#!/usr/bin/env python3
#
# Copyright (c) 2025-present The Bitcoin Core developers
# Distributed under the MIT software license, see the accompanying
# file COPYING or http://www.opensource.org/licenses/mit-license.php.

"""
Check that all source files have proper copyright headers.
"""

import re
import sys
import fnmatch
from subprocess import check_output

# Excludes for specific files
EXCLUDE = [
   # Autogenerated:
   'src/qt/bitcoinstrings.cpp',
   'src/chainparamsseeds.h',
   # Other external copyrights:
   'src/test/fuzz/FuzzedDataProvider.h',
   'src/tinyformat.h',
   'src/bench/nanobench.h',
   'contrib/devtools/clang-format-diff.py',
   # Python init files:
   '*__init__.py',
   # Devtools bitcoin-tidy example:
   'contrib/devtools/bitcoin-tidy/example_nontrivial-threadlocal.cpp',
   # contrib files:
   'contrib/devtools/check-deps.sh',
   'contrib/macdeploy/gen-sdk.py',
   'contrib/verify-binaries/test.py',
]

EXCLUDE_DIRS = [
   # git subtrees
   "src/crypto/ctaes/",
   "src/leveldb/",
   "src/minisketch/",
   "src/secp256k1/",
   "src/crc32c/",
]

# File extensions to include
INCLUDE = ['*.h', '*.cpp', '*.cc', '*.c', '*.mm', '*.py', '*.sh', '*.bash-completion']

# Compile patterns
EXCLUDE_COMPILED = re.compile('|'.join([fnmatch.translate(m) for m in EXCLUDE]))
INCLUDE_COMPILED = re.compile('|'.join([fnmatch.translate(m) for m in INCLUDE]))

# Generic copyright pattern (accepts any copyright holder)
# Matches "Copyright" with optional "(c)", and optional year/year-range
COPYRIGHT_COMPILED = re.compile(r'Copyright', re.IGNORECASE)


def applies_to_file(filename: str) -> bool:
   for excluded_dir in EXCLUDE_DIRS:
       if filename.startswith(excluded_dir):
           return False
   return (EXCLUDE_COMPILED.match(filename) is None and
           INCLUDE_COMPILED.match(filename) is not None)


def get_filenames_to_examine() -> list[str]:
   git_ls_cmd = ['git', 'ls-files', '--full-name']
   all_files = check_output(git_ls_cmd, text=True).splitlines()
   return sorted([f for f in all_files if applies_to_file(f)])


def file_has_copyright(filename: str) -> bool:
   """
   Check if file has a copyright header in the first 10 lines.
   """
   try:
       with open(filename, 'r', encoding='utf-8') as f:
           # Read first 10 lines
           first_lines = ''.join(f.readline() for _ in range(10))
   except (UnicodeDecodeError, IOError):
       # Skip binary files or files that can't be read
       return True
   
   return COPYRIGHT_COMPILED.search(first_lines) is not None


def main():
   exit_code = 0
   files_to_examine = get_filenames_to_examine()
   
   missing_copyright = []
   for filename in files_to_examine:
       if not file_has_copyright(filename):
           missing_copyright.append(filename)
   
   if missing_copyright:
       print(f"The following {len(missing_copyright)} file(s) have no copyright notice:")
       for filename in missing_copyright:
           print(f"  {filename}")
       exit_code = 1
   
   sys.exit(exit_code)


if __name__ == '__main__':
   main()