#!/usr/bin/env python3
# =============================================================================
# Projet : AIR — Accounting Intermediate Representation
# Auteur : Simon-Pierre Boucher
# Contact : contact@spboucher.ai
# Fichier : check_headers.py
# Description : CI gate — fails if any project file is missing the author header.
# =============================================================================
"""Verify that every project file carries the mandatory author header.

Usage: python3 scripts/check_headers.py [root]
Exit code 0 if all files pass, 1 otherwise (with a list of offenders).
"""
from __future__ import annotations

import json
import sys
from pathlib import Path

AUTHOR = "Simon-Pierre Boucher"
CONTACT = "contact@spboucher.ai"

# Extensions checked for a comment header containing author + contact.
COMMENT_EXTS = {
    ".py", ".ts", ".js", ".rs", ".go", ".c", ".h", ".zig",
    ".md", ".yaml", ".yml", ".toml", ".sql", ".sh",
}
SKIP_DIRS = {".git", ".venv", "venv", "node_modules", "__pycache__",
             ".pytest_cache", ".mypy_cache", ".hypothesis", ".claude", "dist", "build"}
SKIP_FILES = {".gitignore", "LICENSE", "py.typed"}
HEADER_WINDOW = 15  # header must appear within the first N lines


def has_header(path: Path) -> bool:
    if path.suffix == ".json":
        try:
            data = json.loads(path.read_text(encoding="utf-8"))
        except (json.JSONDecodeError, UnicodeDecodeError):
            return False
        return isinstance(data, dict) and CONTACT in str(data.get("_author", ""))
    try:
        head = "".join(
            path.read_text(encoding="utf-8").splitlines(keepends=True)[:HEADER_WINDOW]
        )
    except UnicodeDecodeError:
        return True  # binary-ish file, not subject to header rule
    return AUTHOR in head and CONTACT in head


def main() -> int:
    root = Path(sys.argv[1]) if len(sys.argv) > 1 else Path(__file__).resolve().parent.parent
    offenders: list[Path] = []
    for path in sorted(root.rglob("*")):
        if not path.is_file():
            continue
        if any(part in SKIP_DIRS for part in path.parts):
            continue
        if path.name in SKIP_FILES:
            continue
        if path.suffix not in COMMENT_EXTS and path.suffix != ".json":
            continue
        if not has_header(path):
            offenders.append(path.relative_to(root))
    if offenders:
        print("Missing author header (Simon-Pierre Boucher / contact@spboucher.ai):")
        for p in offenders:
            print(f"  - {p}")
        return 1
    print("check_headers: OK — all files carry the author header.")
    return 0


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