| OLD | NEW |
| (Empty) |
| 1 // Copyright (c) 2012 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 "chrome/renderer/spellchecker/spellcheck.h" | |
| 6 | |
| 7 #include <stddef.h> | |
| 8 | |
| 9 #include <memory> | |
| 10 #include <utility> | |
| 11 | |
| 12 #include "base/files/file_path.h" | |
| 13 #include "base/files/file_util.h" | |
| 14 #include "base/macros.h" | |
| 15 #include "base/message_loop/message_loop.h" | |
| 16 #include "base/path_service.h" | |
| 17 #include "base/run_loop.h" | |
| 18 #include "base/strings/string_number_conversions.h" | |
| 19 #include "base/strings/utf_string_conversions.h" | |
| 20 #include "build/build_config.h" | |
| 21 #include "chrome/renderer/spellchecker/hunspell_engine.h" | |
| 22 #include "chrome/renderer/spellchecker/spellcheck_language.h" | |
| 23 #include "components/spellcheck/common/spellcheck_common.h" | |
| 24 #include "components/spellcheck/common/spellcheck_result.h" | |
| 25 #include "testing/gtest/include/gtest/gtest.h" | |
| 26 #include "third_party/WebKit/public/platform/WebVector.h" | |
| 27 #include "third_party/WebKit/public/web/WebTextCheckingCompletion.h" | |
| 28 #include "third_party/WebKit/public/web/WebTextCheckingResult.h" | |
| 29 | |
| 30 #define TYPOGRAPHICAL_APOSTROPHE L"\x2019" | |
| 31 | |
| 32 namespace { | |
| 33 const int kNoOffset = 0; | |
| 34 const int kNoTag = 0; | |
| 35 | |
| 36 base::FilePath GetHunspellDirectory() { | |
| 37 base::FilePath hunspell_directory; | |
| 38 if (!PathService::Get(base::DIR_SOURCE_ROOT, &hunspell_directory)) | |
| 39 return base::FilePath(); | |
| 40 | |
| 41 hunspell_directory = hunspell_directory.AppendASCII("third_party"); | |
| 42 hunspell_directory = hunspell_directory.AppendASCII("hunspell_dictionaries"); | |
| 43 return hunspell_directory; | |
| 44 } | |
| 45 | |
| 46 } // namespace | |
| 47 | |
| 48 // TODO(groby): This needs to be a BrowserTest for OSX. | |
| 49 class SpellCheckTest : public testing::Test { | |
| 50 public: | |
| 51 SpellCheckTest() { | |
| 52 ReinitializeSpellCheck("en-US"); | |
| 53 } | |
| 54 | |
| 55 void ReinitializeSpellCheck(const std::string& language) { | |
| 56 spell_check_.reset(new SpellCheck()); | |
| 57 InitializeSpellCheck(language); | |
| 58 } | |
| 59 | |
| 60 void UninitializeSpellCheck() { | |
| 61 spell_check_.reset(new SpellCheck()); | |
| 62 } | |
| 63 | |
| 64 bool InitializeIfNeeded() { | |
| 65 return spell_check()->InitializeIfNeeded(); | |
| 66 } | |
| 67 | |
| 68 void InitializeSpellCheck(const std::string& language) { | |
| 69 base::FilePath hunspell_directory = GetHunspellDirectory(); | |
| 70 EXPECT_FALSE(hunspell_directory.empty()); | |
| 71 base::File file( | |
| 72 spellcheck::GetVersionedFileName(language, hunspell_directory), | |
| 73 base::File::FLAG_OPEN | base::File::FLAG_READ); | |
| 74 #if defined(OS_MACOSX) | |
| 75 // TODO(groby): Forcing spellcheck to use hunspell, even on OSX. | |
| 76 // Instead, tests should exercise individual spelling engines. | |
| 77 spell_check_->languages_.push_back(new SpellcheckLanguage()); | |
| 78 spell_check_->languages_.front()->platform_spelling_engine_.reset( | |
| 79 new HunspellEngine); | |
| 80 spell_check_->languages_.front()->Init(std::move(file), language); | |
| 81 #else | |
| 82 spell_check_->AddSpellcheckLanguage(std::move(file), language); | |
| 83 #endif | |
| 84 } | |
| 85 | |
| 86 ~SpellCheckTest() override {} | |
| 87 | |
| 88 SpellCheck* spell_check() { return spell_check_.get(); } | |
| 89 | |
| 90 bool CheckSpelling(const std::string& word, int tag) { | |
| 91 return spell_check_->languages_.front() | |
| 92 ->platform_spelling_engine_->CheckSpelling(base::ASCIIToUTF16(word), | |
| 93 tag); | |
| 94 } | |
| 95 | |
| 96 bool IsValidContraction(const base::string16& word, int tag) { | |
| 97 return spell_check_->languages_.front()->IsValidContraction(word, tag); | |
| 98 } | |
| 99 | |
| 100 static void FillSuggestions( | |
| 101 const std::vector<std::vector<base::string16>>& suggestions_list, | |
| 102 std::vector<base::string16>* optional_suggestions) { | |
| 103 SpellCheck::FillSuggestions(suggestions_list, optional_suggestions); | |
| 104 } | |
| 105 | |
| 106 #if !defined(OS_MACOSX) | |
| 107 protected: | |
| 108 void TestSpellCheckParagraph( | |
| 109 const base::string16& input, | |
| 110 const std::vector<SpellCheckResult>& expected) { | |
| 111 blink::WebVector<blink::WebTextCheckingResult> results; | |
| 112 spell_check()->SpellCheckParagraph(input, | |
| 113 &results); | |
| 114 | |
| 115 EXPECT_EQ(results.size(), expected.size()); | |
| 116 size_t size = std::min(results.size(), expected.size()); | |
| 117 for (size_t j = 0; j < size; ++j) { | |
| 118 EXPECT_EQ(results[j].decoration, blink::WebTextDecorationTypeSpelling); | |
| 119 EXPECT_EQ(results[j].location, expected[j].location); | |
| 120 EXPECT_EQ(results[j].length, expected[j].length); | |
| 121 } | |
| 122 } | |
| 123 #endif | |
| 124 | |
| 125 private: | |
| 126 std::unique_ptr<SpellCheck> spell_check_; | |
| 127 base::MessageLoop loop; | |
| 128 }; | |
| 129 | |
| 130 // A fake completion object for verification. | |
| 131 class MockTextCheckingCompletion : public blink::WebTextCheckingCompletion { | |
| 132 public: | |
| 133 MockTextCheckingCompletion() | |
| 134 : completion_count_(0) { | |
| 135 } | |
| 136 | |
| 137 void didFinishCheckingText( | |
| 138 const blink::WebVector<blink::WebTextCheckingResult>& results) override { | |
| 139 completion_count_++; | |
| 140 last_results_ = results; | |
| 141 } | |
| 142 | |
| 143 void didCancelCheckingText() override { | |
| 144 completion_count_++; | |
| 145 } | |
| 146 | |
| 147 size_t completion_count_; | |
| 148 blink::WebVector<blink::WebTextCheckingResult> last_results_; | |
| 149 }; | |
| 150 | |
| 151 // Operates unit tests for the content::SpellCheck::SpellCheckWord() function | |
| 152 // with the US English dictionary. | |
| 153 // The unit tests in this function consist of: | |
| 154 // * Tests for the function with empty strings; | |
| 155 // * Tests for the function with a valid English word; | |
| 156 // * Tests for the function with a valid non-English word; | |
| 157 // * Tests for the function with a valid English word with a preceding | |
| 158 // space character; | |
| 159 // * Tests for the function with a valid English word with a preceding | |
| 160 // non-English word; | |
| 161 // * Tests for the function with a valid English word with a following | |
| 162 // space character; | |
| 163 // * Tests for the function with a valid English word with a following | |
| 164 // non-English word; | |
| 165 // * Tests for the function with two valid English words concatenated | |
| 166 // with space characters or non-English words; | |
| 167 // * Tests for the function with an invalid English word; | |
| 168 // * Tests for the function with an invalid English word with a preceding | |
| 169 // space character; | |
| 170 // * Tests for the function with an invalid English word with a preceding | |
| 171 // non-English word; | |
| 172 // * Tests for the function with an invalid English word with a following | |
| 173 // space character; | |
| 174 // * Tests for the function with an invalid English word with a following | |
| 175 // non-English word, and; | |
| 176 // * Tests for the function with two invalid English words concatenated | |
| 177 // with space characters or non-English words. | |
| 178 // A test with a "[ROBUSTNESS]" mark shows it is a robustness test and it uses | |
