Change Element::isReadOnlyFormControl to Element::shouldMatchReadOnlySelector/shouldMatchReadWriteSelector or HTMLFormControlElement::readOnly

• Component: DOM
• Status: RESOLVED FIXED

This bug is preparation of bug 92473, :read-only and :read-write selector should match none-isFormisFormControlElement.

*** Bug 92607 has been marked as a duplicate of this bug. ***

Created attachment 155224 [details]
Patch 1

Comment on attachment 155224 [details]
Patch 1

Could you review this patch?
Thanks in advance.

[Kent Tamura]

Comment on attachment 155224 [details]
Patch 1

View in context: https://bugs.webkit.org/attachment.cgi?id=155224&action=review

> Source/WebCore/dom/Element.h:354
> +    virtual bool shouldMatchReadWriteSelector() const { return false; }

We should replace isReadOnlyFormControl() with shouldMatchReadWriteSelector in Element.h.
If we have other isReadOnlyFormControl() callsites, we should remove them before this patch.

> Source/WebCore/html/HTMLInputElement.cpp:1479
> +bool HTMLInputElement::shouldMatchReadWriteSelector() const
> +{
> +    return m_inputType->shouldMatchReadWriteSelector();
> +}

We have no reasons to forward it to InputType::shouldMatchReadWriteSelector() for now.
This change just wastes the code size and CPU time.

Element::isReadOnlyFormControl() is used for :read-only and :read-write CSS pseudo class matching. We would like to rename to shouldMatchReadWriteSelector() to make intention of this function explicitly.

Created attachment 155431 [details]
Patch 2

Comment on attachment 155431 [details]
Patch 2

Could you review this patch?
Thanks in advance.

[Kent Tamura]

Comment on attachment 155431 [details]
Patch 2

View in context: https://bugs.webkit.org/attachment.cgi?id=155431&action=review

> Source/WebCore/dom/Element.h:353
> +    virtual bool shouldMatchReadWriteSelector() const { return false; }

nit: we had better avoid to add the function body here for a virtual function. It wastes compile-time resources.

Created attachment 155432 [details]
Patch 3

Comment on attachment 155432 [details]
Patch 3

Clearing flags on attachment: 155432

Committed r124171: <http://trac.webkit.org/changeset/124171>

All reviewed patches have been landed.  Closing bug.

[WebKit Review Bot]

Re-opened since this is blocked by 92712

Created attachment 155452 [details]
Patch 4

Comment on attachment 155452 [details]
Patch 4

Could you review this patch?
Thanks in advance.

= Changes since the last patch =
* Introduce Element::shouldMatchReadOnlySelector, because !shouldMatchReadWriteSelector isn't equal to read only control.
** We found this by failure of fast/css/square-button-appearance.html. This test uses "div" element for rendering square button.

[Kent Tamura]

Comment on attachment 155452 [details]
Patch 4

Looks good.

Comment on attachment 155452 [details]
Patch 4

Clearing flags on attachment: 155452

Committed r124180: <http://trac.webkit.org/changeset/124180>

All reviewed patches have been landed.  Closing bug.