r4619 - trunk/src/target/OM-2007.2/applications/openmoko-dialer2/src/phone-kit
erin_yueh at docs.openmoko.org
erin_yueh at docs.openmoko.org
Tue Sep 2 10:47:22 CEST 2008
Author: erin_yueh
Date: 2008-09-02 10:47:21 +0200 (Tue, 02 Sep 2008)
New Revision: 4619
Modified:
trunk/src/target/OM-2007.2/applications/openmoko-dialer2/src/phone-kit/moko-sms.c
Log:
bug fixed for decoding GSM 7 bit, trac #1591 (Erin Yueh)
Modified: trunk/src/target/OM-2007.2/applications/openmoko-dialer2/src/phone-kit/moko-sms.c
===================================================================
--- trunk/src/target/OM-2007.2/applications/openmoko-dialer2/src/phone-kit/moko-sms.c 2008-09-02 08:26:48 UTC (rev 4618)
+++ trunk/src/target/OM-2007.2/applications/openmoko-dialer2/src/phone-kit/moko-sms.c 2008-09-02 08:47:21 UTC (rev 4619)
@@ -288,10 +288,140 @@
message = NULL;
switch (sms->payload.coding_scheme) {
case ALPHABET_DEFAULT :
- g_debug ("Decoding 7-bit ASCII message:");
- message = g_malloc0 (GSMD_SMS_DATA_MAXLEN);
- unpacking_7bit_character (&sms->payload, message);
- break;
+ {
+ gint i;
+ gint l;
+ gchar *gsmdefault;
+ gchar *dest;
+ g_debug ("Decoding GSM 7-bit default alphabet message:");
+ gsmdefault = g_malloc0 (GSMD_SMS_DATA_MAXLEN + 1);
+ l = unpacking_7bit_character (&sms->payload, gsmdefault);
+ message = g_malloc0 (1 + 3 * l);
+ dest = message;
+ for (i = 0; i < l; i++) {
+ /* Decoding based on the mapping at
+ * http://unicode.org/Public/MAPPINGS/ETSI/GSM0338.TXT
+ */
+ switch (gsmdefault[i]) {
+ case 0x00: // COMMERCIAL AT
+ *(dest++) = '@'; break;
+ case 0x01: // POUND SIGN
+ *(dest++) = 0xc2; *(dest++) = 0xa3; break;
+ case 0x02: // DOLLAR SIGN
+ *(dest++) = '$'; break;
+ case 0x03: // YEN SIGN
+ *(dest++) = 0xc2; *(dest++) = 0xa5; break;
+ case 0x04: // LATIN SMALL LETTER E WITH GRAVE
+ *(dest++) = 0xc3; *(dest++) = 0xa8; break;
+ case 0x05: // LATIN SMALL LETTER E WITH ACUTE
+ *(dest++) = 0xc3; *(dest++) = 0xa9; break;
+ case 0x06: // LATIN SMALL LETTER U WITH GRAVE
+ *(dest++) = 0xc3; *(dest++) = 0xb9; break;
+ case 0x07: // LATIN SMALL LETTER I WITH GRAVE
+ *(dest++) = 0xc3; *(dest++) = 0xac; break;
+ case 0x08: // LATIN SMALL LETTER O WITH GRAVE
+ *(dest++) = 0xc3; *(dest++) = 0xb2; break;
+ case 0x09: // LATIN SMALL LETTER C WITH CEDILLA
+ *(dest++) = 0xc3; *(dest++) = 0xa7; break;
+ case 0x0b: // LATIN CAPITAL LETTER O WITH STROKE
+ *(dest++) = 0xc3; *(dest++) = 0x98; break;
+ case 0x0c: // LATIN SMALL LETTER O WITH STROKE
+ *(dest++) = 0xc3; *(dest++) = 0xb8; break;
+ case 0x0e: // LATIN CAPITAL LETTER A WITH RING ABOVE
+ *(dest++) = 0xc3; *(dest++) = 0x85; break;
+ case 0x0f: // LATIN SMALL LETTER A WITH RING ABOVE
+ *(dest++) = 0xc3; *(dest++) = 0xa5; break;
+ case 0x10: // GREEK CAPITAL LETTER DELTA
+ *(dest++) = 0xce; *(dest++) = 0x94; break;
+ case 0x11: // LOW LINE
+ *(dest++) = '_'; break;
