rds95/charset.py

import unicodedata
# This script prepares the RDS charset conversion function

# RDS code to UCS-2
# The data comes directly from the rdscharset.pdf, which comes from the R22_039_1 standard
charset = {
    0x0A: 0x0A,
    0x0D: 0x0D,
    0x1F: 0x1F,
    0x20: 0x20,
    0x21: 0x21,
    0x22: 0x22,
    0x23: 0x23,
    0x24: 0xA4,
    0x25: 0x25,
    0x26: 0x26,
    0x27: 0x27,
    0x28: 0x28,
    0x29: 0x29,
    0x2A: 0x2A,
    0x2B: 0x2B,
    0x2C: 0x2C,
    0x2D: 0x2D,
    0x2E: 0x2E,
    0x2F: 0x2F,
    0x30: 0x30,
    0x31: 0x31,
    0x32: 0x32,
    0x33: 0x33,
    0x34: 0x34,
    0x35: 0x35,
    0x36: 0x36,
    0x37: 0x37,
    0x38: 0x38,
    0x39: 0x39,
    0x3A: 0x3A,
    0x3B: 0x3B,
    0x3C: 0x3C,
    0x3D: 0x3D,
    0x3E: 0x3E,
    0x3F: 0x3F,
    0x40: 0x40,
    0x41: 0x41,
    0x42: 0x42,
    0x43: 0x43,
    0x44: 0x44,
    0x45: 0x45,
    0x46: 0x46,
    0x47: 0x47,
    0x48: 0x48,
    0x49: 0x49,
    0x4A: 0x4A,
    0x4B: 0x4B,
    0x4C: 0x4C,
    0x4D: 0x4D,
    0x4E: 0x4E,
    0x4F: 0x4F,
    0x50: 0x50,
    0x51: 0x51,
    0x52: 0x52,
    0x53: 0x53,
    0x54: 0x54,
    0x55: 0x55,
    0x56: 0x56,
    0x57: 0x57,
    0x58: 0x58,
    0x59: 0x59,
    0x5A: 0x5A,
    0x5B: 0x5B,
    0x5C: 0x5C,
    0x5D: 0x5D,
    0x5E: 0x2015,
    0x5F: 0x005F,
    0x60: 0x2016,
    0x61: 0x61,
    0x62: 0x62,
    0x63: 0x63,
    0x64: 0x64,
    0x65: 0x65,
    0x66: 0x66,
    0x67: 0x67,
    0x68: 0x68,
    0x69: 0x69,
    0x6A: 0x6A,
    0x6B: 0x6B,
    0x6C: 0x6C,
    0x6D: 0x6D,
    0x6E: 0x6E,
    0x6F: 0x6F,
    0x70: 0x70,
    0x71: 0x71,
    0x72: 0x72,
    0x73: 0x73,
    0x74: 0x74,
    0x75: 0x75,
    0x76: 0x76,
    0x77: 0x77,
    0x78: 0x78,
    0x79: 0x79,
    0x7A: 0x7A,
    0x7B: 0x7B,
    0x7C: 0x7C,
    0x7D: 0x7D,
    0x7E: 0x203E,
    0x80: 0xE1,
    0x81: 0xE0,
    0x82: 0xE9,
    0x83: 0xE8,
    0x84: 0xED,
    0x85: 0xEC,
    0x86: 0xF3,
    0x87: 0xF2,
    0x88: 0xFA,
    0x89: 0xF9,
    0x8A: 0xD1,
    0x8B: 0xC7,
    0x8C: 0x15E,
    0x8D: 0xDF,
    0x8E: 0xA1,
    0x8F: 0x132,
    0x90: 0xE2,
    0x91: 0xE4,
    0x92: 0xEA,
    0x93: 0xEB,
    0x94: 0xEE,
    0x95: 0xEF,
    0x96: 0xF4,
    0x97: 0xF6,
    0x98: 0xFB,
    0x99: 0xFC,
    0x9A: 0xF1,
    0x9B: 0xE7,
    0x9C: 0x15F,
    0x9D: 0x11F,
    0x9E: 0x131,
    0x9F: 0x133,
    0xA0: 0xAA,
    0xA1: 0x3B1,
    0xA2: 0xA9,
    0xA3: 0x2030,
    0xA4: 0x11E,
    0xA5: 0x11B,
    0xA6: 0x148,
    0xA7: 0x151,
    0xA8: 0x3C0,
    0xA9: 0x20AC,
    0xAA: 0xA3,
    0xAB: 0x24,
    0xAC: 0x2190,
    0xAD: 0x2191,
    0xAE: 0x2192,
    0xAF: 0x2193,
    0xB0: 0xBA,
    0xB1: 0xB9,
    0xB2: 0xB2,
    0xB3: 0xB3,
    0xB4: 0xB1,
    0xB5: 0x130,
    0xB6: 0x144,
    0xB7: 0x171,
    0xB8: 0xB5,
    0xB9: 0xBF,
    0xBA: 0xF7,
    0xBB: 0xB0,
    0xBC: 0xBC,
    0xBD: 0xBD,
    0xBE: 0xBE,
    0xBF: 0xA7,
    0xC0: 0xC1,
    0xC1: 0xC0,
    0xC2: 0xC9,
    0xC3: 0xC8,
    0xC4: 0xCD,
    0xC5: 0xCC,
    0xC6: 0xD3,
    0xC7: 0xD2,
    0xC8: 0xDA,
    0xC9: 0xD9,
    0xCA: 0x158,
    0xCB: 0x10C,
    0xCC: 0x160,
    0xCD: 0x17D,
    0xCE: 0xD0,
    0xCF: 0x13F,
    0xD0: 0xC2,
    0xD1: 0xC4,
    0xD2: 0xCA,
    0xD3: 0xCB,
    0xD4: 0xCE,
    0xD5: 0xCF,
    0xD6: 0xD4,
    0xD7: 0xD6,
    0xD8: 0xDB,
    0xD9: 0xDC,
    0xDA: 0x159,
    0xDB: 0x10D,
    0xDC: 0x161,
    0xDD: 0x17E,
    0xDE: 0x111,
    0xDF: 0x140,
    0xE0: 0xC3,
    0xE1: 0xC5,
    0xE2: 0xC6,
    0xE3: 0x152,
    0xE4: 0x177,
    0xE5: 0xDD,
    0xE6: 0xD5,
    0xE7: 0xD8,
    0xE8: 0xDE,
    0xE9: 0x14A,
    0xEA: 0x154,
    0xEB: 0x106,
    0xEC: 0x15A,
    0xED: 0x179,
    0xEE: 0x166,
    0xEF: 0xF0,
    0xF0: 0xE3,
    0xF1: 0xE5,
    0xF2: 0xE6,
    0xF3: 0x153,
    0xF4: 0x175,
    0xF5: 0xFD,
    0xF6: 0xF5,
    0xF7: 0xF8,
    0xF8: 0x0FE,
    0xF9: 0x14B,
    0xFA: 0x155,
    0xFB: 0x107,
    0xFC: 0x15B,
    0xFD: 0x17A,
    0xFE: 0x167}

