[PE] Start working on Progress Events

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

[PE] Start working on Progress Events

Jungkee Song
Hi all,

I came to start working on Progress Events spec to move it towards REC. Because the spec is already a CR, I am planning to focus on satisfying the exit criteria to ship it. Please see inline comments and questions.

Jungkee


> From: Arthur Barstow [mailto:[hidden email]]
> Sent: Thursday, November 15, 2012 10:51 PM
>
> On 11/15/12 3:11 AM, ext Jungkee Song wrote:
> > Hi Art, Charles and Anne,
> >
> > At this stage, it will be of great help if you give me some comments on
> any issues, concerns, expected actions, etc.
>
> Since the spec is already a CR (with exit criteria
> <http://www.w3.org/TR/2011/CR-progress-events-20110922/#crec>), some
> questions ...
>
> * Are there any significant differences between the CR and Anne's WHATWG
> spec? If yes, what are they and should they postponed to v.next?


As I've gone through it, there's no significant change. There are only a few minor ones including term (octets to bytes) and xref (to event definition) things.


> * What is the implementation status of the CR? Are there at least two
> independent implementations that can be tested?


This is my question at the moment. Can anyone share implementation data for this spec?


> * Are the tests in the test suite sufficient to test the CR
> <http://w3c-test.org/webapps/ProgressEvents/tests/submissions/Ms2ger/>?
> If not, what is the plan to fill the gaps?


I will scope it out.


> BTW, I have a relatively strong preference to have this conversation  on
> public-webapps so please feel free to copy any part of what I say above
> to that list.
>
> -Thanks, Art



Reply | Threaded
Open this post in threaded view
|

[PE] interface MessageEvent : ProgressEvent

Jungkee Song
Hi,

While mulling over the usage of PE event, I came up with a suggestion. As the data attribute of message event delivers structured objects including File Blob and ArrayBuffer objects, authors might need a primitive way of checking the progress of message load.

I think the MessageEvent interface derived from ProgressEvent interface would solve many such use cases. It seems we already have candidates: server-sent events, web socket, cross-document messaging, channel messaging, web worker.

WDYT?

[Constructor(DOMString type, optional MessageEventInit eventInitDict)]
interface MessageEvent : ProgressEvent {
  readonly attribute any data;
  readonly attribute DOMString origin;
  readonly attribute DOMString lastEventId;
  readonly attribute (WindowProxy or MessagePort)? source;
  readonly attribute MessagePort[]? ports;
};


Jungkee

> -----Original Message-----
> From: Jungkee Song [mailto:[hidden email]]
> Sent: Friday, November 16, 2012 11:53 AM
> To: [hidden email]
> Subject: [PE] Start working on Progress Events
>
> Hi all,
>
> I came to start working on Progress Events spec to move it towards REC.
> Because the spec is already a CR, I am planning to focus on satisfying the
> exit criteria to ship it. Please see inline comments and questions.
>
> Jungkee
>
>
> > From: Arthur Barstow [mailto:[hidden email]]
> > Sent: Thursday, November 15, 2012 10:51 PM
> >
> > On 11/15/12 3:11 AM, ext Jungkee Song wrote:
> > > Hi Art, Charles and Anne,
> > >
> > > At this stage, it will be of great help if you give me some comments
> on
> > any issues, concerns, expected actions, etc.
> >
> > Since the spec is already a CR (with exit criteria
> > <http://www.w3.org/TR/2011/CR-progress-events-20110922/#crec>), some
> > questions ...
> >
> > * Are there any significant differences between the CR and Anne's WHATWG
> > spec? If yes, what are they and should they postponed to v.next?
>
>
> As I've gone through it, there's no significant change. There are only a
> few minor ones including term (octets to bytes) and xref (to event
> definition) things.
>
>
> > * What is the implementation status of the CR? Are there at least two
> > independent implementations that can be tested?
>
>
> This is my question at the moment. Can anyone share implementation data
> for this spec?
>
>
> > * Are the tests in the test suite sufficient to test the CR
> > <http://w3c-test.org/webapps/ProgressEvents/tests/submissions/Ms2ger/>?
> > If not, what is the plan to fill the gaps?
>
>
> I will scope it out.
>
>
> > BTW, I have a relatively strong preference to have this conversation  on
> > public-webapps so please feel free to copy any part of what I say above
> > to that list.
> >
> > -Thanks, Art
>



Reply | Threaded
Open this post in threaded view
|

Re: [PE] interface MessageEvent : ProgressEvent

Anne van Kesteren-4
On Fri, Nov 16, 2012 at 12:48 AM, Jungkee Song <[hidden email]> wrote:
> I think the MessageEvent interface derived from ProgressEvent interface would solve many such use cases. It seems we already have candidates: server-sent events, web socket, cross-document messaging, channel messaging, web worker.
>
> WDYT?

I'm struggling to see how this makes sense. Sure, we might want
progress indication for some of these, but why would you use the
MessageEvent interface in that? How does that work?


--
http://annevankesteren.nl/