[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Some redundant tests in locking code?

From: Ben Collins-Sussman <sussman_at_collab.net>
Date: 2005-04-07 19:11:03 CEST

On Apr 7, 2005, at 11:44 AM, Julian Foad wrote:
> It is trying to delete a path, having assumed to begin with that the
> path is just a file. If that fails, maybe the path was a directory
> and so more lock tokens are needed. That "if" is trying to say, "if
> (an error occured that might be resolved by sending all the lock
> tokens held under this directory)". However, it doesn't appear to be
> checking the most appropriate set of error codes. As far as I can
> tell, neither SVN_ERR_FS_BAD_LOCK_TOKEN nor
> SVN_ERR_FS_LOCK_OWNER_MISMATCH could be resolved just by sending more
> tokens, so there's no point in them being in the list. And
> SVN_ERR_FS_PATH_ALREADY_LOCKED only applies to creating new locks, not
> using them.
> Please could somebody tell me whether I'm understanding this correctly?

Ah, you're correct, thanks for noticing. This can definitely be
tightened up. Can you change it to look only for the errors that will
actually happen?

To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Apr 7 19:13:38 2005

This is an archived mail posted to the Subversion Dev mailing list.