of the spatial arrangement of the data; second, they are used for specific reading of the data values at specific locations ("T" for "table" use); third, they are used for "boundary" reading where the differences between regions are mostly taken into account and not the values themselves (the "B" for "boundary" use). For each use or reading mode, they define a normalized error index and propose a composite error index which serves for an optimization algorithm in order to obtain the best classification. Now, to the point. The error indices are roughly given by: 1) T error: some average measure of the classification error of the data values with respect to the classified values. For example, in TeX notation: $$ \sum_{i=1}^L ( v_i - m_i )^2 $$ for values $v_i$ indexed by 1..L and $m_i$ being the average of the v values within the class to which $v_i$ belongs. T error relies on the data distribution only, and can be implemented effortlessly for all the methods we have been discussing and implementing (k-means, quantile, even breaks,...) 2) O error: some average on the *overall* visualization error. This is calculated by something like $$ \sum_{i=1}^L |v_i - m_i|*A_i $$ where v and m are the same as the previous ones, and A_i is the area of the i-th polygon. This is an average or total error *volume*, which takes into account the fact that larger polygons are visually prominent over smaller ones. That means that in order to calculate an O-type index error, we need to access the area of the polygons corresponding to the data. Hints on doing this elegantly? The ideal would be to have a classifier class to which an array of areas is passed, instead of accessing the shape info from the classifier. 3) B error: this is the tricky one. The boundary error measures the average of differences in v vs. m values between *neighboring* polygons. In order to implement this, we would need some representation of the layer's topology, ideally an adjacency matrix for all the polygons. How can you get an adjacency matrix from a polygon layer? Finally, given these measures, we could use any of a set of optimization techniques to get a good classification in the sense of each or the overall indices. I am currently implementing T-type classifiers to supplement quantile and even spacing. I have somewhere a genetic algorithm implementation which could be used to optimize classifications based on any of the error measures (although I reckon a properly-tuned simulated annealing could perform better). But I have to be able to compute O-type and B-type error indices, for which I need acces to a layer's geometry and topology. Any hints are welcome! Daniel. -- Daniel Calvelo Aros -- Direcci?n General de Informaci?n Agraria -- Ministerio de Agricultura del Per? -- (51-1) 424-9001 From bh at intevation.de Fri Dec 19 15:38:58 2003 From: bh at intevation.de (Bernhard Herzog) Date: Fri, 19 Dec 2003 15:38:58 +0100 Subject: [Thuban-list] es and fr translations for 1.0 In-Reply-To: <20031216191601.M88061@minag.gob.pe> (dcalvelo@agrario.minag.gob.pe's message of "Tue, 16 Dec 2003 14:19:09 -0500") References: <20031216191601.M88061@minag.gob.pe> Message-ID: <6qy8t8g9dp.fsf@salmakis.intevation.de> "Daniel" writes: > Thubanions, > > Attached are .po files for ES and FR as of 1.0rc1. Thanks. I've just checked them in. > my setup produced UTF-8 po files, which shouldn't be a problem > according to BH. Yes they work fine. Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From bh at intevation.de Mon Dec 22 19:28:43 2003 From: bh at intevation.de (Bernhard Herzog) Date: Mon, 22 Dec 2003 19:28:43 +0100 Subject: [Thuban-list] Thuban 1.0 Translation Status In-Reply-To: <6qy8tdys4g.fsf@salmakis.intevation.de> (Bernhard Herzog's message of "Tue, 16 Dec 2003 11:27:43 +0100") References: <6qy8tdys4g.fsf@salmakis.intevation.de> Message-ID: <6qy8t4btb8.fsf@salmakis.intevation.de> New translation status: de.po: 362 translated es.po: 362 translated fr.po: 362 translated it.po: 287 translated, 30 fuzzy, 45 untranslated ru.po: 268 translated, 18 fuzzy, 76 untranslated Bernhard Herzog writes: > It would be nice if we could make the final release including updated > translations before Christmas, which means next Monday. It won't be released today after all, but tomorrow. I have most things ready, what's missing is some more testing and packaging. So, if somebody still has some translations to submit, tonight is the last chance to get it into 1.0.0. Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From napo at itc.it Tue Dec 23 11:18:13 2003 From: napo at itc.it (Maurizio Napolitano) Date: Tue, 23 Dec 2003 11:18:13 +0100 Subject: [Thuban-list] Thuban 1.0 Translation Status In-Reply-To: <6qy8t4btb8.fsf@salmakis.intevation.de> References: <6qy8tdys4g.fsf@salmakis.intevation.de> <6qy8t4btb8.fsf@salmakis.intevation.de> Message-ID: <1072174693.25581.28.camel@toriamos> Sorry! But for the 19 December i was busy for the university, so, yesterday, i have endend my work of traslation. The traslation is based on Thuban 1.0rc1. On my work of traslation i have found some little errors. The are some messagges write on the source code (i think) or not linked to the .po file. The "errors" (for the traslation) - Select the command "Rename" under the menu "Map" - The program show a modal window with the title "Map Title" This string is not present on the thuban.pot file - Open a shape file - Select the command "Join" under the menu "Table", The default voice is "Select..." This string is not present on the thuban.pot file I think there are others strings like this but is'nt a very problem. In attachment the italian traslation ;) Merry Christmans! -------------- next part -------------- A non-text attachment was scrubbed... Name: it.po.bz2 Type: application/x-bzip