[Hps-analysis] Trident cuts

[Rafayel Paremuzyan]

After today's analysis meeting, as Matt asked for Trident cuts,

- I require to have only 1 e+, and at least 1 e-
- Each e-and e+ should have chi2/NDF < 5 // remove bad tracks
- P_e- < 0.87 // Anti Fee cut
- e- and e+ have opposite sign tanLambda  // Opposite sectors

Determination of e- and e+ is done in a skim, that I am using.
In a bit more detail you can find it in the following document, but 
below the e-mail
I will try to describe it in a concise way.

https://confluence.slac.stanford.edu/download/attachments/195856171/Triden_Selection.pdf?version=1&modificationDate=1463766871000&api=v2

Rafo


======================= Description of Trident particle skim 
=================
== Loop over clusters
==== Throw ones that are outside of [30-52] time window
==== determine indexes of top and bottom clusters
== End of loop ever clusters


== Loop over GBL tracks
===== determine indexes of top and bottom tracks
== End of loop over tracks

// Since now we know indexes of top and bottom cluster, we can loop over 
them

==Loop over top cluster
===Loop over bottom clusters
====== determine delt_t and Esum,
====== check if these two clusters are coincident \\
====== if, yes then clusters are added in to the list of "good 
top(bottom)" clusters
===End of Loop over bottom clusters
==End Loop over top cluster

// Top Matching
==Loop over "Good Top clusters"
===Loop over top tracks

====== Check whether Top clusters and top tracks are matched
====== if yes then adds the track into the list of "good matched top 
tracks" and cluster in the list of "good matched top tracks"
====== if there is already another track matched to this cluster then 
check which one has the smallest chi2/NDf, and remove the other

=== End of Loop over top tracks
== End of Loop over "Good Top clusters"


// Bottom Matching
// Same thing as for Top Matching