Mauro,
Apparently i've missed that point earlier, sorry about that.
I can suggest to implement some kind a callback scheme.
Suppose upiu add a callabck pointer to sdp negotiator which will resid
in pahapi and will call phcodec routines to obtain the info.
Thanks
Vadim
P.S. Btw ilbc codec support is built as plugin and is not residing
anymore in phcodec.c
Le 15 avr. 09 à 17:00, Mauro Sergio Ferreira Brasil a écrit :
Hi Vadim!
I've pointed this dependency while I was implementing the patch, and
I was affraid it to be reject at that time.
Sorry, but your silence on further messages make me believe this
dependency was ok to you.
Anyway the reason to it is quite simple.
The "mode" parameter of "fmtp" attribute for iLBC codec must be
negotiated between the parts (caller and callee). And this is
especific of iLBC codec.
With this in mind I've concluded two things:
1- The way this parameter is negotiatied (as others parameters that
could exists or arise with time) must be a responsibility of the
codec itself. That's the reason I've changed the phcodec structure
to count with a new method responsible for this logic. Here we are
limited to "phapi" layer and no dependency was created so far;
2- IMO the result of this negotiation must be explicitly indicated
via the SDP sent to the other part. That was the point were the
dependency was created;
I agree with you that this is not a desirable dependency on the
project, but it was the only one way I found to get it done
considering the premises I've pointed above.
Please let me know if you agree with the premises and if you have a
sugestion of how can we get this done by other way.
Thanks and best regards,
Mauro.
Vadim Lebedev escreveu:
Mauro Sergio Ferreira Brasil wrote:
Hi Vadim!
Do you have any news about the patch (sent by email at 01/12/2009) ?
I've got the latest revision of "qutecom-2.2" repository today and
it still doesn't count with the last changes I've sent you.
Please let me know if and when they will be applied, ok ?
Thanks and best regards,
Mauro.
Mauro,
Sorry for the long silence.
I'v tried to integrate your patch into QuteCom some time ago and
encountered problems and got sick and later forgot to email to you
about it.
So i've retried the integration today with following results:
libosip2.diff applys OK (i've did some added otpimizations though)
eXosip.diff does not apply cleanly - for example the pacth for
exosip2.h seems to be inverted
once adjusted the compilation fails on Linux because of you're use
of TRUE and FALSE ids instead of 1 and 0.
I saw that in ow_osip /osip_negotiation.c inroduced dependency on
phcodec.h which is BIG NO NO with repsect of layering structure.
(eXosip should not depend on phapi).
I didn't investigate phapi.diff as i am stuck with exosip
Thanks
Vadim
Mauro Sérgio Ferreira Brasil escreveu:
Hi Vadim!
Do you have any news about the last patch I've sent you ?
If I'm not wrong, you said it will be applied on a branch with
other correlated changes...
In fact, I would like to know whether you have an idea of when
this patch will be made part of current Qutecom codes.
Or, maybe you can tell me if you have already tested them.
Please don't misunderstand me.
It's not my intention to hurry you up. I'm just curious about it.
Thanks and best regards,
Mauro.
--
At.,
<mime-attachment.jpeg>
Technology and Quality on Information
Mauro Sérgio Ferreira Brasil
Coordenador de Projetos e Analista de Sistemas
+ [email protected]
: www.tqi.com.br
( + 55 (34)3291-1700
( + 55 (34)9971-2572
--
At.,
<mime-attachment.jpeg>
Technology and Quality on Information
Mauro Sérgio Ferreira Brasil
Coordenador de Projetos e Analista de Sistemas
+ [email protected]
: www.tqi.com.br
( + 55 (34)3291-1700
( + 55 (34)9971-2572
--
At.,
<CMMI_2.jpg>
Technology and Quality on Information
Mauro Sérgio Ferreira Brasil
Coordenador de Projetos e Analista de Sistemas
+ [email protected]
: www.tqi.com.br
( + 55 (34)3291-1700
( + 55 (34)9971-2572
_______________________________________________
QuteCom-dev mailing list
[email protected]
http://lists.qutecom.org/mailman/listinfo/qutecom-dev
