On Friday 2007 October 05, Bill Lear wrote:
> I have a few changes I would like to see in this script, ones that I think
> would make it generally more useful. I don't have a clean patch, though,
> so should I just submit suggestions to you directly, Andy?
Me; or the list. Or both.
I'm happy to try to accommodate any suggestions. I'm happy if it is useful to
anyone other than just me :-)
--- SNIP THIS BIT IF YOU'RE NOT INTERESTED IN ITS BUGS ---
The big fault in it as it stands is that it doesn't try to reorder the refs
being updated to the most logical form. For example:
O --- * --- A (ref2)
\
B (ref1)
Let's say that both ref1 and ref2 were originally at O and this push has moved
them to these new locations. The email hook gets sent this information like
this:
refs/heads/ref1 O B
refs/heads/ref2 O A
The hook iterates through this list, for each ref update it shows only the
commits introduced by the change that aren't already included in an existing
ref. This is the problem, ref2 introduced "*" and "A" and ref1
introduced "B", ideally then the two emails would show
ref2 updated from O to A
new revs: *, A
ref1 updated from O to B
new revs: B
But because ref1 is alphabetically before ref1, what you get is:
ref1 updated from O to B
new revs: *, A, B
ref2 updated from O to A
new revs: <none>
I can't say I know what the answer is; nor even what the correct output should
be. If anyone has opinions on this, I'll be glad to hear them.
Andy
--
Dr Andy Parkins, M Eng (hons), MIET
andyparkins@xxxxxxxxx
-
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
