11 p.m.
Change log from v1:
- Simplify cbs_decode_text() for more clarity
- Fix M3 issue in previous patch
- Add unit test to check padding character is correctly handled
Guillaume Zajac (1):
unit: Add padding character decoding test
Philippe Nunes (1):
smsutil: <CR> is not necessarily a padding character in CBS message
src/smsutil.c | 142 ++++++++++++++++++++++++++++++++++++-------------------
unit/test-sms.c | 58 ++++++++++++++++++++++
2 files changed, 152 insertions(+), 48 deletions(-)
--
1.7.5.4
11 p.m.
New subject: [PATCH_v2 1/2] smsutil: <CR> is not necessarily a padding character in CBS message
From: Philippe Nunes <philippe.nunes(a)linux.intel.com>
---
src/smsutil.c | 142 +++++++++++++++++++++++++++++++++++++-------------------
1 files changed, 94 insertions(+), 48 deletions(-)
diff --git a/src/smsutil.c b/src/smsutil.c
index a541964..347c9d2 100644
--- a/src/smsutil.c
+++ b/src/smsutil.c
@@ -3996,6 +3996,96 @@ gboolean iso639_2_from_language(enum cbs_language lang, char
*iso639)
return FALSE;
}
+static unsigned char *decode_7bits_sms_charset(int taken, int *bufsize,
+ unsigned char *buf,
+ const guint8 *ud,
+ gboolean iso639)
+{
+ unsigned char unpacked[CBS_MAX_GSM_CHARS];
+ long written;
+ int max_chars;
+ int i, j;
+ int buf_size = *bufsize;
+
+ max_chars = sms_text_capacity_gsm(CBS_MAX_GSM_CHARS, taken);
+
+ unpack_7bit_own_buf(ud + taken, 82 - taken, taken, FALSE, max_chars,
+ &written, 0, unpacked);
+
+ i = iso639 ? 3 : 0;
+
+ /*
+ * CR can be used as a padding character, which means
+ * we can safely discard everything afterwards
+ */
+
+ for (; i < written; i++, buf_size++) {
+ if (unpacked[i] == '\r') {
+ /*
+ * check if this is a padding character
+ * or if it is a wanted <CR>
+ */
+ for (j = i + 1; j < written; j++)
+ if (unpacked[j] != '\r')
+ break;
+
+ if (j == written)
+ break;
+ }
+
+
+ buf[buf_size] = unpacked[i];
+ }
+
+ *bufsize = buf_size;
+
+ return buf;
+}
+
+static unsigned char *decode_default_sms_charset(int taken, int *bufsize,
+ unsigned char *buf,
+ const guint8 *ud,
+ gboolean iso639)
+{
+ int num_ucs2_chars = (82 - taken) >> 1;
+ int i = taken;
+ int max_offset = taken + num_ucs2_chars * 2;
+ int buf_size = *bufsize;
+
+ /*
+ * It is completely unclear how UCS2 chars are handled
+ * especially across pages or when the UDH is present.
+ * For now do the best we can.
+ */
+ if (iso639) {
+ i += 2;
+ num_ucs2_chars -= 1;
+ }
+
+ while (i < max_offset) {
+ if (ud[i] == 0x00 && ud[i+1] == '\r') {
+ int j = i + 2;
+
+ for (; j < max_offset; j = j + 2)
+ if (ud[j] != 0x00 || ud[j + 1] != '\r')
+ break;
+
+ if (j == max_offset)
+ break;
+ }
+
+ buf[buf_size] = ud[i];
+ buf[buf_size + 1] = ud[i+1];
+
+ buf_size += 2;
+ i += 2;
+ }
+
+ *bufsize = buf_size;
+
+ return buf;
+}
+
char *cbs_decode_text(GSList *cbs_list, char *iso639_lang)
{
GSList *l;
@@ -4087,30 +4177,8 @@ char *cbs_decode_text(GSList *cbs_list, char *iso639_lang)
taken = sms_udh_iter_get_udh_length(&iter) + 1;
if (charset == SMS_CHARSET_7BIT) {
- unsigned char unpacked[CBS_MAX_GSM_CHARS];
- long written;
- int max_chars;
- int i;
-
- max_chars =
- sms_text_capacity_gsm(CBS_MAX_GSM_CHARS, taken);
-
- unpack_7bit_own_buf(ud + taken, 82 - taken,
- taken, FALSE, max_chars,
- &written, 0, unpacked);
-
- i = iso639 ? 3 : 0;
-
- /*
- * CR is a padding character, which means we can
- * safely discard everything afterwards
- */
- for (; i < written; i++, bufsize++) {
- if (unpacked[i] == '\r')
- break;
-
- buf[bufsize] = unpacked[i];
- }
+ decode_7bits_sms_charset(taken, &bufsize, buf, ud,
+ iso639);
/*
* It isn't clear whether extension sequences
@@ -4120,30 +4188,8 @@ char *cbs_decode_text(GSList *cbs_list, char *iso639_lang)
* the check here since the specification isn't clear
*/
} else {
- int num_ucs2_chars = (82 - taken) >> 1;
- int i = taken;
- int max_offset = taken + num_ucs2_chars * 2;
-
- /*
- * It is completely unclear how UCS2 chars are handled
- * especially across pages or when the UDH is present.
