Automatic unicode conversion

Automatic unicode conversion SearchSearch
Author Message
orbyone
Posted on Wednesday, April 02, 2003 - 11:19 am:   

Hi,

I noticed that when I use &Text=SomeEnglishText, the message is send using the GSM character set, but if I replace any one character of the message with e.g. a greek character, the message is automatically converted to Unicode and sent correctly to the mobile phone.
However, this may not be desirable in all applications (uhm, mine..). The unicode conversion results in a 80-characters/message SMS, so if I want to send a big message, NowSMS splits it into twice the number of messages. Since uppercase greek characters exist in the GSM character set, is there any way to disable the unicode conversion and send the message using that character set?

Thanks in advance
Bryce Norwood - NowSMS Support (Bryce)
Posted on Tuesday, April 08, 2003 - 12:02 am:   

Hi orbyone,

Specifically what character are you using that is triggering this behaviour?

We do checking against the GSM 7-bit character set but maybe there is a particular character that is confusing us?

I usually test this function by using the accented e characters (èé).

I guess I should try the few Greek characters that appear in that table as well ...

Ok. I've tried them, and you're right, the message gets sent out in Unicode format when it doesn't need to go out in that format.

I'll get this submitted as a bug report. I don't have your e-mail address, so if you want to follow-up with me directly by sending an e-mail to nowsms@now.co.uk, I can let you know if we have a fix that you could try prior to the next release.

-bn
Bryce Norwood - NowSMS Support (Bryce)
Posted on Tuesday, April 08, 2003 - 03:41 pm:   

Hi again orbyone,

I can confirm what you're seeing here.

There are 10 greek characters that are part of the standard 7-bit GSM character set.

If NowSMS sees any of these 10 characters in an outbound SMS message, it sets the message for Unicode encoding, when it could actually encode the character in the standard 7-bit GSM character set.

Similarly, on receiving a message that contains one of these 10 characters encoded in the 7-bit character set, NowSMS is not decoding the character correctly.

We have identified the problem, and are currently testing a fix, which will be included in the next release (after v4.11).

If you'd like to help us test this change, please contact me at nowsms@now.co.uk.

-bn