Re: Fix EMS + concat message, cleanup gnokii.c

[Pawel Kot]

On Sat, 8 Jun 2002, Pavel Machek wrote:

> > Again. Look at the SMS layout:
> >
> > .... | A | .... | B | ... | C | D.... | E .... |
> > A is User Data Header Indicator. It's just one bit.
> > B is User Data Length. It contains the length of the User Data field.
> > There are following cases:
> >  - User Data is coded using Default Alphabet (7bit), then UDL is number of
> > the septets in the UserData plus User Data Header (including the
> > alignment)
> >  - User Data is coded 8bit, then UDH is number of the octets in the UD +
> > UDH
> >  - User Data is cuded using UCS2, then as 8bit.
> >
> > Now, if A is set, C denotes User Data Header length, and D is User Data
> > Header (it may contain many parts). If A is not set, C and D are not
> > present.
> >
> > D is User Data. I belive all picutres belong to User Data.
>   ~-- E?

Yep. Sorry.

> No, in EMS case, bitmap is encoded in "D", that's "User Data Header",
> not "User Data", AFAICS.

So in C is the large value covering all the message? Is User Data empty in
that case?

> > There's a common way (or few common ways) to encode the bitmap. There's
> > some boundary around some of them needed for some SMS schemas. What I
> > would like is to move image packing into the boundary into the gsm-sms.c,
> > not in the gsm-bitmaps.
>
> I could do that, but it would make code with pretty ugly dependencies.

No, it would fix the dependencies. Look, now you do:
SMS:
 - call bitmaps functions
BITMAPS:
 - call SMS functions
I want:
SMS
 - call bitmaps functions
BITMAPS:
 - DO NOT call SMS functions

Proposition. Please wait until Monday evening (hope to finish them
tomorrow, but MOday is safe) for my changes, and then will discuss this
issue again (hope no discussion will be needed :-)). Is it acceptable for
you?

pkot
-- 
mailto:address@hidden :: mailto:address@hidden
http://kt.linuxnews.pl/ :: Kernel Traffic po polsku