Re: [PATCH v3 0/3] add merge.renames config setting

From: Elijah Newren <newren@gmail.com>
To: Ben Peart <Ben.Peart@microsoft.com>
Cc: "git@vger.kernel.org" <git@vger.kernel.org>,
	"peff@peff.net" <peff@peff.net>,
	"gitster@pobox.com" <gitster@pobox.com>,
	"pclouds@gmail.com" <pclouds@gmail.com>,
	"vmiklos@frugalware.org" <vmiklos@frugalware.org>,
	Kevin Willford <kewillf@microsoft.com>,
	"Johannes.Schindelin@gmx.de" <Johannes.Schindelin@gmx.de>,
	"eckhard.s.maass@googlemail.com" <eckhard.s.maass@googlemail.com>
Subject: Re: [PATCH v3 0/3] add merge.renames config setting
Date: Thu, 26 Apr 2018 15:08:07 -0700	[thread overview]
Message-ID: <CABPp-BFh=gL6RnbST2bgtynkij1Z5TMgAr1Via5_VyteF5eBMg@mail.gmail.com> (raw)
In-Reply-To: <20180426205202.23056-1-benpeart@microsoft.com>

Hi Ben,

On Thu, Apr 26, 2018 at 1:52 PM, Ben Peart <Ben.Peart@microsoft.com> wrote:
> This is a complete rewrite based on the feedback from earlier patches.

Thanks for pushing forward on this.

> Update the documentation to better indicate command line options that override
> various config settings related to merge.
>
> Add a new config merge.renames setting to to control the rename detection
> behavior of merge.  This setting will default to the value of diff.renames.
>
> Also adds logic so that when rename detection is turned off, the aggressive
> flag is passed to read_tree() so that it can auto resolve more cases that would
> have been handled by rename detection.
>
> For the repro that I have been using this drops the merge time from ~1 hour to
> ~5 minutes and the unmerged entries goes down from ~40,000 to 1.
>
> Helped-by: Kevin Willford <kewillf@microsoft.com>
> Reviewed-by: Johannes Schindelin <johannes.schindelin@gmx.de>
> Signed-off-by: Ben Peart <Ben.Peart@microsoft.com>
>
> Base Ref: master

We may need to figure out how to coordinate amongst a few topics.
Looking over your patches, there are going to be a few conflicts with
en/rename-directory-detection-reboot, so this won't apply to pu.
Martin's series to introduce clear_unpack_trees_porcelain()[1], which
he was waiting to submit until mine went through, will also conflict
with this, if he uses the changes I suggested for the handling in
merge-recursive[2].  These aren't major conflicts, but I'm just
flagging it.

[1] https://public-inbox.org/git/cover.1524545557.git.martin.agren@gmail.com/
[2] https://public-inbox.org/git/20180424162939.20956-1-newren@gmail.com/