Script for analysing the age of PRs on a given GitHub repo

gh_analysis.py

from argparse import ArgumentParser, Namespace
from github import Github, PullRequest
from pandas import DataFrame


def process_pr(pull_request: PullRequest) -> dict:

    pr_row = {
        "title": pull_request.title,
        "number": pull_request.number,
        "created": pull_request.created_at,
        "updated": pull_request.updated_at,
    }

    comments = pull_request.get_comments(sort="update", direction="desc")

    pr_row["comments"] = comments.totalCount

    if comments.totalCount:
        pr_row["latest_comment"] = comments[0].updated_at

    return pr_row


def read_pull_requests(gh: Github, repository: str) -> DataFrame:

    repo = gh.get_repo(repository)
    pull_requests = repo.get_pulls()

    print(f"Found {pull_requests.totalCount} PRs in repository {repository}")

    if pull_requests.totalCount:
        output: list[dict] = []

        for i, pull_request in enumerate(pull_requests):
            percent_complete = ((i+1) / pull_requests.totalCount) * 100
            print(f"({percent_complete:.2f}%) Processing PR {pull_request.number}")
            output.append(process_pr(pull_request))

    return DataFrame(output)


def create_parser() -> ArgumentParser:

    parser = ArgumentParser("Script for analysing PRs in a GitHub Repository")

    parser.add_argument("repo", help="The path ({owner}/{repo}) to the repository to analyse.")

    parser.add_argument("-t", "--token", required=False, help="The GitHub access token to use")

    return parser


if __name__ == "__main__":

    ARGS: Namespace = create_parser().parse_args()

    if ARGS.token is not None:
        GH = Github(ARGS.token)
    else:
        print("No access token provided, this will use the public API rate limit (60 rph)")
        GH = Github()

    FLINK_PRS: DataFrame = read_pull_requests(GH, ARGS.repo)
    FLINK_PRS.to_csv(f"raw_pull_request_data_{ARGS.repo.replace("/", "-")}.csv")