[Rivet] flip3() in FourVector and FourMomentum

Andy Buckley andy.buckley at cern.ch
Thu Oct 1 21:52:48 BST 2015


On 01/10/15 08:37, Leif Lönnblad wrote:
> On 2015-10-01 00:27, Andy Buckley wrote:
>> I've added a "flip3()" method to FourVector and FourMomentum,
>> which does just that.
>
> Do you have any other use cases for this? If we can't find a
> reasonable name that describes what it does, maybe it is an
> indication that it is not really needed and that you should actually
> do it explicitly in MissingMomentum. Sorry, that was probably not
> very useful ;o)

(Back on list...)

Hi Leif,

I'm not convinced -- I think the lack of name (modulo Ben's nice
suggestion) says more about my limited lexicon than restricted 
usefulness of the physics concept!

I think there is a usability case for certain 3-vector operations (the 
ones which don't introduce ambiguity or inconsistency about how the E 
term should be updated) are useful to have directly on the 4-vector 
classes. 3-rotations of 4-vectors are already possible via the (barely 
used) LorentzTransformation class, and I think the only other such 
natural operation is this reflection/parity/reversal. Without a helper 
method, the user has to manually fix up the components somehow, which is 
not exactly hard, but is also easy not to think about (as I 
demonstrated). And making simple things easy is definitely Rivet 
philosophy as I've tried to implement it.

Alternative views welcome -- it looks like we'll have another day or so 
until the analysis integration is finished, so this can have the 
discussion that API changes deserve.

Thanks,
Andy

-- 
Dr Andy Buckley, Lecturer / Royal Society University Research Fellow
Particle Physics Expt Group, University of Glasgow


More information about the Rivet mailing list