From: Junio C Hamano <junkio@cox.net>
Date: Sun, 01 May 2005 21:18:13 GMT
Subject: Re: [PATCH] Really fix git-merge-one-file-script this time.
Message-ID: <7vsm16sl0q.fsf@assigned-by-dhcp.cox.net>
In-Reply-To: <Pine.LNX.4.58.0505011312080.2296@ppc970.osdl.org>

>>>>> "LT" == Linus Torvalds <torvalds@osdl.org> writes:

>> Linus, have you decided to like or dislike the behaviour of
>> git-merge-one-file-script touching the work tree in some cases
>> but not in other cases?

LT> I don't care about the _successful_ merge, since a successful merge is 
LT> basically always followed by a "git-checkout-cache -f -a" anyway (and 
LT> update-cache + remove now-stale files etc).

Let me make sure I understand you correctly before I go
further.  In the above sentence, do you mean a merge that is
_successful_ by merges that:

    (1) "git-remove-tree -m" considers them trivial;

or
    (2) in addition to (1), "git-merge-one-file-script"
        considers them trivial.  That is, only one-side changes
        or removes or adds it, or both sides adds it
        identically, or merge/diff3 merges without conflict.

Which one?

LT> Sounds sane.

LT> On the other hand, I think it would actually be easier to just make your 
LT> "magic SHA1" be just another "stage".

I am undecided on this one but having worked on the diff side, I
think the magic SHA1 of 0{40} would match better with what diff
needs to do.  It would mean "cache is stale, look in the work
tree" to them.