Hi, Of course, you're right. I already sent a patched version of jts to jukka to test this option. JTS is currently in a migration process from sourceforge to locationtech on github, and I suspect that this is not the best moment send a patch, but I'll folow the ticket I just wrote.
Michaël Le 18/02/2016 11:09, edgar.sol...@web.de a écrit : > Mike, > > while converting the wkb's might be a workaround, wouldn't it be cleaner and > more helpful, to check the JTS sources and develop a patch for them? > > ..ede > > On 17.02.2016 23:58, Michaël Michaud wrote: >> You are right. I will deactivate reading of ZM geometries. I will be better >> to have nothing. >> >> I get the same behaviour as yuo from geometry collections. I'll try to have >> a deeper look. >> Also the yellow message sometimes show up for other geometries with M if the >> previous import already throwed an exception... >> >> Michaël >> >> Le 17/02/2016 23:41, Rahkonen Jukka (MML) a écrit : >>> Hi, >>> >>> I got basically same results. I do not agree with simple geometries and ZM, >>> for example linestingZM (1 2 3 200, 4 5 6 200) becomes >>> >>> LINESTRING (1 2 3, 200 4 5). And I have something odd with 2D >>> geometrycollection. OJ reads the feature but it does not show on the map >>> and there is a message at lower left corner “Unknown WKB type 233”. >>> However, I can copy the WKT from attribute table and paste it into another >>> layer and then the geometry becomes visible. >>> >>> I agree that JTS is the right place for the fix. There may be something to >>> look in the WKT part of JTS as well. For example GDAL has just been changed >>> to output ISO style WKT like “POINT Z(1 2 3)”. Read more from >>> https://trac.osgeo.org/gdal/wiki/rfc61_support_for_measured_geometries and >>> https://trac.osgeo.org/gdal/wiki/MeasuredGeometriesInDrivers >>> >>> I wonder how well JUMP JML format supports XYZ, XYM, and XYZM geometries. >>> >>> -Jukka Rahkonen- >>> >>> *Lähettäjä:*Michaël Michaud [mailto:m.michael.mich...@orange.fr] >>> *Lähetetty:* 18. helmikuuta 2016 0:14 >>> *Vastaanottaja:* jump-pilot-devel@lists.sourceforge.net >>> *Aihe:* Re: [JPP-Devel] OJ reads XYZ in PostGIS style EWKB, not in OGC >>> style WKB >>> >>> Hi Jukka, >>> >>> Here are some results (thanks for the test database). >>> >>> - Simple geometries 2D, Z, M, ZM are read correctly (Z, M and ZM are >>> interpreted the same way = 3D coordinates) >>> - MultiGeometries and GeometryCollections are read correctly in 2D, but are >>> not read if there is Z, M or ZM. >>> >>> Explanation : I made a simple patch to change the geometry type in wkb >>> (byte 1 to 4). Unfortunately, in geometry collections, geometry type is >>> repeated in front of each single geometry. Patching spatialite driver to >>> change them all is not reasonable (it means re-implementing a significative >>> part of the wkb parser). >>> I will let it as is and try to work with Martin Davis to have it included >>> fully in JTS in the future. >>> >>> Michaël >>> >>> Le 17/02/2016 16:49, Rahkonen Jukka (MML) a écrit : >>> >>> Sorry Michaël, I noticed that I used the version of OpenJUMP with >>> a hacked JTS instead of the version with patched spatialite driver >>> (r4823). My observations below are useless. >>> >>> -Jukka- >>> >>> *Lähettäjä:*Rahkonen Jukka (MML) >>> [mailto:jukka.rahko...@maanmittauslaitos.fi] >>> *Lähetetty:* 17. helmikuuta 2016 17:34 >>> *Vastaanottaja:* OpenJump develop and use >>> *Aihe:* Re: [JPP-Devel] OJ reads XYZ in PostGIS style EWKB, not in >>> OGC style WKB >>> >>> Hi Michaël, >>> >>> Here is a test database that should contain one feature of each >>> geometrytype which are valid in GeoPackage: >>> >>> http://latuviitta.org/downloads/gpkg_geometrytype_test.gpkg >>> >>> Just the byte order that comes from my Windows/Intel, >>> unfortunately. You can create the same database with SQL script >>> http://latuviitta.org/downloads/make_gpkg.txtbut it requires >>> mod_spatialite with all the dependencies. Script can also be run >>> from the SQL window of spatialite-gui but it may also help to just >>> have a look at the script with bare eyes because all the >>> geometries which are written into GeoPackage are there as WKT. >>> >>> I had time for only a quick test. To start testing was by the way >>> amazingly easy because with the new Spatialite datastore I could >>> select all the layers at the same time and open them into an OJ >>> project. >>> >>> Geometries with Z seem to be mostly readable except >>> GEOMETRYCOLLECTION Z. It looks like most or all ZM geometries >>> behave badly. >>> >>> -Jukka- >>> >>> *Lähettäjä:*Michaël Michaud [mailto:m.michael.mich...