.git-hook-commit-msg

#!/usr/bin/env python3
"""Git commit-msg hook.

Reads the activity (scielo-tools-NNNN) from the branch name -- anywhere in the
name -- and:

  1. validates the message format: <type>(<optional scope>): <message>
     allowed types: feature, fix, chore, unittest
  2. injects the activity prefix into the first line: [scielo-tools-NNNN] ...

Automatic Git commits (merge/revert/squash) are skipped.
Uses the standard library only for portability across Linux and macOS.
"""

import re
import subprocess
import sys

ALLOWED_TYPES = ("feature", "fix", "chore", "unittest")

BRANCH_ACTIVITY_RE = re.compile(r"scielo-tools-(\d+)", re.IGNORECASE)
EXISTING_PREFIX_RE = re.compile(r"^\s*\[scielo-tools-\d+\]\s*", re.IGNORECASE)
MESSAGE_FORMAT_RE = re.compile(
    r"^(?:%s)(?:\([a-z0-9-]+\))?: .+" % "|".join(ALLOWED_TYPES)
)

AUTO_COMMIT_PREFIXES = (
    "Merge branch ",
    "Merge pull request ",
    "Merge remote-tracking ",
    'Revert "',
)
SQUASH_MERGE_RE = re.compile(r"^Merge .+ into ")


def fail(message):
    print(message, file=sys.stderr)
    sys.exit(1)


def is_auto_commit(first_line):
    if first_line.startswith(AUTO_COMMIT_PREFIXES):
        return True
    return bool(SQUASH_MERGE_RE.match(first_line))


def current_branch():
    for cmd in (
        ["git", "symbolic-ref", "--short", "HEAD"],
        ["git", "rev-parse", "--abbrev-ref", "HEAD"],
    ):
        try:
            result = subprocess.run(cmd, capture_output=True, text=True)
        except OSError:
            return None
        if result.returncode == 0:
            branch = result.stdout.strip()
            if branch:
                return branch
    return None


def main():
    if len(sys.argv) < 2:
        fail("commit-msg hook: message file path not provided.")

    msg_path = sys.argv[1]
    with open(msg_path, "r", encoding="utf-8") as handle:
        lines = handle.read().splitlines()

    if not lines:
        fail("Empty commit message.")

    first_line = lines[0].rstrip("\r")

    if is_auto_commit(first_line):
        sys.exit(0)

    branch = current_branch()
    if not branch or branch == "HEAD":
        fail(
            "Could not determine the current branch (detached HEAD?).\n"
            "The branch must contain the activity scielo-tools-NNNN."
        )

    match = BRANCH_ACTIVITY_RE.search(branch)
    if not match:
        fail(
            "Branch without an activity.\n"
            "The branch name must contain scielo-tools-NNNN "
            "(e.g. feature/scielo-tools-1234/my-feature).\n"
            "Current branch: %s" % branch
        )

    activity = "scielo-tools-%s" % match.group(1)

    body = EXISTING_PREFIX_RE.sub("", first_line).strip()

    if not MESSAGE_FORMAT_RE.match(body):
        fail(
            "Invalid commit message.\n"
            "Expected format: <type>(<optional scope>): <message>\n"
            "Allowed types: %s\n"
            "Examples:\n"
            "  chore: format readme\n"
            "  fix(auth): fix authentication when token is invalid"
            % ", ".join(ALLOWED_TYPES)
        )

    lines[0] = "[%s] %s" % (activity, body)

    with open(msg_path, "w", encoding="utf-8") as handle:
        handle.write("\n".join(lines) + "\n")

    sys.exit(0)


if __name__ == "__main__":
    main()