Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!apple!usc!sdd.hp.com!spool.mu.edu!uunet!mcsun!unido!ira.uka.de!fauern!NewsServ!rommel
From: rommel@Informatik.TU-Muenchen.DE (Kai-Uwe Rommel)
Newsgroups: comp.sources.d
Subject: RCS merge script ???
Message-ID: <1991Feb27.200448.14036@newsserv.informatik.tu-muenchen.de>
Date: 27 Feb 91 20:04:48 GMT
Sender: news@newsserv.informatik.tu-muenchen.de
Organization: Technische Universitaet Muenchen, Germany
Lines: 18

I am a bit confused about the merge script of RCS (applies to both 4.3
and 5.5). This shell script is supposed to merge the differences between
file2 and file3 into file1 (three arguments to it), am I right?

Now, I see that this shell script calls diff two times and then calls
diff3 with five arguments: the two files with the output of the previous
two diff calls and the three input files. The output is an ed script and
is piped to ed to apply it to file1.

Can anyone please explain what this means, especially why diff is called
first two times and what diff3 should do with diff's two output files?

If I understood the purpose of merge correctly, I think I can achieve
this by simply calling diff3 -E on the three input files when I have GNU
diff3 and pipe the result to ed. Is this Ok?

Thanks for any hint,
Kai Uwe Rommel