#POSTGIS IRC Log - 2009-04-27

For logs after Feb 3, 2007, all times are GMT-8. Prior logs are GMT-9.
Back to Logs
06:54:22 CIA-34: strk * r2408 /trunk/source/ ( 5 files in 2 dirs ): Quadtree's NodeBase: check port sync, add port info, drop unneeded pointers and virtuals, document objects ownership.
06:57:00 CIA-34: strk * r2409 /trunk/source/ ( 2 files in 2 dirs ): DoubleBits didn't change in JTS, update port info accordingly
07:27:00 CIA-34: strk * r2410 /trunk/source/ ( 2 files in 2 dirs ): Sync to JTS-1.10, reduce heap allocations and pointers usage. Includes a bugfix in collectStats.
07:35:49 CIA-34: strk * r2411 /trunk/source/ ( 2 files in 2 dirs ): Port info and check, doxygen.
07:43:46 CIA-34: strk * r2412 /trunk/source/ ( headers/geos/index/quadtree/Node.h index/quadtree/Node.cpp ): Port info for quadtree::Node
07:48:15 CIA-34: strk * r2413 /trunk/source/ ( headers/geos/index/quadtree/Root.h index/quadtree/Root.cpp ): Port info for quadtree::root
08:05:13 CIA-34: robe * r4025 /trunk/doc/xsl/postgis_gardentest.sql.xsl:
08:05:13 CIA-34: revise to better guarantee a deteriministic sort so can diff compare between
08:05:13 CIA-34: 1.3.5, 1.3.6, 1.4 ( still needs work ). Also exclude curved tests and functions
08:05:13 CIA-34: not supported in 1.3.5 since it curve crashes 1.3.5 thus making it not
08:05:13 CIA-34: comparable.
11:03:10 CIA-34: kneufeld * r4026 /trunk/doc/faq.xml: Test commit to test autobuild
11:05:28 CIA-34: kneufeld * r4027 /branches/1.3/TODO: Test commit to test autobuild
11:17:59 CIA-34: kneufeld * r4028 /branches/1.3/doc/postgis.xml: test commit
11:18:15 CIA-34: strk * r2414 /trunk/source/index/quadtree/NodeBase.cpp: Fix memory leak in NodeBase::remove, exposed by unit testing
11:20:53 CIA-34: kneufeld * r4029 /trunk/doc/postgis.xml: test commit
11:22:10 hobu: knuefeld here?
11:34:53 CIA-34: strk * r2415 /trunk/source/ ( 4 files in 2 dirs ): Refactor signatures to make ownership transfers more explicit. Fixed another leak in Node::insertNode.
11:38:33 CIA-34: strk * r2416 /trunk/tests/unit/algorithm/RobustLineIntersectorTest.cpp: Fix leak in testcase
11:41:08 hobu: kneufeld: try a test commit
11:41:13 hobu: I have to run in 10 min
11:41:19 kneufeld: k
11:42:12 kneufeld: mmm seems to be hanging
11:42:54 CIA-34: kneufeld * r4030 /trunk/doc/template.xml: test commit
11:43:03 : * hobu tries /usr/bin/wget http://office.refractions.net:1500/job/WKTRaster-trunk-build/build?token=build_it
11:43:05 sigq: Title: WKTRaster-trunk-build [Hudson] ( at office.refractions.net:1500 )
11:43:25 kneufeld: yes, that seemed to work
11:43:29 CIA-34: strk * r2417 /trunk/tests/unit/capi/GEOSGeomToWKTTest.cpp: Plug last one. All heap blocks were freed -- no leaks are possible.
11:43:37 hobu: but now you can't commit?