Size: 7819 bytes Desc: not available Url : http://www.intevation.de/pipermail/thuban-list/attachments/20031223/4175606c/it.po.bz2 From napo at itc.it Tue Dec 23 11:29:38 2003 From: napo at itc.it (Maurizio Napolitano) Date: Tue, 23 Dec 2003 11:29:38 +0100 Subject: [Thuban-list] Errors on Thuban 1.0rc1 Message-ID: <1072175378.25581.41.camel@toriamos> I have compiled the wxwindows and the wxPython package on my laptop (before i have used the RPM made for Mandrake) and now i have resolved the problem to load a ESRI Shape file (before this operation Thuban went in crash). Now i have others problems :( The first is by use the command "Projections" under the menu "Map". The errors is very "crazy": my XFree go in crash!!! Probably i need to compile the proj libraries from the sources without use the RPM package for Mandrake. The second i always the problem to use the "Classification" to display a layer. I go to the properties of a layer and the label "Classification" is disable. On the right of the label "Field" the space i "white" but, if i go on the right of the label with the mouse and double click the program show me the combobox with the all fields. Why? Look this pictures (in italian language) to understand my problems. http://sra.itc.it/people/napolitano/images/Errore1.png http://sra.itc.it/people/napolitano/images/Errore2.png PS: i can select the field but ... i can't change the draw of the layer From bh at intevation.de Tue Dec 23 12:39:34 2003 From: bh at intevation.de (Bernhard Herzog) Date: Tue, 23 Dec 2003 12:39:34 +0100 Subject: [Thuban-list] Thuban 1.0 Translation Status References: <6qy8tdys4g.fsf@salmakis.intevation.de> <6qy8t4btb8.fsf@salmakis.intevation.de> <1072174693.25581.28.camel@toriamos> Message-ID: <6qzndj226h.fsf@salmakis.intevation.de> Maurizio Napolitano writes: > Sorry! But for the 19 December i was busy for the university, so, > yesterday, i have endend my work of traslation. > The traslation is based on Thuban 1.0rc1. > On my work of traslation i have found some little errors. > The are some messagges write on the source code (i think) or not > linked to the .po file. > > The "errors" (for the traslation) > - Select the command "Rename" under the menu "Map" > - The program show a modal window with the title "Map Title" > This string is not present on the thuban.pot file The corresponding strings in the other rename dialogs (layer and table) aren't translated either. Seems to be a typical cut&paste error. I've marked them for translation now. > - Open a shape file > - Select the command "Join" under the menu "Table", > The default voice is "Select..." > This string is not present on the thuban.pot file I've marked this one too. > In attachment the italian traslation ;) Thanks. We've decided to go ahead with the 1.0.0 release instead of e.g. making a new release candidate. The release will then have the known bug of a few missing translations but that can be easily remedied by a 1.0.1 in January or so. Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From bh at intevation.de Tue Dec 23 12:51:47 2003 From: bh at intevation.de (Bernhard Herzog) Date: Tue, 23 Dec 2003 12:51:47 +0100 Subject: [Thuban-list] Errors on Thuban 1.0rc1 References: <1072175378.25581.41.camel@toriamos> Message-ID: <6qr7yv21m4.fsf@salmakis.intevation.de> Maurizio Napolitano writes: > I have compiled the wxwindows and the wxPython package on > my laptop Which version? Did you use any unusual configure options? I've used ../configure --with-gtk --prefix=$INSTALL_PREFIX --enable-debug --enable-debug_gdb --enable-resources --enable-prologio > Look this pictures (in italian language) to understand my > problems. > http://sra.itc.it/people/napolitano/images/Errore1.png > http://sra.itc.it/people/napolitano/images/Errore2.png You seem to be using a non-standard theme. IIRC wxGTK has (or at least used to have) problems with theme engines, so maybe that is part of the problem. Which theme are you using? I've just tested Thuban with a few theme engines but I didn't see any problems so far. Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From bh at intevation.de Tue Dec 23 15:59:42 2003 From: bh at intevation.de (Bernhard Herzog) Date: Tue, 23 Dec 2003 15:59:42 +0100 Subject: [Thuban-list] Thuban 1.0.0 released Message-ID: <6qn09j1swx.fsf@salmakis.intevation.de> As a christmas present, we finally present ... ... Thuban 1.0.0! Changes since 1.0rc1 include mostly translation updates but also a few minor bug fixes. I'm unable to update the website at the moment unfortunately, but you can get the usual files from ftp://intevation.de/thuban/. The dependencies are the same as for the release candidate. We didn't have time to do e.g. a Python 2.3 based release but maybe that can follow in January or for 1.0.1. If you like Thuban, please consider supporting it with a contribution to our tip-jar: http://thuban.intevation.org/mini-payments.html Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/ From bh at intevation.de Wed Dec 24 12:48:37 2003 From: bh at intevation.de (Bernhard Herzog) Date: Wed, 24 Dec 2003 12:48:37 +0100 Subject: [Thuban-list] Thuban 1.0.0 released In-Reply-To: <6qn09j1swx.fsf@salmakis.intevation.de> (Bernhard Herzog's message of "Tue, 23 Dec 2003 15:59:42 +0100") References: <6qn09j1swx.fsf@salmakis.intevation.de> Message-ID: <6qu13q5td6.fsf@thetis.intevation.de> Bernhard Herzog writes: > ... Thuban 1.0.0! [...] > I'm unable to update the website at the moment unfortunately, The download page at least has been updated now. > but you can get the usual files from ftp://intevation.de/thuban/. The source archives were missing but they're available too now. Bernhard -- Intevation GmbH http://intevation.de/ Sketch http://sketch.sourceforge.net/ Thuban http://thuban.intevation.org/