+ case 0x12: // GREEK CAPITAL LETTER PHI
+ *(dest++) = 0xce; *(dest++) = 0xa6; break;
+ case 0x13: // GREEK CAPITAL LETTER GAMMA
+ *(dest++) = 0xce; *(dest++) = 0x93; break;
+ case 0x14: // GREEK CAPITAL LETTER LAMDA
+ *(dest++) = 0xce; *(dest++) = 0x9b; break;
+ case 0x15: // GREEK CAPITAL LETTER OMEGA
+ *(dest++) = 0xce; *(dest++) = 0xa9; break;
+ case 0x16: // GREEK CAPITAL LETTER PI
+ *(dest++) = 0xce; *(dest++) = 0xa0; break;
+ case 0x17: // GREEK CAPITAL LETTER PSI
+ *(dest++) = 0xce; *(dest++) = 0xa8; break;
+ case 0x18: // GREEK CAPITAL LETTER SIGMA
+ *(dest++) = 0xce; *(dest++) = 0xa3; break;
+ case 0x19: // GREEK CAPITAL LETTER THETA
+ *(dest++) = 0xce; *(dest++) = 0x98; break;
+ case 0x1a: // GREEK CAPITAL LETTER XI
+ *(dest++) = 0xce; *(dest++) = 0x9e; break;
+ case 0x1b: // Escape character
+ switch (gsmdefault[++i]) {
+ case 0x0a: // FORM FEED
+ *(dest++) = 0x0c; break;
+ case 0x14: // CIRCUMFLEX ACCENT
+ *(dest++) = '^'; break;
+ case 0x28: // LEFT CURLY BRACKET
+ *(dest++) = '{'; break;
+ case 0x29: // RIGHT CURLY BRACKET
+ *(dest++) = '}'; break;
+ case 0x2f: // REVERSE SOLIDUS
+ *(dest++) = '\\'; break;
+ case 0x3c: // LEFT SQUARE BRACKET
+ *(dest++) = '['; break;
+ case 0x3d: // TILDE
+ *(dest++) = '~'; break;
+ case 0x3e: // RIGHT SQUARE BRACKET
+ *(dest++) = ']'; break;
+ case 0x40: // VERTICAL LINE
+ *(dest++) = '|'; break;
+ case 0x65: // EURO SIGN
+ *(dest++) = 0xe2; *(dest++) = 0x82;
+ *(dest++) = 0xac; break;
+ default: // NBSP (for compatibility)
+ *(dest++) = 0xc2; *(dest++) = 0xa0;
+ i--; // Do not consume next character
+ }
+ break;
+ case 0x1c: // LATIN CAPITAL LETTER AE
+ *(dest++) = 0xc3; *(dest++) = 0x86; break;
+ case 0x1d: // LATIN SMALL LETTER AE
+ *(dest++) = 0xc3; *(dest++) = 0xa6; break;
+ case 0x1e: // LATIN SMALL LETTER SHARP S
+ *(dest++) = 0xc3; *(dest++) = 0x9f; break;
+ case 0x1f: // LATIN CAPITAL LETTER E WITH ACUTE
+ *(dest++) = 0xc3; *(dest++) = 0x89; break;
+ case 0x24: // CURRENCY SIGN
+ *(dest++) = 0xc2; *(dest++) = 0xa4; break;
+ case 0x40: // INVERTED EXCLAMATION MARK
+ *(dest++) = 0xc2; *(dest++) = 0xa1; break;
+ case 0x5b: // LATIN CAPITAL LETTER A WITH DIAERESIS
+ *(dest++) = 0xc3; *(dest++) = 0x84; break;
+ case 0x5c: // LATIN CAPITAL LETTER O WITH DIAERESIS
+ *(dest++) = 0xc3; *(dest++) = 0x96; break;
+ case 0x5d: // LATIN CAPITAL LETTER N WITH TILDE
+ *(dest++) = 0xc3; *(dest++) = 0x91; break;
+ case 0x5e: // LATIN CAPITAL LETTER U WITH DIAERESIS
+ *(dest++) = 0xc3; *(dest++) = 0x9c; break;
+ case 0x5f: // SECTION SIGN
+ *(dest++) = 0xc2; *(dest++) = 0xa7; break;
+ case 0x60: // INVERTED QUESTION MARK
+ *(dest++) = 0xc2; *(dest++) = 0xbf; break;
+ case 0x7b: // LATIN SMALL LETTER A WITH DIAERESIS
+ *(dest++) = 0xc3; *(dest++) = 0xa4; break;
+ case 0x7c: // LATIN SMALL LETTER O WITH DIAERESIS
+ *(dest++) = 0xc3; *(dest++) = 0xb6; break;