11:44:52 CIA-34: kneufeld * r4031 /branches/1.3/doc/README: test commit
11:45:08 kneufeld: great ... seems to work now
11:45:12 kneufeld: thanx Howard
11:45:31 hobu: sure thing
12:33:31 sigq: geosfeed: Ticket #249 ( defect created ): GEOS, Shapely, DLLs, and memory <http://trac.osgeo.org/geos/ticket/249>
12:34:38 sigq: geosfeed: geos.patch attached to ticket #249 <http://trac.osgeo.org/geos/attachment/ticket/249/geos.patch>
12:37:04 robe3: I guess I missed all the fun
13:02:16 sigq: geosfeed: Ticket #249 ( defect updated ): GEOS, Shapely, DLLs, and memory <http://trac.osgeo.org/geos/ticket/249#comment:1>
13:06:40 pramsey: robe3: the fun isn't continuous, it's intermittent
13:11:52 pramsey: seen kneufeld 's regression post, robe3 ?
13:12:04 pramsey: looks like the prj export stuff is firing an error?
13:12:22 kneufeld: : )
13:13:20 robe3: no
13:13:28 robe3: Will check it out
13:13:35 pramsey: see, irc rocks : )
13:13:42 : * pramsey cracks the whip
13:14:21 robe3: I'm old fashioned -- I still like email better
13:15:25 robe3: hmm where is your regression post Kevin : )
13:16:15 kneufeld: users
13:16:37 robe3: ah okay
13:16:43 kneufeld: just that I see errors in the log files in 8.3 and 8.4
13:16:53 kneufeld: but the tests claim to pass ok
13:18:26 robe3: how did you generate this error? I'll check my tmp to see if I have such files
13:19:46 robe3: I did test on 8.4 1.3.6 and 1.4 but haven't tested the loader dumper yet on that. the discrepancies I thought I was seeing between my 1.3.5 and 1.3.6 I think were because my tests did not define a sort
13:20:26 robe3: Once I forced a sort most came out about the same except for Union I think -- but those appeared to be sorting differences in the internal multipolygon structure
13:20:54 kneufeld: I think the loader works ok
13:21:20 kneufeld: The loader goes looking in the spatial_ref_sys table, though, doesn't it?
13:22:13 kneufeld: That file is never actually loaded in the regression tests
13:22:25 kneufeld: I'm thinking that could be cause of the error in the logs
13:23:38 robe3: yes
13:23:39 sigq: geosfeed: Ticket #249 ( defect updated ): GEOS, Shapely, DLLs, and memory <http://trac.osgeo.org/geos/ticket/249#comment:2>
13:24:06 robe3: It should skip over it and just not dump a prj file though
13:24:18 robe3: So it should be more like a warning than an actual error
13:30:58 kneufeld: Ah... k
13:39:11 robe3: Okay I checked my install and I get the same dumper.err -- you think we should change to not throw an error if SRID is -1
13:39:41 robe3: It seems to generate the dumper files fine though as expected. Though I guess that should read WARNING rather than ERROR
13:40:00 kneufeld: I think warning would be acceptable
13:41:32 robe3: actually its the dumper not the loader right
13:41:56 robe3: I get a loader.err file but it just has creating table implicit index in that file
13:42:06 kneufeld: ummm could be ... i didn't really look into it yet
13:42:39 robe3: Well your message doesn't make sense with laoder. Loader doesn't check spatial_ref_sy
13:43:36 kneufeld: right .. .it's the dumper that's in error
13:57:59 CIA-34: robe * r4032 /branches/1.3/doc/README: typo
14:00:54 CIA-34: robe * r4033 /branches/1.3/loader/pgsql2shp.c: change empty table error to just a warning
14:03:35 CIA-34: robe * r4034 /trunk/loader/pgsql2shp.c: change error about unknown spatial ref to a warning.
14:04:23 CIA-34: robe * r4035 /branches/1.3/loader/pgsql2shp.c: add another sentence
14:13:18 robe2: Did we ever figure out what causes this -- /tmp/pgis_reg_11624/regress_log:ERROR: table "loadedshp" does not exist
14:13:32 robe2: Shouldn'tthe loader have created that table?