Re: [hybi] WebSocket subprotocol parameters

[Julian Reschke <[e-mail]>]

On 2014-01-17 10:24, Tobias Oberstein wrote:
> The "Sec-WebSocket-Extensions" allows for parameters for each element, eg (from "permessage-deflate"):
>
> Sec-WebSocket-Extensions: permessage-deflate; client_max_window_bits; server_max_window_bits=10, permessage-deflate; client_max_window_bits
>
> Similar seems to be disallowed for _subprotocols_.
>
> RFC6455 says:
>
> """
> The elements that comprise this value
> MUST be non-empty strings with characters in the range U+0021 to
> U+007E not including separator characters as defined in
> [RFC2616] and MUST all be unique strings.  The ABNF for the
> value of this header field is 1#token, where the definitions of
> constructs and rules are as given in [RFC2616].
> """
>
> and RFC2616 says:
>
> """
>         token          = 1*<any CHAR except CTLs or separators>
>         separators     = "(" | ")" | "<" | ">" | "@"
>                        | "," | ";" | ":" | "\" | <">
>                        | "/" | "[" | "]" | "?" | "="
>                        | "{" | "}" | SP | HAT
> """
>
> So
>
> Sec-WebSocket-Protocol: wamp; version=2; serialization=msgpack; batched, wamp; version=1
>
> is disallowed, right?
>
> This is unfortunate, since "simulating parameters" like
>
> Sec-WebSocket-Protocol: wamp.2.msgpack.batched, wamp
>
> is a hack, and more so, RFC6455 requires that all subprotocol name be registered via the " WebSocket Subprotocol Name Registry".
>
> This would mean registering all subprotocol variants:
>
> wamp
> wamp.2.json
> wamp.2.msgpack
> wamp.2.json.batched
> wamp.2.msgpack.batched
> ..
>
> Was there any good reason disallowing parameters for "Sec-WebSocket-Protocol"?

I think that was discussed and the decision was made on purpose. For 
details you'll have to visit the mailing list archives though.

> If not, could this be fixed via errata for RFC6455?

Um, no. This is not a bug but a change request (with the potential of 
breaking existing code).

Best regards, Julian