[PATCH 3 of 3 v2] graft: apply get-only changes in-memory

[Pierre-Yves David]

On Mon, Dec 03, 2012 at 10:28:47PM +0000, David Schleimer wrote:

> Regarding the reviewability.  The code that actually moved is the bit that
> shows up as unchanged in the middle.  Doing that in a prior changeset helps a
> little, but most of the diff still ends up as a single large insertion.  I'll
> resend with it split up into more pieces, but I don't think I can get rid of
> the giant hunk.

Ok, thanks

> > Applying this speedup to trivial filemerge without conflict seems possible
> > too.
> 
> Yeah, it's absolutely possible.  The reason I haven't done it is that it's
> quite a bit more complex, and this diff is makes graft good enough for our
> needs.

I was noting this as possible future extension of this work, not requesting this to be done right now.

> > Same here, This change to extra is not necessary until you use in memory
> > context but that would work to apply that before.
> 
> When I pull the code movement into a separate patch, this appears as a
> separate hunk.  I think that's sufficient to make the patch readable, and
> that sending too many patches can be as hard to review as too few so I'm not
> going to split it out.

Seems fairs

-- 
Pierre-Yves David

http://www.logilab.fr/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://selenic.com/pipermail/mercurial-devel/attachments/20121204/2bd3dd0d/attachment.pgp>