OLD | NEW |
1 // Copyright (c) 2006-2008 The Chromium Authors. All rights reserved. | 1 // Copyright (c) 2006-2008 The Chromium Authors. All rights reserved. |
2 // Use of this source code is governed by a BSD-style license that can be | 2 // Use of this source code is governed by a BSD-style license that can be |
3 // found in the LICENSE file. | 3 // found in the LICENSE file. |
4 | 4 |
5 #include "base/basictypes.h" | |
6 #include "base/string_piece.h" | 5 #include "base/string_piece.h" |
7 #include "base/string_util.h" | |
8 #include "base/sys_string_conversions.h" | 6 #include "base/sys_string_conversions.h" |
9 #include "testing/gtest/include/gtest/gtest.h" | 7 #include "testing/gtest/include/gtest/gtest.h" |
10 | 8 |
11 #ifdef WCHAR_T_IS_UTF32 | 9 #ifdef WCHAR_T_IS_UTF32 |
12 static const std::wstring kSysWideOldItalicLetterA = L"\x10300"; | 10 static const std::wstring kSysWideOldItalicLetterA = L"\x10300"; |
13 #else | 11 #else |
14 static const std::wstring kSysWideOldItalicLetterA = L"\xd800\xdf00"; | 12 static const std::wstring kSysWideOldItalicLetterA = L"\xd800\xdf00"; |
15 #endif | 13 #endif |
16 | 14 |
17 TEST(SysStrings, SysWideToUTF8) { | 15 TEST(SysStrings, SysWideToUTF8) { |
(...skipping 43 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
61 std::string utf8_null("a"); | 59 std::string utf8_null("a"); |
62 utf8_null.push_back(0); | 60 utf8_null.push_back(0); |
63 utf8_null.push_back('b'); | 61 utf8_null.push_back('b'); |
64 | 62 |
65 std::wstring expected_null(L"a"); | 63 std::wstring expected_null(L"a"); |
66 expected_null.push_back(0); | 64 expected_null.push_back(0); |
67 expected_null.push_back('b'); | 65 expected_null.push_back('b'); |
68 | 66 |
69 EXPECT_EQ(expected_null, SysUTF8ToWide(utf8_null)); | 67 EXPECT_EQ(expected_null, SysUTF8ToWide(utf8_null)); |
70 } | 68 } |
71 | |
72 // We assume the test is running in a UTF8 locale. | |
73 #if defined(OS_LINUX) | |
74 TEST(SysStrings, SysWideToNativeMB) { | |
75 using base::SysWideToNativeMB; | |
76 EXPECT_EQ("Hello, world", SysWideToNativeMB(L"Hello, world")); | |
77 EXPECT_EQ("\xe4\xbd\xa0\xe5\xa5\xbd", SysWideToNativeMB(L"\x4f60\x597d")); | |
78 | |
79 // >16 bits | |
80 EXPECT_EQ("\xF0\x90\x8C\x80", SysWideToNativeMB(kSysWideOldItalicLetterA)); | |
81 | |
82 // Error case. When Windows finds a UTF-16 character going off the end of | |
83 // a string, it just converts that literal value to UTF-8, even though this | |
84 // is invalid. | |
85 // | |
86 // This is what XP does, but Vista has different behavior, so we don't bother | |
87 // verifying it: | |
88 //EXPECT_EQ("\xE4\xBD\xA0\xED\xA0\x80zyxw", | |
89 // SysWideToNativeMB(L"\x4f60\xd800zyxw")); | |
90 | |
91 // Test embedded NULLs. | |
92 std::wstring wide_null(L"a"); | |
93 wide_null.push_back(0); | |
94 wide_null.push_back('b'); | |
95 | |
96 std::string expected_null("a"); | |
97 expected_null.push_back(0); | |
98 expected_null.push_back('b'); | |
99 | |
100 EXPECT_EQ(expected_null, SysWideToNativeMB(wide_null)); | |
101 } | |
102 | |
103 // We assume the test is running in a UTF8 locale. | |
104 TEST(SysStrings, SysNativeMBToWide) { | |
105 using base::SysNativeMBToWide; | |
106 EXPECT_EQ(L"Hello, world", SysNativeMBToWide("Hello, world")); | |
