[Rivet] Rivet analysis for CMS UE in DY

[Frank Siegert]

Hi Hendrik, Albert,

>> All of these are related to an API change of the ZFinder when going
>> from Rivet 1.5 to 1.6 a bit over a year ago. I can deal with the
>> transition of that if it's the only problem that is left in the
>> analysis.
>
> sounds good. I've attached a half cleaned up version of the code for
> you. If you can fix the ZFinder stuff, then I'll take care of all the
> division-by-zero and NaN problems, as well as the manual bin division.

I'm attaching a version where I have cleaned up the analyze method to
make use of the current ZFinder API and also of more projection logic
in general (instead of for loops). Although I'm pretty sure it's doing
what it should, I have to admit that I have only tested its
compilation so far and would hope that Albert & Co. can give the
cleaned up version (once your remaining changes are in) a quick round
of validation to make sure it's still producing the same results as
expected.

Cheers,
Frank

PS: Note that the original file had DOS/Windows line breaks, which I
have converted to Unix ones.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CMS_2012_I1107658.cc
Type: application/octet-stream
Size: 10574 bytes
Desc: not available
URL: <http://www.hepforge.org/lists-archive/rivet/attachments/20121008/21d964f9/attachment.obj>