#!/usr/bin/env python3
import argparse
import re
import subprocess
import sys
from pathlib import Path


SHORTCODE_PREFIX = re.compile(r"^:[a-z0-9_+-]+:\s+")
EMOJI_PREFIX = re.compile(r"^[\u2600-\u27BF\U0001F300-\U0001FAFF]\s+")


def strip_prefix(message: str) -> str:
    if SHORTCODE_PREFIX.match(message):
        return SHORTCODE_PREFIX.sub("", message, count=1)
    if EMOJI_PREFIX.match(message):
        return EMOJI_PREFIX.sub("", message, count=1)
    return message


def validate_message(message: str) -> list[str]:
    lines = [line.rstrip("\n") for line in message.splitlines()]
    cleaned_lines = [line for line in lines if line and not line.startswith("#")]
    if not cleaned_lines:
        return ["Commit message must not be empty."]

    subject = cleaned_lines[0]
    errors: list[str] = []

    if not SHORTCODE_PREFIX.match(subject) and not EMOJI_PREFIX.match(subject):
        errors.append("Commit subject must start with a gitmoji shortcode or emoji.")

    body = "\n".join(cleaned_lines)
    normalized = strip_prefix(subject) + ("\n" + "\n".join(cleaned_lines[1:]) if len(cleaned_lines) > 1 else "")

    try:
        normalized.encode("ascii")
    except UnicodeEncodeError:
        errors.append("Commit message must be written in English ASCII text after the gitmoji prefix.")

    if not strip_prefix(subject).strip():
        errors.append("Commit subject must include an English summary after the gitmoji prefix.")

    if re.search(r"[\u4e00-\u9fff]", body):
        errors.append("Commit message must not contain Chinese characters.")

    return errors


def read_message_file(path: Path) -> str:
    return path.read_text(encoding="utf-8")


def run_git(*args: str) -> list[str]:
    result = subprocess.run(
        ["git", *args],
        capture_output=True,
        text=True,
        check=False,
    )
    if result.returncode != 0:
        raise RuntimeError(result.stderr.strip() or "git command failed")
    return [line.strip() for line in result.stdout.splitlines() if line.strip()]


def commit_messages_from_range(rev_range: str) -> list[tuple[str, str]]:
    if ".." in rev_range:
        shas = run_git("rev-list", rev_range)
    else:
        shas = [rev_range]

    messages: list[tuple[str, str]] = []
    for sha in shas:
        message = subprocess.run(
            ["git", "log", "--format=%B", "-n", "1", sha],
            capture_output=True,
            text=True,
            check=False,
        )
        if message.returncode != 0:
            raise RuntimeError(message.stderr.strip() or "git log failed")
        messages.append((sha, message.stdout.strip()))
    return messages


def parse_args() -> argparse.Namespace:
    parser = argparse.ArgumentParser(description="Validate commit message format.")
    parser.add_argument("--file", help="path to commit message file")
    parser.add_argument("--rev-range", help="git revision range or single commit")
    return parser.parse_args()


def main() -> int:
    args = parse_args()

    if bool(args.file) == bool(args.rev_range):
        print("Use exactly one of --file or --rev-range.")
        return 2

    failures: list[str] = []

    if args.file:
        message = read_message_file(Path(args.file))
        errors = validate_message(message)
        if errors:
            failures.extend(errors)
    else:
        for sha, message in commit_messages_from_range(args.rev_range):
            errors = validate_message(message)
            for error in errors:
                failures.append(f"{sha[:12]}: {error}")

    if failures:
        print("Commit message validation failed:")
        for failure in failures:
            print(f"  - {failure}")
        print("\nExpected format example:")
        print("  :sparkles: add hook templates and CI guardrails")
        return 1

    print("Commit message validation passed.")
    return 0


if __name__ == "__main__":
    raise SystemExit(main())