| 179 // grammatically incorrect string. | |
| 180 // TODO(groby): Please feel free to add more tests. | |
| 181 TEST_F(SpellCheckTest, SpellCheckStrings_EN_US) { | |
| 182 static const struct { | |
| 183 // A string to be tested. | |
| 184 const wchar_t* input; | |
| 185 // An expected result for this test case. | |
| 186 // * true: the input string does not have any invalid words. | |
| 187 // * false: the input string has one or more invalid words. | |
| 188 bool expected_result; | |
| 189 // The position and the length of the first invalid word. | |
| 190 int misspelling_start; | |
| 191 int misspelling_length; | |
| 192 } kTestCases[] = { | |
| 193 // Empty strings. | |
| 194 {L"", true}, | |
| 195 {L" ", true}, | |
| 196 {L"\xA0", true}, | |
| 197 {L"\x3000", true}, | |
| 198 | |
| 199 // A valid English word "hello". | |
| 200 {L"hello", true}, | |
| 201 // A valid Chinese word (meaning "hello") consisting of two CJKV | |
| 202 // ideographs | |
| 203 {L"\x4F60\x597D", true}, | |
| 204 // A valid Korean word (meaning "hello") consisting of five hangul | |
| 205 // syllables | |
| 206 {L"\xC548\xB155\xD558\xC138\xC694", true}, | |
| 207 // A valid Japanese word (meaning "hello") consisting of five Hiragana | |
| 208 // letters | |
| 209 {L"\x3053\x3093\x306B\x3061\x306F", true}, | |
| 210 // A valid Hindi word (meaning ?) consisting of six Devanagari letters | |
| 211 // (This word is copied from "http://b/issue?id=857583".) | |
| 212 {L"\x0930\x093E\x091C\x0927\x093E\x0928", true}, | |
| 213 // A valid English word "affix" using a Latin ligature 'ffi' | |
| 214 {L"a\xFB03x", true}, | |
| 215 // A valid English word "hello" (fullwidth version) | |
| 216 {L"\xFF28\xFF45\xFF4C\xFF4C\xFF4F", true}, | |
| 217 // Two valid Greek words (meaning "hello") consisting of seven Greek | |
| 218 // letters | |
| 219 {L"\x03B3\x03B5\x03B9\x03AC" L" " L"\x03C3\x03BF\x03C5", true}, | |
| 220 // A valid Russian word (meaning "hello") consisting of twelve Cyrillic | |
| 221 // letters | |
| 222 {L"\x0437\x0434\x0440\x0430\x0432\x0441" | |
| 223 L"\x0442\x0432\x0443\x0439\x0442\x0435", true}, | |
| 224 // A valid English contraction | |
| 225 {L"isn't", true}, | |
| 226 // A valid English contraction with a typographical apostrophe. | |
| 227 {L"isn" TYPOGRAPHICAL_APOSTROPHE L"t", true}, | |
| 228 // A valid English word enclosed with underscores. | |
| 229 {L"_hello_", true}, | |
| 230 | |
| 231 // A valid English word with a preceding whitespace | |
| 232 {L" " L"hello", true}, | |
| 233 // A valid English word with a preceding no-break space | |
| 234 {L"\xA0" L"hello", true}, | |
| 235 // A valid English word with a preceding ideographic space | |
| 236 {L"\x3000" L"hello", true}, | |
| 237 // A valid English word with a preceding Chinese word | |
| 238 {L"\x4F60\x597D" L"hello", true}, | |
| 239 // [ROBUSTNESS] A valid English word with a preceding Korean word | |
| 240 {L"\xC548\xB155\xD558\xC138\xC694" L"hello", true}, | |
| 241 // A valid English word with a preceding Japanese word | |
| 242 {L"\x3053\x3093\x306B\x3061\x306F" L"hello", true}, | |
| 243 // [ROBUSTNESS] A valid English word with a preceding Hindi word | |
| 244 {L"\x0930\x093E\x091C\x0927\x093E\x0928" L"hello", true}, | |
| 245 // [ROBUSTNESS] A valid English word with two preceding Greek words | |
| 246 {L"\x03B3\x03B5\x03B9\x03AC" L" " L"\x03C3\x03BF\x03C5" | |
| 247 L"hello", true}, | |
| 248 // [ROBUSTNESS] A valid English word with a preceding Russian word | |
| 249 {L"\x0437\x0434\x0440\x0430\x0432\x0441" | |
| 250 L"\x0442\x0432\x0443\x0439\x0442\x0435" L"hello", true}, | |
| 251 | |
| 252 // A valid English word with a following whitespace | |
| 253 {L"hello" L" ", true}, | |
| 254 // A valid English word with a following no-break space | |
| 255 {L"hello" L"\xA0", true}, | |
| 256 // A valid English word with a following ideographic space | |
| 257 {L"hello" L"\x3000", true}, | |
| 258 // A valid English word with a following Chinese word | |
| 259 {L"hello" L"\x4F60\x597D", true}, | |
| 260 // [ROBUSTNESS] A valid English word with a following Korean word | |
| 261 {L"hello" L"\xC548\xB155\xD558\xC138\xC694", true}, | |
| 262 // A valid English word with a following Japanese word | |
| 263 {L"hello" L"\x3053\x3093\x306B\x3061\x306F", true}, | |
| 264 // [ROBUSTNESS] A valid English word with a following Hindi word | |
| 265 {L"hello" L"\x0930\x093E\x091C\x0927\x093E\x0928", true}, | |
| 266 // [ROBUSTNESS] A valid English word with two following Greek words | |
| 267 {L"hello" | |
| 268 L"\x03B3\x03B5\x03B9\x03AC" L" " L"\x03C3\x03BF\x03C5", true}, | |
| 269 // [ROBUSTNESS] A valid English word with a following Russian word | |
| 270 {L"hello" L"\x0437\x0434\x0440\x0430\x0432\x0441" | |
| 271 L"\x0442\x0432\x0443\x0439\x0442\x0435", true}, | |
| 272 | |
| 273 // Two valid English words concatenated with a whitespace | |
| 274 {L"hello" L" " L"hello", true}, | |
| 275 // Two valid English words concatenated with a no-break space | |
| 276 {L"hello" L"\xA0" L"hello", true}, | |
| 277 // Two valid English words concatenated with an ideographic space | |
| 278 {L"hello" L"\x3000" L"hello", true}, | |
| 279 // Two valid English words concatenated with a Chinese word | |
| 280 {L"hello" L"\x4F60\x597D" L"hello", true}, | |
| 281 // [ROBUSTNESS] Two valid English words concatenated with a Korean word | |
| 282 {L"hello" L"\xC548\xB155\xD558\xC138\xC694" L"hello", true}, | |
| 283 // Two valid English words concatenated with a Japanese word | |
| 284 {L"hello" L"\x3053\x3093\x306B\x3061\x306F" L"hello", true}, | |
| 285 // [ROBUSTNESS] Two valid English words concatenated with a Hindi word | |
| 286 {L"hello" L"\x0930\x093E\x091C\x0927\x093E\x0928" L"hello" , true}, | |
| 287 // [ROBUSTNESS] Two valid English words concatenated with two Greek words | |
| 288 {L"hello" L"\x03B3\x03B5\x03B9\x03AC" L" " L"\x03C3\x03BF\x03C5" | |
| 289 L"hello", true}, | |
| 290 // [ROBUSTNESS] Two valid English words concatenated with a Russian word | |
| 291 {L"hello" L"\x0437\x0434\x0440\x0430\x0432\x0441" | |
| 292 L"\x0442\x0432\x0443\x0439\x0442\x0435" L"hello", true}, | |
| 293 // [ROBUSTNESS] Two valid English words concatenated with a contraction | |
| 294 // character. | |
| 295 {L"hello:hello", true}, | |
| 296 | |
| 297 // An invalid English word | |
| 298 {L"ifmmp", false, 0, 5}, | |
| 299 // An invalid English word "bffly" containing a Latin ligature 'ffl' | |
| 300 {L"b\xFB04y", false, 0, 3}, | |
| 301 // An invalid English word "ifmmp" (fullwidth version) | |
| 302 {L"\xFF29\xFF46\xFF4D\xFF4D\xFF50", false, 0, 5}, | |
| 303 // An invalid English contraction | |
| 304 {L"jtm'u", false, 0, 5}, | |
| 305 // An invalid English word enclosed with underscores. | |
| 306 {L"_ifmmp_", false, 1, 5}, | |
| 307 | |
| 308 // An invalid English word with a preceding whitespace | |
| 309 {L" " L"ifmmp", false, 1, 5}, | |
| 310 // An invalid English word with a preceding no-break space | |
| 311 {L"\xA0" L"ifmmp", false, 1, 5}, | |
| 312 // An invalid English word with a preceding ideographic space | |
| 313 {L"\x3000" L"ifmmp", false, 1, 5}, | |
| 314 // An invalid English word with a preceding Chinese word | |
| 315 {L"\x4F60\x597D" L"ifmmp", false, 2, 5}, | |
| 316 // [ROBUSTNESS] An invalid English word with a preceding Korean word | |
| 317 {L"\xC548\xB155\xD558\xC138\xC694" L"ifmmp", false, 5, 5}, | |
| 318 // An invalid English word with a preceding Japanese word | |
