[GRASS-user] Different GIS_ERROR_LOG files for different instances of GRASS?

[Rainer M Krug]

These are more and more reasons, to use 6.4 instead.

Thanks a million for all your suggestions,

Rainer


On Wed, Oct 8, 2008 at 10:34 AM, Markus Neteler <neteler at osgeo.org> wrote:
> On Wed, Oct 8, 2008 at 12:53 AM, Glynn Clements
> <glynn at gclements.plus.com> wrote:
>>
>> Markus Neteler wrote:
>>
>>> > I am running simulations which are using GRASS for some calculations,
>>> > and I would like to use GIS_ERROR_LOG as a log file. How can I have
>>> > different GIS_ERROR_LOG for different instances of GRASS, so that each
>>> > instance has it's own log file?
>>>
>>> Best might be to change the name "GIS_ERROR_LOG" which is defined in
>>> lib/gis/error.c
>>>
>>> into a variable.
>>> ?
>>
>> Done in 7.0 (r33740).
>
> Backported to 6.4.svn: r33753
>
>> If set, $GIS_ERROR_LOG should be the absolute path to the log file (a
>> relative path will be interpreted relative to the process' cwd, not
>> the cwd at the point you set the variable). If not set,
>> $HOME/GIS_ERROR_LOG is used instead.
>
> ok
>
>> The previous behaviour of writing to both $HOME/GIS_ERROR_LOG and
>> $GISBASE/GIS_ERROR_LOG has been eliminated; only one log file will be
>> written.
>
> That's not entirely clear to me. Say, lib/init/variables.html may need a fix
> for that.
>
>> As before, the file will only be used if it already exists.
>>
>> Also, the current directory is obtained using getcwd() rather than
>> popen("pwd"), so it should work on Windows.
>
> Perfect, thanks.
>
> Markus
>



-- 
Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation
Biology, UCT), Dipl. Phys. (Germany)

Centre of Excellence for Invasion Biology
Faculty of Science
Natural Sciences Building
Private Bag X1
University of Stellenbosch
Matieland 7602
South Africa