183604 - The DOM code should re-use the "OS" timer for inteval timers

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect, P3)

Description

[Johnny Stenback (:jst)]

Now that the timer code supports re-initializeing existing timers we should take
advantage of that in the DOM timeout code. Patch coming up.

Comment 1

[Johnny Stenback (:jst)]

Attachment: Don't re-create timers for inteval timeouts, re-use the existing one.

Comment 2

[Christopher Aillon (sabbatical, not receiving bugmail)]

Comment on attachment 108274 [details] [diff] [review]
Don't re-create timers for inteval timeouts, re-use the existing one.

Man, a diff -w of some sorts would have really been nice here.

>+      // Reschedule timeout. Ignore any errors since nobody who
>+      // listens them will care anyways.

listens _to_ them

r=caillon

Comment 3

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 108274 [details] [diff] [review]
Don't re-create timers for inteval timeouts, re-use the existing one.

sr=bzbarsky with that comment change.

Comment 4

[Johnny Stenback (:jst)]

Comment changed, I looked at a diff -w of this before attaching the diff, but it
didn't actually help much so I ended up not attaching it. Thanks for the reviews.

Comment 5

[Johnny Stenback (:jst)]

Comment on attachment 108274 [details] [diff] [review]
Don't re-create timers for inteval timeouts, re-use the existing one.

Bah. That didn't work, I don't know what I was thinking...

New patch coming up...

Comment 6

[Johnny Stenback (:jst)]

Attachment: Re-initialize the existing timer for inteval timeouts

This fixes this bug, for real this time. This also eliminates a bunch of silly
assumptions about who owns what based on the existance of a OS timer or not.
This should be far more maintainable from here on. diff -w coming up of this
one.

Comment 7

[Johnny Stenback (:jst)]

Attachment: diff -w

Comment 8

[Christopher Aillon (sabbatical, not receiving bugmail)]

Comment on attachment 108963 [details] [diff] [review]
diff -w

I thought I marked this already with r=caillon.  Anyway, I think that bz is on
vacation now, so unless you wanna wait like a month or so, maybe peterv should
sr this.

Comment 9

[Peter Van der Beken [:peterv]]

Comment on attachment 108963 [details] [diff] [review]
diff -w

sr=peterv with the minor change you wanted to do for isInterval.

Comment 10

[Johnny Stenback (:jst)]

Attachment: Final complete patch with the minor tweak to isInteval's initialization.

Comment 11

[Johnny Stenback (:jst)]

Fix checked in.