| 319 {L"\x3053\x3093\x306B\x3061\x306F" L"ifmmp", false, 5, 5}, | |
| 320 // [ROBUSTNESS] An invalid English word with a preceding Hindi word | |
| 321 {L"\x0930\x093E\x091C\x0927\x093E\x0928" L"ifmmp", false, 6, 5}, | |
| 322 // [ROBUSTNESS] An invalid English word with two preceding Greek words | |
| 323 {L"\x03B3\x03B5\x03B9\x03AC" L" " L"\x03C3\x03BF\x03C5" | |
| 324 L"ifmmp", false, 8, 5}, | |
| 325 // [ROBUSTNESS] An invalid English word with a preceding Russian word | |
| 326 {L"\x0437\x0434\x0440\x0430\x0432\x0441" | |
| 327 L"\x0442\x0432\x0443\x0439\x0442\x0435" L"ifmmp", false, 12, 5}, | |
| 328 | |
| 329 // An invalid English word with a following whitespace | |
| 330 {L"ifmmp" L" ", false, 0, 5}, | |
| 331 // An invalid English word with a following no-break space | |
| 332 {L"ifmmp" L"\xA0", false, 0, 5}, | |
| 333 // An invalid English word with a following ideographic space | |
| 334 {L"ifmmp" L"\x3000", false, 0, 5}, | |
| 335 // An invalid English word with a following Chinese word | |
| 336 {L"ifmmp" L"\x4F60\x597D", false, 0, 5}, | |
| 337 // [ROBUSTNESS] An invalid English word with a following Korean word | |
| 338 {L"ifmmp" L"\xC548\xB155\xD558\xC138\xC694", false, 0, 5}, | |
| 339 // An invalid English word with a following Japanese word | |
| 340 {L"ifmmp" L"\x3053\x3093\x306B\x3061\x306F", false, 0, 5}, | |
| 341 // [ROBUSTNESS] An invalid English word with a following Hindi word | |
| 342 {L"ifmmp" L"\x0930\x093E\x091C\x0927\x093E\x0928", false, 0, 5}, | |
| 343 // [ROBUSTNESS] An invalid English word with two following Greek words | |
| 344 {L"ifmmp" | |
| 345 L"\x03B3\x03B5\x03B9\x03AC" L" " L"\x03C3\x03BF\x03C5", false, 0, 5}, | |
| 346 // [ROBUSTNESS] An invalid English word with a following Russian word | |
| 347 {L"ifmmp" L"\x0437\x0434\x0440\x0430\x0432\x0441" | |
| 348 L"\x0442\x0432\x0443\x0439\x0442\x0435", false, 0, 5}, | |
| 349 | |
| 350 // Two invalid English words concatenated with a whitespace | |
| 351 {L"ifmmp" L" " L"ifmmp", false, 0, 5}, | |
| 352 // Two invalid English words concatenated with a no-break space | |
| 353 {L"ifmmp" L"\xA0" L"ifmmp", false, 0, 5}, | |
| 354 // Two invalid English words concatenated with an ideographic space | |
| 355 {L"ifmmp" L"\x3000" L"ifmmp", false, 0, 5}, | |
| 356 // Two invalid English words concatenated with a Chinese word | |
| 357 {L"ifmmp" L"\x4F60\x597D" L"ifmmp", false, 0, 5}, | |
| 358 // [ROBUSTNESS] Two invalid English words concatenated with a Korean word | |
| 359 {L"ifmmp" L"\xC548\xB155\xD558\xC138\xC694" L"ifmmp", false, 0, 5}, | |
| 360 // Two invalid English words concatenated with a Japanese word | |
| 361 {L"ifmmp" L"\x3053\x3093\x306B\x3061\x306F" L"ifmmp", false, 0, 5}, | |
| 362 // [ROBUSTNESS] Two invalid English words concatenated with a Hindi word | |
| 363 {L"ifmmp" L"\x0930\x093E\x091C\x0927\x093E\x0928" L"ifmmp" , false, 0, 5}, | |
| 364 // [ROBUSTNESS] Two invalid English words concatenated with two Greek words | |
| 365 {L"ifmmp" L"\x03B3\x03B5\x03B9\x03AC" L" " L"\x03C3\x03BF\x03C5" | |
| 366 L"ifmmp", false, 0, 5}, | |
| 367 // [ROBUSTNESS] Two invalid English words concatenated with a Russian word | |
| 368 {L"ifmmp" L"\x0437\x0434\x0440\x0430\x0432\x0441" | |
| 369 L"\x0442\x0432\x0443\x0439\x0442\x0435" L"ifmmp", false, 0, 5}, | |
| 370 // [ROBUSTNESS] Two invalid English words concatenated with a contraction | |
| 371 // character. | |
| 372 {L"ifmmp:ifmmp", false, 0, 11}, | |
| 373 | |
| 374 // [REGRESSION] Issue 13432: "Any word of 13 or 14 characters is not | |
| 375 // spellcheck" <http://crbug.com/13432>. | |
| 376 {L"qwertyuiopasd", false, 0, 13}, | |
| 377 {L"qwertyuiopasdf", false, 0, 14}, | |
| 378 | |
| 379 // [REGRESSION] Issue 128896: "en_US hunspell dictionary includes | |
| 380 // acknowledgement but not acknowledgements" <http://crbug.com/128896> | |
| 381 {L"acknowledgement", true}, | |
| 382 {L"acknowledgements", true}, | |
| 383 | |
| 384 // Issue 123290: "Spellchecker should treat numbers as word characters" | |
| 385 {L"0th", true}, | |
| 386 {L"1st", true}, | |
| 387 {L"2nd", true}, | |
| 388 {L"3rd", true}, | |
| 389 {L"4th", true}, | |
| 390 {L"5th", true}, | |
| 391 {L"6th", true}, | |
| 392 {L"7th", true}, | |
| 393 {L"8th", true}, | |
| 394 {L"9th", true}, | |
| 395 {L"10th", true}, | |
| 396 {L"100th", true}, | |
| 397 {L"1000th", true}, | |
| 398 {L"25", true}, | |
| 399 {L"2012", true}, | |
| 400 {L"100,000,000", true}, | |
| 401 {L"3.141592653", true}, | |
| 402 }; | |
| 403 | |
| 404 for (size_t i = 0; i < arraysize(kTestCases); ++i) { | |
| 405 size_t input_length = 0; | |
| 406 if (kTestCases[i].input != NULL) { | |
| 407 input_length = wcslen(kTestCases[i].input); | |
| 408 } | |
| 409 int misspelling_start; | |
| 410 int misspelling_length; | |
| 411 bool result = spell_check()->SpellCheckWord( | |
| 412 base::WideToUTF16(kTestCases[i].input).c_str(), | |
| 413 kNoOffset, | |
| 414 static_cast<int>(input_length), | |
| 415 kNoTag, | |
| 416 &misspelling_start, | |
| 417 &misspelling_length, NULL); | |
| 418 | |
| 419 EXPECT_EQ(kTestCases[i].expected_result, result); | |
| 420 EXPECT_EQ(kTestCases[i].misspelling_start, misspelling_start); | |
| 421 EXPECT_EQ(kTestCases[i].misspelling_length, misspelling_length); | |
| 422 } | |
| 423 } | |
| 424 | |
| 425 TEST_F(SpellCheckTest, SpellCheckSuggestions_EN_US) { | |
| 426 static const struct { | |
| 427 // A string to be tested. | |
| 428 const wchar_t* input; | |
| 429 // An expected result for this test case. | |
| 430 // * true: the input string does not have any invalid words. | |
| 431 // * false: the input string has one or more invalid words. | |
| 432 bool expected_result; | |
| 433 // The position and the length of the first invalid word. | |
| 434 int misspelling_start; | |
| 435 int misspelling_length; | |
| 436 | |
| 437 // A suggested word that should occur. | |
| 438 const wchar_t* suggested_word; | |
| 439 } kTestCases[] = { | |
| 440 {L"ello", false, 0, 0, L"hello"}, | |
| 441 {L"ello", false, 0, 0, L"cello"}, | |
| 442 {L"wate", false, 0, 0, L"water"}, | |
| 443 {L"wate", false, 0, 0, L"waste"}, | |
| 444 {L"wate", false, 0, 0, L"sate"}, | |
| 445 {L"wate", false, 0, 0, L"ate"}, | |
| 446 {L"jum", false, 0, 0, L"jump"}, | |
| 447 {L"jum", false, 0, 0, L"hum"}, | |
| 448 {L"jum", false, 0, 0, L"sum"}, | |
| 449 {L"jum", false, 0, 0, L"um"}, | |
| 450 {L"alot", false, 0, 0, L"a lot"}, | |
| 451 // A regression test for Issue 36523. | |
| 452 {L"privliged", false, 0, 0, L"privileged"}, | |
| 453 // TODO (Sidchat): add many more examples. | |
| 454 }; | |
| 455 | |
| 456 for (size_t i = 0; i < arraysize(kTestCases); ++i) { | |
| 457 std::vector<base::string16> suggestions; | |
| 458 size_t input_length = 0; | |
| 459 if (kTestCases[i].input != NULL) { | |
| 460 input_length = wcslen(kTestCases[i].input); | |
| 461 } | |
| 462 int misspelling_start; | |
| 463 int misspelling_length; | |
| 464 bool result = spell_check()->SpellCheckWord( | |
| 465 base::WideToUTF16(kTestCases[i].input).c_str(), | |
| 466 kNoOffset, | |
| 467 static_cast<int>(input_length), | |
| 468 kNoTag, | |
| 469 &misspelling_start, | |
| 470 &misspelling_length, | |
| 471 &suggestions); | |
| 472 | |
| 473 // Check for spelling. | |
| 474 EXPECT_EQ(kTestCases[i].expected_result, result); | |
| 475 | |
| 476 // Check if the suggested words occur. | |
| 477 bool suggested_word_is_present = false; | |
| 478 for (int j = 0; j < static_cast<int>(suggestions.size()); j++) { | |
