Reconstructible:

An MC decay to be reconstructible implies that all final state MCParticles are
reconstructible (the final state particles need to be "tagged" in the decay
description).

• A stable charged MCParticle can be reconstructible in one of the following
  categories (checked in the following order in the code)     

Long       the MCParticle has 3r+3phi velo clusters and 1x+1stereo clusters

           in each of the 3 seed stations
Upstream   the MCParticle has 1x+1stereo clusters in each of the 3 seed

           stations and 3 TT clusters
VeloTT     the MCParticle has 3r+3phi velo clusters and 3 TT clusters

• A stable neutral MCParticle is recontructible if it is in the sensitive 
  acceptance (1 cell in) of the ECAL at z of shower max and the have a 

  pt>200Mev at origin.

Reconstructed:

A decay to be reconstructied implies that all final state MCParticles are
reconstructed meaning an associated ProtoParticle has been found (hence it
depends on which type of tracks have been used to make charged ProtoParticles
and which CaloHypothesis have been used to make Neutral ProtoParticles).

A decay is classified as "Long" if all charged MCparticles in the decay
were associated to charged ProtoParticles originating from long tracks.
If neutral particles are present they have to be recontructed as well. 
The summary classification applies only to the charged part of the decay.
In the full break-down table ( working only for some specific decays at the
moment ) two separate tables are provided if the decay consist completely of
charged MCParticles, completely of neutrals, is a mix of charged of neutrals.

Since the reconstructed definition is based on the association between an
MCParticle and a ProtoParticle below if how this done for charged and neutrals.

• A charged ProtoParticle is associated to an MCParticle if the TrStoredTracks
  from which it originates is. This is true when the TrStoredTrack and

  the MCParticle if:

     Long tracks     share at least 70% of the clusters in the Velo

                     and 70% in the Seeding Stations
     Upstream tracks share at least 70% of the clusters in the Seeding Stations

                     and have no more than 1 different cluster in TT out of those

                     used
     VeloTT tracks   share at least 70% of the clusters in the velo and have

                     no more than 1 different cluster in TT out of those used

• A neutral ProtoParticle is associated to an MCParticle if the CaloClusters used
  for the CaloHypthesis from which it originates are.

A CaloCluster in the OODST for TDR production is associated to an MCParticle 

if the MCParticle is entering directly in the ECAL cluster and contributing 

to at least 10% of the cluster energy or if the sum of the energy of its 

descendant entering the cluster is contributing to at least 10% of the cluster

energy.

     This 10% cut was not covering all gammas from merged pi0 and a work around 
     has been put in the efficiecny algorithm to check that for gamma from pi0 
     that are not associated the pi0 would be and the gammas are within a single

     cluster. 

     Starting with data produced with Brunel v18r0 (hence this applies to the

     Global robustness test data) the contribution to the cluster energy was

     reduced to at least 1% or 200 MeV.

  Note that a ProtoParticle can be associated to more than one MCParticle and
  vise-versa.