+ case 0x7d: // LATIN SMALL LETTER N WITH TILDE
+ *(dest++) = 0xc3; *(dest++) = 0xb1; break;
+ case 0x7e: // LATIN SMALL LETTER U WITH DIAERESIS
+ *(dest++) = 0xc3; *(dest++) = 0xbc; break;
+ case 0x7f: // LATIN SMALL LETTER A WITH GRAVE
+ *(dest++) = 0xc3; *(dest++) = 0xa0; break;
+ default: // Untranslated
+ *(dest++) = gsmdefault[i];
+ }
+ }
+ g_free (gsmdefault);
+ break;
+ }
case ALPHABET_8BIT :
/* TODO: Verify: Is this encoding just UTF-8? (it is on my Samsung phone) */
g_debug ("Decoding UTF-8 message:");
@@ -744,10 +874,12 @@
MokoSmsPrivate *priv;
struct lgsm_sms sms;
gint msg_length, c;
- gboolean ascii;
+ glong msg16_length;
+ gboolean gsm7bit;
JanaNote *note;
gchar *dialcode = NULL;
gchar *sub_num = NULL;
+ gunichar2 *message16;
g_assert (self && number && message);
priv = self->priv;
@@ -782,31 +914,115 @@
} else {
strcpy (sms.addr, number);
}
-
/* Set message */
- /* Check if the text is ascii (and pack in 7 bits if so) */
- ascii = TRUE;
- for (c = 0; message[c] != '\0'; c++) {
- if (((guint8)message[c]) > 0x7F) {
- ascii = FALSE;
- break;
- }
- }
+ /* Try to encode to the 7-bit default alphabet, fall back to UTF-8 */
+ message16 = g_utf8_to_utf16 (message, -1, NULL, &msg16_length, NULL);
+ gsm7bit = TRUE;
/* TODO: Multi-part messages using UDH */
msg_length = strlen (message);
- if ((ascii && (msg_length > 160)) || (msg_length > 140)) {
+ gchar *smschars = g_malloc0 (162);
+ gint i = 0;
+ for (c = 0; c < msg16_length; c++) {
+ /* See http://unicode.org/Public/MAPPINGS/ETSI/GSM0338.TXT for details */
+ switch (message16[c]) {
+ case 0x000c: smschars[i++] = 0x1b; smschars[i++] = 0x0a; break;
+ case 0x0024: smschars[i++] = 0x02; break;
+ /* HACK: 0x80 instead of 0x00, avoids string termination */
+ case 0x0040: smschars[i++] = 0x80; break;
+ case 0x005b: smschars[i++] = 0x1b; smschars[i++] = 0x3c; break;
+ case 0x005c: smschars[i++] = 0x1b; smschars[i++] = 0x2f; break;
+ case 0x005d: smschars[i++] = 0x1b; smschars[i++] = 0x3e; break;
+ case 0x005e: smschars[i++] = 0x1b; smschars[i++] = 0x14; break;
+ case 0x005f: smschars[i++] = 0x11; break;
+ case 0x007b: smschars[i++] = 0x1b; smschars[i++] = 0x28; break;
+ case 0x007c: smschars[i++] = 0x1b; smschars[i++] = 0x40; break;
+ case 0x007d: smschars[i++] = 0x1b; smschars[i++] = 0x29; break;
+ case 0x007e: smschars[i++] = 0x1b; smschars[i++] = 0x3d; break;
+ case 0x00a1: smschars[i++] = 0x40; break;
+ case 0x00a3: smschars[i++] = 0x01; break;
+ case 0x00a4: smschars[i++] = 0x24; break;
+ case 0x00a5: smschars[i++] = 0x03; break;
+ case 0x00a7: smschars[i++] = 0x5f; break;
+ case 0x00bf: smschars[i++] = 0x60; break;
+ case 0x00c4: smschars[i++] = 0x5b; break;
+ case 0x00c5: smschars[i++] = 0x0e; break;
+ case 0x00c6: smschars[i++] = 0x1c; break;
+ case 0x00c9: smschars[i++] = 0x1f; break;
+ case 0x00d1: smschars[i++] = 0x5d; break;
