[Rivet] DISLepton broken in 2.6.0?

Andy Buckley andy.buckley at cern.ch
Tue Feb 20 11:43:36 GMT 2018


Aha, I see the problem: the project() method was calling the "PromptFS" projection, which was correct, but the compare() method was recursing to comparison of a non-existent "FS" projection.

That's definitely a bug, but it wasn't activated in my testing because I was only running a single analysis with a DISLepton projection, and the way that the projection handler works means that the compare() method was never called, because there was never an exact type-match. I guess you were running several DIS analyses at the same time, Holger?
The correct fix is either to change the name PromptFS -> FS (which I've done now on the release-2-6-x branch), or change FS -> PromptFS in the compare method. Updating to the release branch is probably easiest. Thanks!
Andy
Dr Andy Buckley, Lecturer / Royal Society University Research Fellow
Particle Physics Experiment Group, University of Glasgow

On Feb 20 2018, at 11:37 am, Andy Buckley <andy.buckley at cern.ch> wrote:
>
> That's weird -- I tested that just before release. No idea how it got mangled :-/
> The correct fix is that the PromptFS should be just called "FS" -- we're only using prompt leptons in the DIS lepton definition now. I'll fix it for 2.6.1.
> Andy
> Dr Andy Buckley, Lecturer / Royal Society University Research Fellow
> Particle Physics Experiment Group, University of Glasgow
>
>
>
>
>
>
>
>
>
>
>
>
> On Feb 19 2018, at 11:52 pm, Holger Schulz <iamholger at googlemail.com> wrote:
> >
> >
> > Hi,
> > it would seem as if the DISLepton projection is not working in the release 2.6.0:
> > > Error in NUDIS_LEPTONISO::init method: No projection 'FS' found for parent 0x7ffd8ccf2540
> > This happens consistently when loading any of the analyses using the DISLepton (e.g. H1_2000_S4129130, H1_1994_S2919893.cc)
> > I can locally fix this by adding the fat line to include/Rivet/Projections/DISLepton.hh
> >
> >
> >
> >
> > DISLepton(){
> > setName("DISLepton");
> > addProjection(Beam(), "Beam");
> > addProjection(FinalState(), "FS");
> > addProjection(PromptFinalState(), "PromptFS");
> > }
> >
> >
> >
> >
> > Not sure if his is correct though.
> > What do you think?
> > Holger
> >
> > _______________________________________________
> > Rivet mailing list
> > Rivet at projects.hepforge.org
> > https://www.hepforge.org/lists/listinfo/rivet
> >
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.hepforge.org/lists-archive/rivet/attachments/20180220/f4adb2c4/attachment.html>


More information about the Rivet mailing list