From: Junio C Hamano <junkio@cox.net>
Date: Sat, 30 Jul 2005 02:11:21 GMT
Subject: Re: [PATCH 3/2] git-merge-cache -q doesn't complain about failing merge program
Message-ID: <7vack582ly.fsf@assigned-by-dhcp.cox.net>
In-Reply-To: <20050729125338.GB21909@pasky.ji.cz>

Petr Baudis <pasky@suse.cz> writes:

> git-merge-cache reporting failed merge program is undesirable for
> Cogito, since it emits its own more appropriate error message in that
> case. However, I want to show other possible git-merge-cache error
> messages. So -q will just silence this particular error.

That description makes it sound more like it is a Cogito
specific hack, which other Porcelains may not benefit from,
meaning they may want to suppress some other errors but this
patch does not give that possibility.

I do not mind about the above too much, but I'll sit on this one
for now just in case if anybody comes up with a different patch
to give a bit cleaner solution.

I wonder how many die()'s we have in our C code.  It _might_ be
cleaner to say (the first parameter being exit(2) parameter):

    die("unable to execute '%s'", pgm);
    die_with(47, "merge program failed");

and have the calling Porcelain, if it wants to, supress error
messages by redirecting 2>/dev/null, and say whatever it wants
to say based on $?.