Bug 142146 – BreakingContext cleanup

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Myles C. Maxfield]

BreakingContext cleanup

[Myles C. Maxfield]

Created attachment 247631 [details]
Patch

[Myles C. Maxfield]

Created attachment 247634 [details]
Patch

[zalan]

Comment on attachment 247631 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=247631&action=review

> Source/WebCore/rendering/line/BreakingContextInlineHeaders.h:1052
>      if (m_lineBreak == m_resolver.position()) {
>          if (!m_lineBreak.renderer() || !m_lineBreak.renderer()->isBR()) {
>              // we just add as much as possible
> -            if (m_blockStyle.whiteSpace() == PRE && !m_current.offset()) {
> -                m_lineBreak.moveTo(m_lastObject, m_lastObject->isText() ? m_lastObject->length() : 0);
> -            } else if (m_lineBreak.renderer()) {
> +            if (m_blockStyle.whiteSpace() == PRE && !m_current.offset())
> +                commitLineBreakAtCurrentWidth(*m_lastObject, m_lastObject->isText() ? m_lastObject->length() : 0);
> +            else if (m_lineBreak.renderer()) {
>                  // Don't ever break in the middle of a word if we can help it.
>                  // There's no room at all. We just have to be on this line,
>                  // even though we'll spill out.
> -                m_lineBreak.moveTo(m_current.renderer(), m_current.offset());
> +                commitLineBreakAtCurrentWidth(*m_current.renderer(), m_current.offset());

So the extra m_width.commit() here (by calling commitLineBreakAtCurrentWidth()) does not change behavior?

[Myles C. Maxfield]

Comment on attachment 247631 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=247631&action=review

>> Source/WebCore/rendering/line/BreakingContextInlineHeaders.h:1052
>> +                commitLineBreakAtCurrentWidth(*m_current.renderer(), m_current.offset());
> 
> So the extra m_width.commit() here (by calling commitLineBreakAtCurrentWidth()) does not change behavior?

No - this function only ever gets called at the very end of LineBreaker::nextLineBreak(), where the LineWidth immediately gets destroyed.

[Myles C. Maxfield]

Committed r180944: <http://trac.webkit.org/changeset/180944>