Re: [PATCH 4/2] Fix parent rewriting in --early-output

From: Paul Mackerras <paulus@samba.org>
Date: Nov 13, 2007, 21:55 UTC
Message-ID: <18234.7494.851156.347578@cargo.ozlabs.ibm.com>
In-Reply-To: <7vbq9yt1te.fsf@gitster.siamese.dyndns.org>

Junio C Hamano writes:

> I suspect that a "-" in place of a commit object name may not be
> enough for your purpose, as the _number_ of parents can later
> change in the later re-output.

I don't mind if a commit that has "-" as one of its parents later turns out to have more parents (i.e. the "-" can stand for zero or more unknown parents). I would be perturbed if a commit that didn't have any "-" in its parent list later turned out to have a different number of parents - but I don't think that's what you're implying, is it?

> I wonder if the presense of "incomplete" on the "Final output"
> line is a good enough indication for that.  That is, until you
> see "Final output: $N done", you will treat the parent
> information as unreliable.

The easiest way for me to handle an unreliable parent is just to ignore it. But I can't ignore all the parents, because then I wouldn't have a graph at all.

In other words, the presence of "incomplete" doesn't give me any clue as to which particular parent ids are reliable. As far as I can see, git log internally knows when a parent id is unreliable (it's one where it had to terminate the history simplification early), so it shouldn't be hard to tell gitk about that. And it would make my job a lot easier.

Paul.