[beports] Re: [RFC] diff file for gzip 1.3.12
- From: Andreas Färber <andreas.faerber@xxxxxx>
- To: beports@xxxxxxxxxxxxx
- Date: Wed, 14 May 2008 23:08:25 +0200
Am 14.05.2008 um 22:44 schrieb scott mc:
Here's a diff for gzip that fixes a BeOS workaround
Acked-by: Andreas Faerber <andreas.faerber@xxxxxx>
You could consider sticking with their previous coding style of using
defined __BEOS__ and maybe drop the spaces inside the parenthesis. But
that's a change the maintainers could also apply themselves if they
are strict about these things. ;)
( i used 'diff
stdbool_h-haiku stdbool_.h > stdbool_.h' to create it, is that the
correct way?).
Not quite, apart from the missing .diff extension above (you do seem
to have it on the attachment), please always use unified diffs (diff -
u, cvs diff -u, default for svn and git diffs). They are better
readable and old style diffs may be rejected.
For our own repository it would be okay then. For upstream patches you
should check whether they have a CVS/SVN/Git/whatever repository and
create the patch using the specific tool, such as cvs diff -u or svn
diff.
Are we submitting these types of fixes, where we put in " if defined
(__BEOS__) && !defined (__HAIKU__) " in place of ifdef __BEOS__ ?
Ingo? I have not submitted the gnulib patch yet due to the open
question of whether or not to drop __BEOS__ completely. If it remains
for the time being, +1 from me.
What about the config.guess when sending updates?
As mentioned earlier, do not yet. Ingo's changes are not declared
final and afaik not merged upstream for libtool & co either.
Andreas
--
BePorts homepage - http://tools.assembla.com/BePorts
List archives: //www.freelists.org/archives/beports
Administrative contact: brecht@xxxxxxxxxxx
Other related posts:
