--- /dev/null
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+## This program is free software; you can redistribute it and/or modify
+## it under the terms of the GNU General Public License as published
+## by the Free Software Foundation; version 2 and higer.
+##
+## Martin Grimme (martin.grimme # gmail.com) 2010
+## Guseynov Alexey (kibergus # gmail.com) 2010
+
+LANG_DE = 0x0
+LANG_EN = 0x1
+LANG_IT = 0x2
+LANG_FR = 0x3
+LANG_ES = 0x4
+LANG_NL = 0x5
+LANG_SE = 0x6
+LANG_DA = 0x7
+LANG_PO = 0x8
+LANG_FI = 0x9
+LANG_NO = 0xa
+LANG_GR = 0xb
+LANG_TR = 0xc
+LANG_UNSPECIFIED = 0xf
+
+
+GSM_DEFAULT_ALPHABET = [
+ u"@",
+ u"\u00a3",
+ u"$",
+ u"\u00a5",
+ u"\u00e8",
+ u"\u00e9",
+ u"\u00f9",
+ u"\u00ec",
+ u"\u00f2",
+ u"\u00c7",
+ u"\n",
+ u"\u00d8",
+ u"\u00f8",
+ u"\r",
+ u"\u00c5",
+ u"\u00e5",
+
+ u"\u0394",
+ u"_",
+ u"\u03a6",
+ u"\u0393",
+ u"\u039b",
+ u"\u03a9",
+ u"\u03a0",
+ u"\u03a8",
+ u"\u03a3",
+ u"\u0398",
+ u"\u039e",
+ u" ",
+ u"\u00c6",
+ u"\u00e6",
+ u"\u00df",
+ u"\u00c9",
+
+ u" ",
+ u"!",
+ u"\"",
+ u"#",
+ u"\u00a4",
+ u"%",
+ u"&",
+ u"'",
+ u"(",
+ u")",
+ u"*",
+ u"+",
+ u",",
+ u"-",
+ u".",
+ u"/",
+
+ u"0",
+ u"1",
+ u"2",
+ u"3",
+ u"4",
+ u"5",
+ u"6",
+ u"7",
+ u"8",
+ u"9",
+ u":",
+ u";",
+ u"<",
+ u"=",
+ u">",
+ u"?",
+
+ u"\u00a1",
+ u"A",
+ u"B",
+ u"C",
+ u"D",
+ u"E",
+ u"F",
+ u"G",
+ u"H",
+ u"I",
+ u"J",
+ u"K",
+ u"L",
+ u"M",
+ u"N",
+ u"O",
+
+ u"P",
+ u"Q",
+ u"R",
+ u"S",
+ u"T",
+ u"U",
+ u"V",
+ u"W",
+ u"X",
+ u"Y",
+ u"Z",
+ u"\u00c4",
+ u"\u00d6",
+ u"\u00d1",
+ u"\u00dc",
+ u"\u00a7",
+
+ u"\u00bf",
+ u"a",
+ u"b",
+ u"c",
+ u"d",
+ u"e",
+ u"f",
+ u"g",
+ u"h",
+ u"i",
+ u"j",
+ u"k",
+ u"l",
+ u"m",
+ u"n",
+ u"o",
+
+ u"p",
+ u"q",
+ u"r",
+ u"s",
+ u"t",
+ u"u",
+ u"v",
+ u"w",
+ u"x",
+ u"y",
+ u"z",
+ u"\u00e4",
+ u"\u00f6",
+ u"\u00f1",
+ u"\u00fc",
+ u"\u00e0"
+]
+
+
+def decode(s, n):
+ """
+ Decodes the given string using the given cell broadcast data coding scheme.
