Cern ROOT - How to read contents from a TTree root file into array

Question

I am trying to read data in from a Cern ROOT TTree file. I have not used root before and I am struggling a bit with this. I am familiar with C++, so can sort the array side of things, but I have been through several tutorial pages online and haven't got anywhere.

How do I read data from a TTree file? I assume it would be done by iterating over the tree's nodes (again, I'm not exactly sure how the file is organized?) inside a for loop?

Here is an example of the sort of reference I have been trying to follow.

https://root.cern.ch/drupal/content/using-macro-read-ttree

As I understand it this is a generic question, but TTree's are not generic? (Since they could contain different variable names, I think?)

So, further information, which I think is important is:

By opening the root object browser TBrowser b - and browsing to the '.root' file, (and double clicking it) I can see that there are 12 variables in the file. For example, one is called 'mass', 'charge', etc...

Hope that's enough info? If not I can provide more.

Answer 1

This page seems to show nicely how this can be done: https://root.cern.ch/root/htmldoc/TTreeReader.html

The shortest usage example for your case would be:

TFile file("filename.root");
TTreeReader reader("treename", &file);
TTreeReaderValue<float> mass(reader, "mass"); // template type must match datatype
TTreeReaderValue<float> charge(reader, "charge"); // name must match branchname
...
while (reader.Next()) {
    // use *mass, *charge, ...
}

In the olden days, there used to be a more manual way of doing the same thing. You had to redirect the branches of the tree to your local variables. This method looks like this:

TTree* tree = (TTree*) file.Get("treename");
float mass, charge, ...;
tree->SetBranchAddress("mass", &mass);
tree->SetBranchAddress("charge", &charge);
...
for (int i = 0, N = tree->GetEntries(); i < N; ++i) {
    tree->GetEntry(i);
    // use mass, charge
}

From the TBrowser you can read off the names of the branches that you need to supply as the second paramter to TTreeReaderValue or SetBranchAddress.

Basically you should think of TTree as a collection of entries (classical trees). Each of the entries is made up of branches (nodes). This is how you read it.