info-cvs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Strange CVS behaviour


From: Mark E. Hamilton
Subject: Re: Strange CVS behaviour
Date: Tue, 28 Mar 2006 13:36:09 -0700
User-agent: Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US; rv:1.7.12) Gecko/20050920

Thorsten Alteholz wrote:
-> file is moved to .#file.version , but file (which contains the conflict
markers >>>>> and <<<<<) is not replaced by clean version from repository

So update -C does work unless there was a conflict detected before. This
is independent of the client (at least CVS 1.11.1p1 and CVSNT 2.0.5 show
this behaviour). Is this a bug or a feature or did I do something wrong?
Any help is greatly appreciated.

I've never seen 'cvs update -C' fail, and if it created the backup version then it must have done something (or why would it have created the backup?)

I suspect that what happened is that someone checked in the file with the conflict markers still in it. I'm not sure how this can happen for a real conflict, but I've seen it.

If I add conflict markers to a clean test file manually and then commit it I get this:

% cvs commit blarg1.txt
cvs commit: warning: file `blarg1.txt' seems to still contain conflict indicators

but the file is still committed. So, clearly the conflict markers can be checked in.

--
----------------
Mark E. Hamilton
Orion International Technologies, Inc.
Sandia National Laboratory, NM.
505-844-7666





reply via email to

[Prev in Thread] Current Thread [Next in Thread]