from pathlib import Path
import re
import sys

from breadtube_bot.unidecode import DISCORD_PATTERN


UNIDECODE_DICT: dict[int, list[str]] = {}


# Reads perl files from https://metacpan.org/pod/Text::Unidecode project
def generate_dict(data_path: Path):
    quote_pattern = re.compile(r'[\'"]([^\'"]*)["\']')
    perl_quote = re.compile(r'q+\{([^\}]*)\}')
    begin_comment_pattern = re.compile(r'^ *#')
    end_comment_pattern = re.compile(r', *#(.*)$')

    def perl_quote_repl(match_object):
        new_char = match_object.group(1).replace('"', '_').replace("'", '_')
        return f"'{new_char}'"

    for file_path in sorted(data_path.glob('*.pm')):
        content = file_path.read_text().strip()
        if 'make_placeholder_map' in content:
            continue
        section = int(f'0{file_path.stem}', 16)
        content = ''.join([end_comment_pattern.sub(',', line).replace('~', '-') for line in content.splitlines()
                           if line and begin_comment_pattern.match(line) is None])
        content = perl_quote.sub(perl_quote_repl, content)
        replace_chars: list[str] = []
        for index, char in enumerate(quote_pattern.findall(content)):
            if char == '[?]' or (section == 0 and index <= 0x80):  # noqa: PLR2004
                replace_chars.append('_')
            else:
                replace_chars.append(DISCORD_PATTERN.sub('_', char).lower())
        assert len(replace_chars) == 256, f'Wrong size for {file_path.name}: {len(replace_chars)}'  # noqa: PLR2004
        UNIDECODE_DICT[section] = replace_chars


if __name__ == '__main__':
    data_path = Path('data/unidecode')
    if not data_path.exists():
        print(f'No data found at path: {data_path}')
        sys.exit(1)
    if not data_path.is_dir():
        print(f'Path "{data_path}" is not a folder')
        sys.exit(1)

    generate_dict(data_path)
    with Path('breadtube_bot/unidecode_data.py').open(mode='w', encoding='utf-8') as unidecode_file:
        unidecode_file.write('UNIDECODE_DICT: dict[int, list[str]] = {\n')
        for key, value in UNIDECODE_DICT.items():  # noqa: FURB122
            unidecode_file.write(f'    {key}: {value},\n')
        unidecode_file.write('}\n')