+
+ @param s: string to decode
+ @param n: GSM cell broadcast data coding scheme
+ @return: UTF-8 string
+ """
+
+ # separate into nibbles
+ hbits = (n & 0xf0) >> 4
+ lbits = (n & 0x0f)
+
+ if (hbits == 0x0):
+ # language
+ return _decode_language(s, lbits)
+
+ elif (0x1 <= hbits <= 0x3):
+ # reserved language
+ return s
+
+ elif (0x4 <= hbits <= 0x7):
+ # general data coding indication
+ return _decode_general_data_coding(s, hbits, lbits)
+
+ elif (0x8 <= hbits <= 0xe):
+ # reserved coding group
+ return s
+
+ elif (hbits == 0xf):
+ # data coding / message handling
+ return s
+
+
+def _decode_language(s, lang):
+
+ return _decode_default_alphabet(s)
+
+
+def _decode_default_alphabet(s):
+
+ # ought to be all in the 7 bit GSM character map
+ # modem is in 8 bit mode, so it makes 7 bit unpacking itself
+ chars = [ GSM_DEFAULT_ALPHABET[ord(c)] for c in s ]
+ u_str = "".join(chars)
+ return u_str.encode("utf-8")
+
+
+def _decode_hex(s):
+
+ return s.decode("hex")
+
+
+def _decode_usc2(s):
+
+ return s.decode("hex").decode("utf-16-be").encode("utf-8")
+
+
+def _decode_general_data_coding(s, h, l):
+
+ is_compressed = (h & 0x2)
+
+ alphabet = (l & 0xc) >> 2
+
+ if (alphabet == 0x0):
+ # default alphabet
+ return _decode_default_alphabet(s)
+
+ elif (alphabet == 0x1):
+ # 8 bit
+ # actually, encoding is user-defined, but let's assume hex'd ASCII
+ # for now
+ return _decode_hex(s)
+
+ elif (alphabet == 0x2):
+ # USC2 (16 bit, BE)
+ return _decode_usc2(s)
+ elif (alphabet == 0x3):
+ # reserved
+ return s
+
+def decode_number(number):
+ dnumber = ""
+ for i in number:
+ if i & 0xf < 10:
+ dnumber += str(int((i & 0xf)))
+ if (i & 0xf0) >> 4 < 10:
+ dnumber += str(int((i & 0xf0) >> 4))
+ return dnumber
+
+def decode_timestamp(timestamp):
+ res = {}
+ res['year'] = str(timestamp[0] & 0xf) + str((timestamp[0] & 0xf0) >> 4)
+ res['month'] = str(timestamp[1] & 0xf) + str((timestamp[1] & 0xf0) >> 4)
+ res['day'] = str(timestamp[2] & 0xf) + str((timestamp[2] & 0xf0) >> 4)
+ res['hour'] = str(timestamp[3] & 0xf) + str((timestamp[3] & 0xf0) >> 4)
+ res['minute'] = str(timestamp[4] & 0xf) + str((timestamp[4] & 0xf0) >> 4)
+ res['second'] = str(timestamp[5] & 0xf) + str((timestamp[5] & 0xf0) >> 4)
+ res['timezone'] = str(timestamp[6] & 0xf) + str((timestamp[6] & 0xf0) >> 4)
+ return res
+
+def decode_pdu (pdumsg):
+ pdu = {}
+ pdu['type'] = int(pdumsg[0])
+ if pdu['type'] & 0x3 == 0x0:
+ pdu['address_len'] = int(pdumsg[1])
+ pdu['type_of_address'] = int(pdumsg[2])
+ base = 3+(pdu['address_len']+1)/2
+ pdu['sender'] = decode_number(pdumsg[3:base])
+ pdu['pid'] = int(pdumsg[base])
+ pdu['dcs'] = int(pdumsg[base+1])
+ pdu['timestamp'] = decode_timestamp (pdumsg[base+2:base+9]);
+ pdu['udl'] = int(pdumsg[base+9])
+ pdu['user_data'] = pdumsg[base+10:len(pdumsg)]
+
+ alphabet = (pdu['dcs'] & 0xc) >> 2
+ if alphabet == 0x0:
+ # This is 7-bit data. Taking of only pdu['udl'] bytes is important because we can't distinguish 0 at the end from @ if only one bit is used in last bite
+ pdu['user_data'] = _decode_default_alphabet(deoctify(pdu['user_data']))[0:pdu['udl']]
+ elif alphabet == 0x1:
+ # actually, encoding is user-defined, but let's assume ASCII
+ pdu['user_data'] = ''.join([chr(i) for i in pdu['user_data']])
+ elif (alphabet == 0x2):
+ # USC2 (16 bit, BE)
+ pdu['user_data'] = (''.join([chr(i) for i in pdu['user_data'][6:len(pdu['user_data'])]])).decode("utf-16-be").encode("utf-8")
+ elif (alphabet == 0x3):
+ # reserved
+ pdu['user_data'] = ''.join([chr(i) for i in pdu['user_data']])
+
+ if pdu['type'] & 0x4 == 0:
+ pdu['part'] = True
+ else:
+ pdu['part'] = False
+ else:
+ # TODO support other types of messages
+ # This is not incoming message
+ # pdu['type'] & 0x3 == 2 means delivery report
+ return None
+
+ return pdu
+
projects / ussd-widget / blobdiff