Date: Sat, 27 Jul 2013 04:11:06 +0000 From: Jan Beich <jbeich@tormail.org> To: Pedro Giffuni <pfg@FreeBSD.org> Cc: FreeBSD Current <freebsd-current@freebsd.org> Subject: Re: [Heads up] BSD-licensed patch becoming the default RSN. Message-ID: <1V2vqT-000JNd-DJ@internal.tormail.org> In-Reply-To: <51F32288.7050701@FreeBSD.org> (Pedro Giffuni's message of "Fri, 26 Jul 2013 20:29:44 -0500") References: <51F2E627.9090907@FreeBSD.org> <1V2ssp-000Nrk-Q8@internal.tormail.org> <51F32288.7050701@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Pedro Giffuni <pfg@FreeBSD.org> writes:
> Now, just some food for thought, but if you are unsure your patch
> applies cleanly, why would you choose to use the -s (silent) option?
Because by default patch(1) is overly verbose. At first, I'm only
interested if a patch applies cleanly, then what files fail to apply.
To fix the patch I just repeat over edit a hunk (or two) and confirm
patch(1) no longer rejects it.
With -Cs giving up is easy at any time. One may not care about
a failed hunk in a man page or prefer to edit a patch as the whole
instead of on per-file (.rej file) basis.
> It would seem to me that some people may want the -s option to be
> truly silent (those paths may be long) and since those .rej files are
> not
> really being created it is consistent not to list them.
If you need -s to be truly silent then you're probably writing a script.
At which point -C being a BSD extension and -s behaving differently from
GNU patch would make more pain than not using them.
A new option may be better e.g.,
-q, --quiet
Do not write anything to standard output. Exit immediately
with non-zero status if any hunk fails to apply.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1V2vqT-000JNd-DJ>