[Webkit-unassigned] [Bug 148963] Implement EventSenderProxy for WebKitTestRunner on iOS

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=148963

--- Comment #7 from Wenson Hsieh <wenson_hsieh at apple.com> ---
Comment on attachment 260776
  --> https://bugs.webkit.org/attachment.cgi?id=260776
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=260776&action=review

>> LayoutTests/fast/forms/disabled-search-input.html:46
>> +    }, 500);
> 
> This is not right. Nothing in regression tests is guaranteed to happen within 500 ms - there are many processes competing for the CPU, and with hyperthreading, some of the virtual cores can be very slow. Furthermore, CPU load increases even further when there is a crash, or even media playing.

Sorry about that -- I don't intend on landing with this timeout here. There's one last major issue I've been investigating regarding WKTR, which is that sending a tap immediately on body load by calling eventSender.mouseDown, eventSender.mouseUp causes us to hit an early return and ignore the tap. I added the timeout here for now just to circumvent this and fix the rest of the issues with eventSender. When I figure out how to best address this, the timeout will not be necessary.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20150909/dd73fa6e/attachment.html>