- * For now do the best we can.
- */
- if (iso639) {
- i += 2;
- num_ucs2_chars -= 1;
- }
-
- while (i < max_offset) {
- if (ud[i] == 0x00 && ud[i+1] == '\r')
- break;
-
- buf[bufsize] = ud[i];
- buf[bufsize + 1] = ud[i+1];
-
- bufsize += 2;
- i += 2;
- }
+ decode_default_sms_charset(taken, &bufsize, buf, ud,
+ iso639);
}
}
--
1.7.5.4
11 p.m.
New subject: [PATCH_v2 2/2] unit: Add padding character decoding test
---
unit/test-sms.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 58 insertions(+), 0 deletions(-)
diff --git a/unit/test-sms.c b/unit/test-sms.c
index e510554..04a8874 100644
--- a/unit/test-sms.c
+++ b/unit/test-sms.c
@@ -1236,6 +1236,10 @@ static const char *cbs2 =
"0110003201114679785E96371A8D46A3D168341A8D46A3D1683"
"41A8D46A3D168341A8D46A3D168341A8D46A3D168341A8D46A3D168341A8D46A3D168"
"341A8D46A3D168341A8D46A3D168341A8D46A3D168341A8D46A3D100";
+static const char *cbs3 =
"001000000111E280604028180E888462C168381E90886442A95"
+ "82E988C66C3E9783EA09068442A994EA8946AC56AB95EB0986C46ABD96EB89C6EC7EBF"
+ "97EC0A070482C1A8FC8A472C96C3A9FD0A8744AAD5AAFD8AC76CB05";
+
static void test_cbs_encode_decode(void)
{
unsigned char *decoded_pdu;
@@ -1389,6 +1393,58 @@ static void test_cbs_assembly(void)
cbs_assembly_free(assembly);
}
+static void test_cbs_padding_character(void)
+{
+ unsigned char *decoded_pdu;
+ long pdu_len;
+ gboolean ret;
+ struct cbs cbs;
+ GSList *l;
+ char iso639_lang[3];
+ char *utf8;
+
+ decoded_pdu = decode_hex(cbs3, -1, &pdu_len, 0);
+
+ g_assert(decoded_pdu);
+ g_assert(pdu_len == (long)strlen(cbs1) / 2);
+ g_assert(pdu_len == 88);
+
+ ret = cbs_decode(decoded_pdu, pdu_len, &cbs);
+
+ g_free(decoded_pdu);
+
+ g_assert(ret);
+
+ g_assert(cbs.gs == CBS_GEO_SCOPE_CELL_IMMEDIATE);
+ g_assert(cbs.message_code == 1);
+ g_assert(cbs.update_number == 0);
+ g_assert(cbs.message_identifier == 0);
+ g_assert(cbs.dcs == 1);
+ g_assert(cbs.max_pages == 1);
+ g_assert(cbs.page == 1);
+
+ l = g_slist_append(NULL, &cbs);
+
+ utf8 = cbs_decode_text(l, iso639_lang);
+
+ g_assert(utf8);
+
+ if (g_test_verbose()) {
+ g_printf("%s\n", utf8);
+ if (iso639_lang[0] == '\0')
+ g_printf("Lang: Unspecified\n");
+ else
+ g_printf("Lang: %s\n", iso639_lang);
+ }
+
+ g_assert(strlen(utf8) == 122);
+ g_assert(strcmp(iso639_lang, "en") == 0);
+
+ g_free(utf8);
+
+ g_slist_free(l);
+}
+
static const char *ranges[] = { "1-5, 2, 3, 600, 569-900, 999",
"0-20, 33, 44, 50-60, 20-50, 1-5, 5, 3, 5",
NULL };
@@ -1679,6 +1735,8 @@ int main(int argc, char **argv)
test_cbs_encode_decode);
g_test_add_func("/testsms/Test CBS Assembly", test_cbs_assembly);
+ g_test_add_func("/testsms/Test CBS Padding Character",
+ test_cbs_padding_character);
g_test_add_func("/testsms/Range minimizer", test_range_minimizer);
g_test_add_func("/testsms/Status Report Assembly", test_sr_assembly);
--
1.7.5.4
3698
days inactive
3698
days old
2 comments
1 participants
participants (1)
-
Guillaume Zajac