[Teammetrics-discuss] Commit Stat.

[Andreas Tille]

On Wed, Oct 12, 2011 at 12:29:40PM +0530, Sukhbir Singh wrote:
> I think it has stopped again.

Yes:

$ time ./commitstat.py -u tille

real    28m57.719s
user    0m18.085s
sys     0m4.212s

-> so fineally it went through as it seems. :-)

# SELECT count(*), project from commitstat group by project;
 count |     project     
-------+-----------------
     5 | python-apps
   324 | debian-live
   357 | pkg-scicomp
  6004 | pkg-games
  1210 | pkg-phototools
  2047 | pkg-grass
  1944 | pkg-perl
   236 | teammetrics
     4 | python-modules
  1978 | pkg-common-lisp
 14746 | pkg-multimedia
  3837 | pkg-java
  1299 | debian-med
    50 | debian-science
(14 Zeilen)

IMHO those numbers should be definitely higher (for pkg-perl, debian-med
and debian-science I'm 100% sure)

Also the graphs do look as if only git commits would be regarded which
somehow fits my feeling for the numbers.  I keep a copy of the graphs
created using render-all-graphs (which you can also perfectly run
yourself on blends.d.n ... and nay be you might be able to fix the
issues it throws) at

   /home/tille/debian-maintain/teammetrics/0_output_20111012

This directory should be accessible for you.  So please let me win the
commit-battle in debian-med (or at least second place behind charles).
:-)

I also noticed that there are some 'name-guest' entries in pkg-perl.
Please try to update the names.

Kind regards

       Andreas.

-- 
http://fam-tille.de