Subject: Re: [change request] pattern for patch filenames
To: grant beattie <grant@NetBSD.org>
From: Luke Mewburn <lukem@NetBSD.org>
Date: 06/29/2004 10:20:38
Content-Type: text/plain; charset=us-ascii
On Tue, Jun 29, 2004 at 09:28:04AM +1000, grant beattie wrote:
| I actually find encoding the target filename in the patch filename a
| bit superfluous -- it's already at the top of the patch file.
IMHO, I think the idea of encoding the source file name in the patch
file name is a good one, given that each current patch-?? file is
only for one source file. This is per the current practice documented
in Packages.txt (section 4.3, paragraphs 2 & 3), which is effectively:
only modify one source file per patch-?? file, and don't split
patches to the one source file across multiple patch-?? files.
| I would support naming patches after their purpose/function which is
| much more useful:
| patches/patch-statvfs_2 (if there are multiple files that need patching=
This particular suggestion conflicts with the current practice (see above).
(All IMHO, as I'm just an occasional pkgsrc maintainer, not a pkgsrc
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (NetBSD)
-----END PGP SIGNATURE-----