Hg sure is weird!

[mike.kupfer at xemacs.org]

>>>>> "MikeS" == Michael Sperber <sperber at deinprogramm.de> writes:

>> Mike, how are the commit notices currently generated?

MikeS> Essentially via hg log -p.

Are they triggered from the "changegroup" hook or the "incoming" hook?

I'm wondering if a changegroup hook like

    parent=`hg parents -r $HG_NODE | grep ^changeset: | head -1 | 
        sed -e s/.*://`
    hg diff -r $parent:tip

would produce more useful results.

I think the only time it would get tripped up with branches is when the
first node in a changegroup is a merge of two nodes that are already in
the parent--which seems like an unlikely scenario.

OTOH, I haven't thought about this much, so I wouldn't be surprised if
I'm missing something...

mike