| 479 if (suggestions.at(j).compare( | |
| 480 base::WideToUTF16(kTestCases[i].suggested_word)) == 0) { | |
| 481 suggested_word_is_present = true; | |
| 482 break; | |
| 483 } | |
| 484 } | |
| 485 | |
| 486 EXPECT_TRUE(suggested_word_is_present); | |
| 487 } | |
| 488 } | |
| 489 | |
| 490 // This test verifies our spellchecker can split a text into words and check | |
| 491 // the spelling of each word in the text. | |
| 492 #if defined(THREAD_SANITIZER) | |
| 493 // SpellCheckTest.SpellCheckText fails under ThreadSanitizer v2. | |
| 494 // See http://crbug.com/217909. | |
| 495 #define MAYBE_SpellCheckText DISABLED_SpellCheckText | |
| 496 #else | |
| 497 #define MAYBE_SpellCheckText SpellCheckText | |
| 498 #endif // THREAD_SANITIZER | |
| 499 TEST_F(SpellCheckTest, MAYBE_SpellCheckText) { | |
| 500 static const struct { | |
| 501 const char* language; | |
| 502 const wchar_t* input; | |
| 503 } kTestCases[] = { | |
| 504 { | |
| 505 // Afrikaans | |
| 506 "af-ZA", | |
| 507 L"Google se missie is om die w\x00EAreld se inligting te organiseer en " | |
| 508 L"dit bruikbaar en toeganklik te maak." | |
| 509 }, { | |
| 510 // Bulgarian | |
| 511 "bg-BG", | |
| 512 L"\x041c\x0438\x0441\x0438\x044f\x0442\x0430 " | |
| 513 L"\x043d\x0430 Google \x0435 \x0434\x0430 \x043e" | |
| 514 L"\x0440\x0433\x0430\x043d\x0438\x0437\x0438\x0440" | |
| 515 L"\x0430 \x0441\x0432\x0435\x0442\x043e\x0432" | |
| 516 L"\x043d\x0430\x0442\x0430 \x0438\x043d\x0444" | |
| 517 L"\x043e\x0440\x043c\x0430\x0446\x0438\x044f " | |
| 518 L"\x0438 \x0434\x0430 \x044f \x043d" | |
| 519 L"\x0430\x043f\x0440\x0430\x0432\x0438 \x0443" | |
| 520 L"\x043d\x0438\x0432\x0435\x0440\x0441\x0430\x043b" | |
| 521 L"\x043d\x043e \x0434\x043e\x0441\x0442\x044a" | |
| 522 L"\x043f\x043d\x0430 \x0438 \x043f\x043e" | |
| 523 L"\x043b\x0435\x0437\x043d\x0430." | |
| 524 }, { | |
| 525 // Catalan | |
| 526 "ca-ES", | |
| 527 L"La missi\x00F3 de Google \x00E9s organitzar la informaci\x00F3 " | |
| 528 L"del m\x00F3n i fer que sigui \x00FAtil i accessible universalment." | |
| 529 }, { | |
| 530 // Czech | |
| 531 "cs-CZ", | |
| 532 L"Posl\x00E1n\x00EDm spole\x010Dnosti Google je " | |
| 533 L"uspo\x0159\x00E1\x0064\x0061t informace z cel\x00E9ho sv\x011Bta " | |
| 534 L"tak, aby byly v\x0161\x0065obecn\x011B p\x0159\x00EDstupn\x00E9 " | |
| 535 L"a u\x017Eite\x010Dn\x00E9." | |
| 536 }, { | |
| 537 // Danish | |
| 538 "da-DK", | |
| 539 L"Googles " | |
| 540 L"mission er at organisere verdens information og g\x00F8re den " | |
| 541 L"almindeligt tilg\x00E6ngelig og nyttig." | |
| 542 }, { | |
| 543 // German | |
| 544 "de-DE", | |
| 545 L"Das Ziel von Google besteht darin, die auf der Welt vorhandenen " | |
| 546 L"Informationen zu organisieren und allgemein zug\x00E4nglich und " | |
| 547 L"nutzbar zu machen." | |
| 548 }, { | |
| 549 // Greek | |
| 550 "el-GR", | |
| 551 L"\x0391\x03C0\x03BF\x03C3\x03C4\x03BF\x03BB\x03AE " | |
| 552 L"\x03C4\x03B7\x03C2 Google \x03B5\x03AF\x03BD\x03B1\x03B9 " | |
| 553 L"\x03BD\x03B1 \x03BF\x03C1\x03B3\x03B1\x03BD\x03CE\x03BD\x03B5\x03B9 " | |
| 554 L"\x03C4\x03B9\x03C2 " | |
| 555 L"\x03C0\x03BB\x03B7\x03C1\x03BF\x03C6\x03BF\x03C1\x03AF\x03B5\x03C2 " | |
| 556 L"\x03C4\x03BF\x03C5 \x03BA\x03CC\x03C3\x03BC\x03BF\x03C5 " | |
| 557 L"\x03BA\x03B1\x03B9 \x03BD\x03B1 \x03C4\x03B9\x03C2 " | |
| 558 L"\x03BA\x03B1\x03B8\x03B9\x03C3\x03C4\x03AC " | |
| 559 L"\x03C0\x03C1\x03BF\x03C3\x03B2\x03AC\x03C3\x03B9\x03BC\x03B5\x03C2 " | |
| 560 L"\x03BA\x03B1\x03B9 \x03C7\x03C1\x03AE\x03C3\x03B9\x03BC\x03B5\x03C2." | |
| 561 }, { | |
| 562 // English (Australia) | |
| 563 "en-AU", | |
| 564 L"Google's mission is to organise the world's information and make it " | |
| 565 L"universally accessible and useful." | |
| 566 }, { | |
| 567 // English (Canada) | |
| 568 "en-CA", | |
| 569 L"Google's mission is to organize the world's information and make it " | |
| 570 L"universally accessible and useful." | |
| 571 }, { | |
| 572 // English (United Kingdom) | |
| 573 "en-GB", | |
| 574 L"Google's mission is to organise the world's information and make it " | |
| 575 L"universally accessible and useful." | |
| 576 }, { | |
| 577 // English (United States) | |
| 578 "en-US", | |
| 579 L"Google's mission is to organize the world's information and make it " | |
| 580 L"universally accessible and useful." | |
| 581 }, { | |
| 582 // Spanish | |
| 583 "es-ES", | |
| 584 L"La misi\x00F3n de " | |
| 585 // L"Google" - to be added. | |
| 586 L" es organizar la informaci\x00F3n mundial " | |
| 587 L"para que resulte universalmente accesible y \x00FAtil." | |
| 588 }, { | |
| 589 // Estonian | |
| 590 "et-EE", | |
| 591 // L"Google'ile " - to be added. | |
| 592 L"\x00FClesanne on korraldada maailma teavet ja teeb selle " | |
| 593 L"k\x00F5igile k\x00E4ttesaadavaks ja kasulikuks.", | |
| 594 }, { | |
| 595 // Persian | |
| 596 "fa", | |
| 597 L"\x0686\x0647 \x0637\x0648\x0631 \x0622\x06cc\x0627 \x0634\x0645\x0627 " | |
| 598 L"\x0627\x06cc\x0631\x0627\x0646\x06cc \x0647\x0633\x062a\x06cc\x062f" | |
| 599 }, { | |
| 600 // Faroese | |
| 601 "fo-FO", | |
| 602 L"Google er at samskipa alla vitan \x00ED heiminum og gera hana alment " | |
| 603 L"atkomiliga og n\x00FDtiliga." | |
| 604 }, { | |
| 605 // French | |
| 606 "fr-FR", | |
| 607 L"Google a pour mission d'organiser les informations \x00E0 " | |
| 608 L"l'\x00E9\x0063helle mondiale dans le but de les rendre accessibles " | |
| 609 L"et utiles \x00E0 tous." | |
| 610 }, { | |
| 611 // Hebrew | |
| 612 "he-IL", | |
| 613 L"\x05D4\x05DE\x05E9\x05D9\x05DE\x05D4 \x05E9\x05DC Google " | |
| 614 L"\x05D4\x05D9\x05D0 \x05DC\x05D0\x05E8\x05D2\x05DF " | |
| 615 L"\x05D0\x05EA \x05D4\x05DE\x05D9\x05D3\x05E2 " | |
| 616 L"\x05D4\x05E2\x05D5\x05DC\x05DE\x05D9 " | |
| 617 L"\x05D5\x05DC\x05D4\x05E4\x05D5\x05DA \x05D0\x05D5\x05EA\x05D5 " | |
| 618 L"\x05DC\x05D6\x05DE\x05D9\x05DF " | |
| 619 L"\x05D5\x05E9\x05D9\x05DE\x05D5\x05E9\x05D9 \x05D1\x05DB\x05DC " | |
| 620 L"\x05D4\x05E2\x05D5\x05DC\x05DD. " | |
| 621 // Two words with ASCII double/single quoation marks. | |
| 622 L"\x05DE\x05E0\x05DB\x0022\x05DC \x05E6\x0027\x05D9\x05E4\x05E1" | |
| 623 }, { | |
| 624 // Hindi | |
| 625 "hi-IN", | |
| 626 L"Google \x0915\x093E \x092E\x093F\x0936\x0928 " | |
| 627 L"\x0926\x0941\x0928\x093F\x092F\x093E \x0915\x0940 " | |
| 628 L"\x091C\x093E\x0928\x0915\x093E\x0930\x0940 \x0915\x094B " | |
| 629 L"\x0935\x094D\x092F\x0935\x0938\x094D\x0925\x093F\x0924 " | |
| 630 L"\x0915\x0930\x0928\x093E \x0914\x0930 \x0909\x0938\x0947 " | |
| 631 L"\x0938\x093E\x0930\x094D\x0935\x092D\x094C\x092E\x093F\x0915 " | |
| 632 L"\x0930\x0942\x092A \x0938\x0947 \x092A\x0939\x0941\x0901\x091A " | |
| 633 L"\x092E\x0947\x0902 \x0914\x0930 \x0909\x092A\x092F\x094B\x0917\x0940 " | |
| 634 L"\x092C\x0928\x093E\x0928\x093E \x0939\x0948." | |
| 635 }, { | |
| 636 // Hungarian | |
| 637 "hu-HU", | |
| 638 L"A Google azt a k\x00FCldet\x00E9st v\x00E1llalta mag\x00E1ra, " | |
| 639 L"hogy a vil\x00E1gon fellelhet\x0151 inform\x00E1\x0063i\x00F3kat " | |
| 640 L"rendszerezze \x00E9s \x00E1ltal\x00E1nosan el\x00E9rhet\x0151v\x00E9, " | |
| 641 L"illetve haszn\x00E1lhat\x00F3v\x00E1 tegye." | |
| 642 }, { | |
| 643 // Croatian | |
| 644 "hr-HR", | |
| 645 // L"Googleova " - to be added. | |
| 646 L"je misija organizirati svjetske informacije i u\x010Diniti ih " | |
| 647 // L"univerzalno " - to be added. | |
| 648 L"pristupa\x010Dnima i korisnima." | |
| 649 }, { | |
| 650 // Indonesian | |
