[PATCH 3 of 3] revset: two new revsets, copies and renames

Martin von Zweigbergk martinvonz at google.com
Mon Apr 8 13:10:09 EDT 2019

On Sun, Apr 7, 2019 at 5:50 PM Jordi GutiƩrrez Hermoso <jordigh at octave.org>

> On Sun, 2019-04-07 at 09:48 +0900, Yuya Nishihara wrote:
> > So they are quite similar to -r 'file("set:copied()")', and we do include
> > renames in copies in general.
> Hm, but we don't have a renamed() fileset, do we? How do you find renames?

Yes, I don't think there is one.

> > I think these predicates are useful, but it's confusing to handle copies
> > and renames separately only in revset.
> I agree, consistency is important, but I don't know how to implement a
> fileset or revset for finding renames.

Hmm, this patch has a revset for finding renames... :)

Anyway, I agree with Yuya that it will be confusing if the copied() fileset
includes copies and the copies() revset doesn't. Perhaps we can add
arguments to both of them? Something like
copied(renames=True,purecopies=True)? I don't think you need to implement
such support for the copied() fileset, but I'd like to have a plan for how
to eventually make them consistent.

> Mercurial-devel mailing list
> Mercurial-devel at mercurial-scm.org
> https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.mercurial-scm.org/pipermail/mercurial-devel/attachments/20190408/d841fba7/attachment.html>

More information about the Mercurial-devel mailing list