After declaring the Root FullRecon operational earlier this week, I
discovered that wasn't quite so. My EventSource class to read Root files
had problems. I also pursued re-establishing our event header class,
Event, which we had abandoned in favour of separately readable branches.
It turns out Root has a feature which allows creation of sub-branches
automatically, so I have been (vainly) attempting to use this feature.
This would allow us to have the Event class as header, but its contents
would be broken up into separate branches. Unfortunately, if I use the
split option, the pointer to my Event class never gets set when reading
the events back in. For the moment, I am leaving split disabled (so our
Event classes look much like they did before Rene came to talk to us!)
on the hope that it can be sorted out. I'm in touch with Rene about it.
This has been pretty frustrating!
Otherwise, the FullRecon seems functional now: TrackSmear and
ClusterCheater are doing their things. Next step in dev is to make the
appropriate track, cluster classes inherit from the 4-Vector class in
prep for the jet/thrust finders.
Also, David Williams has expressed interest in joining our merry band
(at an undetermined level); we will open discussions next week. I am
leaning to suggesting he write an interface to the FNAL MCFAST track
fitting package.
Richard
--
Richard Dubois
SLD, Stanford Linear Accelerator Center
[log in to unmask]
http://www.slac.stanford.edu/~richard/
650-926-3824
650-926-2923 (FAX)
|