Re: patch for empy message and UDH > 140

At 11:53 AM 5/6/03 +0200, Stipe Tolj wrote:
>Andreas Fink wrote:
>
> > I agree that SMSBox should be bullet proof, whatever you pass to it.
>
>agreed, all smsbox (or other boxes that connect to bearerbox and
>interact with user-scope) should be bullet proof.
>
>So please excessive testing mode on. ;)


this is a problem we run into as well.. its a tricky one, as we dont want 
Kannel to have to know about the various forms of UDH, esp. with EMS et al 
in mind.
Then there is the issue that some EMS messages are all UDH and no msgdata.
If one completely removes the checking of the length of the UDH against the 
first byte then there is the problem of bufferoveruns and assumptions about 
lengths of udh and msgdata in the smsc modules.

In order for Kannel to split all messages succesfully it will have to know 
about EMS at least.
We almost need an extra sendsms param to tell Kannel not to attempt to 
split a particular message.

As far as the patch is concerned we are running with something very 
similar, but it is far from perfect.

Essentially I reckon that Kannel should accept any udh/msgdata combination, 
along with a flag to disable concat/split on a per msg basis. Whether udh 
or msgdata are required and their relative lengths is really something the 
smsc modules should deal with.

Nisan


>Stipe
>
>[email protected]
>-------------------------------------------------------------------
>Wapme Systems AG
>
>Vogelsanger Weg 80
>40470 Düsseldorf
>
>Tel: +49-211-74845-0
>Fax: +49-211-74845-299
>
>E-Mail: [email protected]
>Internet: http://www.wapme-systems.de
>-------------------------------------------------------------------
>wapme.net - wherever you are