+ case 0x00d6: smschars[i++] = 0x5c; break;
+ case 0x00d8: smschars[i++] = 0x0b; break;
+ case 0x00dc: smschars[i++] = 0x5e; break;
+ case 0x00df: smschars[i++] = 0x1e; break;
+ case 0x00e0: smschars[i++] = 0x7f; break;
+ case 0x00e4: smschars[i++] = 0x7b; break;
+ case 0x00e5: smschars[i++] = 0x0f; break;
+ case 0x00e6: smschars[i++] = 0x1d; break;
+ case 0x00e7: smschars[i++] = 0x09; break;
+ case 0x00e8: smschars[i++] = 0x04; break;
+ case 0x00e9: smschars[i++] = 0x05; break;
+ case 0x00ec: smschars[i++] = 0x07; break;
+ case 0x00f1: smschars[i++] = 0x7d; break;
+ case 0x00f2: smschars[i++] = 0x08; break;
+ case 0x00f6: smschars[i++] = 0x7c; break;
+ case 0x00f8: smschars[i++] = 0x0c; break;
+ case 0x00f9: smschars[i++] = 0x06; break;
+ case 0x00fc: smschars[i++] = 0x7e; break;
+ /* Greek characters have the same mapping as capital Latin characters where
+ * they both have the same form.
+ */
+ case 0x0391: smschars[i++] = 0x41; break;
+ case 0x0392: smschars[i++] = 0x42; break;
+ case 0x0393: smschars[i++] = 0x13; break;
+ case 0x0394: smschars[i++] = 0x10; break;
+ case 0x0395: smschars[i++] = 0x45; break;
+ case 0x0396: smschars[i++] = 0x5a; break;
+ case 0x0397: smschars[i++] = 0x48; break;
+ case 0x0398: smschars[i++] = 0x19; break;
+ case 0x0399: smschars[i++] = 0x49; break;
+ case 0x039a: smschars[i++] = 0x4b; break;
+ case 0x039b: smschars[i++] = 0x14; break;
+ case 0x039c: smschars[i++] = 0x4d; break;
+ case 0x039d: smschars[i++] = 0x4e; break;
+ case 0x039e: smschars[i++] = 0x1a; break;
+ case 0x039f: smschars[i++] = 0x4f; break;
+ case 0x03a0: smschars[i++] = 0x16; break;
+ case 0x03a1: smschars[i++] = 0x50; break;
+ case 0x03a3: smschars[i++] = 0x18; break;
+ case 0x03a4: smschars[i++] = 0x54; break;
+ case 0x03a5: smschars[i++] = 0x55; break;
+ case 0x03a6: smschars[i++] = 0x12; break;
+ case 0x03a7: smschars[i++] = 0x58; break;
+ case 0x03a8: smschars[i++] = 0x17; break;
+ case 0x03a9: smschars[i++] = 0x15; break;
+ case 0x20ac: smschars[i++] = 0x1b; smschars[i++] = 0x65; break;
+ default:
+ {
+ gunichar2 d = message16[c];
+ if (d == 0x000a || d == 0x000d ||
+ (d >= 0x0020 && d < 0x005b) || (d >= 0x0061 && d < 0x0080))
+ smschars[i++] = (gchar) d;
+ else
+ gsm7bit = FALSE;
+ }
+ }
+ if (i > 160 || !gsm7bit) break;
+ }
+ if ((i > 160 && gsm7bit) || (msg_length > 140 && !gsm7bit)) {
*error = g_error_new (PHONE_KIT_SMS_ERROR, PK_SMS_ERROR_MSG_TOOLONG,
"Message too long");
+ g_free (smschars);
return FALSE;
}
- if (ascii) {
- packing_7bit_character (message, &sms);
- } else {
+ if (gsm7bit) {
+ packing_7bit_character (smschars, &sms);
+ }
+ else {
sms.alpha = ALPHABET_8BIT;
strcpy ((gchar *)sms.data, message);
+ sms.length = msg_length;
}
- sms.length = msg_length;
+ g_free (smschars);
/* Send message */
lgsm_sms_send (handle, &sms);
More information about the commitlog
mailing list
