From: Philip Oakley <philipoakley@iee.org>
Date: Tue, 21 Feb 2017 22:40:06 GMT
Subject: Re: Git bisect does not find commit introducing the bug
Message-ID: <2C3983FC41474DCC880C1E53149DFBE0@PhilipOakley>
In-Reply-To: <CAMX8fZUR4h0T3hBmh6Z_0=f5LHFJetOi-vO_JJgnYPB7KjU_bg@mail.gmail.com>

From: "Alex Hoffman" <spec@gal.ro>
>> isn't that spelt `--ancestry-path` ?
>> (--ancestry-path has it's own issues such as needing 
>> an --first-parent-show
>> option, but that's possibly a by the by)
>
> Indeed it is spelled `--ancestry-path`. And interestingly enough you
> may use it multiple times with the wanted effect in our case (e.g when
> the user has multiple good commit and a single bad commit before
> running the bisect itself).

> Also it is `--first-parent` (not `--first-parent-show`),

My spelling was deliberate ;-)

If you use the currently coded --first-parent with a properly relevant 
commit for --ancestry-path then you get nothing. The purpose of 
ancestry-path is to find the ancestry chain that deviates from being a 
first-parent traversal [1], but the first-parent want to hold the walk to 
just the first parent chain - a contradiction.

Adding the -show at the end is my attempt to indicate that it is for the 
second aspect, that of selecting which commits to show/use.

I had an initial discussion back at [2], but failed then too.

> but I do not understand why do we need this
> option? What kind of issues does `--ancestry-path` have?
>
> Best regards,
> VG
>
--
Philip

[1] \git\Documentation\technical\api-revision-walking.txt ["two 
diff_options, one for path limiting, another for output"]
[2] 
https://public-inbox.org/git/2FA1998250474E76A386B82AD635E56A@PhilipOakley/