table = {}

import struct
for rds_code, ucs2 in charset.items():
    utf8 = struct.pack("!H", ucs2).decode("utf-16-be").encode("utf-8")

    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, level=0):
    ind = "\t" * indent
    code_lines = [f"{ind}switch(*str) {{"]

    for byte, child in node.items():
        if isinstance(child, dict):
            code_lines.append(f"{ind}\tcase (char)0x{byte:02x}:")
            code_lines.append(f"{ind}\t\tstr++;")
            code_lines.extend(generate_switch(child, indent + 2, level + 1))
            code_lines.append(f"{ind}\t\tbreak;")
        else:
            comment = unicodedata.name(struct.pack("!H", charset[child]).decode("utf-16-be"), "N/A")
            code_lines.append(f"{ind}\tcase (char)0x{byte:02x}: output[i] = (char)0x{child:02x}; break; // {comment}")
    code_lines.append(f"{ind}\tdefault: output[i] = (char)0x3f; break; // {unicodedata.name("?")}") # Question mark

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

data = generate_switch(table, 2)

lines = []
lines_end = []
with open("src/lib.c", "r", encoding="utf-8") as f:
    while True:
        pos = f.tell()
        line = f.readline()
        if f.tell() == pos: break
        lines.append(line.rstrip())
        if "// CHARSET START" in line: break
    while True:
        pos = f.tell()
        line = f.readline()
        if f.tell() == pos: break
        if "// CHARSET END" in line:
            lines_end.append(line.rstrip())
            break
    while True:
        pos = f.tell()
        line = f.readline()
        if f.tell() == pos: break
        lines_end.append(line.rstrip())

with open("src/lib.c", "w", encoding="utf-8") as f:
    for line in lines + data + lines_end:
        f.write(line + "\n")