= GDAL Maintenance Reports by Chaitanya kumar CH = === Feb 24, 2010 to Mar 23, 2010 === #3434: Fixed the shapefile driver related to system specific directory seperator. #3436: Added the INTERLEAVE value to the IMAGE_STRUCTURE domain for ENVIDataset. Also fixed the overflow check. #3462: Reviewed a fix to gdal_contour algorithm. Need to write a test script for further testing. #3473: Caught a bug in GDALRasterizeLayers() that was revealed by a bug in the script tiff_write.py. #3467: Applied a patch to fix a bug in OGRPolygon::PointOnSurface() and OGRPolygon::Centroid() #3385: Applied a patch to fix problems caused by a call to unixODBC in some architectures. #3480: Fixed the PG driver to differentiate system tables correctly. === Jan 2, 2010 to Feb 22, 2010 === #3354: Fixed some bugs and added multi-band initialization to gdal_merge.py #3391: Fixed the mitab driver to create an OGRFeatureDefn for a newly created layer. #3394: Fixed SQLite driver to report correct transaction capability. #3402: Tested CPLError calls with a static string argument. I had a problem, probably related to linking wrong libraries. #3335: Transaction support to CopyLayer() #3396: Fixed a bug in GDALRasterizeLayers Added some features to gdalbuildvrt and the VRT format to facilitate a user-defined background colour. === Oct 28, 2009 to Jan 1, 2010 === #3148: Patch to make frmts/gtiff/tifvsi.cpp compile under VC++ 9.0 #3154: Configure snippet is not truly able to manage both netcdf3 and netcdf4 #3166: can't make large/64-bit netCDF file #2270: Patch for HDF5 block size recognition #2948: GML read failure for GDAL 1.6+ #3213: ESRI ArcInfo ASCII driver not Mac-compatible #3242: Northwood/VerticalMapper driver isn't big endian compatible #2991: OGR PGeo driver's GetGeometryRef() returns NULL for lines/polygons with Z coordinates #1918: PGEO: Missing geometry types (ie. type 9/19/29) #3100: Type 51 of geometry in Personal Geodatabase/shape #2842: Unable to delete colormap from HFA #2221: OGR SQL limits table name and column name character set #1999: Slowness of CPLReadLine() function on Windows #1367: PGEO driver should not force MDB Tools on Unix platforms #3259: GeoTiff driver problem with JPEG Encoded Images #3260: like in ticket #2643 where shape type 50 PGeodatabases are not recognized, 51 seems to be the same issue #1540: Minor memory leaks in gmlreader.cpp #3202: BURN_VALUE_FROM option for GDALRasterizeGeometries #2304: Failed test in ogr_mysql.py #3247: Shapefile driver truncated field names, sometimes causing duplicates #3237: Northwood/VerticalMapper Driver Contribution #3161: Enhancements proposal for gdal_merge.py #3236: Ogr2ogr stops appending to .shp file at 2 GB #3129: Contour orientation Compiler warnings in MS Windows === Sep 25, 2009 to Oct 27, 2009 === #3156: Fixed the problem with xerces library's changes between versions 2 and 3. Also found and fixed a bug that caused segmentation fault in ili driver. #2270: Tested and applied the patch for HDF5 block size recognition #3161: Reviewed proposal for gdal_merge.py #3129: Looked into the norms of contour orientation and gone through the code to get a feel of the algorithm. #3166: Researched netcdf and hdf specifications for the size issue. #3154: Configure snippet is not truly able to manage both netcdf3 and netcdf4 Read up on autoconf gdal-dev === Sep 9, 2009 to Sep 24, 2009 === #2930 Cleanup Win32 Compile Warnings #2798 Remove ESRI style spatial indexes from Shapefile on update via OGR. Prepared a patch. Still needs some debugging and testing. #3148 Patch to avoid VC++ compiler errors. Installed VC++ 9.0 alongside VC++ 8.0 on my Win32 machine. Got them both working! woohoo!! gdal-dev === Jul 18, 2009 to Sep 8, 2009 === #3083 Tested geos configuration with various permutations. #3112 Fixed the 'pixel validity mask' issue in TSX COSAR driver with suggestions from petar. #906 Tested NetCDF with a reportedly problematic dataset. #2893 Researched on the NetCDF format. #2857 Fixed the .xsd writer for date types and also added support for precision support for decimal data types. #2852 OGR2OGR MySQL to Shapefile Truncating Decimal Places #2930 Cleanup Win32 Compile Warnings #2991 Applied the