Produce GitHub "scoreboard" highlighting users with many stale branches and pull requests

github_scoreboard.py

import os
from collections import defaultdict
from datetime import datetime
from datetime import timedelta

import pandas as pd
from github import Github
from tqdm import tqdm

# Replace 'YOUR_GITHUB_TOKEN' with your actual GitHub token
GITHUB_TOKEN = os.environ["GITHUB_TOKEN"]

# Replace 'recoord' with the actual organization name
ORG_NAME = "recoord"

# Replace 'sunstone' with the actual repository name
REPO_NAME = "sunstone"


def count_branches_and_pull_requests(repo):
    branches = repo.get_branches()
    pull_requests = repo.get_pulls(state="open")

    branch_counts = defaultdict(int)
    stale_counts = defaultdict(int)
    pr_counts = defaultdict(int)

    for branch in tqdm(branches, desc="Processing branches", total=branches.totalCount):
        branch_owner = branch.commit.author.login
        branch_date = branch.commit.commit.committer.date
        branch_counts[branch_owner] += 1
        if branch_date < datetime.now() - timedelta(days=90):
            stale_counts[branch_owner] += 1

    for pr in tqdm(pull_requests, desc="Processing pull requests", total=pull_requests.totalCount):
        pr_owner = pr.user.login
        pr_counts[pr_owner] += 1

    all_users = set().union(branch_counts).union(stale_counts).union(pr_counts)
    counts = {
        user: {
            "branches": branch_counts[user],
            "stale_branches": stale_counts[user],
            "pull_requests": pr_counts[user],
        }
        for user in all_users
    }
    return counts


def create_dataframe(branch_owners):
    data = []
    for owner, counts in branch_owners.items():
        branches_count = counts["branches"]
        pr_count = counts["pull_requests"]
        stale_branches_count = counts["stale_branches"]
        data.append([owner, stale_branches_count, branches_count, pr_count])

    df = pd.DataFrame(data, columns=["User", "Stale Branches", "Branches", "Pull Requests"])
    df.set_index("User", inplace=True)
    df.sort_values(by=["Stale Branches", "Branches", "Pull Requests"], ascending=False, inplace=True)

    # Add medals to top 3 users
    top_3_users = df.index[:3]
    medals = ["🥇", "🥈", "🥉"]
    df.index = df.index.where(
        ~df.index.isin(top_3_users), [f"{medals[i]} {user}" for i, user in enumerate(top_3_users)]
    )

    return df


def main():
    g = Github(GITHUB_TOKEN)
    org = g.get_organization(ORG_NAME)
    repo = org.get_repo(REPO_NAME)

    branch_owners = count_branches_and_pull_requests(repo)
    df = create_dataframe(branch_owners)

    print(df)


if __name__ == "__main__":
    main()

Had a little fun coding with ChatGPT 😂

                 Stale Branches  Branches  Pull Requests
User                                                    
🥇 rasmusroeg                 3        11              2
🥈 tobiasolo                  3         7              0
🥉 AdriaLlopar                3         6              3
pprie                         2         4              0
vladd                         1        13              2
elgehelge                     1         4              2
ilya-ve                       1         1              0
alexflorez                    0         6              0
hesh                          0         2              1