Re: [PATCH v4 1/2] diff-files --raw: show correct post-image of intent-to-add files

From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Jul 1, 2020, 09:52 UTC
Message-ID: <nycvar.QRO.7.76.6.2007011149000.56@tvgsbejvaqbjf.bet>
In-Reply-To: <xmqqeeq33t2q.fsf@gitster.c.googlers.com>

Hi Junio,

On Thu, 25 Jun 2020, Junio C Hamano wrote:

Show 11 quoted lines

> "Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com>
> writes:
>
> > From: Johannes Schindelin <johannes.schindelin@gmx.de>
> >
> > The documented behavior of `git diff-files --raw` is to display
> >
> > 	[...] 0{40} if creation, unmerged or "look at work tree".
>
> "on the right hand (i.e. postimage) side" is missing, which is
> crucial in this description.

True. Sorry for omitting that.

Show 5 quoted lines

> > This happens for example when showing modified, unstaged files.
>
> Modified I would understand.  We notice that the contents on the
> work tree is different from what is in the index, and we tell the
> consumer "look at work tree".  I do not think you meant "unstaged",

Right. I probably should have written "[...] when showing files with unstaged modifications".

Show 9 quoted lines

> though.  If truly removed from the index, diff-files won't even know
> about the path.  You probably meant "removed from the working tree",
> but 0{40} in that case means totally different thing (i.e. it would
> be a (D)eletion record, so 0{40} on the postimage side is a filler,
> not even "look at work tree").  What would make more sense to
> describe might be
>
> 	This happens for paths that are modified, or unmodified but
> 	stat-dirty.

Yes, but that includes more than I wanted to describe because the stat-dirty does not matter for intent-to-add files, and I wanted to point out the analogy between unstaged modifications and intent-to-add-files.

I noticed that you merged this commit into `next` already, so I am assuming that you do not want me to send a fifth iteration ;-)

Ciao, Dscho

Show 14 quoted lines

> Either case, we tell the consumer to check the "work tree".
>
> > For intent-to-add files, we used to show the empty blob's hash instead.
> > In c26022ea8f5 (diff: convert diff_addremove to struct object_id,
> > 2017-05-30), we made that worse by inadvertently changing that to the
> > hash of the empty tree.
> >
> > Let's make the behavior consistent with modified files by showing
> > all-zero values also for intent-to-add files.
>
> Well described.
>
> Thanks.
>