I agree with Jo. We want one, and only one, way to define equality, and
thats with URL equality rules.
-Jonathan R.
---
Jonathan D. Rosenberg 72 Eagle Rock Ave.
Chief Scientist First Floor
dynamicsoft East Hanover, NJ 07936
[EMAIL PROTECTED] FAX: (973) 952-5050
http://www.cs.columbia.edu/~jdrosen PHONE: (973) 952-5000
http://www.dynamicsoft.com
> -----Original Message-----
> From: Shail Bhatnagar [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, February 08, 2001 2:13 PM
> To: [EMAIL PROTECTED]
> Subject: [Sip-implementors] [Fwd: Re: [SIP] Second try - Branch
> parameter question]
>
>
> Just want to hear comments about this.
> (original question and Jo's reply are quoted)
>
> Shail
>
>
> -------- Original Message --------
> Subject: Re: [SIP] Second try - Branch parameter question
> Date: Thu, 08 Feb 2001 12:02:56 -0500
> From: Shail Bhatnagar <[EMAIL PROTECTED]>
> To: Jo Hornsby <[EMAIL PROTECTED]>
> CC: [EMAIL PROTECTED]
> References: <001b01c091ec$03d2f440$[EMAIL PROTECTED]>
>
> Jo, I tend to think that doing URL match for addr-spec of
> From/To headers and concatenating normalized
> URL components even for hash computation,
> is an unnecessary overhead. Practically speaking,
> every proxy/ua will save the unparsed form
> of From/To as it came in the first time (or
> as it sent out the first time) and could do a
> simple string match on the addr-spec. (I thought
> URLs cannot have white spaces ??)
> I don't find the URL matching rules very useful
> for the addr-spec of From and To headers, but
> would be forced to use them.
>
> Shail
>
>
>
>
> Jo Hornsby wrote:
> >
> > Shail Bhatnagar wrote:
> > > (a) When a proxy computes the cryptographic hash of
> > > From, To, Call-ID, CSeq.seqnum and Request-URI,
> > > can it use the unparsed From and To or normalized
> > > form of their parsed components.
> > > Does the loop detection scheme in the bis-02
> > > version assumes that From header of a looped
> > > request or a spiralled request is byte-by-byte
> > > same as the From header of the original incoming
> > > request ? (Same question for To header)
> >
> > I would say that, in general, byte-by-byte is far too
> > fragile. Some sort of normalised form sounds like a
> > good way to go.
> >
> > > (b) When using the To header, should the proxy include
> > > the To tag in the hash computation ? If it did, then
> > > ACK will map to a different transaction than the
> > > corresponding INVITE ??
> >
> > Indeed. I would just go on the To's URI.
> >
> > HTH,
> >
> > - Jo.
>
> _______________________________________________
> This list is for continuing development of the SIP protocol.
> The sip-implementer's list is the place to discuss implementation,
> and to receive advice on understanding existing sip.
> To subscribe to it, send mail to [EMAIL PROTECTED] with
> "subscribe sip-implementors" in the body.
> _______________________________________________
> Sip-implementors mailing list
> [EMAIL PROTECTED]
> http://lists.cs.columbia.edu/mailman/listinfo/sip-implementors
>
_______________________________________________
Sip-implementors mailing list
[EMAIL PROTECTED]
http://lists.cs.columbia.edu/mailman/listinfo/sip-implementors
