[PATCH memctx-cache V2] memctx: always use cache for filectxfn

[Yuya Nishihara]

On Mon, 12 Jun 2017 16:52:36 +0900, FUJIWARA Katsunori wrote:
> At Sun, 11 Jun 2017 18:20:26 -0700,
> Sean Farley wrote:
> > 
> > # HG changeset patch
> > # User Sean Farley <sean at farley.io>
> > # Date 1497135618 25200
> > #      Sat Jun 10 16:00:18 2017 -0700
> > # Branch memctx-cache
> > # Node ID df6f0855d09ede751ec95e98fa840291f03c69e9
> > # Parent  7e9d0d8ff938dcf8ca193c17db5321a05a48e718
> > memctx: always use cache for filectxfn
> > 
> > I don't see a downside to doing this unless I'm missing something.
> > Thanks to foozy for correcting my previous bad logic.
> 
> LGTM :-)

Yeah. Queued per review, thanks.

> > diff --git a/mercurial/context.py b/mercurial/context.py
> > index 7ce34af..818f173 100644
> > --- a/mercurial/context.py
> > +++ b/mercurial/context.py
> > @@ -2104,17 +2104,17 @@ class memctx(committablectx):
> >          if branch is not None:
> >              self._extra['branch'] = encoding.fromlocal(branch)
> >          self.substate = {}
> >  
> >          if isinstance(filectxfn, patch.filestore):
> > -            self._filectxfn = memfilefrompatch(filectxfn)
> > +            filectxfn = memfilefrompatch(filectxfn)
> >          elif not callable(filectxfn):
> >              # if store is not callable, wrap it in a function
> > -            self._filectxfn = memfilefromctx(filectxfn)

One possible downside would be there might be a cache leak by e.g. ctx->p1/p2
chain, but I couldn't come up with a real example.