Re: serious bug: a succession of "svn up" can yield a working copy with local changes

Hi Stephen,

On 2009-08-05 16:13:56 +0200, Stephen Butler wrote:
> I believe the question is: how did the files get from CVS into
> Subversion? Your text-base file contains the string
>
> $Id: lplain.bst,v 4.0 2000/01/31 18:11:53 vlefevre Exp $
>
> which must have been generated by CVS (note the ",v" appended to the
> filename).

Well, I've never used CVS for my personal files. The file comes in fact
from RCS (but as CVS is based on RCS, this looks similar).

I simply did the following in my SVN working copy:

$ cp <the_file_in_question> lplain.bst
$ svn add lplain.bst
$ svn ci

Note: at that time, I didn't add an Id keyword. I added the keyword
in some later revision (with "svn ps", as usual).

> Creating a new file with that string in it, importing into a
> Subversion repository, and setting its svn:keywords property has
> the desired affect. The file in the working copy contains a
> Subversion-style value:
>
> $Id: alpha 2 2009-08-05 13:28:53Z steve $
>
> and its text-base contains the unexpanded keyword:
>
> $Id$

Has this *always* been like that? It doesn't seem so.

> So it appears that your file is part of a repository generated
> by a CVS-to-Subversion tool. Can you find out which one?

No, I didn't use such a tool. I did all my modifications with the
svn client.

-- 
Vincent Lefèvre <vincent@vinc17.org> - Web: <http://www.vinc17.org/>
100% accessible validated (X)HTML - Blog: <http://www.vinc17.org/blog/>
Work: CR INRIA - computer arithmetic / Arenaire project (LIP, ENS-Lyon)
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2380482