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

Re: Proposal: Improved locking implementation for fsfs

From: Peter N. Lundblad <peter_at_famlundblad.se>
Date: 2004-12-22 00:17:05 CET

On Mon, 20 Dec 2004, Max Bowsher wrote:

> Brian W. Fitzpatrick wrote:
> > Lock storage path:
> >
> > repos/db/locks/d3b07384d113edec/c157a79031e1c40f/72f8c9bac968893a
> >
> > Lockfiles can contain multiple lock representations, so collisions
> > will not be an issue (think of the end file as a hash bucket).
>
> Could problems arise if a commit turns a locked file into a directory of the
> same name?
>
I don't think so. The lock file will be deleted when the file is deleted
(I assume). The commit can't lock any files below the new directory. I
don't see a problem the other way round either. When a directory is
deleted, the subtree will also be deleted during commit finalization.

Hmm, but what about locking a non-existent path? That's not a problem in
our own client, but say a DAV client locks foo/bar/dir/file, where foo/bar
is a file already. I assume it has to delete foo/bar first.

Thanks,
//Peter

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Wed Dec 22 00:16:37 2004

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

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.