[webkit-reviews] review granted: [Bug 191333] PointerEvents should not require touch event listeners to be registered : [Attachment 354947] Patch
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Thu Nov 15 09:52:19 PST 2018
Dean Jackson <dino at apple.com> has granted Antoine Quint <graouts at apple.com>'s
request for review:
Bug 191333: PointerEvents should not require touch event listeners to be
registered
https://bugs.webkit.org/show_bug.cgi?id=191333
Attachment 354947: Patch
https://bugs.webkit.org/attachment.cgi?id=354947&action=review
--- Comment #3 from Dean Jackson <dino at apple.com> ---
Comment on attachment 354947
--> https://bugs.webkit.org/attachment.cgi?id=354947
Patch
View in context: https://bugs.webkit.org/attachment.cgi?id=354947&action=review
> Source/WebCore/dom/EventNames.h:362
> + std::array<std::reference_wrapper<const AtomicString>, 9>
touchEventNames() const;
Rename this to touchAndPointerEventNames
> Source/WebKit/UIProcess/WebPageProxy.cpp:2327
> + updateTrackingType(m_touchEventTracking.touchStartTracking,
names.pointerdownEvent);
> + updateTrackingType(m_touchEventTracking.touchMoveTracking,
names.pointermoveEvent);
> + updateTrackingType(m_touchEventTracking.touchEndTracking,
names.pointerupEvent);
Maybe m_touchEventTracking should be renamed as well.
> LayoutTests/pointerevents/utils.js:10
> + if (arguments.length !== 2 && arguments.length !== 3) {
> + console.error(`target_test expected 2 or 3 arguments but got
${arguments.length}.`);
> + return;
> + }
> +
> + const impliedOptions = arguments.length == 2;
> +
Don't use the horrible arguments object. Instead, do a ..args rest parameter.
> LayoutTests/pointerevents/utils.js:54
> + inputType : "hand",
Strange this isn't "finger"
> LayoutTests/pointerevents/utils.js:106
> + events = [events];
Wouldn't this already take a list?
> LayoutTests/pointerevents/utils.js:107
> + return this._run(`uiController.sendEventStream('${JSON.stringify({
events })}')`);
nice. You could probably Array-ify the events inside the ${} though.
More information about the webkit-reviews
mailing list