Markus R. wrote:
Hi,
I have a problem sending special characters in mobile terminated
messages through Kannel.
The Setup: Debian (Sarge), Kannel 1.4.1 (from .deb packages), libxml
2.6.16-7
SMSC: Vodafone Germany, EMI
The Problem: I want to send the following test message: ABCDäöüÄÖÜß
Each time I send this message I receive ABCDäößÄÖ§ñ on the mobile phone.
Here are my requests to kannel:
ISO-8859-1 encoded:
http://SERVER:PORT/cgi-bin/sendsms?user=USER&password=PASS&to=RECIPIENT&text=ABCD%e4%f6%fc%c4%d6%dc%df&charset=ISO-8859-1&from=SHORTNUMBER
smsbox.log says:
2007-04-03 11:23:49 [762] [3] INFO: sendsms
sender:<smsSender:SHORTNUMBER> (IPADDRESS) to:<RECIPIENT>
msg:<ABCDäöüÄÖÜß>
This at least tells me that the characters are transmitted to kannel
correctly but on the mobile I receive ABCDäößÄÖ§ñ
If I send the following UTF-8 encoded request:
http://SERVER:PORT/cgi-bin/sendsms?user=USER&password=PASS&to=RECIPIENT&text=ABCD%c3%a4%c3%b6%c3%bc%c3%84%c3%96%c3%9c%c3%9f&charset=UTF-8&from=SHORTNUMBER
smsbox.log says:
2007-04-03 11:29:12 [762] [3] INFO: sendsms
sender:<smsSender:SHORTNUMBER> (IPADDRESS) to:<RECIPIENT>
msg:<ABCDäöüÃ<84>Ã<96>Ã<9C>Ã<9F>>
The Mobile still receives ABCDäößÄÖ§ñ
It is interesting that at least ö, ä, Ö and Ä pass through correctly.
How can I find out, why ü, Ü and ß get garbled?
now, I recall all this for Vodafone D2 years ago. I added the alt-charset things
for Vodafone D2 here to the EMI/UCP module of Kannel.
See user's guide section for UCP smsc group:
http://www.kannel.org/download/kannel-userguide-snapshot/userguide.html#AEN1382
where you find:
alt-charset - number - Defines which character conversion kludge may be used for
this specific link. Currently implemented alternative charsets are defined in
"alt_charsets.h" and new ones can be added.
and in alt_charsets.h:
/*
* for CMG's EMI/UCP, operators may use NCR (national representation
* codes), like ISO 21 German for the german "Umlauts".
* i.e. Vodafone D2 uses this charset kludges.
*/
#define EMI_NRC_ISO_21 3
this means setting:
group = smsc
smsc = emi
...
alt-charset = 3
...
should do the kludge to use ISO 21 German representation for UCP/EMI connection
and will map the ISO latin1 values internally. ;)
Stipe
-------------------------------------------------------------------
Kölner Landstrasse 419
40589 Düsseldorf, NRW, Germany
tolj.org system architecture Kannel Software Foundation (KSF)
http://www.tolj.org/ http://www.kannel.org/
mailto:st_{at}_tolj.org mailto:stolj_{at}_kannel.org
-------------------------------------------------------------------
