git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Uwe Zeisberger <zeisberg@informatik.uni-freiburg.de>
To: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Cc: git@vger.kernel.org
Subject: Re: [PATCH] Use diff3 instead of merge in merge-recursive.
Date: Fri, 20 Oct 2006 23:11:21 +0200	[thread overview]
Message-ID: <20061020211121.GB3024@cepheus.pub> (raw)
In-Reply-To: <Pine.LNX.4.63.0610181135120.14200@wbgn013.biozentrum.uni-wuerzburg.de>

Johannes Schindelin wrote:
> Hi Uwe,
> 
> On Wed, 18 Oct 2006, Uwe Zeisberger wrote:
> 
> > If no error occurs, merge (from rcs 5.7) is nothing but:
> > 
> > 	diff3 -E -am -L label1 -L label2 -L label3 file1 file2 file3 > tmpfile
> > 	cat tmpfile > file1
> 
> Interesting.

> I wonder if we could streamline the code such that index_fd 
> is called directly on the output of diff3? Of course, the result has to be 
> removed when the call to diff3 fails.
I thought about that, too.  But my primary intention was to get rid of
'merge', because the Solaris boxes I use from time to time lack merge,
but have (GNU) diff3[1].  I already had a mental note to look into that.

If Linus is right that there are systems that have merge but lack diff3,
then a combined approach is maybe the best?  That is, try diff3 and if
that is missing, try merge.  (Or the other way round if you prefer.)

OK, I looked a bit deeper into rcs, and it seems to handle the BSD diff3
case.  So Linus might be right.

BTW, merge -p sends the merged result to stdout instead of overwriting
the first file given.  That is

	merge -p -L label1 -L label2 -L label3 file1 file2 file3

and (GNU)

	diff3 -E -am -L label1 -L label2 -L label3 file1 file2 file3

are exactly equivalent.
So if that option of merge is old enough, these are the candidates for
the "combined approach" (see above).

> > I didn't made any timing tests or further tests for correctness, but I
> > hope Johannes still has the framework from the time when he converted
> > the Python script to C?  
> > 
> > @Johannes: If so, could you test this patch?
> 
> I have to dig a little where I have it, but I think I can give it a try in 
> a few hours (imagine this lyrics to the melody of the day job blues).
Seems to be a long blues because you didn't sent any results. :-(

Best regards
Uwe

[1] They also have a version of diff3 (I guess from BSD) that is not
suited to be used for merging, at least rcs' merge cannot use it.

-- 
Uwe Zeisberger

If a lawyer and an IRS agent were both drowning, and you could only save
one of them, would you go to lunch or read the paper?

  reply	other threads:[~2006-10-20 21:05 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-10-18  8:59 [PATCH] Use diff3 instead of merge in merge-recursive Uwe Zeisberger
2006-10-18  9:35 ` Jakub Narebski
2006-10-18 10:04   ` Johannes Schindelin
2006-10-18 15:53   ` Linus Torvalds
2006-10-19  6:31     ` Daniel Barkalow
2006-10-18  9:38 ` Johannes Schindelin
2006-10-20 21:11   ` Uwe Zeisberger [this message]
2006-10-21  0:06     ` Johannes Schindelin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20061020211121.GB3024@cepheus.pub \
    --to=zeisberg@informatik.uni-freiburg.de \
    --cc=Johannes.Schindelin@gmx.de \
    --cc=git@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).