Bug report #4667: GRASS python modules stopped to work - QGIS Application - QGIS Issue Tracking

Bug report #4667

GRASS python modules stopped to work

Added by Giovanni Manghi almost 13 years ago. Updated over 10 years ago.

Status:	Closed
Priority:	Low
Assignee:	-
Category:	GRASS
Affected QGIS version:	master	Regression?:	No
Operating System:		Easy fix?:	No
Pull Request or Patch supplied:	No	Resolution:
Crashes QGIS or corrupts data:	No	Copied to github as #:	14559

Description

The message is

Cannot start module r.external.all.py
command: /usr/share/qgis/grass/scripts//r.external.all.py --interface-description

Cannot start module v.out.ogr.pg.py
command: /usr/share/qgis/grass/scripts//v.out.ogr.pg.py --interface-description

Cannot start module qgis.v.kernel.rast.py
command: /usr/share/qgis/grass/scripts//qgis.v.kernel.rast.py --interface-description

Cannot start module db.connect-login.pg.py
command: /usr/share/qgis/grass/scripts//db.connect-login.pg.py --interface-description

Launching them from the GRASS console works as expected, as the native module GRASS GUI opens and works ok.

History

#1 Updated by Giuseppe Sucameli almost 13 years ago

Affected QGIS version set to master
Crashes QGIS or corrupts data set to No

"*.py --interface-descriptor" returns 1 (instead of 0) so that error message is shown.

Those GRASS plugin py modules just call the grass.parser(), so I think this could be a GRASS bug:
the grass.parser() called from python bindings exits with a wrong code (i.e. 1) after displaying the interface description.

On Ubuntu 11.04, QGis master, GRASS 6.4.1

#2 Updated by Giuseppe Sucameli almost 13 years ago

Status changed from Open to Closed
% Done changed from 0 to 100

Fixed in changeset fdd98947d763af5023cbd70fb43750f7611ba8f0.

#3 Updated by Giuseppe Sucameli almost 13 years ago

Here's the related ticket on GRASS trac: https://trac.osgeo.org/grass/ticket/1540

#4 Updated by Giuseppe Sucameli almost 13 years ago

% Done changed from 100 to 50
Priority changed from Normal to Low
Assignee set to Giuseppe Sucameli
Status changed from Closed to Reopened

martinl fixed it on the GRASS side (see http://trac.osgeo.org/grass/changeset/50240).
His fix may be added in GRASS 6.4.2 (is frozen for release yet, so cross your fingers).

After that fix is backported to GRASS 6.4 branch I have to make my workaround on the QGis side working for GRASS < 6.4.2 only (or to remove it at all) to handle the retcode correctly and avoid problems in the future.