Add utf_string_conversions to base namespace.

chrome/browser/search_engines/template_url_unittest.cc

Index: chrome/browser/search_engines/template_url_unittest.cc
diff --git a/chrome/browser/search_engines/template_url_unittest.cc b/chrome/browser/search_engines/template_url_unittest.cc
index 3273fa5b4cdf170d5f8bb751ec1a43b5adb87405..42a86accb5e50dde2d1fe40f119cb406a778ee58 100644
--- a/chrome/browser/search_engines/template_url_unittest.cc
+++ b/chrome/browser/search_engines/template_url_unittest.cc
@@ -243,13 +243,14 @@ TEST_F(TemplateURLTest, URLRefTermToWide) {
   } to_wide_cases[] = {
     {"hello+world", ASCIIToUTF16("hello world")},
     // Test some big-5 input.
-    {"%a7A%A6%6e+to+you", WideToUTF16(L"\x4f60\x597d to you")},
+    {"%a7A%A6%6e+to+you", base::WideToUTF16(L"\x4f60\x597d to you")},
     // Test some UTF-8 input. We should fall back to this when the encoding
     // doesn't look like big-5. We have a '5' in the middle, which is an invalid
     // Big-5 trailing byte.
-    {"%e4%bd%a05%e5%a5%bd+to+you", WideToUTF16(L"\x4f60\x35\x597d to you")},
+    {"%e4%bd%a05%e5%a5%bd+to+you",
+     base::WideToUTF16(L"\x4f60\x35\x597d to you")},
     // Undecodable input should stay escaped.
-    {"%91%01+abcd", WideToUTF16(L"%91%01 abcd")},
+    {"%91%01+abcd", base::WideToUTF16(L"%91%01 abcd")},
     // Make sure we convert %2B to +.
     {"C%2B%2B", ASCIIToUTF16("C++")},
     // C%2B is escaped as C%252B, make sure we unescape it properly.
@@ -351,7 +352,7 @@ TEST_F(TemplateURLTest, ReplaceArbitrarySearchTerms) {
     const std::string url;
     const std::string expected_result;
   } test_data[] = {
-    { "BIG5",  WideToUTF16(L"\x60BD"),
+    { "BIG5",  base::WideToUTF16(L"\x60BD"),
       "http://foo/?{searchTerms}{inputEncoding}",
       "http://foo/?%B1~BIG5" },
     { "UTF-8", ASCIIToUTF16("blah"),