Hi chi, maybe is good that I re-iterate our mail-etiquette
(or at least the intention of it) just in everybody's interest.
For a message like yours it is sufficient to mail it once, only to the
blast_anaware list. But don't worry, I understand there could be some
confusion at the beginning. All the people you want to reach are on _anaware.
The same people are also on blast_shift, but that should be used only for
shift summaries/scheduling. This helps the web page, the mailbox and later
the retrieval of info.
Again, the subdetectors lists are there, it's up to the individual contributors
and all the interested people to subscribe (http://blast.lns.mit.edu/BlastTalk)
In excess to the local users/workers, BlastTalk also contains *all* the people in
the collaboration, including people in far away places. So the idea is that we
should use BlastTalk sparingly (very general announcements) and use the local lists
to help organize the local discussions.
There is also no need to send to multiple lists. Again, every local user
(bates/campus) is already in _anaware, _shifts and Talk. Finally, all
messages go to the web which is a nice way to store and consult them
But thanks for your message anyways,
Regards
-- tancredi
________________________________________________________________________________
Tancredi Botto, phone: +1-617-253-9204 fax: +1-617-253-9599
research scientist MIT/Bates, 21 Manning Av Middleton MA, 01949
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
On Mon, 12 Aug 2002, zhangchi wrote:
>
> Hi blasters:
>
> As an attempt initiated by Tancredi to unify and make easy the cuts people
> are using in analysis of blast data, I checked in a new class TBLCut, and
> a few of its subclasses, namely TBLHid(Hit ID), TOF_HIT, CC_HIT, into
> BlastLib2. They are compiled into libBlast.so so if you loas the lib, they
> are available for use.
>
> It is only a premitive type, does not do much yet. I will be adding more
> into it. Right now at least I wish it could save people some typing by
> doing those "atl%dt>300" things automatically.
>
> Also I would like to welcome any suggestions.
>
> Chi
>
>
> I will check in later a script containing sample usages of the
> classes. Before that, please refer to the following few lines if you
> really were trying to use the class:
>
> .L libBlast.so;
>
> TFile *run = new TFile("commis-923-925.root","r");
> TTree *tree = (TTree*)run->Get("ntp");
>
> TBLCut::Init(tree) // tell TBLCut which tree you want to analyze
>
> //TBLHid::Init("HitIDCalib") // optional, the initialization is already
> // done when libBlast.so is loaded
>
> tree->Draw("ttl12t", TOF_HIT('l', 13)); // a clean plot with garbages cut out
>
> // TBLCut supports basic logic manipulations +/||, */&&. !, -.
> tree->Draw("ttl12t", TOF_HIT('l', 13) && TOF_HIT('r', 3)) // a coinc plot
>
> TOF_HIT l_12('l', 12); // create a cut
>
> TBLCut cut1 = l_12, cut2(cut1); // supports copy and assignment
>
> cut1.Examine(); // tells you what the cut is
>
> TBLCut cut3 = cut1 * cut2; // same thing as cut1 && cut2
> TBLCut cut4 = cut1 + cut2; // smae thing as cut1 || cut2
> TBLCut cut5("cut5", "atl12t > 1000 && ttl9b < 3000") // create a generic cut
>
> tree->GetEntry(10);
> cout<<l_12<<endl; // this casts the cut into a double according as if
> // current event passed the cut or not, for this event, it is 0
> // it can then be cast into a bool to control you program flow.
>
> tree->GetEntry(11);
> cout<<TOF_HIT('l', 12)<<endl; // this creates a cut on the fly and uses
> // it, for this event, return value is 1;
>
>
> // subscription [i] is to evaluate the cut on the ith entry in the tree;
> for (int i=0; i<50; i++)
> if (l_12[i])
> cout<<"event "<<i<<" passed the cut<<endl;
>
> // unfortunatly subscription does not work with cuts created on the fly,
> // use the function Eval(int)
> for (int i=0; i<50; i++)
> if (TOF_HIT('l', 12).Eval(i))
> cout<<"event"<<i<<" passed the cut"<<endl;
>
> Enjoy.
>
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:28 EST