Excerpts from Ryan McElroy's message of 2017-03-29 11:30:50 +0100: > On 3/29/17 4:31 AM, Jun Wu wrote: > > # HG changeset patch > > # User Jun Wu <qu...@fb.com> > > # Date 1490758201 25200 > > # Tue Mar 28 20:30:01 2017 -0700 > > # Node ID 5c6ebe1cd99e9557b61ef51ed88e335ac87df8da > > # Parent 31073077267b5e330925d48098dd6e0fd28bdd8e > > revlog: use raw revision for rawsize > > > > diff --git a/mercurial/revlog.py b/mercurial/revlog.py > > --- a/mercurial/revlog.py > > +++ b/mercurial/revlog.py > > @@ -439,5 +439,5 @@ class revlog(object): > > return l > > > > - t = self.revision(self.node(rev)) > > + t = self.revision(self.node(rev), raw=True) > > return len(t) > > size = rawsize > > > > Having looked at the code, this is inside the "rawsize" function so this > is an obviously correct fix. I have marked pre-reviewed in patchwork. > > Is there anywhere this gets exposed today? Maybe a test would be nice in > the future.
Since the index will handle most cases, it seems not easy to get exposed unless index returns 0. _______________________________________________ Mercurial-devel mailing list Mercurial-devel@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel