Re: [PATCH] Allow transparent slaves to handle REPORT?

On Oct 12, 2007 11:14 AM, Jack Repenning <jrepenning@collab.net> wrote:
> It sounds to me, therefore, as if "[locking] would never get reported
> properly" may be restated as "locking would be broken." They would,
> I think you're saying, eventually learn that they both modified the
> same file (at someone's commit), but the "modify concurrently" would
> already be done.

Yes.

Actually, thinking about it some more - I think we could try to be
cute - have the post-lock and post-unlock hooks trigger a push to the
slaves of the lock database. This way the slaves would indeed have
the correct locking information stored locally.

We could use the bindings (or svnadmin??) to set/release the locks on
the slaves directly - that's probably a 'cleaner' way to do it. So, I
think we have a way for the repository admins to solve this if they
want to.

The only question that remains is about the merge-tracking info. -- justin

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Fri Oct 12 20:27:27 2007