Hi,
I am asking for a [remote] colleague (and have no windows accessible here right now):
The GRASS_BATCH_JOB environment variable seems to work fine in linux, but
not in windows. He sets it and when he starts grass, on windows it
opens the normal grass console and does nothing (that is, it does not run
the batch job).
Any way around this problem? It is for the integration of GRASS with another
software, so ideally it should also become possible on Windows…
Markus
Markus Neteler wrote:
The GRASS_BATCH_JOB environment variable seems to work fine
in linux, but not in windows. He sets it and when he starts
grass, on windows it opens the normal grass console and does
nothing (that is, it does not run the batch job).Any way around this problem? It is for the integration of
GRASS with another software, so ideally it should also
become possible on Windows...
which GRASS installer?
the osgeo4w one uses its own grass.bat which bypasses init.sh (and thus
GRASS_BATCH_JOB logic) AFAIK. the stand-alone installer uses init.sh
AFAIR but it may have a few layers of .bat before it which complicates
things. (I am unable to test wingrass right now as well)
Hamish
ps-
-----Inline Attachment Follows-----
has anything changed re. mailing list attachment-stripping policy recently?
On Thu, Jul 23, 2009 at 4:45 AM, Hamish<hamish_b@yahoo.com> wrote:
Markus Neteler wrote:
The GRASS_BATCH_JOB environment variable seems to work fine
in linux, but not in windows. He sets it and when he starts
grass, on windows it opens the normal grass console and does
nothing (that is, it does not run the batch job).Any way around this problem? It is for the integration of
GRASS with another software, so ideally it should also
become possible on Windows...which GRASS installer?
Have to ask...
the osgeo4w one uses its own grass.bat which bypasses init.sh (and thus
GRASS_BATCH_JOB logic) AFAIK. the stand-alone installer uses init.sh
AFAIR but it may have a few layers of .bat before it which complicates
things. (I am unable to test wingrass right now as well)
mhh, ok. Anyone else with a suggestion?
ps-
-----Inline Attachment Follows-----
has anything changed re. mailing list attachment-stripping policy recently?
AFAIR I didn't touch anything over the last months and Mailman is
version 2.1.5 as before.
Markus
The GRASS_BATCH_JOB environment variable seems to work fine
in linux, but not in windows. He sets it and when he starts
grass, on windows it opens the normal grass console and does
nothing (that is, it does not run the batch job).Any way around this problem? It is for the integration of
GRASS with another software, so ideally it should also
become possible on Windows...which GRASS installer?
Have to ask...
the osgeo4w one uses its own grass.bat which bypasses init.sh (and thus
GRASS_BATCH_JOB logic) AFAIK. the stand-alone installer uses init.sh
AFAIR but it may have a few layers of .bat before it which complicates
things. (I am unable to test wingrass right now as well)mhh, ok. Anyone else with a suggestion?
I haven't used GRASS_BATCH_JOB myself but it should to be ok in the
"with Msys" startup option for the stand-alone installer. (The other
startup option follows the osgeo4w method). The osgeo4w version can be
modified fairly easily to work as well since it includes the necessary
grass## script and init.sh.
All you need to do is fix the GISBASE variable in
/osgeo4w/apps/grass/bin/grass##, start msys and run the grass startup
script.
To get it working for natively in windows (without msys) would require
an addition to the init.bat file.
PS.
I haven't released a new stand-alone package for a while since no
changes to the windows parts seem to have been made recently. However,
I can put together an updated grass-6.4.0svn or brand new grass-7.0
installer (working on my machine) whenever there is a need.
Colin wrote:
PS.
I haven't released a new stand-alone package for a while
since no changes to the windows parts seem to have been made
recently. However, I can put together an updated grass-6.4.0svn
or brand new grass-7.0 installer (working on my machine)
whenever there is a need.
well the fixing of #654 is a major bug in RC5, but Maris's
spaces-in-path-names fixes to gis.m has not yet been backported
to relbr64 and similar changes to some C modules/libs has not be
backported to relbr64+trunk yet. I would hope to have those
reviewed, ported, and released ASAP but have reached the limit
of what I can do with that.
https://trac.osgeo.org/grass/log/grass/branches/develbranch_6/gui/tcltk/gis.m
(+ more, see ticket & svn log)
thanks,
Hamish