[PATCH 2 of 3] move store walking from streamclone.py into store.py

Matt Mackall mpm at selenic.com
Wed Jul 23 01:58:27 CDT 2008


On Wed, 2008-07-23 at 06:13 +0000, Dirkjan Ochtman wrote:
> Adrian Buehlmann <adrian <at> cadifra.com> writes:
> > Is there anything left I can do for patch 1 and 2,
> > so that at least these two patches could be included
> > or should this work be dropped together with patch 3?
> 
> Patch 1 and 2 look good to me, too, but I have one minor improvement that I
> think would make sense: instead of having storefiles() catch LockHeld and
> LockUnavailable and returning a sentinel value to indicate whether that call
> succeeded, why not just let them throw some exception? I'm currently on the
> fence about whether it should just raise the lock exceptions to the caller (just
> streamclone, I think), or catch the lock exceptions and raise something else.
> That would also allow for eliminating the first element of the return tuple, I
> think. Does that make sense?

I think so. I'm really trying to move the code away from returning
complex tuples.

-- 
Mathematics is the supreme nostalgia of our time.



More information about the Mercurial-devel mailing list