Re: bisect-helper: we do not bisect --objects

From: Christian Couder <christian.couder@gmail.com>
Date: Mar 3, 2017, 23:04 UTC
Message-ID: <CAP8UFD0tzH+QiNM2BMhe9qcKdD0rP9T=Ry94=EPV+=d+7cdLfQ@mail.gmail.com>
In-Reply-To: <xmqqvarq9vzo.fsf@gitster.mtv.corp.google.com>

On Fri, Mar 3, 2017 at 5:16 PM, Junio C Hamano <gitster@pobox.com> wrote:

Show 22 quoted lines

> Ever since "bisect--helper" was introduced in 1bf072e366
> ("bisect--helper: implement "git bisect--helper"", 2009-03-26),
> after setting up the "rev-list $bad --not $good_ones" machinery, the
> code somehow prepared to mark the trees and blobs at the good boundary
> as uninteresting, only when --objects option was given.  This was kept
> across a bit of refactoring done by 2ace9727be ("bisect: move common
> bisect functionality to "bisect_common"", 2009-04-19) and survives
> to this day.
>
> However, "git bisect" does not care about tree/blob object
> reachability at all---it purely works at the commit DAG level and
> nobody passes (and nobody should pass) "--objects" option to the
> underlying rev-list machinery.  Remove the dead code.
>
> Signed-off-by: Junio C Hamano <gitster@pobox.com>
> ---
>
>  * Christian, do you recall what we were thinking when we added this
>    mark_edges_uninteresting() call in this program?  If you don't,
>    don't worry--this was done more than 8 years ago.  I am just
>    being curious and also a bit being cautious in case I am missing
>    something.

I think I just copy pasted the code from cmd_rev_list() in builtin-rev-list.c and probably didn't realize that revs->tree_objects would always be false.

Thanks for spotting this and removing the dead code.