[Webkit-unassigned] [Bug 183621] Add support for the `inputmode` attribute

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=183621

--- Comment #13 from Daniel Bates <dbates at webkit.org> ---
Comment on attachment 347820
  --> https://bugs.webkit.org/attachment.cgi?id=347820
Patch for landing

View in context: https://bugs.webkit.org/attachment.cgi?id=347820&action=review

> Source/WebCore/ChangeLog:8
> +        Added the inputmode attribute to the HTMLInputElement and HTMLTextAreaElement IDL

It is an unwritten convention that we reference the spec. and version/date of the spec. in the ChangeLog when adding new web exposed functionality. One example of such a ChangeLog (though it does not include a date) is <https://trac.webkit.org/changeset/234680/>. This records the spec you referenced when writing this patch.

> Source/WebCore/html/HTMLTextAreaElement.idl:62
> +    attribute DOMString inputMode;

As per <https://html.spec.whatwg.org/multipage/interaction.html#dom-inputmode> this should be a reflected attribute. Please add "[Reflect]" before "attribute".

> Source/WebCore/html/HTMLTextFormControlElement.cpp:678
> +String HTMLTextFormControlElement::inputMode() const

How did you come to the decision to return a String instead of returning an const AtomicString&? I mean, returning a String causes unnecessary ref-count churn.

> Source/WebCore/html/HTMLTextFormControlElement.cpp:702
> +void HTMLTextFormControlElement::setInputMode(const String& value)
> +{
> +    setAttributeWithoutSynchronization(inputmodeAttr, value);
> +}

This function is unnecessary when the attribute is marked "[Reflect]" in the IDL.

-- 
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/20180822/01240dd2/attachment.html>