Bug#648357: grass: segfault with -l flag in r.proj

[Sebastian P. Luque]

On Fri, 11 Nov 2011 13:53:45 +0100,
"Francesco P. Lovergine" <frankie at debian.org> wrote:

> On Thu, Nov 10, 2011 at 01:11:12PM -0600, Sebastian Luque wrote:
>> Package: grass Version: 6.4.1-1 Severity: normal

>> Dear Maintainer,

>> Doing 'r.proj -l location=ANY_LOCATION' gives a segmentation fault in
>> that program, even though grass session is not affected.

> It does not seem so even if having to specify an explicit existing
> mapset seems annoying and counter-intuitive (but it is not the only
> case):

[...]

I get the same results when specifying the mapset argument, but if I
don't supply it, then I see:

GRASS 6.4.1 > r.proj -l location=spearfish60
Segmentation fault

If the mapset argument is mandatory, shouldn't the help page show this
argument without the surrounding brackets under the SYNOPSIS section?

-- 
Seb