Bryce Pasechnik <[EMAIL PROTECTED]> wrote: > Hey everyone, I posted a message regarding this awhile ago, but didn't > understand enough at the time to properly formulate my question. > > The problem is this, a number of the SMSC's we're connected to send us data in > different character sets. Sometime it is the latin iso standard character set, > other times its the GSM character set, and othertimes its in unicode. Whats > even stranger, is that these three character sets can come from the SAME phone > through the SAME SMSC depending on what characters I send our number. I.E. if > I just send plain text, it is often just the iso standard latin set. If I send > a semi special character like _ it seems to be the GSM character set. And if > I use an extended character like a [ or ] square bracket, the entire SMS will > arrive in unicode! This is all fine and dandy as I'm able to view the > data_coding variable via a CGI variable passed onto my webserver. However, > I'm wondering if there is a way to get KANNEL to do some of the conversion > work for me. I.e. can I get it to convert down to the latin character set for > me? Or must I do that myself? If I have to do it myself, it means I need to > have several converters capable of going GSM -> latin, unicode -> latin, and > possibly others.
Which SMSC module are you using ? It should always deliver latin1 for latin or GSM alphabet - it should convert GSM to latin1. If you want to convert unicode to latin1 or to utf8, add mo-recode=true to smsbox (see userguide to confirm this), and smsbox will try to recode your ucs2 message to latin1, then to utf8 if it fails.
