Re: [PATCH] Remove restriction on notes ref base

From: Jeff King <peff@peff.net>
To: Shawn Pearce <spearce@spearce.org>
Cc: Johan Herland <johan@herland.net>, Kenny Root <kroot@google.com>,
	git@vger.kernel.org, Jonathan Nieder <jrnieder@gmail.com>,
	Thomas Rast <trast@student.ethz.ch>
Subject: Re: [PATCH] Remove restriction on notes ref base
Date: Tue, 2 Nov 2010 10:29:31 -0400	[thread overview]
Message-ID: <20101102142931.GA31394@sigill.intra.peff.net> (raw)
In-Reply-To: <AANLkTinN1UXSmkxOg59pT_xVd2eWS0Ms2sgAweLv7hbg@mail.gmail.com>

On Tue, Nov 02, 2010 at 07:11:45AM -0700, Shawn O. Pearce wrote:

> I didn't want to use refs/notes/bad-commits because its not really an
> annotation you would be looking at with git log.  But we do want to
> have a log of who banned particular SHA-1s from entering the
> repository, and being able to push that branch from a workstation to
> the server is a convenient way to edit that list of banned SHA-1s.

FWIW, I use refs/notes/cache/textconv/* to store textconv caches, and
there is no problem. Unless somebody specifically configures
GIT_NOTES_REF or notes.displayRef to see them, log should never look at
them.

So I think you are being overly cautious. That being said:

> I think the docs are correct, and the code is buggy.  If the user
> asked us to edit refs/meta/bad-commits, we should.  If the user asked
> us to edit refs/heads/my-branch... well, they asked us to edit it.
> :-)

I agree. This part of git is unlike most other parts, which tend to DWYM
with a partial ref (by prepending refs/heads, refs/tags, etc), but still
accept a full ref if the user really feels like organizing their refs
differently.

> A better safety measure might be to sniff the ref's contents and see
> what it is.  If the top level directory has a number of non-note like
> entries, we should abort editing the branch.  Its not common for users
> to name their directories "02" and "fe".

Agreed.

-Peff