Re: [Gnu-arch-users] cscvs log parsing failure

[Harald Meland]

[Florian Weimer]

> On Sun, Sep 28, 2003 at 04:03:39PM +0200, Harald Meland wrote:
>
>> This proves my assumption false, I'm afraid, which means my attempted
>> fix is wrong.  However, I'm at a loss as to what a proper fix would
>> be. :-(
>
> Probably, there isn't any. 8-(
>
> Maybe you could examine the revision number and check if it is out of
> sequence.  This is still random guessing, but it would be a bit more
> reliable.

I (sort of) tried that already, but it got quite nasty -- do we have
any guarantees about the order CVS will report revisions in?  From my
endeavour with Mailman's CVS, I came to the following conclusions:

 * All trunk revisions are listed before any revisions on other
   branches.

 * All revisions on a single branch appear consecutively.

 * Revisions on a single branch are listed newest-to-oldest.  Note,
   though, that there might be gaps in the revision numbers (1.42
   might follow immediately after 1.1).

 * There is no guarantee that all files will have any particular
   numeric revision (e.g. 1.1) in common; some files might start their
   history at e.g. 2.0.

 * A numerically smaller non-trunk branch will always appear before a
   numerically larger one (with the same number of components).

> Otherwise, the only fix is a patch to CVS to properly escape log
> messages if necessary.

... or bite the bullet and accept that we have to edit the output of
"cvs -q log" before passing it to "cscvs cache -f
/name/of/fixed/log/file".

>> As a potentially useful data point, what cvs version are you using
>> (both client- and server-side, if applicable)?
>
> 1.12.1 on the client.  I don't know how to query the server version.

That might depend on how strict the access control has been set up on
your server; for SF (which houses the Mailman CVS), this works:

  $ ssh cvs.sf.net cvs -v
  
  Concurrent Versions System (CVS) 1.11.1p1 (client/server)
  [...]

My CVS client is also at 1.12.1 (to be even more specific, it's
Debian's 1.12.1-4).
-- 
Harald