Changes are:

* base::IsValidCodepoint() now returns false on non-character code points.
* base::IsStringUTF8() now uses ICU library (removed old Mozilla implementation).
* Removed base::IsStringWideUTF8() (was unused and confusing)
* file_util::ReplaceIllegalCharactersInPath() now treats Unicode replacement character (U+FFFD) as invalid.
* Associated unit tests updated.

BUG=2759
BUG=30662
TEST=base_unittests --gtest_filter=StringUtilTest.IsStringUTF8
TEST=base_unittests --gtest_filter=UTFStringConversionsTest.*
TEST=base_unittests --gtest_filter=FileUtilICUTestReplaceIllegalCharactersInPathTest
Review URL: http://codereview.chromium.org/548017

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@36459 0039d316-1c4b-4281-b951-d872f2087c98

1 files changed, 10 insertions, 10 deletions

diff --git a/base/utf_string_conversions_unittest.cc b/base/utf_string_conversions_unittest.cc
index 6ba0b5b..f68c593 100644
--- a/base/utf_string_conversions_unittest.cc
+++ b/base/utf_string_conversions_unittest.cc
@@ -91,8 +91,8 @@ TEST(UTFStringConversionsTest, ConvertUTF8ToWide) {
   } convert_cases[] = {
     // Regular UTF-8 input.
     {"\xe4\xbd\xa0\xe5\xa5\xbd", L"\x4f60\x597d", true},
-    // Non-character is passed through.
-    {"\xef\xbf\xbfHello", L"\xffffHello", true},
+    // Non-character is rejected.
+    {"\xef\xbf\xbfHello", L"\xfffdHello", false},
     // Truncated UTF-8 sequence.
     {"\xe4\xa0\xe5\xa5\xbd", L"\xfffd\x597d", false},
     // Truncated off the end.
@@ -105,10 +105,10 @@ TEST(UTFStringConversionsTest, ConvertUTF8ToWide) {
     // The result will either be in UTF-16 or UTF-32.
 #if defined(WCHAR_T_IS_UTF16)
     {"A\xF0\x90\x8C\x80z", L"A\xd800\xdf00z", true},
-    {"A\xF4\x8F\xBF\xBEz", L"A\xdbff\xdffez", true},
+    {"A\xF4\x8F\xBF\xBEz", L"A\xfffdz", false},
 #elif defined(WCHAR_T_IS_UTF32)
     {"A\xF0\x90\x8C\x80z", L"A\x10300z", true},
-    {"A\xF4\x8F\xBF\xBEz", L"A\x10fffez", true},
+    {"A\xF4\x8F\xBF\xBEz", L"A\xfffdz", false},
 #endif
   };
 
@@ -148,9 +148,9 @@ TEST(UTFStringConversionsTest, ConvertUTF16ToUTF8) {
     {L"\x4f60\x597d", "\xe4\xbd\xa0\xe5\xa5\xbd", true},
     // Test a non-BMP character.
     {L"\xd800\xdf00", "\xF0\x90\x8C\x80", true},
-    // Non-characters are passed through.
-    {L"\xffffHello", "\xEF\xBF\xBFHello", true},
-    {L"\xdbff\xdffeHello", "\xF4\x8F\xBF\xBEHello", true},
+    // Non-characters are rejected.
+    {L"\xffffHello", "\xef\xbf\xbdHello", false},
+    {L"\xdbff\xdffeHello", "\xef\xbf\xbdHello", false},
     // The first character is a truncated UTF-16 character.
     {L"\xd800\x597d", "\xef\xbf\xbd\xe5\xa5\xbd", false},
     // Truncated at the end.
@@ -180,9 +180,9 @@ TEST(UTFStringConversionsTest, ConvertUTF32ToUTF8) {
     {L"\x4f60\x597d", "\xe4\xbd\xa0\xe5\xa5\xbd", true},
     // Test a non-BMP character.
     {L"A\x10300z", "A\xF0\x90\x8C\x80z", true},
-    // Non-characters are passed through.
-    {L"\xffffHello", "\xEF\xBF\xBFHello", true},
-    {L"\x10fffeHello", "\xF4\x8F\xBF\xBEHello", true},
+    // Non-characters are rejected.
+    {L"\xffffHello", "\xEF\xBF\xBDHello", false},
+    {L"\x10fffeHello", "\xEF\xBF\xBDHello", false},
     // Invalid Unicode code points.
     {L"\xfffffffHello", "\xEF\xBF\xBDHello", false},
     // The first character is a truncated UTF-16 character.