Without looking into the proposed changes: libzrtp does not use IPv4 or
IPv6 address elements. It only extends the AVPQueue and uses RTPPacket
to access RTP header fields.

The same is true for the SRTP functions: no direct use of IPV4 or
IPV6.

However, a double check is always good :-) .

Regards,
Werner

David Sugar wrote:
> Temporarily we could introduce an internal generic address object in
> ccrtp itself, and have wrappers representing the existing API that
> accept the existing InetAddress/IPV6Address objects, and then translate
> into the internal generic address type before calling the "real"
> function.  The IPV6 accepting methods that translate could be wrapped in
> CCXX_IPV6 defines.  This would also make the stack work for both ipv4
> and ipv6 without breaking any existing external code, but we should look
> at how that would effect libzrtp.
> 
> Federico Montesino Pouzols wrote:
>>
>>> I checked in my changes for ipv4/ipv6 outbound packet handling, which
>>> went very well, and thought I would wait for you to comment.
>>
>> I would definitely do whatever requires the less amount of work. I wonder
>> if introducing a common base class for v4 and v6 addresses in cc++, and
>> using that for addresses lists would make it easier to solve the
>> incoming queue issue.
>>
>>
>>
>>
>> _______________________________________________
>> Ccrtp-devel mailing list
>> [email protected]
>> http://lists.gnu.org/mailman/listinfo/ccrtp-devel
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Ccrtp-devel mailing list
> [email protected]
> http://lists.gnu.org/mailman/listinfo/ccrtp-devel



_______________________________________________
Ccrtp-devel mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/ccrtp-devel

Reply via email to