Date: Mon, 3 Feb 2003 22:36:29 -0800 (PST) From: Doug Barton <DougB@FreeBSD.org> To: Garance A Drosihn <drosih@rpi.edu> Cc: Dag-Erling Smorgrav <des@ofug.org>, freebsd-hackers@FreeBSD.org Subject: Re: Modifying mergemaster behavior Message-ID: <20030203221928.S11438@12-234-22-23.pyvrag.nggov.pbz> In-Reply-To: <p05200f4cba64ed4989d3@[128.113.24.47]> References: <200301281844.h0SIie3G086935@repoman.freebsd.org> <p05200f24ba5c8412dcd1@[128.113.24.47]> <200301281908.h0SJ8OoY028183@apollo.backplane.com> <p05200f26ba5c85ee4c77@[128.113.24.47]> <20030201235429.G77226@12-234-22-23.pyvrag.nggov.pbz> <xzpy94xa6fb.fsf@flood.ping.uio.no> <20030203091128.Y4113@12-234-22-23.pyvrag.nggov.pbz> <p05200f4cba64ed4989d3@[128.113.24.47]>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 3 Feb 2003, Garance A Drosihn wrote:
> I happen to be updating my system tonight, so when it came to the
> mergemaster step I first modified the script.
Cool... contrary to popular opinion, I do like UI feedback. :)
> I added:
> -I '$FreeBSD:.*$'
> to the 'diff ${DIFF_FLAG}' command in diff_loop, and it seems to have
> worked the way I wanted it to work.
How did you want it to work? (This isn't a rhetorical question.) I also
think it's unlikely that you'd run into a version-string only diff in a
recent current -> recent current upgrade. It's these that are the chief
problem I'm concerned about.
> From my skimming of the script, I do not see how this change would cause
> problems for any other processing. However, it's possible I'm missing
> something.
What do you think mm should do if the only diff between two files is the
cvs ID? This comes up inside a branch if a change is checked in, then
backed out. Of course, it happens a lot when upgrading from 4 to 5 for
example.
> If so, my gut reaction is that if a lot of FreeBSD users are requesting
> this, then we should figure out how to make it happen -- one way or
> another.
The request I actually get is that if the only diff is the cvs ID, that mm
should do <X various things> with it. Most commonly, just install the file
(which I'm opposed to on principle, and by fundamental mm design). Trying
to deal rationally with this scenario is why I'd have to add at least one
more diff to the process.
1. Are the cvs id's different?
2. If so, is that the only difference? <--- new
3. If cvs ID is not the only diff, display that to the user.
> I suppose we should also change the 'absolute diff first' part to
> include that, or people will be asked about merging a file, only
> to be shown a file which apparently has zero changes.
I'm sorry, I don't understand exactly what you're getting at here.
> I'd also note that the separator line (=====) is usually scrolled
> off the window for me. Not much point to a good eye-catching line
> if the line is usually not on the screen.
That's actually a good thing. The ==== is for those cases where a diff is
smaller than one screenful. It was requested by users to give more visual
definition to that scenario, and also make logs of mm sessions easier to
parse. Passing it to PAGER when the diff already fills a page is a waste
of screen space.
Doug
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030203221928.S11438>