298313 - nsIWebNavigation.stop(STOP_CONTENT|STOP_ALL) doesn't cancel pending JavaScript timeouts

Status: NEW

(Core :: DOM: Navigation, defect)

Description

[Giorgio Maone [:ma1]]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; it-IT; rv:1.7.8) Gecko/20050511 Firefox/1.0.4
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; it-IT; rv:1.7.8) Gecko/20050511 Firefox/1.0.4 ***AND*** Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b2) Gecko/20050620 Firefox/1.0+

According to specs (
http://lxr.mozilla.org/seamonkey/source/docshell/base/nsIWebNavigation.idl#239
), nsIWebNavigation.STOP_CONTENT and nsIWebNavigation.STOP_ALL flags, passed to
the nsIWebNavigation.stop() method, should cause pending JavaScript timeouts to
be cancelled. This doesn't happen in branch nor in trunk (even though in trunk a
cancelPendingEvents() is called on the UI event queue by nsWebNavigation.stop()
for STOP_CONTENT flag).

Reproducible: Always

Steps to Reproduce:
1. Open a page with a recurring Javascript timeout (e.g.
http://software.informaction.com/webnavtest.html )
2. Invoke stop(2) and/or stop(3) on the WebNavigation object hosting that page
3. check if timeout is running

Actual Results:  
The JavaScript timeout callback is executed again after the stop() calls

Expected Results:  
The callback should have not be invoked after the stop() calls

Comment 1

[Boris Zbarsky [:bzbarsky]]

What behavior do we actually want here?  Brendan keeps saying we shouldn't be stopping stuff in various places, but maybe he means not calling Stop() at all?

Comment 2

[Colin Blake]

What behavior? How about what's documented at
http://lxr.cenzic.com/mozilla/source/docshell/base/nsIWebNavigation.idl#259
(killing all pending javascript timeouts).

Comment 3

[Colin Blake]

Sorry about that internal URL in comment #2. I meant to say
http://lxr.mozilla.org/seamonkey/source/docshell/base/nsIWebNavigation.idl#260

Comment 4

[Boris Zbarsky [:bzbarsky]]

Yeah, I'm fine with that happening.  Like I said, I think brendan just means we should just be calling Stop() in fewer places.

I'll be happy to review a patch to do this.

Comment 5

[BugBot [:suhaib / :marco/ :calixte]]

The bug assignee is inactive on Bugzilla, so the assignee is being reset.