patch after some coding style adjustments and test it with various datasets. #2798 Remove ESRI style spatial indexes from Shapefile on update via OGR. Prepared a patch. Still needs some debugging and testing. === May 9, 2009 to Jul 17, 2009 === Added some GML3 support to GML driver. Worked with Paolo Cavallini and Andrea Peri. #2667 Incorrect retrieving NUMBER width and precision for Oracle #2727 ogrinfo returns 'Corrupt data' error for wkt of length 1023 characters #2852 OGR2OGR MySQL to Shapefile Truncating Decimal Places #2930 Cleanup Win32 Compile Warnings #2948 GML read failure for GDAL 1.6+ #2991 OGR PGeo driver's GetGeometryRef() returns NULL for lines/polygons with Z coordinates #2993 Out of memory reading small filtered VRT #2995 ogr2ogr: issue with filename of Pgeo type #3012: PPM Driver 16bit Support Byte Order Decision Questionable #3045 Memory leak when exporting to MapInfo === Apr 22, 2009 to May 8, 2009 === #2955 docs missing for the c wrappers to the the geos functions #2956 Keyword causing format unrecognized (PDS) #2958 rgb2pct using external color palette #2960 Lanczos and !CubicSpline resampler buffer error on 2x2 images #1681 Add cast to (unsigned char) in all isspace() calls #2852 OGR2OGR MySQL to Shapefile Truncating Decimal Places #2857 ogr2ogr, shp > gml: incomplete schema #2798 Remove ESRI style spatial indexes from Shapefile on update via OGR === Apr 01, 2009 to Apr 17, 2009 === #2930 Cleanup Win32 Compile Warnings[[br]] Categerized the compiler warnings. Looked into the code. Researched for possibilities of improvements. ''Needs discussion.'' #2938 xsi:schemaLocation attribut in GML output #2945 postgres driver failes to create geometry index if layer name starts with a number #2942 external libz header is not found when setting the --with-libz=ARG switch #2875 patch for ogr feature style to make it more useable and one bugfix #2948 GML read failure for GDAL 1.6+ Mapserver ticket http://trac.osgeo.org/mapserver/ticket/2965 Present status:[[br]] ''Suspending my work from now "Apr 17, 19:30:28 2009 IST(+5:30)" to "Apr 21, 2009" on account of my sister's wedding on Sunday.[[br]] Will resume work on the morning of Apr 22, 2009 IST.'' === Mar 09, 2009 to Mar 31, 2009 === #2873 Projection Warning only printed for debug mode (ISIS 2/3, PDS) #2194 gdal_contour always uses field id's of 0 and 1 for id and elevation #2879 sqlite metadata=no option causes error #2493 GDAL thinks string field is an integer field in GML layer gdal-dev and IRC === Feb 07, 2009 to Mar 08, 2009 === #1017 [OGR-GML] fids written by the GML driver are not read back correctly[[br]]Updated patch to handle very long fids and added the test ogr_gml_6() for this ticket. #2493 GDAL thinks string field is an integer field in GML layer[[br]] Modified ParseXSD() to handle attribute values with a prefix.[[br]] Added a test script ogr_gml_7() for this ticket. #2825 Buffer overflow detected in ogr2ogr (included in FWTools-2.0.6) #2852 OGR2OGR MySQL to Shapefile Truncating Decimal Places #2857 ogr2ogr, shp > gml: incomplete schema #2879 sqlite metadata=no option causes error #2194 gdal_contour always uses field id's of 0 and 1 for id and elevation #2879 sqlite metadata=no option causes error #2843 HDF4 Update In Place Support[[br]] Researched HDF4 and driver implementation procedure. gdal-dev and IRC === Jan 18, 2009 to Feb 05, 2009 === #2817 bsb kap noaa #2486 IReadBlock failure in gdaladdo #2782 Failure reading BSB image #2776 e00 file not working with avcbin, but ok with avc driver #2798 Remove ESRI style spatial indexes from Shapefile on update via OGR #1017 [OGR-GML] fids written by the GML driver are not read back correctly gdal-dev and IRC. === Work done till Jan 17, 2009. === * Familiarized myselves with the GDAL/OGR project through the developer's view point. * Researched on AVCE00 specification for the bug in #2776 and gome through the source code. * Researched on noaa RNC BSB format. * Minimum of 3 Reference points are to be specified. * No mention of a maximum limit for the number of Reference points. * Possibility of overflow or loss of precision while calculating * Identified the MAX_GCP bug. #2777 * Suggested a patch for #2777