Ticket #6647: convert_domains.py

import sys
import re
import io


def idna(s):
    try:
        return s.encode('idna').decode('ascii')
    except UnicodeError:
        return s


def convert_domains(domains):
    result = []
    for domain in domains:
        if domain.startswith('~'):
            converted = '~' + idna(domain[1:])
        else:
            converted = idna(domain)
        if domain == converted or converted not in domains:
            result.append(converted)
    return result


def strip(s):
    return re.sub(r'\s', '', s)


for filename in sys.argv[1:]:
    with io.open(filename, 'r+', encoding='utf-8') as file:
        lines = file.readlines()
        i = 0
        while i < len(lines):
            line = lines[i]
            if not line.lstrip().startswith('!'):
                m = re.search(r'^(.*?)#[@?]?#', line)
                if m:
                    domains = ','.join(convert_domains(m.group(1).split(',')))
                    converted = domains + line[m.end(1):]
                else:
                    converted = re.sub(
                        r'^([\s@|]*(?:[a-zA-Z]*://)?)([^*^/?#$\r\n]+)',
                        lambda m: m.group(1) + idna(m.group(2)),
                        line
                    )
                    converted = re.sub(
                        r'(\$(?:\w*,)?domain=)([^,\r\n]+)',
                        lambda m: m.group(1) + '|'.join(convert_domains(m.group(2).split('|'))),
                        converted
                    )

                if converted != line and strip(converted) in map(strip, lines):
                    del lines[i]
                    continue

                lines[i] = converted
            i += 1

        file.seek(0)
        file.writelines(lines)
        file.truncate()