474952 - get rid 200 mlsec timeout in mochitests addA11yLoadEvent function

Status: RESOLVED WORKSFORME

(Core :: Disability Access APIs, defect)

Description

[alexander :surkov (:asurkov)]

Technically when we fire document is loaded then we should be ready for the work. So if I replace 200 on 0 then two mochitests fails

test_events_caret_move - it uses addLoadEvent instead of addA11yLoadEvent, but caret events are fired after document load so these tests fails. If I start use addA11yLoadEvent then two first tests fails (tests for -1 caret offset)
test_mutation_events - I suspect tests for display properties are fail, even document is loaded we aren't ready to work with some display stuffs.

Comment 1

[alexander :surkov (:asurkov)]

taking, I hope to investigate the problem

Comment 2

[David Bolter [:davidb] (NeedInfo me for attention)]

Regarding test_events_caret_move, it seems to me that the editable areas are not truly ready when the doc has loaded (and not busy) -- otherwise why should tacking <=200 msecs on after the doc is not busy have an affect?

Comment 3

[David Bolter [:davidb] (NeedInfo me for attention)]

By the way, we should maybe reduce 200 to 0 for the exsiting, and do the aFunc.call() itself on a timeout (of 200).

Comment 4

[(no longer active)]

I haven't investigated the HTML editor code closely yet, but I'll let you know if I find out anything interesting in this regard.

Comment 5

[David Bolter [:davidb] (NeedInfo me for attention)]

Thanks Ehsan.

Alexander, regarding comment 3 - this was my thinking:
https://bug510128.bugzilla.mozilla.org/attachment.cgi?id=414772

Comment 6

[alexander :surkov (:asurkov)]

(In reply to comment #5)
> Thanks Ehsan.
> 
> Alexander, regarding comment 3 - this was my thinking:
> https://bug510128.bugzilla.mozilla.org/attachment.cgi?id=414772

Yeah, that's good idea I think. Either you're right or not or somewhere in the middle :) Any it's worth to try.

Comment 7

[Trevor Saunders (:tbsaunde)]

It looks like at some point this got changed since the set time out in AddA11yLoadEvent() in trunk looks to be 0 now.  So calling this worksforme