[PATCH 03 of 10] histedit: use cg?unpacker.apply() instead of changegroup.addchangegroup()

[Augie Fackler]

> On Oct 14, 2015, at 14:10, Gregory Szorc <gregory.szorc at gmail.com> wrote:
> 
> On Wed, Oct 14, 2015 at 10:34 AM, Augie Fackler <raf at durin42.com <mailto:raf at durin42.com>> wrote:
> # HG changeset patch
> # User Augie Fackler <augie at google.com <mailto:augie at google.com>>
> # Date 1444770861 14400
> #      Tue Oct 13 17:14:21 2015 -0400
> # Node ID a2cf25ebbe3d6eaa5e2dcddd355ff8384dc35028
> # Parent  2c2ad8c31a6d64c9186a7260d82aef79c61c75ac
> histedit: use cg?unpacker.apply() instead of changegroup.addchangegroup()
> 
> diff --git a/hgext/histedit.py b/hgext/histedit.py
> --- a/hgext/histedit.py
> +++ b/hgext/histedit.py
> @@ -163,7 +163,6 @@ import sys
>  from mercurial import cmdutil
>  from mercurial import discovery
>  from mercurial import error
> -from mercurial import changegroup
>  from mercurial import copies
>  from mercurial import context
>  from mercurial import exchange
> @@ -823,8 +822,7 @@ def _histedit(ui, repo, state, *freeargs
>                  backupfile = repo.join(state.backupfile)
>                  f = hg.openpath(ui, backupfile)
>                  gen = exchange.readbundle(ui, f, backupfile)
> -                changegroup.addchangegroup(repo, gen, 'histedit',
> -                                        'bundle:' + backupfile)
> +                gen.apply(repo, 'histedit', 'bundle:' + backupfile)
> 
> FWIW, this code is already wrong because it fails to handle bundle2. But this is probably orthogonal to this series (we need a high level "apply this bundle" API). 

Yep. I'm trying to avoid scope creep on this refactor - totally agree that bundle application needs to be its own thing distinct from changegroups.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://selenic.com/pipermail/mercurial-devel/attachments/20151014/93b7dc61/attachment.html>