W3C home > Mailing lists > Public > public-webapps@w3.org > October to December 2009

Re: [EventSource] feedback from implementors

From: Ian Hickson <ian@hixie.ch>
Date: Tue, 13 Oct 2009 07:22:58 +0000 (UTC)
To: Per-Erik Brodin <per-erik.brodin@ericsson.com>
Cc: "Michael A. Puls II" <shadow2531@gmail.com>, Anne van Kesteren <annevk@opera.com>, public-webapps@w3.org
Message-ID: <Pine.LNX.4.62.0910130548420.25383@hixie.dreamhostps.com>
On Thu, 8 Oct 2009, Per-Erik Brodin wrote:
> Ian Hickson wrote:
> > On Fri, 18 Sep 2009, Per-Erik Brodin wrote:
> >  >
> >  > When parsing an event stream, allowing carriage return, carriage return
> >  > line feed, and line feed to denote line endings introduces unnecessary
> >  > ambiguity into the spec. For example, the sequence "\r\r\n\n" could be
> >  > interpreted as three or four line endings.
> > 
> > I've clarified the stream parsing format to define the above strictly as
> > three line endings (not counting the end of the line).
> 
> That should take care of the ambiguity in stream parsing, but "not 
> followd by a U+000D CARRIAGE RETURN (LF)" should probably read "not 
> followed by a U+000A LINE FEED (LF)".

Fixed.


> > Similarly, no 'close' event is dispatched, no 'peanut' event is 
> > dispatched, and the disk doesn't get reformatted. :-)
> 
> I figured the disk doesn't get reformatted because that wouldn't be nice 
> to users, but are you sure about the peanut event? Seriously though, 
> sometimes it could be nice to have a "must not ..". Maybe it's not 
> motivated here.

I try to avoid listing what should _not_ happen because then people start 
thinking there's a reason I haven't listed some other crazy thing that I 
didn't list. :-)


> > > Finally, it could be useful to be able to reset the reconnection 
> > > time to the user agent default value by sending the retry field only 
> > > and leave out the value similar to how you reset the last event id.
> > 
> > What's the use case?
> 
> Take the stock ticker as an example. When the stock market closes the 
> server logic knows that there won't be any new events for a number of 
> hours and so it can send the corresponding reconnection time and close 
> the connection. If the client is still running by the time the market 
> opens, it will reconnect, and the server can now reset the reconnection 
> time to a time that is convenient for the user agent (which is the user 
> agent default value, unknown to the server).

That's an interesting use case, I will grant you that. It's probably not 
critical enough to justify adding right away though. I've noted it for a 
future version.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Tuesday, 13 October 2009 07:12:14 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 18:49:34 GMT