Changes to do with primary particle definition as implemented

[cholmcc]

in AliStack::IsPhysicalPrimary

• The code of AliStack::IsPhysicalPrimary was changed to conform with
  the definition adopted by ALICE in a recent public note:

A primary particle is a particle with a mean proper lifetime
tau larger than 1cm/c, which is either a) produced directly in the
interaction, or b) from decays of particles with tau smaller
than 1cm/c, restricted to decay chains leading to the interaction.

See also ALICE-PUBLIC-2017-005 at

 https://cds.cern.ch/record/2270008

Please note, that the definition is approved by the Physics Board

• The old implementation of AliStack::IsPhysicalPrimary is still
  available as AliStack::IsPhysicalPrimaryOld to allow testing etc.

• Various bugs in the old implementation (now AliStack::IsPhysicalPrimaryOld)
  and in AliStack::IsSecondaryFromWeakDecay were corrected. Specifically
  these bugs had to do with extracting the heaviest flavour or a
  particle (to see if a particle was heavy flavour or strange)

  To get the heaviest flavour of a particle with PID pdg the old code
  tried to extract the heaviest flavour quark by doing

  f = floor(|pdg| / 10^{floor(log10(|pdg|)})

  Value of f corresponds to the heaviest quark in the particle
  (1: d, 2: u, 3: s, 4: c, 5: b, 6; t) For most particles
  this is fine, e.g.,

  • pi+: pdg=211 -> f=2
  • K+: pdg=311 -> f=3
  • K0(S): pdg=310 -> f=3
  • K0(L): pdg=130 -> f=1(!)
  • p pdg=2212 -> f=2
  • B+: pdg=521 -> f=5
  • D+: pdg=411 -> f=4

  while for others it will give the wrong result

  • D(0)* pdg=10421 -> f=1 (should be 4)
  • K(1)+ pdg=10323 -> f=1 (should be 3)
  • f(0)(500) pdg=9000221 -> f=9(!) (should be 2)

  and the leptons are really messed up

  • e- pdg=11 -> f=1
  • mu- pdg=13 -> f=1
  • tau- pdg=15 -> f=1

  The definition of the PDG code conventions state that each
  PDG code is formatted as

n nr nL nq1 nq2 nq3 nJ

where nq1, nq2, and nq3 encode the quark content.
For mesons nq1=0. So a better way to extract the heaviest
flavour is to take the absolute value of the PDG code, modulo
10,000 and then, depending on whether the value is larger than
999 (baryons) extract the 4th or 3rd digit

    if (|pdg|<100 && |pdg|>6) return 0; // No quark content
    m = |pdg| % 10000; // get nq1 nq2 nq3 nJ
    if (m == K(0)L) return 3; // special for K0L
    f = (m > 999) ? m / 1000 : m/100;

Note, for nuclei the contvention is to encode as

10LZZZAAAI

where L is the number of Lambda's in the nucleus. So we
should check if this number is non-zero and in that case
get 3 (strange), otherwiese 2 (up). I.e., the full
pseudo-code is

    if (|pdg|>1000000000)
      f = (|pdg|/10000000) % 10) != 0 ? 3 : 2;
    else if (|pdg|<100 && |pdg|>6)
      f = 0; // No quark content
    else
      m = |pdg| % 10000; // get nq1 nq2 nq3 nJ
      if (m == K(0)L) return 3; // special for K0L
      f = (m > 999) ? m / 1000 : m/100;

This algorithm is implemented in the service function
HeaviestFlavour in the scope of AliStack.cxx

• Other small efficiency fixes.

[alibuild]

729f879: approval required: 1 of @jgrosseo (Jan Fiete Grosse-Oetringhaus), @chiarazampolli (Chiara Zampolli), @qgp (Jochen Klein), @dberzano (Dario Berzano), @sawenzel (Sandro Christian Wenzel), @ktf (Giulio Eulisse), @shahor02 (Ruben Shahoyan)

Comment with +1 to approve and allow automatic merging,or with +test to run tests only. Please comment on the pull request: click here and comment at the bottom of the page.

[dberzano]

+1
thanks for specifying that this is approved by the PB :-)

[alibuild]

729f879: approved: will be automatically merged on successful tests

[amorsch]

The while-loop in AliStack::IsPhysicalPrimary can go back to the colliding particles which are stable.
In this case the particle is falsely flagged as a secondary. This happens quite frequently. We have to switch back to the old version of the code until this is fixed.