[GRASS-dev] [GRASS-PSC] too many branches => retirement GRASS6.5.svn (=develbranch6)

Glynn Clements glynn at gclements.plus.com
Mon Apr 7 14:13:56 PDT 2014


Moritz Lennert wrote:

> In my understanding we were still quite far away from solving the python 
> on windows issues.

That depends upon what you consider the "issue" to be.

If the user already has a compatible version of Python and the
required libraries correctly installed, there isn't any issue, AFAIK.

If you consider the issue to be that the system's Python installation
matters, then we will always be quite far away from solving it,
because running a script from non-GRASS code via e.g. 
system("script.py") will always depend upon the system's handling of
the .py extension, and we can't change that.

The best that can be achieved in that regard is to add hard-coded
handling for Python scripts to those execution vectors which we
control as we encounter them, i.e. never-ending maintenance (aka
Whac-a-Mole), and just give up on the ones we don't control (i.e. 
GRASS will have to remain some form of "walled garden" on Windows).

Note that this isn't entirely specific to Windows. We'll start
encountering similar issues once Linux distributions start installing
Python 3.x as /usr/bin/python (although that's slightly easier to work
around, due to the #!/usr/bin/env trick).

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


More information about the grass-dev mailing list