Add functions to get the length or compute a substring of UTF8 and UTF16

encoded strings.

BUG=none
TEST=Added new unit tests for each.

Review URL: http://codereview.chromium.org/3133028

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@57051 0039d316-1c4b-4281-b951-d872f2087c98

1 files changed, 95 insertions, 0 deletions

diff --git a/base/i18n/char_iterator_unittest.cc b/base/i18n/char_iterator_unittest.cc
new file mode 100644
index 0000000..4fe7ebb
--- /dev/null
+++ b/base/i18n/char_iterator_unittest.cc
@@ -0,0 +1,95 @@
+// Copyright (c) 2010 The Chromium Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#include "base/i18n/char_iterator.h"
+
+#include "base/utf_string_conversions.h"
+#include "testing/gtest/include/gtest/gtest.h"
+
+TEST(CharIteratorsTest, TestUTF8) {
+  std::string empty("");
+  base::UTF8CharIterator empty_iter(&empty);
+  ASSERT_TRUE(empty_iter.end());
+  ASSERT_EQ(0, empty_iter.array_pos());
+  ASSERT_EQ(0, empty_iter.char_pos());
+  ASSERT_FALSE(empty_iter.Advance());
+
+  std::string str("s\303\273r");  // [u with circumflex]
+  base::UTF8CharIterator iter(&str);
+  ASSERT_FALSE(iter.end());
+  ASSERT_EQ(0, iter.array_pos());
+  ASSERT_EQ(0, iter.char_pos());
+  ASSERT_EQ('s', iter.get());
+  ASSERT_TRUE(iter.Advance());
+
+  ASSERT_FALSE(iter.end());
+  ASSERT_EQ(1, iter.array_pos());
+  ASSERT_EQ(1, iter.char_pos());
+  ASSERT_EQ(251, iter.get());
+  ASSERT_TRUE(iter.Advance());
+
+  ASSERT_FALSE(iter.end());
+  ASSERT_EQ(3, iter.array_pos());
+  ASSERT_EQ(2, iter.char_pos());
+  ASSERT_EQ('r', iter.get());
+  ASSERT_TRUE(iter.Advance());
+
+  ASSERT_TRUE(iter.end());
+  ASSERT_EQ(4, iter.array_pos());
+  ASSERT_EQ(3, iter.char_pos());
+
+  // Don't care what it returns, but this shouldn't crash
+  iter.get();
+
+  ASSERT_FALSE(iter.Advance());
+}
+
+TEST(CharIteratorsTest, TestUTF16) {
+  string16 empty = UTF8ToUTF16("");
+  base::UTF16CharIterator empty_iter(&empty);
+  ASSERT_TRUE(empty_iter.end());
+  ASSERT_EQ(0, empty_iter.array_pos());
+  ASSERT_EQ(0, empty_iter.char_pos());
+  ASSERT_FALSE(empty_iter.Advance());
+
+  // This test string contains 4 characters:
+  //   x
+  //   u with circumflex - 2 bytes in UTF8, 1 codeword in UTF16
+  //   math double-struck A - 4 bytes in UTF8, 2 codewords in UTF16
+  //   z
+  string16 str = UTF8ToUTF16("x\303\273\360\235\224\270z");
+  base::UTF16CharIterator iter(&str);
+  ASSERT_FALSE(iter.end());
+  ASSERT_EQ(0, iter.array_pos());
+  ASSERT_EQ(0, iter.char_pos());
+  ASSERT_EQ('x', iter.get());
+  ASSERT_TRUE(iter.Advance());
+
+  ASSERT_FALSE(iter.end());
+  ASSERT_EQ(1, iter.array_pos());
+  ASSERT_EQ(1, iter.char_pos());
+  ASSERT_EQ(251, iter.get());
+  ASSERT_TRUE(iter.Advance());
+
+  ASSERT_FALSE(iter.end());
+  ASSERT_EQ(2, iter.array_pos());
+  ASSERT_EQ(2, iter.char_pos());
+  ASSERT_EQ(120120, iter.get());
+  ASSERT_TRUE(iter.Advance());
+
+  ASSERT_FALSE(iter.end());
+  ASSERT_EQ(4, iter.array_pos());
+  ASSERT_EQ(3, iter.char_pos());
+  ASSERT_EQ('z', iter.get());
+  ASSERT_TRUE(iter.Advance());
+
+  ASSERT_TRUE(iter.end());
+  ASSERT_EQ(5, iter.array_pos());
+  ASSERT_EQ(4, iter.char_pos());
+
+  // Don't care what it returns, but this shouldn't crash
+  iter.get();
+
+  ASSERT_FALSE(iter.Advance());
+}