diff options
| author | jschuh@chromium.org <jschuh@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-01-16 16:40:38 +0000 |
|---|---|---|
| committer | jschuh@chromium.org <jschuh@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-01-16 16:40:38 +0000 |
| commit | 548a6c0f30c3dcf374cc06be48f02a06da5c1d19 (patch) | |
| tree | 57a3406e6d51b44774b9fa37cde1810a1d76a47c /base/utf_string_conversion_utils.h | |
| parent | b78e6049b83a6a221b6c7b47c229b337e3abce43 (diff) | |
| download | chromium_src-548a6c0f30c3dcf374cc06be48f02a06da5c1d19.zip chromium_src-548a6c0f30c3dcf374cc06be48f02a06da5c1d19.tar.gz chromium_src-548a6c0f30c3dcf374cc06be48f02a06da5c1d19.tar.bz2 | |
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
Diffstat (limited to 'base/utf_string_conversion_utils.h')
| -rw-r--r-- | base/utf_string_conversion_utils.h | 11 |
1 files changed, 6 insertions, 5 deletions
diff --git a/base/utf_string_conversion_utils.h b/base/utf_string_conversion_utils.h index a8a76c5..3fcb689 100644 --- a/base/utf_string_conversion_utils.h +++ b/base/utf_string_conversion_utils.h @@ -12,11 +12,12 @@ namespace base { inline bool IsValidCodepoint(uint32 code_point) { - // Excludes the surrogate code points ([0xD800, 0xDFFF]) and - // codepoints larger than 0x10FFFF (the highest codepoint allowed). - // Non-characters and unassigned codepoints are allowed. - return code_point < 0xD800u || - (code_point >= 0xE000u && code_point <= 0x10FFFFu); + // Excludes non-characters (U+FDD0..U+FDEF, and all codepoints ending in + // 0xFFFE or 0xFFFF), surrogate code points (U+D800..U+DFFF), and codepoints + // larger than U+10FFFF (the highest codepoint allowed). + return code_point < 0xD800u || (code_point >= 0xE000u && + code_point < 0xFDD0u) || (code_point > 0xFDEFu && + code_point <= 0x10FFFFu && (code_point & 0xFFFEu) != 0xFFFEu); } // ReadUnicodeCharacter -------------------------------------------------------- |