Opened 8 years ago
Last modified 4 years ago
#3079 new task
shapelib/DBF driver: re-sync to GDAL/OGR needed
Reported by: | neteler | Owned by: | |
---|---|---|---|
Priority: | major | Milestone: | 7.6.2 |
Component: | Database | Version: | svn-trunk |
Keywords: | dbf, shapelib | Cc: | volter |
CPU: | Unspecified | Platform: | Unspecified |
Description
The DBF driver is based on lib/external/shapelib which has been derived from Shapelib provided through GDAL/OGR (ogr/ogrsf_frmts/).
The last sync was done in 2008, time to update again.
Change History (14)
comment:1 by , 8 years ago
comment:2 by , 8 years ago
Milestone: | 7.2.0 → 7.4.0 |
---|---|
Type: | defect → task |
comment:3 by , 8 years ago
Done in trunk r69890. Unfortunately, more fixes were now needed to make it work outside GDAL, see updated README.
follow-up: 5 comment:4 by , 8 years ago
I've just resync'ed shapelib CVS from GDAL and make various fixes to compile in standalone mode (likely similar to yours). See http://lists.maptools.org/pipermail/shapelib/2016-December/000627.html
Looking at r69890, the following is incorrect:
+#define CPL_IGNORE_RET_VAL_INT(ret_val_int) return
The purpose of this macro is to make the compiler not warn about the fact that we don't verify the return value of a function. The correct dummy implementation should be :
#define CPL_IGNORE_RET_VAL_INT(x) x
comment:5 by , 8 years ago
Replying to rouault:
I've just resync'ed shapelib CVS from GDAL and make various fixes to compile in standalone mode (likely similar to yours). See http://lists.maptools.org/pipermail/shapelib/2016-December/000627.html
Looking at r69890, the following is incorrect:
+#define CPL_IGNORE_RET_VAL_INT(ret_val_int) return
I noticed that too after comparing GDAL shapelib with standalone shapelib. CPL_IGNORE_RET_VAL_INT is used only once, so I have already replaced in r69892
CPL_IGNORE_RET_VAL_INT(DBFFlushRecord( psDBF ));
with
DBFFlushRecord( psDBF );
as in standalone shapelib. Thanks for reviewing the changes!
comment:6 by , 8 years ago
The new SHAPELIB 1.4 version is out:
http://lists.maptools.org/pipermail/shapelib/2016-December/000631.html
Time to resync fixes into GRASS GIS or can we get rid of this clone?
comment:7 by , 8 years ago
Cc: | added |
---|
comment:10 by , 6 years ago
Milestone: | 7.4.1 → 7.4.2 |
---|
comment:11 by , 6 years ago
Milestone: | 7.4.2 → 7.6.0 |
---|
Milestone for this ticket should always be trunk.
Should this be done before the 7.2 release ?