Re: Encoding in atgen.c

[Hugo Haas]

* "Burbure, Pavan \(MED\)" <address@hidden> [2004-04-22 05:49:02+0200]:
>> Interesting. After selecting UCS2, you got an error when Gnokii did
>> AT+CPBS="ME". I wonder if this had to be UCS2 encoded.
>> 
>> We have seen:
>> - T610: everything between double-quotes is UCS2 encoded.
>> - Pawel's Nokia: only the name field in a phonebook entry is UCS2
>> encoded.
>> 
>> I wonder if your Philips has the name field as well as the memory type
>> encoded in UCS2, and still has the phone number in ASCII, like the
>> Nokia.
>
> Oh, btw, here is the log with your patch. Note that the gnokii
> output is junk!

So actually, selecting the ME memory type encode "ME" in UCS2 didn't
work better.

> 1. Name: CustomerCare
> Number: à¢^
> Group id: 0

Yes, as on the Nokia, the phone number is received in ASCII.

> Any idea how I can read the phone memory? I plan to look into
> the communication of the windows app that I got with the phone.

That I don't know as it doesn't seem to be an encoding problem like it
was on my T610.

Regarding the encoding issue I am seeing, it seems that only the T610
encodes everything in UCS2. Actually, the T68 test worked fine with
and without the patch. I wonder if we can apply the patch for all
Ericssons.

Pawel, what are your thoughts? I am not sure that a simple extra
encoding operation warrants redifining completely the functions that
need it in ateric.c.

We could maybe check whether it is an Ericsson before doing the extra
character encoding/decoding.

Regards,

Hugo

-- 
Hugo Haas - http://larve.net/people/hugo/