@orange.fr] >>> *Lähetetty:* 17. helmikuuta 2016 14:48 >>> *Vastaanottaja:* jump-pilot-devel@lists.sourceforge.net >>> <mailto:jump-pilot-devel@lists.sourceforge.net> >>> *Aihe:* Re: [JPP-Devel] OJ reads XYZ in PostGIS style EWKB, not in >>> OGC style WKB >>> >>> Hi Jukke, >>> >>> Just patched spatialite driver in OpenJUMP. >>> Can you try different geometry types, and if possible, with >>> different byte order ? >>> Not sure what happens wih geometry M and geometry ZM. >>> >>> Michaël >>> >>> Le 16/02/2016 13:40, Rahkonen Jukka (MML) a écrit : >>> >>> Hi, >>> >>> I can confirm that with the patched jts jar I can read XYZ >>> geometries from GeoPackage database. Points, lines, and >>> polygons tested. It would be very nice to get even a patched >>> temporary read-only support for OGC style XYZ geometries into >>> OpenJUMP soon without a need to wait until JTS will support them. >>> >>> -Jukka Rahkonen- >>> >>> Michaël Michaud wrote: >>> >>> Thanks Jukka, >>> >>> I was not aware of this difference between postgis wkb and ogc >>> wkb. >>> Now, it is clear why I've never had problem with postgis 3D >>> wkb and why there is problem with geopackage driver. >>> >>> We'll have to make some suggestion to MD for JTS, but we have >>> some options there. >>> JTS can manage 2D coordinates (very well), 3D coordinates >>> (just holding information), and even 4D geometries (poorly = >>> 4th dimension is generally lost after geometry processing) >>> JTS has not clear semantic for Z, M, ZM. >>> >>> In the attached prototype, I made the simplest patch I could do : >>> if geometry type is > 1000, consider it is a 3D geometry (may >>> be buggy for geometryZM which is 4D) >>> another option would be : >>> if geometry type is > 1000 and < 2000, consider it is a 3D >>> geometry and throw an error over 2000 >>> >>> For wkb writer, it is another problem as we cannot write >>> postgis wkb AND ogc wkb at the same time. >>> We'll have to add an option to the writer. We have to discuss >>> that with MD. >>> >>> Please, let me know if the attached proto resolve the problem >>> (you have to add .jar and drop it in /lib folder in >>> replacement of jts-1.14.0.jar). >>> >>> As official JTS-1.14 has just been released and 1.15 is >>> probably far ahead, we can also circumvent the problem by >>> changing the wkb byte array in the geopackage reader so that >>> it conforms to postgis wkb. It would need to extract and >>> transform the four geometry type bytes. Not too difficult, but >>> a bit convoluted >>> >>> Michaël >>> >>> Le 11/02/2016 11:26, Rahkonen Jukka (MML) a écrit : >>> >>> Hi, >>> >>> Pasting this thread >>> >>> https://lists.osgeo.org/pipermail/geos-devel/2013-December/006757.htmlfor >>> making it easier to find from the OpenJUMP archives what >>> is the difference between XYZ geometries as they appear >>> in OGC WKT and in the PostGIS EWKB. Adding also some >>> keywords from the original error: >>> >>> unknown wkb type 235 >>> >>> It seems that the fix requires enhancement to JTS >>> WKBReader/Writer for making it possible to the OGC style >>> XYZ, XYM, and XYZM WKB. >>> >>> Quotation from JTS 1.14 source file >>> >>> https://sourceforge.net/p/jts-topo-suite/code/HEAD/tree/trunk/jts/java/src/com/vividsolutions/jts/io/WKBWriter.java >>> >>> * This implementation also supports the <b>Extended WKB</b> >>> >>> * standard. Extended WKB allows writing 3-dimensional >>> coordinates >>> >>> * and including the geometry SRID value. >>> >>> * The presence of 3D coordinates is signified >>> >>> * by setting the high bit of the <tt>wkbType</tt> word. >>> >>> * The presence of an SRID is signified >>> >>> * by setting the third bit of the <tt>wkbType</tt> word. >>> >>> * EWKB format is upward compatible with the original SFS >>> WKB format. >>> >>> If you will write to JTS list you could perhaps suggest to >>> at least add here and there in the code and javadocs a >>> clear notice telling that only 2D OGC