Expose InlineTextBoxes in the accessibility tree.

LayoutTests/platform/linux/accessibility/inline-text-word-boundaries-expected.txt

Index: LayoutTests/platform/linux/accessibility/inline-text-word-boundaries-expected.txt
diff --git a/LayoutTests/platform/linux/accessibility/inline-text-word-boundaries-expected.txt b/LayoutTests/platform/linux/accessibility/inline-text-word-boundaries-expected.txt
new file mode 100644
index 0000000000000000000000000000000000000000..4aff0299ac7efb3261c65f8353421e9849823624
--- /dev/null
+++ b/LayoutTests/platform/linux/accessibility/inline-text-word-boundaries-expected.txt
@@ -0,0 +1,51 @@
+Ace ten twenty-one. xxxxxxxxxxxxxxxxxxx
+
+Tests that we can compute word boundaries for any character index from the accessibility tree.
+
+On success, you will see a series of "PASS" messages, followed by "TEST COMPLETE".
+
+
+Accessible text: "Ace ten twenty-one. xxxxxxxxxxxxxxxxxxx"
+Character 0: 'A' word=(0, 3): 'Ace'
+Character 1: 'c' word=(0, 3): 'Ace'
+Character 2: 'e' word=(0, 3): 'Ace'
+Character 3: ' ' word=(0, 3): 'Ace'
+Character 4: 't' word=(4, 7): 'ten'
+Character 5: 'e' word=(4, 7): 'ten'
+Character 6: 'n' word=(4, 7): 'ten'
+Character 7: ' ' word=(4, 7): 'ten'
+Character 8: 't' word=(8, 14): 'twenty'
+Character 9: 'w' word=(8, 14): 'twenty'
+Character 10: 'e' word=(8, 14): 'twenty'
+Character 11: 'n' word=(8, 14): 'twenty'
+Character 12: 't' word=(8, 14): 'twenty'
+Character 13: 'y' word=(8, 14): 'twenty'
+Character 14: '-' word=(8, 14): 'twenty'
+Character 15: 'o' word=(15, 18): 'one'
+Character 16: 'n' word=(15, 18): 'one'
+Character 17: 'e' word=(15, 18): 'one'
+Character 18: '.' word=(15, 18): 'one'
+Character 19: ' ' word=(15, 18): 'one'
+Character 20: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 21: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 22: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 23: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 24: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 25: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 26: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 27: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 28: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 29: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 30: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 31: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 32: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 33: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 34: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 35: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 36: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 37: 'x' word=(20, 38): 'xxxxxxxxxxxxxxxxxx'
+Character 38: 'x' word=(38, 39): 'x'

[aboxhall, 2013/09/06 17:19:21]

Why is this x on its own?

[dmazzoni, 2013/09/06 20:25:27]

That's on purpose. The style of the element is set to "width: 8em; word-wrap:
break-word;" and there are enough x's that it wraps to the next line. This tests
that when a word is split between multiple lines it's treated as two separate
words from the perspective of this API - which I think is correct behavior given
how word APIs are used by assistive technology. Otherwise the user wouldn't
realize the word wrapped.

[aboxhall, 2013/09/06 20:51:12]

Ahh got it. How about replacing the x's with "averylongwordthatwillwrap" - or is
the word break detection smart enough to detect the words?

+PASS successfullyParsed is true
+
+TEST COMPLETE
+