Re: Confusion about diffing branches

From: Junio C Hamano <gitster@pobox.com>
Date: Aug 27, 2007, 17:20 UTC
Message-ID: <7vmywczyfp.fsf@gitster.siamese.dyndns.org>
In-Reply-To: <20070827170504.GH4680@thunk.org>

Theodore Tso <tytso@mit.edu> writes:

Show 9 quoted lines

> The bigger inconsistency is with git-format-patch, which *does* take a
> set of commits, and where "git-format-patch a" does operates on a very
> different set of commits than what what "git-rev-list a" returns.  I
> understand that for backwards compatibility "git-format-patch a" is
> equivalent to "git-format-patch a..HEAD", but what it means is that
> all other ways of specifying sets of commits work with
> git-format-patch, *except* if you want to specify all commits from the
> beginning of time to a particular head (i.e., "git-rev-list a") is
> something that you simply can not possibly do using git-format-patch.

Heh, I really hate when people say this.

If you wanted to, you could:

	$ git format-patch a a

The original syntax

	$ git format-patch upstream

may have been a mistake and maybe

	$ git format-patch ^upstream

or even

	$ git format-patch --since upstream

might have been better. It is (1) too late to change now, and (2) for too small or perhaps negative a gain.

The reason why I say (2) is because _I_ think it is far more common and frequent to want to get "patches the other guy does not have" than "everything since nothingness up to this point".

Oops.  I think I have a solution.

	$ git format-patch a a

does not do _ROOT_ commit.  So you have to say

	$ git format-patch --root a a

for the above example to work. Why not tweak the option parser so that:

	$ git format-patch --root a

to do what you want? Without --root and with a single positive commit, it can keep doing the traditional "what I did since I forked from that guy's history".

Hmm...