Re: r1295006 and openhub

Stefan Sperling wrote on Sun, Oct 16, 2016 at 23:22:22 +0200:
> Reading our openhub stats for fun I noticed that the stats
> are totally screwed. It seems they stop scanning our history
> at r1295006 (the infamous commit that changed our trunk's copy ID).
> https://www.openhub.net/p/subversion/commits?sort=oldest
>
> It seems openhub have a problem distinguishing branches and copies :-)
>
> Should we bother them about this? Some contributors are severely
> underrepresented.

*sigh* The cat is out of the bag. That commit was no accident; it was
a plot by me to inflate my openhub stats. ;-)

Seriously, though, my apologies to those contributors underrepresented in
consequence of that commit.

> I wonder if this problem affects other projects as well.

It would also affect projects that replace trunk with a branch, with
«svnmucc rm trunk cp HEAD branches/2.0 trunk» (or, if the project devs
are unaware of svnmucc, by doing the same thing in two commits).

I would say that yes, dropping them a line to the effect of "You don't
handle certain histories correctly" would be worthwhile. Do we have
a contact at ohloh/openhub/blackduck?

(I assume I ought to volunteer for writing that email, so I do.)

Daniel
Received on 2016-10-17 00:38:34 CEST