Bug 217832

Created attachment 411584 [details]
Patch

Created attachment 411611 [details]
Patch

Created attachment 411664 [details]
Patch

Comment on attachment 411664 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=411664&action=review

> Source/WebCore/layout/inlineformatting/InlineFormattingContextGeometry.cpp:411
>      auto lineBoxLogicalHeight = lineBox.logicalHeight();
> -    auto lineLogicalHeight = InlineLayoutUnit { };
> -    if (rootStyle.lineHeight().isNegative()) {
> -        // Negative line height value means the line height is driven by the content.
> -        auto usedLineSpacing = [&] {
> -            auto logicalTopWithLineSpacing = InlineLayoutUnit { };
> -            auto logicalBottomWithLineSpacing = lineBoxLogicalHeight;
> -            for (auto& inlineLevelBox : lineBox.inlineLevelBoxList()) {
> -                if (auto lineSpacing = inlineLevelBox->lineSpacing()) {
> -                    // FIXME: check if line spacing is distributed evenly.
> -                    logicalTopWithLineSpacing = std::min(logicalTopWithLineSpacing, inlineLevelBox->logicalTop() - *lineSpacing / 2);
> -                    logicalBottomWithLineSpacing = std::max(logicalBottomWithLineSpacing, inlineLevelBox->logicalBottom() + *lineSpacing / 2);
> -                }
> -            }
> -            return -logicalTopWithLineSpacing + (logicalBottomWithLineSpacing - lineBoxLogicalHeight);
> -        };
> -        lineLogicalHeight = lineBox.logicalHeight() + usedLineSpacing();
> -    } else
> -        lineLogicalHeight = rootStyle.computedLineHeight();
> -
> +    auto lineLogicalHeight = lineBox.logicalHeight();
>      auto logicalRect = InlineRect { lineContent.logicalTopLeft, lineContent.lineLogicalWidth, lineLogicalHeight};
>      // Inline tree height is all integer based.
>      auto lineBoxOffset = floorf((lineLogicalHeight - lineBoxLogicalHeight) / 2);

lineBoxOffset == 0?

(In reply to Antti Koivisto from comment #4)
> Comment on attachment 411664 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=411664&action=review
> 
> > Source/WebCore/layout/inlineformatting/InlineFormattingContextGeometry.cpp:411
> >      auto lineBoxLogicalHeight = lineBox.logicalHeight();
> > -    auto lineLogicalHeight = InlineLayoutUnit { };
> > -    if (rootStyle.lineHeight().isNegative()) {
> > -        // Negative line height value means the line height is driven by the content.
> > -        auto usedLineSpacing = [&] {
> > -            auto logicalTopWithLineSpacing = InlineLayoutUnit { };
> > -            auto logicalBottomWithLineSpacing = lineBoxLogicalHeight;
> > -            for (auto& inlineLevelBox : lineBox.inlineLevelBoxList()) {
> > -                if (auto lineSpacing = inlineLevelBox->lineSpacing()) {
> > -                    // FIXME: check if line spacing is distributed evenly.
> > -                    logicalTopWithLineSpacing = std::min(logicalTopWithLineSpacing, inlineLevelBox->logicalTop() - *lineSpacing / 2);
> > -                    logicalBottomWithLineSpacing = std::max(logicalBottomWithLineSpacing, inlineLevelBox->logicalBottom() + *lineSpacing / 2);
> > -                }
> > -            }
> > -            return -logicalTopWithLineSpacing + (logicalBottomWithLineSpacing - lineBoxLogicalHeight);
> > -        };
> > -        lineLogicalHeight = lineBox.logicalHeight() + usedLineSpacing();
> > -    } else
> > -        lineLogicalHeight = rootStyle.computedLineHeight();
> > -
> > +    auto lineLogicalHeight = lineBox.logicalHeight();
> >      auto logicalRect = InlineRect { lineContent.logicalTopLeft, lineContent.lineLogicalWidth, lineLogicalHeight};
> >      // Inline tree height is all integer based.
> >      auto lineBoxOffset = floorf((lineLogicalHeight - lineBoxLogicalHeight) / 2);
> 
> lineBoxOffset == 0?
was testing the review process :)
Will get rid of that offset concept in a follow-up patch.

Committed r268642: <https://trac.webkit.org/changeset/268642>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 411664 [details].

<rdar://problem/70407284>