[GRASS-dev] Two new strange behaviours in GRASS 6.4.1SVN

Markus Neteler neteler at osgeo.org
Fri Jan 14 08:07:30 EST 2011


On Fri, Jan 14, 2011 at 1:24 PM, Benjamin Ducke
<benjamin.ducke at oxfordarch.co.uk> wrote:
> Dear list,
>
> I noticed two strange new things in 6.4.1svn that may
> both be related to the recent GUI changes.
>
> 1. I configured and compiled GRASS explicitely without
> Tcl/Tk and without Python support. The configure script
> correctly reported a "No" for both options.
> Yet if I just run e.g. "r.out.gdal" (i.e. with no options
> or flags), I get:
>
> python: can't open file '/usr/local/grass-6.4.1svn/etc/wxpython/gui_modules/menuform.py': [Errno 2] No such file or directory

(no idea unfortunately)

> Whereas I would expect the synopsis to be displayed in
> a CLI-only environment. And I really didn't want GRASS
> to interact with my system's Python at all.
>
> 2. The --help flag is only interpreted correctly depending
> on its position on the CLI. This:
>
>  r.out.gdal --help input=aspect output=aspect.tiff
>
> correctly displays the module synopsis, but this:
>
>  r.out.gdal input=aspect output=aspect.tiff --help
>
> gives:
>
>  Sorry, <-> is not a valid flag
>  Sorry, <h> is not a valid flag
>  Sorry, <e> is not a valid flag
>  Sorry, <p> is not a valid flag
>
> which is unfortunated, because I often type a CLI up
> to the point I can remember it and then add --help to
> get a reminder of the rest.

Same problem here! It would be good to have that back, it
must be a change in lib/gis/parser.c causing this. Looks like
this?
http://trac.osgeo.org/grass/changeset/45025/grass/branches/releasebranch_6_4/lib/gis/parser.c

Markus


More information about the grass-dev mailing list