OK so everybody is for an at3. Would the best approach be to copy the AT2 methodology ?
I have to say that any changes to pdu encoding, etc. would have to be done in all three files, AT, AT2 and AT3. I have again attatched my AT retry loging patch, which is small and simple. I ask that this be my first patch of the CVS. Once given access to the CVS tree, I will then undertake smsc_at3.c. Is this OK with the current maintainer group ? I would just like to re-iterate. Please correct me if I'm wrong.... I believe AT2 encompasses the original AT methodology, however with vastly improved operation. Surely we can give up the original smsc_at.c file for the AT2 approach which lifted operation from 20% success to more then 80% success and with more debugging will head towards 100%. I believe AT3 should - as andreas said, be a combination of both SM memory and memoryless operation ... this would be ideal for speed. I would also like to point out that all mobiles have SM memory and we should not be afraid of taking advantage of it. It offers the key to 100% reliability. So again : Please accept my attatched patch which is small and simple. I will undertake to produce AT3 - that is smsc_at3.c will be produced by me in return for access to the CVS tree. I would however like to offer for consideration to altering the original smsc_at.c to operate thru memory and let smsc_at2.c encompas direct messaging. I don't believe there is a problem with throwing away smsc_at.c however you have my word that I will not take this approach. thanks Matt On 12 October 2001, Andreas Fink wrote: > one comment about Matt's approach to use SIM storage: > > While debugging AT2 over the last few days I discovered that Nokia > phones have a big weakness which make them only really work well if > using SIM storage. > > While testing I sent messages to myself. For some reason the message > went out but my code didnt detect it. So I've seen > > AT+CMGS=... > > <pdu> > > And the message was sent and my code was waiting for "OK". In this > moment the phone's display was saying "there is no space left to > store the incoming message". In other words, the message I've just > sent already came back to the phone but couldnt be delivered to the > AT interface as it was busy still with the sending request. With the > Siemens modems this was never the case as they simply gave an > inidcation of a waiting message. So its obviously that Nokiapones > will never work reliably with the way we send & receive messages. So > for those cases the use of SIM storage is useful. The nokia phones > also do not support packet mode. Packet mode is a different SMS > access method compared to AT commands whereas you switch to packet > mode and you receive packets with headers and trailers and checksums > which you have to aknowledge (as it is also the case with most > protocols like EMI/UCP or CIMD2). As using the AT commands often is > kludgy and modems interpret standards slightly different, i was > looking into using packet mode in AT2 (or maybe AT3...). My Nokia > 6210 however doesnt understand the command to switch to packet mode. > > So, yes we want SIM storage support. But we should make it > configurable. I agree with Stipe's comment of having temporarely > smsc_at3.c would be advisable. I would then come up with at4 for > packet mode :-)... > > But one day we will end up with at5 incorporating all of the features > under one hat. > > -- > > Andreas Fink > Fink-Consulting > > ------------------------------------------------------------------ > Tel: +41-61-6932730 Fax: +41-61-6932729 Mobile: +41-79-2457333 > Address: A. Fink, Schwarzwaldallee 16, 4058 Basel, Switzerland > E-Mail: [EMAIL PROTECTED] Homepage: http://www.finkconsulting.com > ------------------------------------------------------------------ > Something urgent? Try http://www.smsrelay.com/ Nickname afink -- Matt For electronic musicians ... Vector Bass : http://mffmvectorbass.sourceforge.net/ For developers ... 3D Audio Library : http://mffm3daudiolib.sourceforge.net/ Multimedia Time Code : http://mffmtimecode.sourceforge.net/
? patch.retry Index: gw/smsc_at.c =================================================================== RCS file: /home/cvs/gateway/gw/smsc_at.c,v retrieving revision 1.42 diff -u -r1.42 smsc_at.c --- gw/smsc_at.c 2001/09/19 15:59:37 1.42 +++ gw/smsc_at.c 2001/10/11 04:28:59 @@ -382,6 +382,12 @@ debug("AT", 0, "send command status: %d", ret); retries--; } + if (retries<0){ + debug("AT", 0, "AT: Retries exceeded \nSMS send failure :"); + debug("AT", 0, "Phone number: %s", +octstr_get_cstr(msg->sms.receiver)); + debug("AT", 0, "SMS data: %s", +octstr_get_cstr(msg->sms.msgdata)); + } + } } return ret;