From: Martin von Zweigbergk <martinv...@google.com>
Date: Monday, October 17, 2016 at 4:39 PM
To: Durham Goode <dur...@fb.com>, "email@example.com"
Subject: Re: [PATCH] treemanifest: fix bad argument order to treemanifestctx
On Mon, Oct 17, 2016 at 4:34 PM Durham Goode
On 10/17/16, 4:13 PM, "Martin von Zweigbergk"
># HG changeset patch
># User Martin von Zweigbergk
># Date 1476745932 25200
># Mon Oct 17 16:12:12 2016 -0700
># Node ID b36a81cd4015b9742d1fbb0d5f94207e7a400cdb
># Parent 8a864844d5a0c34bdb24d2e098a0cd339e32e020
>treemanifest: fix bad argument order to treemanifestctx
>Found by running tests with _treeinmem (both of them) modified to be
>diff -r 8a864844d5a0 -r b36a81cd4015 mercurial/manifest.py
>--- a/mercurial/manifest.py Wed Oct 12 21:33:45 2016 +0200
>+++ b/mercurial/manifest.py Mon Oct 17 16:12:12 2016 -0700
>@@ -1386,7 +1386,7 @@
> # Need to perform a slow delta
> revlog = self._revlog
> r0 = revlog.deltaparent(revlog.rev(self._node))
>- m0 = treemanifestctx(revlog, revlog.node(r0), dir=self._dir).read()
>+ m0 = treemanifestctx(revlog, self._dir, revlog.node(r0)).read()
> m1 = self.read()
> md = treemanifest(dir=self._dir)
> for f, ((n0, fl0), (n1, fl1)) in m0.diff(m1).iteritems():
Looks good to me. What would be required to get some _treeinmem code coverage
in the tests?
Twice the runtime? :-) But seriously, I suspect that's reason enough that we
don't want it enabled by default. And I don't know what subset would be useful
to run it on either.
But it was much less broken than I expected, actually. This was all that was
needed. I expected it to be broken before your series even started, but it
wasn't. There is one failing test case (test-rebase-newancestor.t) that I have
not understood why it's failing. It scares me a bit that I still don't know why
Just doing one run of test-manifest.t with it on would probably be enough to
catch most issues. What’s the exact command you used to run the tests? So I
can run it myself before sending more patches.
Mercurial-devel mailing list