Re: ws: and wss: schemes

Ian Hickson wrote:
> ...
>> I meant Section 3.1, which essentially is useless, as it replicates 
>> what's said in the ABNF in the registration template.
> 
> The ABNF doesn't say how you parse the URI, it says how you check if it's 
> valid. Section 3.1 doesn't say how you parse the URI, it says how you 

Actually, unless it's ambiguous, an ABNF *does* define how to parse.

> apply the algorithm from [WebAddresses] in a way that extracts the fields 
> you need to use Web Sockets.

As far as I can tell, it's pretty stand-alone, except for a reference to 
the "resolve the URL" algorithm, which doesn't seem to make sense here.

>>>> I hear that by specifying an algorithm you want to exclude certain
>>>> standard things like fragments, and include error handling; but I think
>>>> ABNF + prose would be much easier to understand.
>>> Please send such feedback to Larry; I am no longer editing those algorithms.
>> I'm still talking about WebSockets, Part 3.1.
> 
> I have no idea how ABNF could possibly be used in section 3.1. Could you 
> show an example of what you mean?

Just use the same ABNF, state which parts define the resource name (and 
what it means), and be done with it.

>>>> Furthermore, fragment identifiers are orthogonal to the URI scheme, 
>>>> see 
>>>> <http://greenbytes.de/tech/webdav/rfc3986.html#rfc.section.3.5.p.2>:
>>>>
>>>> "Fragment identifier semantics are independent of the URI scheme and 
>>>> thus cannot be redefined by scheme specifications."
>>> I've no idea to what you are referring here. Where are fragment 
>>> identifiers even mentioned in the Web Socket protocol spec?
>> You did mention them on IRC 
>> (<http://krijnhoetmer.nl/irc-logs/whatwg/20090904#l-1007>):
>>
>>> # [23:26] <Hixie> annevk3: and i want the frag-id case to be invalid 
>>> before conversion
>> What I'm trying to explain is you can't make frag-ids "invalid", even by 
>> the way you specify the parsing.
> 
> Ok.
> 
> I encourage you to review the actual spec, and not my comments on IRC.

That conversation seemed to explain why you are including all that 
specification text that IMHO is useless. I was just pointing out that, 
if the reason is to exclude fragment IDs, this is not going to have the 
effect you seem to intend.

Best regards, Julian

Received on Thursday, 17 September 2009 10:36:09 UTC