From cf49ba2362de69e4fd45aa1209bf9893ba95888a Mon Sep 17 00:00:00 2001
From: Shengfa <3363396+k7z45@users.noreply.github.com>
Date: Tue, 20 Oct 2020 14:52:30 -0500
Subject: [PATCH] [atoms] Fix getText atom for unicode charater middle of word
(#8736)
* [atoms] Fix getText atom for unicode charater middle of word
The change to getText atom by commit c065ddaeb6c5cf91ebf797c242128604e38b79e1
does not handle case when unicode character is in the middle
of a word, and unicode character will be incorrectly capitalized
(see https://bugs.chromium.org/p/chromedriver/issues/detail?id=3611).
The problem is \b mark the boundary between word character and
unicode character as a boundary. This is fixed by explicitly using
unicode flag and specifying unicode character and unicode symbol.
* [atoms] Add meta tag for charset in test html
Add charset for utf-8 in text_test.html
---
javascript/atoms/dom.js | 2 +-
javascript/atoms/test/text_test.html | 29 ++++++++++++++++++++++++++++
2 files changed, 30 insertions(+), 1 deletion(-)
diff --git a/javascript/atoms/dom.js b/javascript/atoms/dom.js
index 68dbdacc163ee..3e683184e94f1 100644
--- a/javascript/atoms/dom.js
+++ b/javascript/atoms/dom.js
@@ -1175,7 +1175,7 @@ bot.dom.appendVisibleTextLinesFromTextNode_ = function(textNode, lines,
}
if (textTransform == 'capitalize') {
- text = text.replace(/(^|\s|\b)(\S)/g, function() {
+ text = text.replace(/(^|[^\d\p{L}\p{S}])([\p{Ll}|\p{S}])/gu, function() {
return arguments[1] + arguments[2].toUpperCase();
});
} else if (textTransform == 'uppercase') {
diff --git a/javascript/atoms/test/text_test.html b/javascript/atoms/test/text_test.html
index 4698ac5db46db..5dc0d1ccbaa4e 100644
--- a/javascript/atoms/test/text_test.html
+++ b/javascript/atoms/test/text_test.html
@@ -18,6 +18,7 @@
text_test.html
+