Date: Fri, 26 Mar 2004 04:27:41 +0000 From: Dan Foster <dsf@globalcrossing.net> To: "David O'Brien" <obrien@FreeBSD.org> Cc: freebsd-gnats-submit@FreeBSD.org Subject: Re: ports/63192: mutt change breaks vim syntax highlighting Message-ID: <20040326042741.GA16229@gblx.net> In-Reply-To: <20040324181628.GB6200@dragon.nuxi.com> References: <20040324174642.GB18920@anyware12.anyware> <20040324181628.GB6200@dragon.nuxi.com>
next in thread | previous in thread | raw e-mail | index | archive | help
I'd like to revisit this because I've come across some more information
that paints things in a different light.
Specifically, what changed mutt's behavior for the temp file name is
patch 'patch-mktemp' supplied by the FreeBSD ports maintainer for mutt.
I just checked the mutt 1.4 and 1.5 source trees (multiple versions),
and in the unmodified mutt sources, mutt-<version/muttlib.c has this
function:
void mutt_mktemp (char *s)
{
snprintf (s, _POSIX_PATH_MAX, "%s/mutt-%s-%d-%d", NONULL (Tempdir),
NONULL(Hostname), (int) getpid (), Counter++);
unlink (s);
}
The FreeBSD patch -- which doesn't seem to have come from the mutt
developers -- has this modification:
--- muttlib.c.orig Mon Feb 9 08:25:28 2004
+++ muttlib.c Mon Feb 9 08:32:46 2004
@@ -656,7 +656,8 @@
void mutt_mktemp (char *s)
{
- snprintf (s, _POSIX_PATH_MAX, "%s/mutt-%s-%d-%d", NONULL (Tempdir),
NONULL(Hostname), (int) getpid (), Counter++);
+ snprintf (s, _POSIX_PATH_MAX, "%s/mutt-%s-XXXXXXXX", NONULL(Tempdir),
NONULL(Hostname));
+ mktemp (s);
unlink (s);
}
My point is that if mutt's behavior changed due to a FreeBSD-specific
decision, rather than a change made by the mutt developers, then a
FreeBSD-specific patch (for the syntax highlighting autocommand) should
also be committed to vim to keep mutt and vim in sync.
A reasonable person couldn't expect the vim team to accept a patch for
supporting a change in runtime.vim due to a change in mutt that was not
committed by the mutt developers and is specific only to FreeBSD.
This appears to be why vim developers will not commit the patch, and
they seem to be on solid ground with that position given the reasoning.
So I would, therefore, like to politely request reconsideration of the
proposed vim patch for inclusion to the FreeBSD vim port or to have the
mutt patch-mktemp patch backed out.
Either change would be sufficient to keep the two in sync and remain
consistent with policy regarding local changes.
-Dan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040326042741.GA16229>