Hi
The thing is that Kannel says that its UTF-8 encoded message but the
same time message content is not converted and is still iso-8859-1.
Thats where the bug is.
Illimar
Hi,
this changed because Kannel has UTF-8 as default charset by now.
SMSC module handle this transparently for you and sendsms interface sends
UTF-8 message to you.
Thanks,
Alex
Am 20.07.2009 um 12:32 schrieb Illimar Reinbusch:
Hi
Im currently using "cvs-20090506" version (with custom modifications
for keyword handling) in my servers, earlier i was using cvs-20080808.
I started noticing that with new version some of my messages got
wrong charset.
I traced error back to the urltrans.c file, where i found the
following :
"cvs-20090506" version
...cut...
switch (request->sms.coding) {
case DC_UNDEF:
case DC_7BIT:
octstr_append(result, octstr_imm("UTF-8"));
break;
...cut...
"cvs-20080808" version
...cut...
switch (request->sms.coding) {
case DC_UNDEF:
case DC_7BIT:
octstr_append(result, octstr_imm("ISO-8859-1"));
break;
...cut...
Messages that came in thru SMSC connections were standard GSM charset
(SMPP data coding 0 or data coding 3) but if i use
charset=%C in sms-service and my php script detects the charset using
this parameter i got information that message is UTF-8 but it
actually was ISO-8859-1 because no internal recoding was done.
Any ideas why this was changed in new version?
Illimar