| 651 "id-ID", | |
| 652 L"Misi Google adalah untuk mengelola informasi dunia dan membuatnya " | |
| 653 L"dapat diakses dan bermanfaat secara universal." | |
| 654 }, { | |
| 655 // Italian | |
| 656 "it-IT", | |
| 657 L"La missione di Google \x00E8 organizzare le informazioni a livello " | |
| 658 L"mondiale e renderle universalmente accessibili e fruibili." | |
| 659 }, { | |
| 660 // Lithuanian | |
| 661 "lt-LT", | |
| 662 L"\x201EGoogle\x201C tikslas \x2013 rinkti ir sisteminti pasaulio " | |
| 663 L"informacij\x0105 bei padaryti j\x0105 prieinam\x0105 ir " | |
| 664 L"nauding\x0105 visiems." | |
| 665 }, { | |
| 666 // Latvian | |
| 667 "lv-LV", | |
| 668 L"Google uzdevums ir k\x0101rtot pasaules inform\x0101" | |
| 669 L"ciju un padar\x012Bt to univers\x0101li pieejamu un noder\x012Bgu." | |
| 670 }, { | |
| 671 // Norwegian | |
| 672 "nb-NO", | |
| 673 // L"Googles " - to be added. | |
| 674 L"m\x00E5l er \x00E5 organisere informasjonen i verden og " | |
| 675 L"gj\x00F8re den tilgjengelig og nyttig for alle." | |
| 676 }, { | |
| 677 // Dutch | |
| 678 "nl-NL", | |
| 679 L"Het doel van Google is om alle informatie wereldwijd toegankelijk " | |
| 680 L"en bruikbaar te maken." | |
| 681 }, { | |
| 682 // Polish | |
| 683 "pl-PL", | |
| 684 L"Misj\x0105 Google jest uporz\x0105" L"dkowanie \x015Bwiatowych " | |
| 685 L"zasob\x00F3w informacji, aby sta\x0142y si\x0119 one powszechnie " | |
| 686 L"dost\x0119pne i u\x017Cyteczne." | |
| 687 }, { | |
| 688 // Portuguese (Brazil) | |
| 689 "pt-BR", | |
| 690 L"A miss\x00E3o do " | |
| 691 #if !defined(OS_MACOSX) | |
| 692 L"Google " | |
| 693 #endif | |
| 694 L"\x00E9 organizar as informa\x00E7\x00F5" | |
| 695 L"es do mundo todo e " | |
| 696 #if !defined(OS_MACOSX) | |
| 697 L"torn\x00E1-las " | |
| 698 #endif | |
| 699 L"acess\x00EDveis e \x00FAteis em car\x00E1ter universal." | |
| 700 }, { | |
| 701 // Portuguese (Portugal) | |
| 702 "pt-PT", | |
| 703 L"O " | |
| 704 #if !defined(OS_MACOSX) | |
| 705 L"Google " | |
| 706 #endif | |
| 707 L"tem por miss\x00E3o organizar a informa\x00E7\x00E3o do " | |
| 708 L"mundo e " | |
| 709 #if !defined(OS_MACOSX) | |
| 710 L"torn\x00E1-la " | |
| 711 #endif | |
| 712 L"universalmente acess\x00EDvel e \x00FAtil" | |
| 713 }, { | |
| 714 // Romanian | |
| 715 "ro-RO", | |
| 716 L"Misiunea Google este de a organiza informa\x021B3iile lumii \x0219i de " | |
| 717 L"a le face accesibile \x0219i utile la nivel universal." | |
| 718 }, { | |
| 719 // Russian | |
| 720 "ru-RU", | |
| 721 L"\x041C\x0438\x0441\x0441\x0438\x044F Google " | |
| 722 L"\x0441\x043E\x0441\x0442\x043E\x0438\x0442 \x0432 " | |
| 723 L"\x043E\x0440\x0433\x0430\x043D\x0438\x0437\x0430\x0446\x0438\x0438 " | |
| 724 L"\x043C\x0438\x0440\x043E\x0432\x043E\x0439 " | |
| 725 L"\x0438\x043D\x0444\x043E\x0440\x043C\x0430\x0446\x0438\x0438, " | |
| 726 L"\x043E\x0431\x0435\x0441\x043F\x0435\x0447\x0435\x043D\x0438\x0438 " | |
| 727 L"\x0435\x0435 " | |
| 728 L"\x0434\x043E\x0441\x0442\x0443\x043F\x043D\x043E\x0441\x0442\x0438 " | |
| 729 L"\x0438 \x043F\x043E\x043B\x044C\x0437\x044B \x0434\x043B\x044F " | |
| 730 L"\x0432\x0441\x0435\x0445." | |
| 731 // A Russian word including U+0451. (Bug 15558 <http://crbug.com/15558>) | |
| 732 L"\x0451\x043B\x043A\x0430" | |
| 733 }, { | |
| 734 // Serbo-Croatian (Serbian Latin) | |
| 735 "sh", | |
| 736 L"Google-ova misija je da organizuje sve informacije na svetu i " | |
| 737 L"u\x010dini ih univerzal-no dostupnim i korisnim." | |
| 738 }, { | |
| 739 // Serbian | |
| 740 "sr", | |
| 741 L"\x0047\x006f\x006f\x0067\x006c\x0065\x002d\x043e\x0432\x0430 " | |
| 742 L"\x043c\x0438\x0441\x0438\x0458\x0430 \x0458\x0435 \x0434\x0430 " | |
| 743 L"\x043e\x0440\x0433\x0430\x043d\x0438\x0437\x0443\x0458\x0435 " | |
| 744 L"\x0441\x0432\x0435 " | |
| 745 L"\x0438\x043d\x0444\x043e\x0440\x043c\x0430\x0446\x0438\x0458\x0435 " | |
| 746 L"\x043d\x0430 \x0441\x0432\x0435\x0442\x0443 \x0438 " | |
| 747 L"\x0443\x0447\x0438\x043d\x0438 \x0438\x0445 " | |
| 748 L"\x0443\x043d\x0438\x0432\x0435\x0440\x0437\x0430\x043b\x043d\x043e " | |
| 749 L"\x0434\x043e\x0441\x0442\x0443\x043f\x043d\x0438\x043c \x0438 " | |
| 750 L"\x043a\x043e\x0440\x0438\x0441\x043d\x0438\x043c." | |
| 751 }, { | |
| 752 // Slovak | |
| 753 "sk-SK", | |
| 754 L"Spolo\x010Dnos\x0165 Google si dala za \x00FAlohu usporiada\x0165 " | |
| 755 L"inform\x00E1\x0063ie " | |
| 756 L"z cel\x00E9ho sveta a zabezpe\x010Di\x0165, " | |
| 757 L"aby boli v\x0161eobecne dostupn\x00E9 a u\x017Eito\x010Dn\x00E9." | |
| 758 }, { | |
| 759 // Slovenian | |
| 760 "sl-SI", | |
| 761 // L"Googlovo " - to be added. | |
| 762 L"poslanstvo je organizirati svetovne informacije in " | |
| 763 L"omogo\x010Diti njihovo dostopnost in s tem uporabnost za vse." | |
| 764 }, { | |
| 765 // Swedish | |
| 766 "sv-SE", | |
| 767 L"Googles m\x00E5ls\x00E4ttning \x00E4r att ordna v\x00E4rldens " | |
| 768 L"samlade information och g\x00F6ra den tillg\x00E4nglig f\x00F6r alla." | |
| 769 }, { | |
| 770 // Turkish | |
| 771 "tr-TR", | |
| 772 // L"Google\x2019\x0131n " - to be added. | |
| 773 L"misyonu, d\x00FCnyadaki t\x00FCm bilgileri " | |
| 774 L"organize etmek ve evrensel olarak eri\x015Filebilir ve " | |
| 775 L"kullan\x0131\x015Fl\x0131 k\x0131lmakt\x0131r." | |
| 776 }, { | |
| 777 // Ukranian | |
| 778 "uk-UA", | |
| 779 L"\x041c\x0456\x0441\x0456\x044f " | |
| 780 L"\x043a\x043e\x043c\x043f\x0430\x043d\x0456\x0457 Google " | |
| 781 L"\x043f\x043e\x043b\x044f\x0433\x0430\x0454 \x0432 " | |
| 782 L"\x0442\x043e\x043c\x0443, \x0449\x043e\x0431 " | |
| 783 L"\x0443\x043f\x043e\x0440\x044f\x0434\x043a\x0443\x0432\x0430\x0442" | |
| 784 L"\x0438 \x0456\x043d\x0444\x043e\x0440\x043c\x0430\x0446\x0456\x044e " | |
| 785 L"\x0437 \x0443\x0441\x044c\x043e\x0433\x043e " | |
| 786 L"\x0441\x0432\x0456\x0442\x0443 \x0442\x0430 " | |
| 787 L"\x0437\x0440\x043e\x0431\x0438\x0442\x0438 \x0457\x0457 " | |
| 788 L"\x0443\x043d\x0456\x0432\x0435\x0440\x0441\x0430\x043b\x044c\x043d" | |
| 789 L"\x043e \x0434\x043e\x0441\x0442\x0443\x043f\x043d\x043e\x044e " | |
| 790 L"\x0442\x0430 \x043a\x043e\x0440\x0438\x0441\x043d\x043e\x044e." | |
| 791 }, { | |
| 792 // Vietnamese | |
| 793 "vi-VN", | |
| 794 L"Nhi\x1EC7m v\x1EE5 c\x1EE7\x0061 " | |
| 795 L"Google la \x0111\x1EC3 t\x1ED5 ch\x1EE9\x0063 " | |
| 796 L"c\x00E1\x0063 th\x00F4ng tin c\x1EE7\x0061 " | |
| 797 L"th\x1EBF gi\x1EDBi va l\x00E0m cho n\x00F3 universal c\x00F3 " | |
| 798 L"th\x1EC3 truy c\x1EADp va h\x1EEFu d\x1EE5ng h\x01A1n." | |
| 799 }, { | |
| 800 // Korean | |
| 801 "ko", | |
| 802 L"Google\xC758 \xBAA9\xD45C\xB294 \xC804\xC138\xACC4\xC758 " | |
| 803 L"\xC815\xBCF4\xB97C \xCCB4\xACC4\xD654\xD558\xC5EC \xBAA8\xB450\xAC00 " | |
| 804 L"\xD3B8\xB9AC\xD558\xAC8C \xC774\xC6A9\xD560 \xC218 " | |
| 805 L"\xC788\xB3C4\xB85D \xD558\xB294 \xAC83\xC785\xB2C8\xB2E4." | |
| 806 }, { | |
| 807 // Albanian | |
| 808 "sq", | |
| 809 L"Misioni i Google \x00EBsht\x00EB q\x00EB t\x00EB organizoj\x00EB " | |
| 810 L"informacionin e bot\x00EBs dhe t\x00EB b\x00EBjn\x00EB at\x00EB " | |
| 811 L"universalisht t\x00EB arritshme dhe t\x00EB dobishme." | |
| 812 }, { | |
| 813 // Tamil | |
| 814 "ta", | |
| 815 L"Google \x0B87\x0BA9\x0BCD " | |
| 816 L"\x0BA8\x0BC7\x0BBE\x0B95\x0BCD\x0B95\x0BAE\x0BCD " | |
| 817 L"\x0B89\x0BB2\x0B95\x0BBF\x0BA9\x0BCD \x0BA4\x0B95\x0BB5\x0BB2\x0BCD " | |
| 818 L"\x0B8F\x0BB1\x0BCD\x0BAA\x0BBE\x0B9F\x0BC1 \x0B87\x0BA4\x0BC1 " | |
| 819 L"\x0B89\x0BB2\x0B95\x0BB3\x0BBE\x0BB5\x0BBF\x0BAF " | |
