rds95/charset.py

import unicodedata
# This script prepares the RDS charset conversion function

# RDS code to UCS-2
charset = {}

charset[0x24] = 0x00A4 # ¤
charset[0x5E] = 0x2015 # ―
charset[0x60] = 0x2016 # ║
charset[0x7E] = 0x203E # ¯
charset[0x80] = 0x00E1 # á
charset[0x81] = 0x00E0 # à
charset[0x82] = 0x00E9 # é
charset[0x83] = 0x00E8 # è
charset[0x84] = 0x00ED # í
charset[0x85] = 0x00EC # ì
charset[0x86] = 0x00F3 # ó
charset[0x87] = 0x00F2 # ò
charset[0x88] = 0x00FA # ú
charset[0x89] = 0x00F9 # ù
charset[0x8A] = 0x00D1 # Ñ
charset[0x8B] = 0x00C7 # Ç
charset[0x8C] = 0x015E # Ş
charset[0x8D] = 0x00DF # ß
charset[0x8E] = 0x00A1 # ¡
charset[0x8F] = 0x0132 # IJ
charset[0x90] = 0x00E2 # â
charset[0x91] = 0x00E4 # ä
charset[0x92] = 0x00EA # ê
charset[0x93] = 0x00EB # ë
charset[0x94] = 0x00EE # î
charset[0x95] = 0x00EF # ï
charset[0x96] = 0x00F4 # ô
charset[0x97] = 0x00F6 # ö
charset[0x98] = 0x00FB # û
charset[0x99] = 0x00FC # ü
charset[0x9A] = 0x00F1 # ñ
charset[0x9B] = 0x00E7 # ç
charset[0x9C] = 0x015F # ş
charset[0x9D] = 0x011F # ğ
charset[0x9E] = 0x0131 # ı
charset[0x9F] = 0x0133 # ij
charset[0xA0] = 0x00AA # ª
charset[0xA1] = 0x03B1 # α
charset[0xA2] = 0x00A9 # ©
charset[0xA3] = 0x2030 # ‰
charset[0xA4] = 0x011E # Ğ
charset[0xA5] = 0x011B # ĕ
charset[0xA6] = 0x0148 # ň
charset[0xA7] = 0x0151 # ő
charset[0xA8] = 0x03C0 # π
charset[0xA9] = 0x20AC # €
charset[0xAA] = 0x00A3 # ₤
charset[0xAB] = 0x0024 # $
charset[0xAC] = 0x2190 # ←
charset[0xAD] = 0x2191 # ↑
charset[0xAE] = 0x2192 # →
charset[0xAF] = 0x2193 # ↓
charset[0xB0] = 0x00BA # º
charset[0xB1] = 0x00B9 # ¹
charset[0xB4] = 0x00B1 # ±
charset[0xB5] = 0x0130 # İ
charset[0xB6] = 0x0144 # ń
charset[0xB7] = 0x0171 # ű
charset[0xB8] = 0x00B5 # µ
charset[0xB9] = 0x00BF # ¿
charset[0xBA] = 0x00F7 # ÷
charset[0xBB] = 0x00B0 # º
charset[0xBF] = 0x00A7 # §
charset[0xC0] = 0x00C1 # Á
charset[0xC1] = 0x00C0 # À
charset[0xC2] = 0x00C9 # É
charset[0xC3] = 0x00C8 # È
charset[0xC4] = 0x00CD # Í
charset[0xC5] = 0x00CC # Ì
charset[0xC6] = 0x00D3 # Ó
charset[0xC7] = 0x00D2 # Ò
charset[0xC8] = 0x00DA # Ú
charset[0xC9] = 0x00D9 # Ù
charset[0xCA] = 0x0158 # Ř
charset[0xCB] = 0x010C # Č
charset[0xCC] = 0x0160 # Š
charset[0xCD] = 0x017D # Ž
charset[0xCE] = 0x00D0 # Ð
charset[0xCF] = 0x013F # Ŀ
charset[0xD0] = 0x00C2 # Â
charset[0xD1] = 0x00C4 # Ä
charset[0xD2] = 0x00CA # Ê
charset[0xD3] = 0x00CB # Ë
charset[0xD4] = 0x00CE # Î
charset[0xD5] = 0x00CF # Ï
charset[0xD6] = 0x00D4 # Ô
charset[0xD7] = 0x00D6 # Ö
charset[0xD8] = 0x00DB # Û
charset[0xD9] = 0x00DC # Ü
charset[0xDA] = 0x0159 # ř
charset[0xDB] = 0x010D # č
charset[0xDC] = 0x0161 # š
charset[0xDD] = 0x017E # ž
charset[0xDE] = 0x0111 # đ
charset[0xDF] = 0x0140 # ŀ
charset[0xE0] = 0x00C3 # Ã
charset[0xE1] = 0x00C5 # Å
charset[0xE2] = 0x00C6 # Æ
charset[0xE3] = 0x0152 # Œ
charset[0xE4] = 0x0177 # ŷ
charset[0xE5] = 0x00DD # Ý
charset[0xE6] = 0x00D5 # Õ
charset[0xE7] = 0x00D8 # Ø
charset[0xE8] = 0x00DE # Þ
charset[0xE9] = 0x014A # Ŋ
charset[0xEA] = 0x0154 # Ŕ
charset[0xEB] = 0x0106 # Ć
charset[0xEC] = 0x015A # Ś
charset[0xED] = 0x0179 # Ź
charset[0xEE] = 0x0166 # Ŧ
charset[0xEF] = 0x00F0 # ð
charset[0xF0] = 0x00E3 # ã
charset[0xF1] = 0x00E5 # å
charset[0xF2] = 0x00E6 # æ
charset[0xF3] = 0x0153 # œ
charset[0xF4] = 0x0175 # ŵ
charset[0xF5] = 0x00FD # ý
charset[0xF6] = 0x00F5 # õ
charset[0xF7] = 0x00F8 # ø
charset[0xF8] = 0x00FE # þ
charset[0xF9] = 0x014B # ŋ
charset[0xFA] = 0x0155 # ŕ
charset[0xFB] = 0x0107 # ć
charset[0xFC] = 0x015B # ś
charset[0xFD] = 0x017A # ź
charset[0xFE] = 0x0167 # ŧ

