[GRASS-dev] [grass-code I][455] r.proj segfaults after Jul 14 changes

[grass-dev at grass.itc.it]

code I item #455, was opened at 2007-08-08 12:01
Status: Open
Priority: 3
Submitted By: William Kyngesburye (kyngchaos)
Assigned to: Nobody (None)
Summary: r.proj segfaults after Jul 14 changes 
Issue type: module bug
Issue status: None
GRASS version: CVS HEAD
GRASS component: raster
Operating system: MacOS X
Operating system version: 10.4
GRASS CVS checkout date, if applies (YYMMDD): 


Initial Comment:
Something happened in the middle of the r.proj.seg->r.proj change that's causing a segfault on OSX.

GRASS 6.3.cvs (us83-utm16):~ > r.proj loc=world1k map=globe in=globe met=cubic
Input Projection Parameters: +proj=longlat +a=6378137 +rf=298.257223563 +no_defs +towgs84=0.000,0.000,0.000
Input Unit Factor: 1
Output Projection Parameters: +proj=utm +zone=16 +a=6378137 +rf=298.257222101 +no_defs +towgs84=0.000,0.000,0.000
Output Unit Factor: 1
Bus error

Thread 0 Crashed:
0   libSystem.B.dylib  	0x9000b3c0 __vfprintf + 479
1   libSystem.B.dylib  	0x9002989c vsprintf + 476
2   libgrass_gis.dylib 	0x00057cc7 G_message + 57
3   r.proj             	0x00004003 main + 2209
4   r.proj             	0x000024a2 _start + 216
5   r.proj             	0x000023c9 start + 41


My build from Jul 14 works for r.proj.seg, but crashes for r.proj.  I see in the CVS history that on Jul 14 some "Message/Debug cleaning" was done to r.proj, then probably applied to r.proj.seg on Jul 16 as "Synchronize messages with r.proj", and this seems to agree with the vsfprint crash above.

More recent builds (I have one from Jul 30, as well as today), where r.proj.seg is now r.proj, also crash.

In main.c of r.proj.seg, if I replace any G_message(NULL) with what it used to be: G_message(""), it doesn't segfault.  And G_done_msg(NULL) -> G_done_msg("").


----------------------------------------------------------------------

You can respond by visiting: 
http://wald.intevation.org/tracker/?func=detail&atid=204&aid=455&group_id=21