(CANCELED) Improve performance of large option list with CJK text

third_party/WebKit/Source/platform/fonts/Font.h

Index: third_party/WebKit/Source/platform/fonts/Font.h
diff --git a/third_party/WebKit/Source/platform/fonts/Font.h b/third_party/WebKit/Source/platform/fonts/Font.h
index 16924c32eff0fc3da1fa4e4ef27622d5906a471f..19a746219077d0fa7a8d10d6bca8251140349e5d 100644
--- a/third_party/WebKit/Source/platform/fonts/Font.h
+++ b/third_party/WebKit/Source/platform/fonts/Font.h
@@ -84,6 +84,7 @@ public:
     // Glyph bounds will be the minimum rect containing all glyph strokes, in coordinates using
     // (<text run x position>, <baseline position>) as the origin.
     float width(const TextRun&, HashSet<const SimpleFontData*>* fallbackFonts = nullptr, FloatRect* glyphBounds = nullptr) const;
+    float widthForUnbrokenRun(const TextRun&) const;

[drott, 2015/12/16 13:17:39]

This method and its implementation, i.e. new splitting logic, should not be in
Font.h. Rather I think we should have something in CachingWordShaper that allows
as to use the m_singleCharMap of ShapeCache for the CJK case.

If possible, I'd like to avoid "by character" special casing. Could we use
FontDescription.locale() for the distinction of what kind of splitting we do in
the CachingWordShaper?

 
     int offsetForPosition(const TextRun&, float position, bool includePartialGlyphs) const;
     FloatRect selectionRectForText(const TextRun&, const FloatPoint&, int h, int from = 0, int to = -1, bool accountForGlyphBounds = false) const;