v3.2.2 is history - why are we discussing about it???. To answer your
question, there were some problems in the message encoding, especially when
the message length was near the multi-part limit.

The latest v3 library uses a new encoding package (contributed by Jeff
Jongko) which is much more stable.

2008/11/13 nchin <[EMAIL PROTECTED]>

>
> Thank you very much for your kind assistance. The specs you
> recommended has been very useful for me.
>
> During my testings for multi-part messages, I have accidentally
> discover an "unusual behaviour" of SMS Server v3.2.2 I am using now.
> The issue in question as follow: -
>
> 1) Message: encoding 7bit, length exactly = 160 char. Perfect, no
> problem
> 2) Message: encoding 7bit, lenght exactly = 302 char. problem -
> message never got sent, db status field shows "Q"
> 3) Message: encoding 7bit, lenght exactly = 453 char. problem -
> message never got sent, db status field shows "Q"
> 4) Message: encoding 7bit, lenght exactly = 604 char. problem -
> message never got sent, db status field shows "Q"
>
> The problem appears to be happening in multiple of 151 chars, which is
> the max length for 7bit encoding as mention in the specs
> documentation, 3GPP 23.040.
>
> I am not sure if this issue is localised in SMS Server 3.2.2. Would
> you be able to shed some light whether the latest version 3.3.1. has
> the same issue?
>
> Regards
>
>
> On Nov 13, 3:44 am, Thanasis <[EMAIL PROTECTED]> wrote:
> > I think you should read the spec :)
> >
> > From the spec sheet (http://www.3gpp.org/ftp/Specs/archive/23_series/
> > 23.040/23040-601.zip)
> >
> > <<<
> >
> > 9.2.3.24.8      Concatenated short messages, 16-bit reference number
> >
> > This facility is an enhanced variant of the Concatenated Short Message
> > facility (see clause 9.2.3.24.1). The enhancement is a 16-bit
> > reference number, instead of the short 8-bit reference number. The
> > larger reference number reduces the probability that two different
> > concatenated messages are mistakenly sent with identical reference
> > numbers to a receiver. Except for the size of the reference number
> > this facility is identical to the Concatenated Short Message facility
> > (see clause 9.2.3.24.1).
> > In the case of uncompressed 8 bit data, the maximum length of the
> > short message within the TP UD field is 133 (140 - 7) octets.
> > In the case of uncompressed GSM 7 bit default alphabet data, the
> > maximum length of the short message within the TP UD field is 151 (160
> > - 9) characters.
> > In the case of 16 bit uncompressed USC2 data, the maximum length of
> > the short message within the TP UD field is 66 ((140 - 7)/2)
> > characters. A UCS2 character must not be split in the middle; if the
> > length of the User Data Header is odd, the maximum length of the whole
> > TP-UD field is 139 octets.
> >
> >
> >
> > The 3GPP 23.040 series is a very informative spec sheet if you want to
> > dig deeper.
> >
> > On Nov 12, 8:13 pm, nchin <[EMAIL PROTECTED]> wrote:
> >
> >
> >
> > > Sorry for the trouble, I am actually quite confused, hope you guys
> > > could help me:
> >
> > > So far this is what I understood
> > > 1) 7-bits encoding gives 160 chars per sms
> > > 2) 8-bits encoding gives 140 chars per sms
> > > 3) Unicode encoding gives 70 chars per sms
> >
> > > Can someone tell me, how to calculate the number of multi-parts sms
> > > for the three scenario above, please? Since I cannot simple divide
> > > into 160, 140 or 70 for all three scenario respectively because of the
> > > overhead.
> >
> > > Thanasis, you mentioned in your earlier reply that it is about 4 to 6
> > > char, so what should I use to calculate? Kindly accept my apologies if
> > > my question sounds dumb. I really need to be precise to get the exact
> > > number of multi-part.
> >
> > > TIA
> >
> > > On Nov 11, 7:28 pm, Thanasis <[EMAIL PROTECTED]> wrote:
> >
> > > > Unicode/UCS2 is two-bytes per char!
> > > > Unicode max message length is 70 chars - you should take this into
> account
> > > > in your calculations.
> >
> > > > 2008/11/11 Aryo Sanjaya <[EMAIL PROTECTED]>
> >
> > > > > On Mon, Nov 10, 2008 at 12:35 AM, nchin <[EMAIL PROTECTED]>
> wrote:
> >
> > > > >> Aryo, thank you for your assistance.
> >
> > > > >> Correct me if I am wrong, based on your formula, if the encoding
> is
> > > > >> Unicode/UCS2, then the formula should be:-
> > > > >>  if length(string) <= 140 the
> > > > >>    sms_count = 1
> > > > >>  else
> > > > >>    sms_count = floor( length(string) / (140-7) ) + 1
> > > > >>  endif
> >
> > > > >> Am I right? I will be grateful if you could help clarify.
> >
> > > > > Sorry for the late response.
> >
> > > > > I have no idea how concatenated SMS is implemented on Unicode.
> Sorry.
> >
> > > > > But I think the formula will be different, not just subtract it
> with 7
> > > > > chars.
> >
> > > > > As Thanasis said, PDU will get 4-6 bytes per part as an
> information. Since
> > > > > Unicode has different character wide (8 bits instead of 7), so I
> think it
> > > > > will not take 7 chars as on 7-bits SMS. Maybe fewer.
> >
> > > > > It's only my opinion, without any references :)
> >
> > > > > --
> > > > > Best Regards,
> >
> > > > > Aryo Sanjaya- Hide quoted text -
> >
> > > > - Show quoted text -- Hide quoted text -
> >
> > > - Show quoted text -- Hide quoted text -
> >
> > - Show quoted text -
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"SMSLib Users Group" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/SMSLib?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to