Don't populate GlyphBuffer in InlineTextBox::characterWidths

Don't populate GlyphBuffer in InlineTextBox::characterWidths

The current InlineTextBox::characterWidths implementation populates a
GlyphBuffer even though it is never used. Change the implementation to
not pass a GlyphBuffer and thereby avoiding the unnecessary work and
allocation costs.

R=dominik.rottsches@intel.com

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=182843

[eae, 2014-09-27 12:43:37]

eae@chromium.org changed reviewers:
+ dominik.rottsches@intel.com
- eseidel@chromium.org

[eae, 2014-09-27 12:44:00]

Ready for review.

[Dominik Röttsches, 2014-09-29 08:06:55]

https://codereview.chromium.org/608083003/diff/1/Source/core/rendering/Inline...
File Source/core/rendering/InlineTextBox.cpp (right):

https://codereview.chromium.org/608083003/diff/1/Source/core/rendering/Inline...
Source/core/rendering/InlineTextBox.cpp:1244: it.advance(i + 1);
We can do it like this - but alternatively, it looks like we could replace the
characterWidths function to just return the complete advances-array of the glyph
buffer, now that we have the new representation of advances + offsets, couldn't
we? That might be better than doing a copy loop by character index here.

[Dominik Röttsches, 2014-09-29 08:08:47]

What I meant to say: LGTM, but perhaps you could consider the above for further
optimization/code reduction?

[eae, 2014-09-29 13:15:15]

Thanks

https://codereview.chromium.org/608083003/diff/1/Source/core/rendering/Inline...
File Source/core/rendering/InlineTextBox.cpp (right):

https://codereview.chromium.org/608083003/diff/1/Source/core/rendering/Inline...
Source/core/rendering/InlineTextBox.cpp:1244: it.advance(i + 1);
On 2014/09/29 08:06:55, Dominik Röttsches wrote:
> We can do it like this - but alternatively, it looks like we could replace the
> characterWidths function to just return the complete advances-array of the
glyph
> buffer, now that we have the new representation of advances + offsets,
couldn't
> we? That might be better than doing a copy loop by character index here.

Hmm, that might work. I'll look into it.

[eae, 2014-09-29 13:33:00]

The CQ bit was checked by eae@chromium.org

[commit-bot: I haz the power, 2014-09-29 13:33:39]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/608083003/20001

[commit-bot: I haz the power, 2014-09-29 14:58:38]

Committed patchset #2 (id:20001) as 182843