Re: [freesdp-devel] "c=..." parsing...

[Federico Montesino Pouzols]

        Hi, I agree. This is one of the things for which the current
freesdp interface is not complete, and thus a value that should be
decomposed is instead returned as a string. I recall other cases, such
as fsdp_get_media_formats, which returns a string instead of providing
access to an array of integers. Or the m= line port, which can be
specified as '<port>/<number of ports>'.

        We could keep sdp_get_global_conn_address for returning the IP
address string, and add the following functions:

        unsigned char sdp_get_global_conn_address_is_mcast, that says
whether the address is within the multicast range.

        and

        unsigned char sdp_get_global_conn_address_ttl
        unsigned char sdp_get_global_conn_address_count

        Similarly, the formatter should take these parameters as
separate values.

        Also, I would suggest you to follow draft-ietf-mmusic-sdp-new
rather than rfc2327. In fact, for the few things that the draft
differs from the original rfc, freesdp currently follows the draft.

On Sat, Jan 18, 2003 at 12:34:49AM +0000, Carlos Palminha wrote:
> Hi,
> 
> After reading the rfc2327 and using the library i think i found some
> differences...
> 
> In the "c=..." line, when parsing multicast addresses the function
> "sdp_get_global_conn_address" returns a string with ip/ttl... 
> Should this function return only the address and should we implement
> some functions for:
> 1) Checking for multicast addresses
> 2) Return the TTL from the address (in case of multicast)
> 3) Return the "number of addresses"
> 
> The RFC specifies:
> <base multicast address>/<ttl>/<number of addresses>
> 
> []
> C.Palminha
> -- 
> Carlos Palminha
> Engenharia de Desenvolvimento
> INOV - INESC Inovação
> http://www.inov.pt
> 
> 
> 
> 
> _______________________________________________
> freesdp-devel mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/freesdp-devel