[GRASS-dev] WinGRASS71svn using Windows shell (not bourne shell) thogh MSYS is installed
hellik at web.de
Fri Oct 24 02:07:37 PDT 2014
Moritz Lennert wrote
> On 24/10/14 04:20, Vaclav Petras wrote:
>> On Thu, Oct 23, 2014 at 7:08 PM, Glynn Clements
> glynn at .plus
> glynn at .plus
> >> wrote:
>> Vaclav Petras wrote:
>> > I'm afraid we have an issue with what we want. Do we want to remove
>> > in favor of native command line or do we want to have msys
>> > because it is much better then the MS Windows native command line.
>> Nothing stops the user from installing MSys if they want it. In fact,
>> users who are likely to want it may already have it installed, so we
>> shouldn't be installing a second copy (which may conflict with their
>> existing copy; even if it doesn't, they now have to configure and
>> maintain two copies).
>> We shouldn't be forcing it on people. Whether it's "better" isn't the
>> issue. It's not what Windows users will be familiar with. It's not
>> what's going to be described if users search for documentation on
>> "Windows command line".
>> GRASS 6 has to bundle MSys because GRASS 6 won't work without it.
>> shouldn't be the case for GRASS 7.
>> This sounds reasonable. If users want something better then what MS
>> provides they should install it, as you say, or they should install
>> different OS or use Python for scripting, I would say.
I'm still in favour to keep it.
anyway in case of not shipping msys, there are a some places needed to be
adapted in the nsis-script  and the package script .
AFAIR years ago there was some issue with the g.mkfontcap run during
installation which needed msys in %path%.
549 FileWrite $0 'set
View this message in context: http://osgeo-org.1560.x6.nabble.com/WinGRASS71svn-using-Windows-shell-not-bourne-shell-thogh-MSYS-is-installed-tp5168808p5169350.html
Sent from the Grass - Dev mailing list archive at Nabble.com.
More information about the grass-dev