Re: BUG with complex branching, only over neon and serf

Eric Gillespie <epg@pretzelnet.org> writes:

> I think I have a fix for neon; I found where the error comes
> from, anyway, and commenting out what looks like a redundant
> client-side check makes this test run to completion. I still
> have to run davautocheck and get some DAV gurus to look at it, to
> make sure I didn't break something else.

Nope, I just broke commit_tests.py 21: issue 644 attempt to add a
file twice. I had commented out the existence check in
ra_neon/commit.c:commit_add_file, thinking it redundant. But, I
don't understand why that block is necessary. If I remove it, we
should still get an error (from the *fs*) when the path already
exists in the txnroot. Why don't we?

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Sep 13 01:15:44 2007