| 820 L"\x0B85\x0BA3\x0BC1\x0B95\x0B95\x0BCD \x0B95\x0BC2\x0B9F\x0BBF\x0BAF " | |
| 821 L"\x0BAE\x0BB1\x0BCD\x0BB1\x0BC1\x0BAE\x0BCD " | |
| 822 L"\x0BAA\x0BAF\x0BA9\x0BC1\x0BB3\x0BCD\x0BB3 " | |
| 823 L"\x0B9A\x0BC6\x0BAF\x0BCD\x0BAF \x0B89\x0BB3\x0BCD\x0BB3\x0BA4\x0BC1." | |
| 824 }, { | |
| 825 // Tajik | |
| 826 "tg", | |
| 827 L"\x041c\x0438\x0441\x0441\x0438\x044f\x0438 Google \x0438\x043d " | |
| 828 L"\x043c\x0443\x0440\x0430\x0442\x0442\x0430\x0431 " | |
| 829 L"\x0441\x043e\x0445\x0442\x0430\x043d\x0438 " | |
| 830 L"\x043c\x0430\x044a\x043b\x0443\x043c\x043e\x0442\x04b3\x043e\x0438 " | |
| 831 L"\x043c\x0430\x0432\x04b7\x0443\x0434\x0430, \x043e\x0441\x043e\x043d " | |
| 832 L"\x043d\x0430\x043c\x0443\x0434\x0430\x043d\x0438 " | |
| 833 L"\x0438\x0441\x0442\x0438\x0444\x043e\x0434\x0430\x0431\x0430\x0440" | |
| 834 L"\x04e3 \x0432\x0430 \x0434\x0430\x0441\x0442\x0440\x0430\x0441\x0438 " | |
| 835 L"\x0443\x043c\x0443\x043c " | |
| 836 L"\x0433\x0430\x0440\x0434\x043e\x043d\x0438\x0434\x0430\x043d\x0438 " | |
| 837 L"\x043e\x043d\x04b3\x043e \x0430\x0441\x0442." | |
| 838 }, | |
| 839 }; | |
| 840 | |
| 841 for (size_t i = 0; i < arraysize(kTestCases); ++i) { | |
| 842 ReinitializeSpellCheck(kTestCases[i].language); | |
| 843 size_t input_length = 0; | |
| 844 if (kTestCases[i].input != NULL) | |
| 845 input_length = wcslen(kTestCases[i].input); | |
| 846 | |
| 847 int misspelling_start = 0; | |
| 848 int misspelling_length = 0; | |
| 849 bool result = spell_check()->SpellCheckWord( | |
| 850 base::WideToUTF16(kTestCases[i].input).c_str(), | |
| 851 kNoOffset, | |
| 852 static_cast<int>(input_length), | |
| 853 kNoTag, | |
| 854 &misspelling_start, | |
| 855 &misspelling_length, NULL); | |
| 856 | |
| 857 EXPECT_TRUE(result) | |
| 858 << "\"" | |
| 859 << std::wstring(kTestCases[i].input).substr( | |
| 860 misspelling_start, misspelling_length) | |
| 861 << "\" is misspelled in " | |
| 862 << kTestCases[i].language | |
| 863 << "."; | |
| 864 EXPECT_EQ(0, misspelling_start); | |
| 865 EXPECT_EQ(0, misspelling_length); | |
| 866 } | |
| 867 } | |
| 868 | |
| 869 // Verify that our SpellCheck::SpellCheckWord() returns false when it checks | |
| 870 // misspelled words. | |
| 871 TEST_F(SpellCheckTest, MisspelledWords) { | |
| 872 static const struct { | |
| 873 const char* language; | |
| 874 const wchar_t* input; | |
| 875 } kTestCases[] = { | |
| 876 { | |
| 877 // A misspelled word for English | |
| 878 "en-US", | |
| 879 L"aaaaaaaaaa", | |
| 880 }, { | |
| 881 // A misspelled word for Greek. | |
| 882 "el-GR", | |
| 883 L"\x03B1\x03B1\x03B1\x03B1\x03B1\x03B1\x03B1\x03B1\x03B1\x03B1", | |
| 884 }, { | |
| 885 // A misspelled word for Persian. | |
| 886 "fa", | |
| 887 L"\x06cc\x06a9\x06cc\x0634\x0627\x0646", | |
| 888 }, { | |
| 889 // A misspelled word for Hebrew | |
| 890 "he-IL", | |
| 891 L"\x05D0\x05D0\x05D0\x05D0\x05D0\x05D0\x05D0\x05D0\x05D0\x05D0", | |
| 892 }, { | |
| 893 // Hindi | |
| 894 "hi-IN", | |
| 895 L"\x0905\x0905\x0905\x0905\x0905\x0905\x0905\x0905\x0905\x0905", | |
| 896 }, { | |
| 897 // A misspelled word for Russian | |
| 898 "ru-RU", | |
| 899 L"\x0430\x0430\x0430\x0430\x0430\x0430\x0430\x0430\x0430\x0430", | |
| 900 }, | |
| 901 }; | |
| 902 | |
| 903 for (size_t i = 0; i < arraysize(kTestCases); ++i) { | |
| 904 ReinitializeSpellCheck(kTestCases[i].language); | |
| 905 | |
| 906 base::string16 word(base::WideToUTF16(kTestCases[i].input)); | |
| 907 int word_length = static_cast<int>(word.length()); | |
| 908 int misspelling_start = 0; | |
| 909 int misspelling_length = 0; | |
| 910 bool result = spell_check()->SpellCheckWord(word.c_str(), | |
| 911 kNoOffset, | |
| 912 word_length, | |
| 913 kNoTag, | |
| 914 &misspelling_start, | |
| 915 &misspelling_length, | |
| 916 NULL); | |
| 917 EXPECT_FALSE(result); | |
| 918 EXPECT_EQ(0, misspelling_start); | |
| 919 EXPECT_EQ(word_length, misspelling_length); | |
| 920 } | |
| 921 } | |
| 922 | |
| 923 // Since SpellCheck::SpellCheckParagraph is not implemented on Mac, | |
| 924 // we skip these SpellCheckParagraph tests on Mac. | |
| 925 #if !defined(OS_MACOSX) | |
| 926 | |
| 927 // Make sure SpellCheckParagraph does not crash if the input is empty. | |
| 928 TEST_F(SpellCheckTest, SpellCheckParagraphEmptyParagraph) { | |
| 929 std::vector<SpellCheckResult> expected; | |
| 930 TestSpellCheckParagraph(base::UTF8ToUTF16(""), expected); | |
| 931 } | |
| 932 | |
| 933 // A simple test case having no misspellings. | |
| 934 TEST_F(SpellCheckTest, SpellCheckParagraphNoMisspellings) { | |
| 935 const base::string16 text = base::UTF8ToUTF16("apple"); | |
| 936 std::vector<SpellCheckResult> expected; | |
| 937 TestSpellCheckParagraph(text, expected); | |
| 938 } | |
| 939 | |
| 940 // A simple test case having one misspelling. | |
| 941 TEST_F(SpellCheckTest, SpellCheckParagraphSingleMisspellings) { | |
| 942 const base::string16 text = base::UTF8ToUTF16("zz"); | |
| 943 std::vector<SpellCheckResult> expected; | |
| 944 expected.push_back(SpellCheckResult( | |
| 945 SpellCheckResult::SPELLING, 0, 2)); | |
| 946 | |
| 947 TestSpellCheckParagraph(text, expected); | |
| 948 } | |
| 949 | |
| 950 // A simple test case having multiple misspellings. | |
| 951 TEST_F(SpellCheckTest, SpellCheckParagraphMultipleMisspellings) { | |
| 952 const base::string16 text = base::UTF8ToUTF16("zz, zz"); | |
| 953 std::vector<SpellCheckResult> expected; | |
| 954 expected.push_back(SpellCheckResult( | |
| 955 SpellCheckResult::SPELLING, 0, 2)); | |
| 956 expected.push_back(SpellCheckResult( | |
| 957 SpellCheckResult::SPELLING, 4, 2)); | |
| 958 | |
| 959 TestSpellCheckParagraph(text, expected); | |
| 960 } | |
| 961 | |
| 962 // Make sure a relatively long (correct) sentence can be spellchecked. | |
| 963 TEST_F(SpellCheckTest, SpellCheckParagraphLongSentence) { | |
| 964 std::vector<SpellCheckResult> expected; | |
| 965 // The text is taken from US constitution preamble. | |
| 966 const base::string16 text = base::UTF8ToUTF16( | |
| 967 "We the people of the United States, in order to form a more perfect " | |
| 968 "union, establish justice, insure domestic tranquility, provide for " | |
| 969 "the common defense, promote the general welfare, and secure the " | |
| 970 "blessings of liberty to ourselves and our posterity, do ordain and " | |
| 971 "establish this Constitution for the United States of America."); | |
| 972 | |
| 973 TestSpellCheckParagraph(text, expected); | |
| 974 } | |
| 975 | |
| 976 // Make sure all misspellings can be found in a relatively long sentence. | |
| 977 TEST_F(SpellCheckTest, SpellCheckParagraphLongSentenceMultipleMisspellings) { | |
| 978 std::vector<SpellCheckResult> expected; | |
| 979 | |
| 980 // All 'the' are converted to 'hte' in US consitition preamble. | |
| 981 const base::string16 text = base::UTF8ToUTF16( | |
| 982 "We hte people of hte United States, in order to form a more perfect " | |
| 983 "union, establish justice, insure domestic tranquility, provide for " | |
| 984 "hte common defense, promote hte general welfare, and secure hte " | |
| 985 "blessings of liberty to ourselves and our posterity, do ordain and " | |
| 986 "establish this Constitution for hte United States of America."); | |
| 987 | |
| 988 expected.push_back(SpellCheckResult( | |
| 989 SpellCheckResult::SPELLING, 3, 3)); | |
| 990 expected.push_back(SpellCheckResult( | |
| 991 SpellCheckResult::SPELLING, 17, 3)); | |
| 992 expected.push_back(SpellCheckResult( | |