107 EXPECT_EQ(L"\x4f60\x597d", SysNativeMBToWide("\xe4\xbd\xa0\xe5\xa5\xbd")); | |
108 // >16 bits | |
109 EXPECT_EQ(kSysWideOldItalicLetterA, SysNativeMBToWide("\xF0\x90\x8C\x80")); | |
110 | |
111 // Error case. When Windows finds an invalid UTF-8 character, it just skips | |
112 // it. This seems weird because it's inconsistent with the reverse conversion. | |
113 // | |
114 // This is what XP does, but Vista has different behavior, so we don't bother | |
115 // verifying it: | |
116 //EXPECT_EQ(L"\x4f60zyxw", SysNativeMBToWide("\xe4\xbd\xa0\xe5\xa5zyxw")); | |
117 | |
118 // Test embedded NULLs. | |
119 std::string utf8_null("a"); | |
120 utf8_null.push_back(0); | |
121 utf8_null.push_back('b'); | |
122 | |
123 std::wstring expected_null(L"a"); | |
124 expected_null.push_back(0); | |
125 expected_null.push_back('b'); | |
126 | |
127 EXPECT_EQ(expected_null, SysNativeMBToWide(utf8_null)); | |
128 } | |
129 | |
130 static const wchar_t* const kConvertRoundtripCases[] = { | |
131 L"Google Video", | |
132 // "网页 图片 资讯更多 »" | |
133 L"\x7f51\x9875\x0020\x56fe\x7247\x0020\x8d44\x8baf\x66f4\x591a\x0020\x00bb", | |
134 // "Παγκόσμιος Ιστός" | |
135 L"\x03a0\x03b1\x03b3\x03ba\x03cc\x03c3\x03bc\x03b9" | |
136 L"\x03bf\x03c2\x0020\x0399\x03c3\x03c4\x03cc\x03c2", | |
137 // "Поиск страниц на русском" | |
138 L"\x041f\x043e\x0438\x0441\x043a\x0020\x0441\x0442" | |
139 L"\x0440\x0430\x043d\x0438\x0446\x0020\x043d\x0430" | |
140 L"\x0020\x0440\x0443\x0441\x0441\x043a\x043e\x043c", | |
141 // "전체서비스" | |
142 L"\xc804\xccb4\xc11c\xbe44\xc2a4", | |
143 | |
144 // Test characters that take more than 16 bits. This will depend on whether | |
145 // wchar_t is 16 or 32 bits. | |
146 #if defined(WCHAR_T_IS_UTF16) | |
147 L"\xd800\xdf00", | |
148 // ????? (Mathematical Alphanumeric Symbols (U+011d40 - U+011d44 : A,B,C,D,E) | |
149 L"\xd807\xdd40\xd807\xdd41\xd807\xdd42\xd807\xdd43\xd807\xdd44", | |
150 #elif defined(WCHAR_T_IS_UTF32) | |
151 L"\x10300", | |
152 // ????? (Mathematical Alphanumeric Symbols (U+011d40 - U+011d44 : A,B,C,D,E) | |
153 L"\x11d40\x11d41\x11d42\x11d43\x11d44", | |
154 #endif | |
155 }; | |
156 | |
157 | |
158 TEST(SysStrings, SysNativeMBAndWide) { | |
159 for (size_t i = 0; i < arraysize(kConvertRoundtripCases); ++i) { | |
160 std::wstring wide = kConvertRoundtripCases[i]; | |
161 std::wstring trip = base::SysNativeMBToWide(base::SysWideToNativeMB(wide)); | |
162 EXPECT_EQ(wide.size(), trip.size()); | |
163 EXPECT_EQ(wide, trip); | |
164 } | |
165 | |
166 // We assume our test is running in UTF-8, so double check through ICU. | |
167 for (size_t i = 0; i < arraysize(kConvertRoundtripCases); ++i) { | |
168 std::wstring wide = kConvertRoundtripCases[i]; | |
169 std::wstring trip = base::SysNativeMBToWide(WideToUTF8(wide)); | |
170 EXPECT_EQ(wide.size(), trip.size()); | |
171 EXPECT_EQ(wide, trip); | |
172 } | |
173 | |
174 for (size_t i = 0; i < arraysize(kConvertRoundtripCases); ++i) { | |
175 std::wstring wide = kConvertRoundtripCases[i]; | |
176 std::wstring trip = UTF8ToWide(base::SysWideToNativeMB(wide)); | |
177 EXPECT_EQ(wide.size(), trip.size()); | |
178 EXPECT_EQ(wide, trip); | |
179 } | |
180 } | |
181 #endif // OS_LINUX | |
OLD | NEW |