Theoretical Analysis of Git Bisect

Abstract

In this paper, we consider the problem of finding a regression in a version control system (VCS), such as git. The set of versions is modelled by a Directed Acyclic Graph (DAG) where vertices represent versions of the software, and arcs are the changes between different versions. We assume that somewhere in the DAG, a bug was introduced, which persists in all of its subsequent versions. It is possible to query a vertex to check whether the corresponding version carries the bug. Given a DAG and a bugged vertex, the Regression Search Problem consists in finding the first vertex containing the bug in a minimum number of queries in the worst-case scenario. This problem is known to be NP-complete. We study the algorithm used in git to address this problem, known as git bisect. We prove that in a general setting, git bisect can use an exponentially larger number of queries than an optimal algorithm. We also consider the restriction where all vertices have indegree at most 2 (i.e. where merges are made between at most two branches at a time in the VCS), and prove that in this case, git bisect is a 1 log 2 (3/2) -approximation algorithm, and that this bound is tight. We also provide a better approximation algorithm for this case. Finally, we give an alternative proof of the NP-completeness of the Regression Search Problem, via a variation with bounded indegree.

Random Generation of Git Graphs

“…The fact that we forbid merges of multiple feature branches into the main one is not a restriction of the VCS, but is advisable to maintain a clearer and more understandable project history, reduce the risk of conflicts, and enhance traceability and maintainability. This restriction is also discussed in [2].…”

“…By generating diverse graph structures that adhere to the workflow's specifications, we ensure a comprehensive examination of the VCS's behavior according to plausible scenarios. To give a concrete example based on work by one of the authors [2], a random DAG sampler could experimentally check the effectiveness of git bisect, an algorithm that finds the commit where a bug has been introduced.…”