[AVT] Re: Technical comments on draft-ietf-avt-rtp-hdrext-12

[Dave Singer <singer at apple.com>]

At 11:47  +0200 20/06/07, Magnus Westerlund wrote:
> Hi,
>
> Sorry for not having read this in WGLC. However I think there is a 
> serious error in the spec that needs to be fixed before this going 
> further.

Thank you;  I do not claim BNF expertise...

> 7.  BNF Syntax
>
>    The syntax element 'URI-reference' is as defined in [RFC3986], except
>    that only absolute URIs are permitted here.  The syntax element
>    'extmap' is an attribute as defined in [RFC4566].
>    Extensionattributes are not defined here, but by the specification
>    that defines a specific extension name; there may be several.
>
>      digit = "0"/"1"/"2"/"3"/"4"/"5"/"6"/"7"/"8"/"9"
>
>      integer = 1*digit
>
>      space = " "
>
>      extensionname = URI-reference
>
>      direction = "sendonly" / "recvonly" / "sendrecv" / "inactive"
>
>      mapentry = "extmap:" integer ["/" direction]
>
>      mapattrs = [space extensionattributes]
>
>      extmap = mapentry space extensionname mapattrs
>
>
> There are two serious issues with this. First RFC 4234 is not 
> referenced to define that your using that type of BNF.

OK

> Secondly and really seriously, the "extensionattributes" syntax is 
> not defined.

well, it consumes everything to end of line.  we could make it 
tighter, I guess.

> Unless one defines the form of it this exercise is useless as no 
> parsers pre extension can handle and ignore the extension anyway.
>
> I would propose to rewrite this section to the following:
>
> 7.  ABNF Syntax
>
> The below syntax definition uses ABNF according to [RFC4234]. The 
> syntax element 'URI' is as defined in [RFC3986]. The syntax element 
> 'extmap' is an SDP attribute as defined in [RFC4566], i.e "a=" 
> precede the extmap definition.
>
>      extmap = mapentry SP extensionname mapattrs
>
>      extensionname = URI
>
>      URI = <Defined in RFC 3986>
>
>      direction = "sendonly" / "recvonly" / "sendrecv" / "inactive"
>
>      mapentry = "extmap:" 1*5DIGIT ["/" direction]
>
>      mapattrs = *(SP extensionattribute)
>
>      extensionattribute = non-ws-string
>
>      non-ws-string = <Defined in RFC 4566>
>
>      SP = <Defined in RFC 4234>
>
>      DIGIT = <Defined in RFC 4234>
>
> Some comments on this ABNF compared to the original one.
> 1. It limits the number of digits for the temporary identifier to 5, 
> i.e. up to 16 bits supported if this ever is extended.
>
> 2. The extension can now be basically any string as long as it 
> doesn't contain spaces. In addition there might be multiple 
> additions with spaces in between them. The alternative would be to 
> demand a space and then use byte-string (from rfc4566) to allow for 
> any separators.
>
> 3. Uses RFC 4234 base syntax.
>
> 4. Uses URI rule instead of URI-Reference so that relative ones are 
> not possible.
>
>
> Cheers
>
> Magnus Westerlund
>
>
> IETF Transport Area Director & TSVWG Chair
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM/M
> ----------------------------------------------------------------------
> Ericsson AB                | Phone +46 8 4048287
> Torshamsgatan 23           | Fax   +46 8 7575550
> S-164 80 Stockholm, Sweden | mailto: magnus.westerlund at ericsson.com
> ----------------------------------------------------------------------


--
David Singer
Apple/QuickTime

_______________________________________________
Audio/Video Transport Working Group
avt at ietf.org
https://www1.ietf.org/mailman/listinfo/avt