table = {}

import struct
for rds_code, ucs2 in charset.items():
    utf8 = struct.pack("!H", ucs2).decode("utf-16-be").encode()

    node = table
    for i, byte in enumerate(utf8):
        if i == len(utf8) - 1:
            # Last byte: store the RDS code
            node[byte] = rds_code
        else:
            # Intermediate byte: ensure a dict exists
            if byte not in node:
                node[byte] = {}
            node = node[byte]
def generate_switch(node, indent=0):
    """
    Recursively generates nested C switch statements from the table.
    `node` is the current dictionary level.
    `indent` is used for formatting the output.
    """
    ind = "    " * indent
    code_lines = [f"{ind}switch(*str) {{"]

    for byte, child in node.items():
        if isinstance(child, dict):
            code_lines.append(f"{ind}    case (char)0x{byte:02x}:")
            code_lines.append(f"{ind}        str++;")
            code_lines.extend(generate_switch(child, indent + 2))
            code_lines.append(f"{ind}        break;")
        else:
            comment = unicodedata.name(struct.pack("!H", charset[child]).decode("utf-16-be"))
            code_lines.append(f"{ind}    case (char)0x{byte:02x}: new_str[i] = (char)0x{child:02x}; break; // {comment}")
    code_lines.append(f"{ind}    default: new_str[i] = *str; break;")

    code_lines.append(f"{ind}}}")
    return code_lines

data = generate_switch(table)
with open("switch.c", "w", encoding="utf-8") as f:
    for line in data:
        f.write(line + "\n")