W3C home > Mailing lists > Public > public-webapps@w3.org > July to September 2008

ISSUE-38: DOM 3 Events abort/error bubble, ProgressEvent abort/error do not [Progress Events]

From: Web Applications Working Group Issue Tracker <sysbot+tracker@w3.org>
Date: Thu, 10 Jul 2008 11:01:20 +0000 (GMT)
To: public-webapps@w3.org
Message-Id: <20080710110120.E06426B62A@kent.w3.org>

ISSUE-38: DOM 3 Events abort/error bubble, ProgressEvent abort/error do not [Progress Events]


Raised by: Philip J├Ągenstedt
On product: Progress Events

The DOM 3 abort/error events (http://www.w3.org/TR/DOM-Level-3-Events/events.html) bubble, while the progress events by the same name do not bubble.

First, is formally/practically possible to have two separate events with the same name but different bubbles/cancelable flags? I am not aware of any other such cases so it is reasonable to think that there will be implementation problems because the event name can no longer be uniquely mapped to bubbles/cancelable without taking some other information into account.

It's also not obvious what type of event addEventListener("error", <listener>, true) captures if it is attached to an ancenstor of one node which is a target for ProgressEvent-error and another which is a target for Event-error.

Even though it is possible to work around these problems they would be better be solved at the spec level.

Possible ways forward:

1. Change progress event abort/error to bubble. The problem is that this doesn't really make any sense in the context where progress events are supposed to be used though.

2. Change DOM 3 Events abort/error to not bubble. Is this even remotely possible? In what cases do bubbling abort/error matter?

3. Change the name of progress event's abort/error to something else.

Unconfirmed rumors say that the SVG WG is considering changing SVGAbort/SVGError to not bubble so that they are compatible with progress events, so some kind of synchronization with that is probably a good idea.
Received on Thursday, 10 July 2008 11:03:00 UTC

This archive was generated by hypermail 2.3.1 : Friday, 27 October 2017 07:26:11 UTC