base/i18n/case_conversion_unittest.cc

   1 // Copyright (c) 2011 The Chromium Authors. All rights reserved.
   2 // Use of this source code is governed by a BSD-style license that can be
   3 // found in the LICENSE file.
   4
   5 #include "base/i18n/case_conversion.h"
   6 #include "base/i18n/rtl.h"
   7 #include "base/strings/utf_string_conversions.h"
   8 #include "testing/gtest/include/gtest/gtest.h"
   9 #include "third_party/icu/source/i18n/unicode/usearch.h"
  10
  11 namespace base {
  12 namespace i18n {
  13
  14 namespace {
  15
  16 const wchar_t kNonASCIIMixed[] =
  17     L"\xC4\xD6\xE4\xF6\x20\xCF\xEF\x20\xF7\x25"
  18     L"\xA4\x23\x2A\x5E\x60\x40\xA3\x24\x2030\x201A\x7E\x20\x1F07\x1F0F"
  19     L"\x20\x1E00\x1E01";
  20 const wchar_t kNonASCIILower[] =
  21     L"\xE4\xF6\xE4\xF6\x20\xEF\xEF"
  22     L"\x20\xF7\x25\xA4\x23\x2A\x5E\x60\x40\xA3\x24\x2030\x201A\x7E\x20\x1F07"
  23     L"\x1F07\x20\x1E01\x1E01";
  24 const wchar_t kNonASCIIUpper[] =
  25     L"\xC4\xD6\xC4\xD6\x20\xCF\xCF"
  26     L"\x20\xF7\x25\xA4\x23\x2A\x5E\x60\x40\xA3\x24\x2030\x201A\x7E\x20\x1F0F"
  27     L"\x1F0F\x20\x1E00\x1E00";
  28
  29 }  // namespace
  30
  31 // Test upper and lower case string conversion.
  32 TEST(CaseConversionTest, UpperLower) {
  33   const string16 mixed(ASCIIToUTF16("Text with UPPer & lowER casE."));
  34   const string16 expected_lower(ASCIIToUTF16("text with upper & lower case."));
  35   const string16 expected_upper(ASCIIToUTF16("TEXT WITH UPPER & LOWER CASE."));
  36
  37   string16 result = ToLower(mixed);
  38   EXPECT_EQ(expected_lower, result);
  39
  40   result = ToUpper(mixed);
  41   EXPECT_EQ(expected_upper, result);
  42 }
  43
  44 TEST(CaseConversionTest, NonASCII) {
  45   const string16 mixed(WideToUTF16(kNonASCIIMixed));
  46   const string16 expected_lower(WideToUTF16(kNonASCIILower));
  47   const string16 expected_upper(WideToUTF16(kNonASCIIUpper));
  48
  49   string16 result = ToLower(mixed);
  50   EXPECT_EQ(expected_lower, result);
  51
  52   result = ToUpper(mixed);
  53   EXPECT_EQ(expected_upper, result);
  54 }
  55
  56 TEST(CaseConversionTest, TurkishLocaleConversion) {
  57   const string16 mixed(WideToUTF16(L"\x49\x131"));
  58   const string16 expected_lower(WideToUTF16(L"\x69\x131"));
  59   const string16 expected_upper(WideToUTF16(L"\x49\x49"));
  60
  61   std::string default_locale(uloc_getDefault());
  62   i18n::SetICUDefaultLocale("en_US");
  63
  64   string16 result = ToLower(mixed);
  65   EXPECT_EQ(expected_lower, result);
  66
  67   result = ToUpper(mixed);
  68   EXPECT_EQ(expected_upper, result);
  69
  70   i18n::SetICUDefaultLocale("tr");
  71
  72   const string16 expected_lower_turkish(WideToUTF16(L"\x131\x131"));
  73   const string16 expected_upper_turkish(WideToUTF16(L"\x49\x49"));
  74
  75   result = ToLower(mixed);
  76   EXPECT_EQ(expected_lower_turkish, result);
  77
  78   result = ToUpper(mixed);
  79   EXPECT_EQ(expected_upper_turkish, result);
  80
  81   SetICUDefaultLocale(default_locale.data());
  82 }
  83
  84 TEST(CaseConversionTest, FoldCase) {
  85   // Simple ASCII, should lower-case.
  86   EXPECT_EQ(ASCIIToUTF16("hello, world"),
  87             FoldCase(ASCIIToUTF16("Hello, World")));
  88
  89   // Non-ASCII cases from above. They should all fold to the same result.
  90   EXPECT_EQ(FoldCase(WideToUTF16(kNonASCIIMixed)),
  91             FoldCase(WideToUTF16(kNonASCIILower)));
  92   EXPECT_EQ(FoldCase(WideToUTF16(kNonASCIIMixed)),
  93             FoldCase(WideToUTF16(kNonASCIIUpper)));
  94
  95   // Turkish cases from above. This is the lower-case expected result from the
  96   // US locale. It should be the same even when the current locale is Turkish.
  97   const string16 turkish(WideToUTF16(L"\x49\x131"));
  98   const string16 turkish_expected(WideToUTF16(L"\x69\x131"));
  99
 100   std::string default_locale(uloc_getDefault());
 101   i18n::SetICUDefaultLocale("en_US");
 102   EXPECT_EQ(turkish_expected, FoldCase(turkish));
 103
 104   i18n::SetICUDefaultLocale("tr");
 105   EXPECT_EQ(turkish_expected, FoldCase(turkish));
 106
 107   // Test a case that gets bigger when processed.
 108   // U+130 = LATIN CAPITAL LETTER I WITH DOT ABOVE gets folded to a lower case
 109   // "i" followed by U+307 COMBINING DOT ABOVE.
 110   EXPECT_EQ(WideToUTF16(L"i\u0307j"), FoldCase(WideToUTF16(L"\u0130j")));
 111
 112   // U+00DF (SHARP S) and U+1E9E (CAPIRAL SHARP S) are both folded to "ss".
 113   EXPECT_EQ(ASCIIToUTF16("ssss"), FoldCase(WideToUTF16(L"\u00DF\u1E9E")));
 114 }
 115
 116 }  // namespace i18n
 117 }  // namespace base
 118
 119
 120