Use rerun to track the number of "TODOs" in a code base across commits

rerun_commit_todos.py

from pathlib import Path

from git import Repo

import rerun as rr

GIT_REPO = Path("/tmp") / "rerun"


if not GIT_REPO.exists():
    repo = Repo.clone_from("https://github.com/rerun-io/rerun.git", GIT_REPO)
else:
    repo = Repo(GIT_REPO)


origin = repo.remotes.origin
origin.fetch()
origin.pull()


main_branch = repo.heads["main"]
main_branch.checkout()


all_commits = list(reversed(list(repo.iter_commits(main_branch.name))))


rr.init("rerun_example_todo_count", spawn=True)

# Iterate through the list of commits
for i, commit in enumerate(all_commits):
    todo_count = 0
    for blob in commit.tree.traverse():
        if blob.type == 'blob':  # blobs are files
            todo_count += blob.data_stream.read().decode(errors='ignore').count("TODO(")

    #print(f"{commit.hexsha} {commit.author.name} {commit.authored_datetime} {todo_count}")

    rr.set_time_sequence("commit_idx", i)
    rr.set_time_seconds("commit_time", commit.authored_datetime.timestamp())

    rr.log("commit", rr.TextLog(f"{commit.hexsha} - {commit.author.name} - {commit.summary}"))
    rr.log("todos", rr.TimeSeriesScalar(scalar=todo_count))