TBranch

TBranch is a class in the ROOT data analysis framework that represents a branch in a TTree. A TTree is a hierarchical container used to organize large datasets, and each branch stores data for a subset of variables or objects, often corresponding to a specific physical quantity or data structure. Within a branch, leaves (TLeaf) describe the data items and their types, and branches can contain primitive types, arrays, or user-defined objects.

Branches are created when filling a TTree, typically via TTree::Branch, where the branch name, the address of the data, and an optional leaf specification are provided. For reading data from a file, one associates a branch with a memory location using TTree::SetBranchAddress, enabling per-event data retrieval as the tree is iterated. The TBranch class provides access to metadata such as its name, the associated leaves, and the number of entries, as well as I/O operations to load data for each event.

Root I/O and performance are closely tied to branching. Splitting controls how complex objects are stored across branches and leaves, affecting storage size and read efficiency. Users can enable or disable the reading of specific branches to optimize performance when only a subset of data is needed for analysis.

TBranch interacts with other ROOT components, including TTree for organization, TFile for storage, and TLeaf/TBranchElement for data description and composition. It supports various data types, including primitive types, STL containers, and user-defined classes, making it a central element for organizing event-based data in ROOT analyses.