Re: Is this a peg revision bug?

On Wed, Apr 04, 2007 at 10:04:24PM +0100, Malcolm Rowe wrote:
> This is true, but it's impossible to guarantee - the commits of r4 and r6
> are racing with all other commits by any user in the repository.
>

It strikes me that's not very clear. What I meant is that there's no
way to atomically commit a change via a wc that copies a file from the
revision that is HEAD at the time the commit is made (as you typically
want to do if you're renaming a file).

The solution might just be to do all renames and copies via URLs, but
that's not great. A better solution would be for log-tracing to
understand that if a node doesn't change between rX and rY, we can trace
back the history from rZ to the copyfrom rX and 'forward' to rY (which
is actually the same node, I think).

Regards,
Malcolm