| 993 SpellCheckResult::SPELLING, 135, 3)); | |
| 994 expected.push_back(SpellCheckResult( | |
| 995 SpellCheckResult::SPELLING, 163, 3)); | |
| 996 expected.push_back(SpellCheckResult( | |
| 997 SpellCheckResult::SPELLING, 195, 3)); | |
| 998 expected.push_back(SpellCheckResult( | |
| 999 SpellCheckResult::SPELLING, 298, 3)); | |
| 1000 | |
| 1001 TestSpellCheckParagraph(text, expected); | |
| 1002 } | |
| 1003 | |
| 1004 // We also skip RequestSpellCheck tests on Mac, because a system spellchecker | |
| 1005 // is used on Mac instead of SpellCheck::RequestTextChecking. | |
| 1006 | |
| 1007 // Make sure RequestTextChecking does not crash if input is empty. | |
| 1008 TEST_F(SpellCheckTest, RequestSpellCheckWithEmptyString) { | |
| 1009 MockTextCheckingCompletion completion; | |
| 1010 | |
| 1011 spell_check()->RequestTextChecking(base::string16(), &completion); | |
| 1012 | |
| 1013 base::RunLoop().RunUntilIdle(); | |
| 1014 | |
| 1015 EXPECT_EQ(completion.completion_count_, 1U); | |
| 1016 } | |
| 1017 | |
| 1018 // A simple test case having no misspellings. | |
| 1019 TEST_F(SpellCheckTest, RequestSpellCheckWithoutMisspelling) { | |
| 1020 MockTextCheckingCompletion completion; | |
| 1021 | |
| 1022 const base::string16 text = base::ASCIIToUTF16("hello"); | |
| 1023 spell_check()->RequestTextChecking(text, &completion); | |
| 1024 | |
| 1025 base::RunLoop().RunUntilIdle(); | |
| 1026 | |
| 1027 EXPECT_EQ(completion.completion_count_, 1U); | |
| 1028 } | |
| 1029 | |
| 1030 // A simple test case having one misspelling. | |
| 1031 TEST_F(SpellCheckTest, RequestSpellCheckWithSingleMisspelling) { | |
| 1032 MockTextCheckingCompletion completion; | |
| 1033 | |
| 1034 const base::string16 text = base::ASCIIToUTF16("apple, zz"); | |
| 1035 spell_check()->RequestTextChecking(text, &completion); | |
| 1036 | |
| 1037 base::RunLoop().RunUntilIdle(); | |
| 1038 | |
| 1039 EXPECT_EQ(completion.completion_count_, 1U); | |
| 1040 EXPECT_EQ(completion.last_results_.size(), 1U); | |
| 1041 EXPECT_EQ(completion.last_results_[0].location, 7); | |
| 1042 EXPECT_EQ(completion.last_results_[0].length, 2); | |
| 1043 } | |
| 1044 | |
| 1045 // A simple test case having a few misspellings. | |
| 1046 TEST_F(SpellCheckTest, RequestSpellCheckWithMisspellings) { | |
| 1047 MockTextCheckingCompletion completion; | |
| 1048 | |
| 1049 const base::string16 text = base::ASCIIToUTF16("apple, zz, orange, zz"); | |
| 1050 spell_check()->RequestTextChecking(text, &completion); | |
| 1051 | |
| 1052 base::RunLoop().RunUntilIdle(); | |
| 1053 | |
| 1054 EXPECT_EQ(completion.completion_count_, 1U); | |
| 1055 EXPECT_EQ(completion.last_results_.size(), 2U); | |
| 1056 EXPECT_EQ(completion.last_results_[0].location, 7); | |
| 1057 EXPECT_EQ(completion.last_results_[0].length, 2); | |
| 1058 EXPECT_EQ(completion.last_results_[1].location, 19); | |
| 1059 EXPECT_EQ(completion.last_results_[1].length, 2); | |
| 1060 } | |
| 1061 | |
| 1062 // A test case that multiple requests comes at once. Make sure all | |
| 1063 // requests are processed. | |
| 1064 TEST_F(SpellCheckTest, RequestSpellCheckWithMultipleRequests) { | |
| 1065 MockTextCheckingCompletion completion[3]; | |
| 1066 | |
| 1067 const base::string16 text[3] = { | |
| 1068 base::ASCIIToUTF16("what, zz"), | |
| 1069 base::ASCIIToUTF16("apple, zz"), | |
| 1070 base::ASCIIToUTF16("orange, zz") | |
| 1071 }; | |
| 1072 | |
| 1073 for (int i = 0; i < 3; ++i) | |
| 1074 spell_check()->RequestTextChecking(text[i], &completion[i]); | |
| 1075 | |
| 1076 base::RunLoop().RunUntilIdle(); | |
| 1077 | |
| 1078 for (int i = 0; i < 3; ++i) { | |
| 1079 EXPECT_EQ(completion[i].completion_count_, 1U); | |
| 1080 EXPECT_EQ(completion[i].last_results_.size(), 1U); | |
| 1081 EXPECT_EQ(completion[i].last_results_[0].location, 6 + i); | |
| 1082 EXPECT_EQ(completion[i].last_results_[0].length, 2); | |
| 1083 } | |
| 1084 } | |
| 1085 | |
| 1086 // A test case that spellchecking is requested before initializing. | |
| 1087 // In this case, we postpone to post a request. | |
| 1088 TEST_F(SpellCheckTest, RequestSpellCheckWithoutInitialization) { | |
| 1089 UninitializeSpellCheck(); | |
| 1090 | |
| 1091 MockTextCheckingCompletion completion; | |
| 1092 const base::string16 text = base::ASCIIToUTF16("zz"); | |
| 1093 | |
| 1094 spell_check()->RequestTextChecking(text, &completion); | |
| 1095 | |
| 1096 // The task will not be posted yet. | |
| 1097 base::RunLoop().RunUntilIdle(); | |
| 1098 EXPECT_EQ(completion.completion_count_, 0U); | |
| 1099 } | |
| 1100 | |
| 1101 // Requests several spellchecking before initializing. Except the last one, | |
| 1102 // posting requests is cancelled and text is rendered as correct one. | |
| 1103 TEST_F(SpellCheckTest, RequestSpellCheckMultipleTimesWithoutInitialization) { | |
| 1104 UninitializeSpellCheck(); | |
| 1105 | |
| 1106 MockTextCheckingCompletion completion[3]; | |
| 1107 const base::string16 text[3] = { | |
| 1108 base::ASCIIToUTF16("what, zz"), | |
| 1109 base::ASCIIToUTF16("apple, zz"), | |
| 1110 base::ASCIIToUTF16("orange, zz") | |
| 1111 }; | |
| 1112 | |
| 1113 // Calls RequestTextchecking a few times. | |
| 1114 for (int i = 0; i < 3; ++i) | |
| 1115 spell_check()->RequestTextChecking(text[i], &completion[i]); | |
| 1116 | |
| 1117 // The last task will be posted after initialization, however the other | |
| 1118 // requests should be pressed without spellchecking. | |
| 1119 base::RunLoop().RunUntilIdle(); | |
| 1120 for (int i = 0; i < 2; ++i) | |
| 1121 EXPECT_EQ(completion[i].completion_count_, 1U); | |
| 1122 EXPECT_EQ(completion[2].completion_count_, 0U); | |
| 1123 | |
| 1124 // Checks the last request is processed after initialization. | |
| 1125 InitializeSpellCheck("en-US"); | |
| 1126 | |
| 1127 // Calls PostDelayedSpellCheckTask instead of OnInit here for simplicity. | |
| 1128 spell_check()->PostDelayedSpellCheckTask( | |
| 1129 spell_check()->pending_request_param_.release()); | |
| 1130 base::RunLoop().RunUntilIdle(); | |
| 1131 for (int i = 0; i < 3; ++i) | |
| 1132 EXPECT_EQ(completion[i].completion_count_, 1U); | |
| 1133 } | |
| 1134 | |
| 1135 #endif | |
| 1136 | |
| 1137 // Verify that the SpellCheck class keeps the spelling marker added to a | |
| 1138 // misspelled word "zz". | |
| 1139 TEST_F(SpellCheckTest, CreateTextCheckingResultsKeepsMarkers) { | |
| 1140 base::string16 text = base::ASCIIToUTF16("zz"); | |
| 1141 std::vector<SpellCheckResult> spellcheck_results; | |
| 1142 spellcheck_results.push_back( | |
| 1143 SpellCheckResult(SpellCheckResult::SPELLING, 0, 2, base::string16())); | |
| 1144 blink::WebVector<blink::WebTextCheckingResult> textcheck_results; | |
| 1145 spell_check()->CreateTextCheckingResults(SpellCheck::USE_NATIVE_CHECKER, 0, | |
| 1146 text, spellcheck_results, | |
| 1147 &textcheck_results); | |
| 1148 ASSERT_EQ(spellcheck_results.size(), textcheck_results.size()); | |
| 1149 EXPECT_EQ(blink::WebTextDecorationTypeSpelling, | |
| 1150 textcheck_results[0].decoration); | |
| 1151 EXPECT_EQ(spellcheck_results[0].location, textcheck_results[0].location); | |
| 1152 EXPECT_EQ(spellcheck_results[0].length, textcheck_results[0].length); | |
| 1153 } | |
| 1154 | |
| 1155 // Verify that the SpellCheck class replaces the spelling marker added to a | |
| 1156 // contextually-misspelled word "bean" with a grammar marker. | |
| 1157 TEST_F(SpellCheckTest, CreateTextCheckingResultsAddsGrammarMarkers) { | |
