scikit-hep / decaylanguage

Package to parse decay files, describe and convert particle decays between digital representations.
https://decaylanguage.readthedocs.io
BSD 3-Clause "New" or "Revised" License
42 stars 16 forks source link

Support all available Belle II decay models #401

Closed eduardo-rodrigues closed 9 months ago

eduardo-rodrigues commented 9 months ago

FYI @sognetic, this is what I could find in Basf2.

eduardo-rodrigues commented 9 months ago

FYI some files still give parsing errors, namely

['basf2/decfiles/dec/1113042002.dec',
 'basf2/decfiles/dec/1263700001.dec',
 'basf2/decfiles/dec/1263752000.dec',
 'basf2/decfiles/dec/3470012000.dec',
 'basf2/decfiles/dec/3470012001.dec',
 'basf2/decfiles/dec/3470030000.dec',
 'basf2/decfiles/dec/3470030001.dec',
 'basf2/decfiles/dec/BSemiTauonic_BSTD.dec',
 'basf2/decfiles/dec/BSemiTauonic_BSTD_2HDMTYPE2.dec']

I need to look into those, and help from Belle-II colleagues would be appreciated. For example the first file happens to have spurious characters a line after the End statement. That should not be a problem per se, meaning one could imagine a parser that ignores anything after the End statement, but it still is true that those spurious characters are a mistake.

Let's see for the rest of the files ...

eduardo-rodrigues commented 9 months ago

The 2nd file as the keyword PHOTOS twice in a decay line. Again, that is not a major issue in itself, and could easily be supported, but it was likely a copy-paste mistake to have ... PHOTOS PHOTOS ....

The 3rd file has a line finishing with LLSW 0.71 -1.6 -0.5 2.9;sig, and it is not allowed to get non-comment characters after the end-of-line ; (... LLSW 0.71 -1.6 -0.5 2.9;#sig would be OK).

Then files 3470012000.dec and 3470012001.dec in fact aren't dec files, since ASCII tables ;-). They meant to be .dat files as similar ones.

Etc. I will stop here.

I reckon I am going to merge and you see what to do with these files. Maybe worth opening an issue in the Basf2 repo and point to this discussion?

HTH.

sognetic commented 9 months ago

Hi @eduardo-rodrigues , thank you for pointing this out. It seems like we have some legacy issues in the code there which EvtGen doesn't seem to complain about. I'll try to get this fixed in our internal repo so it can be mirrored to the github one soon!