Re: Subversion M4 issue #528 resolved, half of #530 resolved

cmpilato@collab.net writes:

> The ugly fix for this requires us driving our commit-editor
> differently in the local vs DAV cases. Fortunately, we are planning
> to make ra_local's commit schema mimic what DAV is currently doing
> anyway.

I think you hit the nail on the head. The *only* reason the
commit-crawler is sending a 'delete' command before the 'add' command
is because it's trying to build an "accurate" model of the working
copy on the server.

We *do* have a plan to make both ra layers use a "commit against head"
system (issue #463). That's the key.

In such a "new" system, the commit-crawler wouldn't worry about the
'deleted' item at all; it would simply ignore the flag and try to do
an add. Either the add will succeed in the head revision (because the
original item is already deleted), or it won't succeed (maybe the item
came back at some point.)

So I suspect that this Hudson Scenario has always failed over DAV,
ever since M2 last April. We just never noticed it, because we didn't
get the python tests working over DAV until last month. :-(

We could comment test #10 out for now; or we could rewrite the commit
system. Hm.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sat Oct 21 14:36:44 2006