Well-known variations should clamp to the values listed in the @font-face block

[Myles C. Maxfield]

See https://github.com/w3c/csswg-drafts/issues/782

[Myles C. Maxfield]

Created attachment 305824 [details]
Needs tests

[Radar WebKit Bug Importer]

<rdar://problem/31516360>

[Myles C. Maxfield]

Created attachment 306651 [details]
CSSFontFace::m_fontSelectionCapabilities needs to be able to be optional in CSSFontFace::font()

[Myles C. Maxfield]

Created attachment 306856 [details]
Needs tests

[Myles C. Maxfield]

Created attachment 306862 [details]
Patch

[Myles C. Maxfield]

Created attachment 306864 [details]
Patch

[Myles C. Maxfield]

Created attachment 306885 [details]
Patch

[Myles C. Maxfield]

Created attachment 306886 [details]
Patch

[Myles C. Maxfield]

Created attachment 306888 [details]
Patch

[Dean Jackson]

Comment on attachment 306888 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=306888&action=review

> Source/WebCore/platform/graphics/FontSelectionAlgorithm.h:319
> +    uint32_t uniqueValue() const
> +    {
> +        return minimum.rawValue() << 16 | maximum.rawValue();
> +    }
> +

Is that really unique?

[Myles C. Maxfield]

(In reply to Dean Jackson from comment #10)
> Comment on attachment 306888 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=306888&action=review
> 
> > Source/WebCore/platform/graphics/FontSelectionAlgorithm.h:319
> > +    uint32_t uniqueValue() const
> > +    {
> > +        return minimum.rawValue() << 16 | maximum.rawValue();
> > +    }
> > +
> 
> Is that really unique?

This function maps two 16-bit numbers to a unique 32-bit number, so yes.

[WebKit Commit Bot]

Comment on attachment 306888 [details]
Patch

Clearing flags on attachment: 306888

Committed r215287: <http://trac.webkit.org/changeset/215287>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.