Date: Wed, 30 Jul 2008 00:28:36 +0700 From: Max Khon <fjoe@samodelkin.net> To: =?ISO-8859-1?Q?G=E1bor_K=F6vesd=E1n?= <gabor@FreeBSD.org> Cc: Kris Kennaway <kris@FreeBSD.org>, Perforce Change Reviews <perforce@freebsd.org> Subject: Re: PERFORCE change 146209 for review Message-ID: <488F5344.7060807@samodelkin.net> In-Reply-To: <488F4EB8.5010308@FreeBSD.org> References: <200807291601.m6TG1FXh039193@repoman.freebsd.org> <488F4B1A.3000207@FreeBSD.org> <488F4EB8.5010308@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Hello! Gábor Kövesdán wrote: >>> http://perforce.freebsd.org/chv.cgi?CH=146209 >>> >>> Change 146209 by gabor@gabor_server on 2008/07/29 16:01:05 >>> >>> - Just handle some command line options as noop. They seem to be >>> rarely used based on the resources describing them. From now on >>> let's concentrate on the really practical features instead of >>> these ones. >> >> I don't think it's a good idea to "implement" options as NOPs unless >> they really are NOPs. This will just cause silent failure and/or >> script misbehaviour, which may be very hard to track down. > I've been also thinking of this, and I'm still a bit unsure. It would be > bad if scripts failed due to this, but it would be also bad if scripts > didn't run because of a e.g. --side-by-side argument, which rarely (or > never?) makes any difference. I've played with the options a bit and a > lot of them made no difference in the output neither with normal diff > nor with context diff nor with unified diff. Maybe the best way to > investigate what are they for would be to dig deeply in the GNU code, > which is a mess. I haven't even found info about them on Google, thus I > don't think they worth the effort. I think that you should not implement options as NOPs unless they are really NOPs too. /fjoe
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?488F5344.7060807>