[GRASS-dev] QGIS winGRASS blank space problem

[Glynn Clements]

Paul Kelly wrote:

> > a friend told me that with the latest QGIS release candidate,
> > there is a winGRASS problem:
> >
> > $ v.rast.stats vector=CEA_riserve raster=CEA_slope colprefix=delete
> > c:/Programmi/Quantum: c:/Programmi/Quantum: No such file or directory
> 
> v.rast.stats is a shell script, so you are likely to have loads of 
> problems if not finding it nearly impossible to get it working properly on 
> Windows. I notice there are loads of places in it that paths aren't 
> quoted so that is probably the first problem (although to be fair that is 
> also a problem for Unix and isn't a problem for Windows if you don't 
> install software in paths with spaces in them!).

That's certainly a fundamental problem; $TMP and $SQLTMP appear to be
the most likely candidates.

Regardless of Windows-specific issues, failure to quote variables
which contain pathnames is a bug and needs to be fixed.

AFAICT, $SQLTMP is unnecessary. You can just as easily use a subshell
with its output piped directly to db.execute; no need for a temporary
file.

> Then there is the fact that Msys doesn't understand Windows-style paths, 

That's incorrect; MSys itself understands Windows paths, with either a
forward or backward slash.

However, there may be individual scripts which can't handle this.

> In general it's a pleasant surprise if some shell scripts to happen to 
> work OK on Windows, and IMHO shouldn't really be considered a bug if they 
> don't. I feel the fact that QGIS includes a copy of Msys with GRASS gives 
> the "illusion" of GRASS being more compatible with Windows than it really 
> is - the truly native WinGRASS we're working on is a more accurate 
> expression of GRASS's compatiblity with Windows as I see it.

Ultimately, if we're serious about GRASS on Windows, we need to make
it a priority to migrate away from Bourne shell to something more
portable (with Python being the obvious choice).

-- 
Glynn Clements <glynn at gclements.plus.com>