[PATCH 1 of 3] copies: extract an explicit `computechangesetcopie` method from context

Yuya Nishihara yuya at tcha.org
Tue Aug 6 08:54:45 EDT 2019


On Tue, 6 Aug 2019 14:38:23 +0200, Pierre-Yves David wrote:
> 
> 
> On 8/6/19 2:25 PM, Yuya Nishihara wrote:
> > On Tue, 06 Aug 2019 11:50:38 +0200, Pierre-Yves David wrote:
> >> # HG changeset patch
> >> # User Pierre-Yves David <pierre-yves.david at octobus.net>
> >> # Date 1565054260 -7200
> >> #      Tue Aug 06 03:17:40 2019 +0200
> >> # Node ID fc8e461200c7262246ebd610100bcf9a75ded461
> >> # Parent  f95b59ffc307c4549d9640a81d750a99bd75f423
> >> # EXP-Topic extrameta
> >> # Available At https://bitbucket.org/octobus/mercurial-devel/
> >> #              hg pull https://bitbucket.org/octobus/mercurial-devel/ -r fc8e461200c7
> >> copies: extract an explicit `computechangesetcopie` method from context
> > 
> > The idea seems fine, but the last two functions are unrelated to "copies",
> > and I feel the copies module is more like a set of copy-detection algorithms.
> > 
> > Maybe it's better to move to scmutil or new module.
> 
> These two functions (added and removed) returns data that the changesets 
> centric copy tracing algorithm use. This is why I moved them to the 
> `copies` module.

IIRC, filesadded/removed are the sources of the template keywords.

> Do you still want them in the scmutil module ?

Yes, it seems less bad. And I can't think of a name for new module which
will host helper functions of context.py


More information about the Mercurial-devel mailing list