[Clas_offline] inconsistent (?) PART bank information in reconstructed h10 Trees

[Arjun Trivedi]

I have been simulating elastic events using elast_gen 
(packages/aao/elast_gen), which generates PART banks. I notice that the 
information in the generated PART banks is not propagated properly (or 
rather in the manner I would expect it to) to the reconstructed h10 
Trees. Event generators that generate MCTK banks, seem to propagate fine 
to the reconstructed h10 Trees.

For example, after elast_gen event generation, the generated z position 
(as obtained from the PART bank) is centered at -25mm. However, after 
reconstruction, the "zpart" variable in the h10 tree is 0. Even the z 
momentum, "pzpart" is 0. The "qpart" variable resembles the distribution 
I would expect for "pzpart".

The input file that I used for reconstruction is at:
https://jlabsvn.jlab.org/svnroot/clas/users/atrivedi/sim/recsis_e1f.tcl

Is there a way to ensure that the generated PART banks and the 
information contained therein is propagated intact to the reconstructed 
h10 Trees?

Arjun Trivedi