[Rivet] Mailing list oops / 2D histos

Albert Knutsson albert.knutsson at desy.de
Tue Oct 5 12:41:34 BST 2010


  Dear Andy,

thanks for your answer.

As soon as I #include "TH2F.h" in my analysis routine I get

Rivet.AnalysisLoader: WARN  Cannot open RivetAKRIDGETEST.so: 
RivetAKRIDGETEST.so: undefined symbol: _ZN13TVersionCheckC1Ei
'AKRIDGETEST' is not a valid analysis. Available analyses are:

and I can't run the analyser. No complaints when compiling.
Without the TH2F include, running the code work as usual.

Any suggestions?

Cheers,
Albert


On 10/5/10 10:05 AM, Andy Buckley wrote:
> Hi everyone,
>
> Apologies if I've been unusually unresponsive to list emails recently. I
> only just realised that my old Durham address had been bouncing, so I've
> missed Rivet list emails for the last... actually, I don't know how
> long. Since the start of September, I suspect. D'oh: I really thought I
> had updated my list subscription to use a valid email address.
>
> Anyway, apologies for the unintentional silence. I've heard via covert
> channels that there's been discussion about 2D histos in Rivet... yeah,
> if we'd had time to get YODA into action 1+ years ago we could have
> avoided this. A pity that we didn't get user feedback about it being
> needed: we definitely don't want to force people to look elsewhere.
>
> At any rate, ROOT 2D histos can be used within Rivet and the
> binEdges(...) functions can be used to do the autobooking for 1D histos.
> We don't have any autobooking hooks for 2D histos yet, but I can put
> them in for Leif's 2D histos in LWH -- and they will then be ready for
> swap-over to YODA when that's usable. I've said that a lot, but there is
> not so much more work needed: I would personally like to have a clear
> run at that in the next couple of months so if others can pick up the
> slack in the analysis development and core Rivet maintenance it will
> happen faster.
>
> Cheers,
> Andy
>



More information about the Rivet mailing list