[Webkit-unassigned] [Bug 202214] Update API availability for autoplay event

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Oct 7 21:01:56 PDT 2019


https://bugs.webkit.org/show_bug.cgi?id=202214

--- Comment #12 from Wenson Hsieh <wenson_hsieh at apple.com> ---
Comment on attachment 379869
  --> https://bugs.webkit.org/attachment.cgi?id=379869
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=379869&action=review

> Source/WebCore/page/Quirks.cpp:-109
> -    auto host = m_document->topDocument().url().host();
> -    return equalLettersIgnoringASCIICase(host, "netflix.com") || host.endsWithIgnoringASCIICase(".netflix.com");

Just to double check — would we not need this quirk on iOS anymore because it will be handled by Safari?

> Source/WebKit/WebProcess/WebPage/ios/WebPageIOS.mm:848
> +void WebPage::simulateTap(const IntPoint& point, OptionSet<WebEvent::Modifier> modifiers)
> +{
> +    FloatPoint adjustedPoint;
> +    Node* nodeRespondingToClick = m_page->mainFrame().nodeRespondingToClickEvents(point, adjustedPoint);
> +    if (nodeRespondingToClick)
> +        handleSyntheticClick(*nodeRespondingToClick, adjustedPoint, modifiers);
> +}

Hm…I think we generally try to avoid deep testing hooks like this if we can, and mostly limit interaction-related testing hooks such as this to the UI process (commonly, WKContentViewInteraction and WKWebView).

That said, I recall that this was difficult in API tests due to how there are `lastLayerTreeTransactionId` checks throughout WebPageIOS that fail due to `lastLayerTreeTransactionId ` being 0. If that’s the case, could we additionally synthesize touch events by calling into -_webTouchEventsRecognized: (or add a simple hook to do so) before sending the click? I think that /might/ be slightly cleaner than adding a different testing codepath in the web process for simulating clicks.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20191008/7cb25e79/attachment-0001.html>


More information about the webkit-unassigned mailing list