#!/usr/bin/env python3
"""extract_author_census.py

Extract-lane artifact: per-author commit census of the troupe repository.

Source (real tool call): `git -C /root/troupe-standup log --format=%an`
aggregated with a stdlib Counter. No LLM summarization, no fabricated rows.

Output: 2026-11-02-author-census.csv  (rank,author,commits)
"""
import csv
import subprocess
from collections import Counter

REPO = "/root/troupe-standup"
OUT = "/root/troupe-standup/wizard-work/ledger/2026-11-02-author-census.csv"
TOP_N = 20


def main():
    out = subprocess.run(
        ["git", "-C", REPO, "log", "--format=%an"],
        check=True, capture_output=True, text=True,
    ).stdout

    counts = Counter(out.splitlines())
    total = sum(counts.values())
    distinct = len(counts)

    rows = sorted(counts.items(), key=lambda kv: (-kv[1], kv[0]))[:TOP_N]

    with open(OUT, "w", newline="") as fh:
        w = csv.writer(fh)
        w.writerow(["rank", "author", "commits"])
        for i, (author, n) in enumerate(rows, start=1):
            w.writerow([i, author, n])

    print(f"total commits : {total}")
    print(f"distinct authors: {distinct}")
    print(f"rows written  : {len(rows)} -> {OUT}")
    print("top 5:")
    for author, n in rows[:5]:
        print(f"  {n:>5}  {author}")


if __name__ == "__main__":
    main()