Remove AffectedByDrag style flag

• Component: CSS
• Status: RESOLVED FIXED

[Antti Koivisto]

Remove the last remaining user action related AffectedBy flag.

[Antti Koivisto]

Created attachment 393490 [details]
patch

[Simon Fraser (smfr)]

Comment on attachment 393490 [details]
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=393490&action=review

Dragging is under tested, so some manual testing of this is warranted.

> Source/WebCore/dom/Element.h:318
> +    bool isDragged() const { return isUserActionElement() && isUserActionElementDragged(); }

isDragged is weird. isBeingDragged?

> Source/WebCore/dom/Element.h:324
> +    void setDragged(bool);

setIsBeingDragged?

> Source/WebCore/dom/Node.h:612
> +        ChildrenAffectedByForwardPositionalRules = 1 << 1,
> +        DescendantsAffectedByForwardPositionalRules = 1 << 2,
> +        ChildrenAffectedByBackwardPositionalRules = 1 << 3,
> +        DescendantsAffectedByBackwardPositionalRules = 1 << 4,
> +        ChildrenAffectedByPropertyBasedBackwardPositionalRules = 1 << 5,

I would align the = but that's just me.

[Antti Koivisto]

Created attachment 393504 [details]
patch

[WebKit Commit Bot]

Comment on attachment 393504 [details]
patch

Clearing flags on attachment: 393504

Committed r258416: <https://trac.webkit.org/changeset/258416>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Radar WebKit Bug Importer]

<rdar://problem/60427264>

[Darin Adler]

Comment on attachment 393490 [details]
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=393490&action=review

> Source/WebCore/dom/UserActionElementSet.h:58
>      enum class Flag {

: uint8_t maybe?

> Source/WebCore/dom/UserActionElementSet.h:63
> +        IsActive        = 1 << 0,
> +        InActiveChain   = 1 << 1,
> +        IsHovered       = 1 << 2,
> +        IsFocused       = 1 << 3,
> +        IsDragged       = 1 << 4,

I’m like the opposite of Simon. I would not have lined these up, and it’s just me saying that ;)

[Antti Koivisto]

I'm totally in line-up camp, I just forgot to do it with this enum in the original patch!

[Antti Koivisto]

I wish this was a language feature.

[Simon Fraser (smfr)]

(In reply to Antti Koivisto from comment #9)
> I wish this was a language feature.

We should just clang-format everything!

[Antti Koivisto]

Hand-indenting these is not that hard either.

I meant a C++ feature to generate bit enums without the silly 1 << n dance.