[Rivet] WFinder does not keep charged particle cuts

[Frank Siegert]

Hi Holger and Daniel,

On 26/05/11 22:16, Daniel Weyh wrote:
> Hi all,
>
> I'd guess it's in these lines
> ----WFinder.cc-------L103-----------
>      // FS for non-signal bits of the event
>      VetoedFinalState remainingFS;
>      remainingFS.addVetoOnThisFinalState(signalFS);
>      addProjection(remainingFS, "RFS");
> --------------------------------------------
> By now I don't know pretty much about when projections are evaluated.
> But given as far is I (think I have) seen, these lines do a veto on the global event.
> For maintaining cuts made earlier, second line would maybe have to be changed to
> --------------------------------------------
>      VetoedFinalState remainingFS(fs_l);
> --------------------------------------------
> where fs_l is the input final state.

No, this doesn't make sense to me. The remainingFS is supposed to give 
you everything but the identified constituents of the W, such that you 
can use it for jet clustering etc.

The ChargedFinalState passed to one of the WFinder constructors is 
supposed to contain the (cuts on the) leptons for the W reconstruction, 
so it's not related to remainingFinalState at all.

Best,
Frank

>
> But here again the question about inteded behaviour...
>
> Cheers,
> Daniel
>
> PS @ Holger: In our MC_EVENTSHAPES I think I solved the problem by using some veto-juggling... (therefore no recompilation of rivet/Sherpa is needed).
> Look into the git repo (git pull) ;-)
>
> Am 26.05.2011 um 18:03 schrieb Holger Schulz:
>
>> Hi guys,
>>
>> we realised that when constructing a WFinder with a ChargedFinalState
>> as first argument in the constructor, the pT cuts for instance as applied
>> on the ChargedFinalState are ignored when asking for the WFinder's
>> remainingFinalState.
>>
>> Is this the intended behaviour?
>>
>> Cheers,
>> Holger
>> _______________________________________________
>> Rivet mailing list
>> Rivet at projects.hepforge.org
>> http://www.hepforge.org/lists/listinfo/rivet
>
> _______________________________________________
> Rivet mailing list
> Rivet at projects.hepforge.org
> http://www.hepforge.org/lists/listinfo/rivet