[GRASS5] [bug #1969] (grass) r.fill.dir segfaults

[Request Tracker]

this bug's URL: http://intevation.de/rt/webrt?serial_num=1969
-------------------------------------------------------------------------

Subject: r.fill.dir segfaults

Platform: GNU/Linux/i386
grass obtained from: CVS
grass binary for platform: Compiled from Sources
GRASS Version: release_10_04_2003_grass5_0_2 and CVS HEAD

AS reported on the devel-list, here is the bug-report.
reproducable in Spearfish-data:

with floating-point grid
GRASS:~ > r.fill.dir in=elev.float elev=blah dir=dir
Reading map...
Filling sinks...
Determining flow directions for ambiguous cases...
Downward pass 1
Upward pass 1

Downward pass 2
Upward pass 2

Downward pass 3
Found 1897 unresolved areas
wtrshed pass 1
wtrshed pass 2
wtrshed pass 3
wtrshed pass 4
wtrshed pass 5
Speicherzugriffsfehler [segmentation fault]
GRASS:~ >

with integer grid
r.fill.dir in=elevation.dem elev=blah dir=dir
Reading map...
Filling sinks...
Determining flow directions for ambiguous cases...
Downward pass 1
Upward pass 1

Downward pass 2
Upward pass 2

Downward pass 3
Found 1895 unresolved areas
wtrshed pass 1
wtrshed pass 2
wtrshed pass 3
wtrshed pass 4
wtrshed pass 5
Speicherzugriffsfehler [segmentation fault]
GRASS:~ >

Markus found out the following:
<markus>
read mor in: Message-ID: <20030626111418.A20397 at itc.it>
The bug appears in get_min() which is called in
raster/r.fill.dir/ppupdate.c
row 134

The second parameter to get_min() seems to be empty when
it crashes.
</markus>

the whole thread can be found under
Message-Id: <20030626083947.35dc769f.sholl at gmx.net>

A sample location to reproduce this is packaged here:
www.8ung.at/stephn.holl/tmp/sample_r.fill.dir_bug.tgz (1,2 MB)

Stephan Holl

-------------------------------------------- Managed by Request Tracker