input[type='number'] with step="any" suffers from step mismatch.

[Shinya Kawanaka]

This bug is related to https://bugs.webkit.org/show_bug.cgi?id=57723

For this input:
<input min="0" max="100" step="any" value="1.5">
stepping up from renderer makes the value 3.0. I think this should be 2.5.
The input with step="any" should not suffer from step mismatch.

[Shinya Kawanaka]

Created attachment 101446 [details]
Patch

[Kent Tamura]

Comment on attachment 101446 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=101446&action=review

> Source/WebCore/html/HTMLInputElement.cpp:429
> +    // If step is not "any", the value should by aligned to the step.
> +    const AtomicString& stepString = fastGetAttribute(stepAttr);
> +    if (!equalIgnoringCase(stepString, "any")) {

This comment is not helpful. We avoid comments about "what we do here".  This comment should be removed.
Also, this comment has a typo; by -> be.

In this case, you might want to make a function named alignValueForStep(...) for readability.

[Shinya Kawanaka]

Created attachment 101690 [details]
Patch

[Kent Tamura]

Comment on attachment 101690 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=101690&action=review

ok.

> Source/WebCore/html/HTMLInputElement.cpp:428
> +    const AtomicString& stepString = fastGetAttribute(stepAttr);
> +    if (!equalIgnoringCase(stepString, "any"))

nit: The variable 'stepString' is not needed.

[WebKit Review Bot]

Comment on attachment 101690 [details]
Patch

Clearing flags on attachment: 101690

Committed r91549: <http://trac.webkit.org/changeset/91549>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.