Re: BUG REPORT: failed with Chinese charaters (I've fixed it)

[Pawel Kot]

Hi,

On 10/8/06, waterser forrest <address@hidden> wrote:
>   I have a Nokia 3100, which needs the same driver with N6510. I downloaded
> gnokii-0.6.13 and found the the Chinese messages in my cell phone cannot be
> wholly fetched, instead, some of the messages are truncated to only a half.
> I read the code and found that there are 2 bytes seem related with the
> length of the msg block, one of which can be found as `block[3]' in Ln815,
> common/phones/nk6510.c.

Could you please show the sample dumps? Probably you are right but I
cannot achieve it with neither ascii sms nor with Polish alphabet.

>   When I read the raw data of the message, I found that the byte ahead of
> block[3], which can be read as block[2] in the C code, is the correst
> length, and this equation is true:
>
>   block[2] = 2 * block[3]
>
>   One weird thing is that sometimes block[2] == block[3], while sometimes
> block[2] = 2 * block[3]. I guess that block[2] is "How many SYMBOLS are
> there in the message" and block[3] is "How many BYTES". I cannot find the
> origin format document about this, so I'm just guessing.

And the example?

>   I changed block[3] into block[2] at Ln815, and I changed Ln714 of nk6510.c
> like this:
>
>     /*data->raw_sms->dcs = message[5];*/
>     if (message[1] == 2) {
>         data->raw_sms->dcs = message[6];
>     }
>     else {
>         data->raw_sms->dcs = message[5];
>     }

And what is the reason for this?

And please don't get me wrong. I don't say that your findings are
incorrect, I just want to fix the problem in the right way.

take care,
pkot
--
Pawel Kot