import unicodedata

CHARACTERS = (
    # First try printable and common characters to have a readable filename.
    # For each character, the encoding list are just example of encodings able
    # to encode the character (the list is not exhaustive).

    # U+00E6 (Latin Small Letter Ae): cp1252, iso-8859-1
    '\u00E6',
    # U+0130 (Latin Capital Letter I With Dot Above): cp1254, iso8859_3
    '\u0130',
    # U+0141 (Latin Capital Letter L With Stroke): cp1250, cp1257
    '\u0141',
    # U+03C6 (Greek Small Letter Phi): cp1253
    '\u03C6',
    # U+041A (Cyrillic Capital Letter Ka): cp1251
    '\u041A',
    # U+05D0 (Hebrew Letter Alef): Encodable to cp424
    '\u05D0',
    # U+060C (Arabic Comma): cp864, cp1006, iso8859_6, mac_arabic
    '\u060C',
    # U+062A (Arabic Letter Teh): cp720
    '\u062A',
    # U+0E01 (Thai Character Ko Kai): cp874
    '\u0E01',

    # Then try more "special" characters. "special" because they may be
    # interpreted or displayed differently depending on the exact locale
    # encoding and the font.

    # U+00A0 (No-Break Space)
    '\u00A0',
    # U+20AC (Euro Sign)
    '\u20AC',
)

CODE_PAGES = (424, 720, 864, 874, 932, 950, 1006) + tuple(range(1250, 1257+1))
LOCALE_ENCODINGS = [
    'iso-8859-1', 'iso8859_3', 'iso8859_6',
    'mac_arabic', 'mac_farsi'] #, 'utf8']

def is_encodable(character, encoding, errors):
    try:
        encoded = character.encode(encoding, errors)
        decoded = encoded.decode(encoding, errors)
    except UnicodeError:
        return False
    else:
        return True

def check_encoding(encoding, errors):
    for character in CHARACTERS:
        if is_encodable(character, encoding, errors):
            return
    print("No character for encoding %s:%s :-(" % (encoding, errors))

for code_page in CODE_PAGES:
    check_encoding('cp%s' % code_page, 'strict')
for encoding in LOCALE_ENCODINGS:
    check_encoding(encoding, 'surrogateescape')

print("CHARACTERS = (")
for character in sorted(CHARACTERS):
    encodings = []
    for code_page in CODE_PAGES:
        encoding = 'cp%s' % code_page
        if is_encodable(character, encoding, 'strict'):
            encodings.append(encoding)
    for encoding in LOCALE_ENCODINGS:
        if is_encodable(character, encoding, 'surrogateescape'):
            encodings.append(encoding)
    name = unicodedata.name(character).title()
    print("    # U+%04X (%s): Encodable to %s" % (ord(character), name, ', '.join(encodings)))
    print("    '\\u%04X'," % ord(character))
print(")")