[GRASS-dev] Re: on the fly grass location

[Roger Bivand]

Markus Neteler wrote:
> 
> On Sun, Nov 14, 2010 at 9:54 AM, Paolo Cavallini <cavallini at faunalia.it>
> wrote:
>> Hi all.
>> As it has been discussed earlier, having location generated (and
>> destroyed)
>> on the fly would help a lot the casual qgis grass user to approach grass.
>> The idea is to run an analysis by loading (r.external, v.external,
>> v.in.ogr)
>> data, generating a location from it, run, save the results as tif or shp,
>> and destroy the location at the end, all without users noticing it (but
>> explaining for future reuse).
>> The good news is that here at the hackfest we got a voulnteer willing to
>> help with the coding (preferably in Python). Welcome Peter Loewe!
> 
> Great, Peter!
> 
>> Anybody willing to help him with the first steps?
> 
> Sure - note that all ingredients are already there including the
> "demolocation" which
> comes with any GRASS binary package. 
> 

The GRASS R package was for GRASS5, and does not work with GRASS 6. A better
mechanism is provided in the spgrass package as initGRASS(). The code of the
R function sets the environment variables, etc., for a location generated on
the fly (with an .GRASSRC file) in the running R temporary directory. When R
exits, the location goes away.

Roger


Markus Neteler wrote:
> 
> Maybe that's easier discussed over chat?
> 
> Markus
> _______________________________________________
> grass-dev mailing list
> grass-dev at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/grass-dev
> 
> 


-----
Roger Bivand
Economic Geography Section
Department of Economics
Norwegian School of Economics and Business Administration
Helleveien 30
N-5045 Bergen, Norway

-- 
View this message in context: http://osgeo-org.1803224.n2.nabble.com/on-the-fly-grass-location-tp5737110p5737837.html
Sent from the Grass - Dev mailing list archive at Nabble.com.