"""Commit message rules of this repository, in one place.

The local commit-msg hook calls this file, and autorun checks every job's
subject with it when the job list loads. Kept in the tree, the rules hold in
any clone and in CI: a copy in the hook alone left the runner checking only
the shape there.

Usage as a hook: python scripts/commit_rules.py MESSAGE_FILE
"""

from __future__ import annotations

import re
import sys
from pathlib import Path

MAX_SUBJECT = 50
SUBJECT_TYPES = re.compile(r"^(feat|refactor|chore|docs|fix): [a-z0-9]")
REFUSED_WORDS = re.compile(
    r"\b(bug|regression|revert|dead code|broken|wrong|hack|workaround|"
    r"comprehensive|implement)",
    re.IGNORECASE,
)
CO_AUTHOR = re.compile(r"^co-authored-by", re.IGNORECASE | re.MULTILINE)


def problems(message: str) -> list[str]:
    """What this repository refuses in a commit message, first line first."""
    found = []
    subject = message.split("\n", 1)[0]
    if len(subject) > MAX_SUBJECT:
        found.append(f"subject is {len(subject)} characters, {MAX_SUBJECT} at most: {subject}")
    if not SUBJECT_TYPES.match(subject):
        found.append("subject must read 'type: description' (feat, refactor, chore, docs, fix)")
    text = "\n".join(line for line in message.splitlines() if not line.startswith("#"))
    if REFUSED_WORDS.search(text):
        found.append("the message uses a word this repository's history does not carry")
    if CO_AUTHOR.search(message):
        found.append("no Co-Authored-By trailer here")
    return found


def main(argv: list[str]) -> int:
    if len(argv) != 2:
        print("usage: commit_rules.py MESSAGE_FILE", file=sys.stderr)
        return 2
    found = problems(Path(argv[1]).read_text(encoding="utf-8", errors="replace"))
    for problem in found:
        print(f"commit-msg: {problem}", file=sys.stderr)
    return 1 if found else 0


if __name__ == "__main__":
    sys.exit(main(sys.argv))