W3C home > Mailing lists > Public > public-webrtc@w3.org > January 2015

DataChannel clarifications/nits

From: Randell Jesup <randell-ietf@jesup.org>
Date: Thu, 15 Jan 2015 09:25:05 -0500
Message-ID: <54B7CDC1.6090605@jesup.org>
To: public-webrtc@w3.org
Some areas of vagueness that have been pointed out:

MessageEvent (onmessage) is not defined (marked TODO)

It's not specified if onmessage can fire after close() or if it can fire 
after onclose.  Implied in the existing text elsewhere is that onmessage 
can fire after close(), but not after onclose.

In the "closing procedure" for DataChannels, it doesn't explicitly call 
out what happens when the other side calls close(); it is implied by 
"When a RTCDataCHannel object's underlying data transport has been 
closed", and earlier (for close()) "This will eventually render the data 
transport closed".   Something along the lines of clarifying "When that 
happens the user agent MUST..." to "When the underlying data transport 
is torn down the user agent MUST..." and later changing "This will 
eventually render..." and moving those parts to close(), since they only 
apply to close(), not to the other end causing it to shut down in a 
"non-abrupt manner".  Also, the "closing procedure" should also be run 
for abrupt transitions to 'closed'.

That part just generally needs to be cleaned up.

It also speaks to firing "NetworkError", which isn't defined. 'error' is 
defined (though as a TODO).

Nit: in 11 Event Summary, in 'close': "has bee closed" [sic]

-- 
Randell Jesup -- rjesup a t mozilla d o t com
Received on Thursday, 15 January 2015 14:24:50 UTC

This archive was generated by hypermail 2.3.1 : Monday, 23 October 2017 15:19:42 UTC