WKB is supported. >>> >>> -Jukka Rahkonen- >>> >>> >>> >>> >>> >>> >>> ------------------------------------------------------------------------------ >>> >>> Site24x7 APM Insight: Get Deep Visibility into Application >>> Performance >>> >>> APM + Mobile APM + RUM: Monitor 3 App instances at just >>> $35/Month >>> >>> Monitor end-to-end web transactions and take corrective >>> actions now >>> >>> Troubleshoot faster and improve end-user experience. >>> Signup Now! >>> >>> >>> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 >>> >>> >>> >>> >>> >>> _______________________________________________ >>> >>> Jump-pilot-devel mailing list >>> >>> Jump-pilot-devel@lists.sourceforge.net >>> <mailto:Jump-pilot-devel@lists.sourceforge.net> >>> >>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel >>> >>> >>> >>> >>> >>> ------------------------------------------------------------------------------ >>> >>> Site24x7 APM Insight: Get Deep Visibility into Application >>> Performance >>> >>> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month >>> >>> Monitor end-to-end web transactions and take corrective >>> actions now >>> >>> Troubleshoot faster and improve end-user experience. Signup Now! >>> >>> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 >>> >>> >>> >>> >>> _______________________________________________ >>> >>> Jump-pilot-devel mailing list >>> >>> Jump-pilot-devel@lists.sourceforge.net >>> <mailto:Jump-pilot-devel@lists.sourceforge.net> >>> >>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel >>> >>> >>> >>> >>> >>> ------------------------------------------------------------------------------ >>> >>> Site24x7 APM Insight: Get Deep Visibility into Application Performance >>> >>> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month >>> >>> Monitor end-to-end web transactions and take corrective actions now >>> >>> Troubleshoot faster and improve end-user experience. Signup Now! >>> >>> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 >>> >>> >>> >>> >>> _______________________________________________ >>> >>> Jump-pilot-devel mailing list >>> >>> Jump-pilot-devel@lists.sourceforge.net >>> <mailto:Jump-pilot-devel@lists.sourceforge.net> >>> >>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel >>> >>> >>> >>> ------------------------------------------------------------------------------ >>> Site24x7 APM Insight: Get Deep Visibility into Application Performance >>> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month >>> Monitor end-to-end web transactions and take corrective actions now >>> Troubleshoot faster and improve end-user experience. Signup Now! >>> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 >>> >>> >>> _______________________________________________ >>> Jump-pilot-devel mailing list >>> Jump-pilot-devel@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel >> >> >> >> ------------------------------------------------------------------------------ >> Site24x7 APM Insight: Get Deep Visibility into Application Performance >> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month >> Monitor end-to-end web transactions and take corrective actions now >> Troubleshoot faster and improve end-user experience. Signup Now! >> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 >> >> >> >> _______________________________________________ >> Jump-pilot-devel mailing list >> Jump-pilot-devel@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel >> > ------------------------------------------------------------------------------ > Site24x7 APM Insight: Get Deep Visibility into Application Performance > APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month > Monitor end-to-end web transactions and take corrective actions now > Troubleshoot faster and improve end-user experience. Signup Now! > http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 > _______________________________________________ > Jump-pilot-devel mailing list > Jump-pilot-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel > ------------------------------------------------------------------------------ Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 _______________________________________________ Jump-pilot-devel mailing list Jump-pilot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
