Re: [RFC] get_sha1() shorthands for blob/tree objects

From: Linus Torvalds <torvalds@osdl.org>
Date: Apr 19, 2006, 04:14 UTC
Message-ID: <Pine.LNX.4.64.0604182108070.3701@g5.osdl.org>
In-Reply-To: <7v8xq2ciws.fsf@assigned-by-dhcp.cox.net>

On Tue, 18 Apr 2006, Junio C Hamano wrote:

> 
> A small fry in the ointment.  What should the parts that are
> output with --name-only say for such a diff?

I have no idea, I have to say ;)

> Blob references like v0.99.6:git-commit-script are resolved by
> the extended SHA1 interpreter, and all what the caller of
> setup_revisions() can see and feed the diff machinery with has
> are their object names.

Actually, the names are there. The object list has the "->name" field (used to do the name-based sorting), and we actually even fill it in for the stuff we pass in as arguments. See the "add_pending_object()" calls in setup_revisions().

We just don't use them right now. We _could_.

Show 6 quoted lines

> Something like this is a possibility, but is ugly.
> 
>         diff --git a/a2455b0... b/01c73bd...
>         index a2455b0..01c73bd 100644
>         --- a/a2455b0...
>         +++ b/01c73bd...

Yes. But if you look at the object list name (in the "pending_object" thing), you _could_ actually get this to be something like

	diff --git v0.99.6:git-commit-script git-commit.sh
	index a2455b0..01c73bd 100644
	--- v0.99.6:git-commit-script
	+++ git-commit.sh

which would be much prettier, although I'm not saying it's necessarily worth it. I'm just saying that it's _possible_ with the cmd line parsing infrastructure we have now.

		Linus