| 1158 base::string16 text = base::ASCIIToUTF16("I have bean to USA."); | |
| 1159 std::vector<SpellCheckResult> spellcheck_results; | |
| 1160 spellcheck_results.push_back( | |
| 1161 SpellCheckResult(SpellCheckResult::SPELLING, 7, 4, base::string16())); | |
| 1162 blink::WebVector<blink::WebTextCheckingResult> textcheck_results; | |
| 1163 spell_check()->CreateTextCheckingResults(SpellCheck::USE_NATIVE_CHECKER, 0, | |
| 1164 text, spellcheck_results, | |
| 1165 &textcheck_results); | |
| 1166 ASSERT_EQ(spellcheck_results.size(), textcheck_results.size()); | |
| 1167 EXPECT_EQ(blink::WebTextDecorationTypeGrammar, | |
| 1168 textcheck_results[0].decoration); | |
| 1169 EXPECT_EQ(spellcheck_results[0].location, textcheck_results[0].location); | |
| 1170 EXPECT_EQ(spellcheck_results[0].length, textcheck_results[0].length); | |
| 1171 } | |
| 1172 | |
| 1173 // Verify that the SpellCheck preserves the original apostrophe type in the | |
| 1174 // checked text, regardless of the type of apostrophe the browser returns. | |
| 1175 TEST_F(SpellCheckTest, CreateTextCheckingResultsKeepsTypographicalApostrophe) { | |
| 1176 base::string16 text = base::WideToUTF16( | |
| 1177 L"Ik've havn" TYPOGRAPHICAL_APOSTROPHE L"t ni'n" | |
| 1178 TYPOGRAPHICAL_APOSTROPHE L"out-s I've I" TYPOGRAPHICAL_APOSTROPHE | |
| 1179 L"ve"); | |
| 1180 std::vector<SpellCheckResult> spellcheck_results; | |
| 1181 | |
| 1182 // All typewriter apostrophe results. | |
| 1183 spellcheck_results.push_back(SpellCheckResult(SpellCheckResult::SPELLING, 0, | |
| 1184 5, base::UTF8ToUTF16("I've"))); | |
| 1185 spellcheck_results.push_back(SpellCheckResult( | |
| 1186 SpellCheckResult::SPELLING, 6, 6, base::UTF8ToUTF16("haven't"))); | |
| 1187 spellcheck_results.push_back(SpellCheckResult( | |
| 1188 SpellCheckResult::SPELLING, 13, 10, base::UTF8ToUTF16("in'n'out's"))); | |
| 1189 | |
| 1190 // Replacements that differ only by apostrophe type should be ignored. | |
| 1191 spellcheck_results.push_back(SpellCheckResult(SpellCheckResult::SPELLING, 24, | |
| 1192 4, base::UTF8ToUTF16("I've"))); | |
| 1193 spellcheck_results.push_back(SpellCheckResult(SpellCheckResult::SPELLING, 29, | |
| 1194 4, base::UTF8ToUTF16("I've"))); | |
| 1195 | |
| 1196 // All typographical apostrophe results. | |
| 1197 spellcheck_results.push_back( | |
| 1198 SpellCheckResult(SpellCheckResult::SPELLING, 0, 5, | |
| 1199 base::WideToUTF16(L"I" TYPOGRAPHICAL_APOSTROPHE L"ve"))); | |
| 1200 spellcheck_results.push_back(SpellCheckResult( | |
| 1201 SpellCheckResult::SPELLING, 6, 6, | |
| 1202 base::WideToUTF16(L"haven" TYPOGRAPHICAL_APOSTROPHE L"t"))); | |
| 1203 spellcheck_results.push_back(SpellCheckResult( | |
| 1204 SpellCheckResult::SPELLING, 13, 10, base::WideToUTF16( | |
| 1205 L"in" TYPOGRAPHICAL_APOSTROPHE L"n" TYPOGRAPHICAL_APOSTROPHE L"out" | |
| 1206 TYPOGRAPHICAL_APOSTROPHE L"s"))); | |
| 1207 | |
| 1208 // Replacements that differ only by apostrophe type should be ignored. | |
| 1209 spellcheck_results.push_back( | |
| 1210 SpellCheckResult(SpellCheckResult::SPELLING, 24, 4, | |
| 1211 base::WideToUTF16(L"I" TYPOGRAPHICAL_APOSTROPHE L"ve"))); | |
| 1212 spellcheck_results.push_back( | |
| 1213 SpellCheckResult(SpellCheckResult::SPELLING, 29, 4, | |
| 1214 base::WideToUTF16(L"I" TYPOGRAPHICAL_APOSTROPHE L"ve"))); | |
| 1215 | |
| 1216 blink::WebVector<blink::WebTextCheckingResult> textcheck_results; | |
| 1217 spell_check()->CreateTextCheckingResults(SpellCheck::USE_NATIVE_CHECKER, 0, | |
| 1218 text, spellcheck_results, | |
| 1219 &textcheck_results); | |
| 1220 | |
| 1221 static const wchar_t* kExpectedReplacements[] = { | |
| 1222 L"I've", | |
| 1223 L"haven" TYPOGRAPHICAL_APOSTROPHE L"t", | |
| 1224 L"in'n" TYPOGRAPHICAL_APOSTROPHE L"out's", | |
| 1225 L"I've", | |
| 1226 L"haven" TYPOGRAPHICAL_APOSTROPHE L"t", | |
| 1227 L"in'n" TYPOGRAPHICAL_APOSTROPHE L"out" TYPOGRAPHICAL_APOSTROPHE L"s", | |
| 1228 }; | |
| 1229 | |
| 1230 ASSERT_EQ(arraysize(kExpectedReplacements), textcheck_results.size()); | |
| 1231 for (size_t i = 0; i < arraysize(kExpectedReplacements); ++i) { | |
| 1232 EXPECT_EQ(base::WideToUTF16(kExpectedReplacements[i]), | |
| 1233 textcheck_results[i].replacement) | |
| 1234 << "i=" << i << "\nactual: \"" | |
| 1235 << base::string16(textcheck_results[i].replacement) << "\""; | |
| 1236 } | |
| 1237 } | |
| 1238 | |
| 1239 // Checks some words that should be present in all English dictionaries. | |
| 1240 TEST_F(SpellCheckTest, EnglishWords) { | |
| 1241 static const struct { | |
| 1242 const char* input; | |
| 1243 bool should_pass; | |
| 1244 } kTestCases[] = { | |
| 1245 // Issue 146093: "Chromebook" and "Chromebox" not included in spell-checking | |
| 1246 // dictionary. | |
| 1247 {"Chromebook", true}, | |
| 1248 {"Chromebooks", true}, | |
| 1249 {"Chromebox", true}, | |
| 1250 {"Chromeboxes", true}, | |
| 1251 {"Chromeblade", true}, | |
| 1252 {"Chromeblades", true}, | |
| 1253 {"Chromebase", true}, | |
| 1254 {"Chromebases", true}, | |
| 1255 // Issue 94708: Spell-checker incorrectly reports whisky as misspelled. | |
| 1256 {"whisky", true}, | |
| 1257 {"whiskey", true}, | |
| 1258 {"whiskies", true}, | |
| 1259 // Issue 98678: "Recency" should be included in client-side dictionary. | |
| 1260 {"recency", true}, | |
| 1261 {"recencies", false}, | |
| 1262 // Issue 140486 | |
| 1263 {"movie", true}, | |
| 1264 {"movies", true}, | |
| 1265 }; | |
| 1266 | |
| 1267 static const char* const kLocales[] = { "en-GB", "en-US", "en-CA", "en-AU" }; | |
| 1268 | |
| 1269 for (size_t j = 0; j < arraysize(kLocales); ++j) { | |
| 1270 ReinitializeSpellCheck(kLocales[j]); | |
| 1271 for (size_t i = 0; i < arraysize(kTestCases); ++i) { | |
| 1272 size_t input_length = 0; | |
| 1273 if (kTestCases[i].input != NULL) | |
| 1274 input_length = strlen(kTestCases[i].input); | |
| 1275 | |
| 1276 int misspelling_start = 0; | |
| 1277 int misspelling_length = 0; | |
| 1278 bool result = spell_check()->SpellCheckWord( | |
| 1279 base::ASCIIToUTF16(kTestCases[i].input).c_str(), | |
| 1280 kNoOffset, | |
| 1281 static_cast<int>(input_length), | |
| 1282 kNoTag, | |
| 1283 &misspelling_start, | |
| 1284 &misspelling_length, NULL); | |
| 1285 | |
| 1286 EXPECT_EQ(kTestCases[i].should_pass, result) << kTestCases[i].input << | |
| 1287 " in " << kLocales[j]; | |
| 1288 } | |
| 1289 } | |
| 1290 } | |
| 1291 | |
| 1292 // Checks that NOSUGGEST works in English dictionaries. | |
| 1293 TEST_F(SpellCheckTest, NoSuggest) { | |
| 1294 static const struct { | |
| 1295 const char* input; | |
| 1296 const char* suggestion; | |
| 1297 const char* locale; | |
| 1298 bool should_pass; | |
| 1299 } kTestCases[] = { | |
| 1300 {"suckerbert", "cocksucker", "en-GB", true}, | |
| 1301 {"suckerbert", "cocksucker", "en-US", true}, | |
| 1302 {"suckerbert", "cocksucker", "en-CA", true}, | |
| 1303 {"suckerbert", "cocksucker", "en-AU", true}, | |
| 1304 {"suckerbert", "cocksuckers", "en-GB", true}, | |
| 1305 {"suckerbert", "cocksuckers", "en-US", true}, | |
| 1306 {"suckerbert", "cocksuckers", "en-CA", true}, | |
| 1307 {"suckerbert", "cocksuckers", "en-AU", true}, | |
| 1308 {"Batasunaa", "Batasuna", "ca-ES", true}, | |
| 1309 {"pornoo", "porno", "it-IT", true}, | |
| 1310 {"catass", "catas", "lt-LT", true}, | |
| 1311 {"kuracc", "kurac", "sl-SI", true}, | |
| 1312 {"pittt", "pitt", "sv-SE", true}, | |
| 1313 }; | |
| 1314 | |
| 1315 size_t test_cases_size = arraysize(kTestCases); | |
| 1316 for (size_t i = 0; i < test_cases_size; ++i) { | |
| 1317 ReinitializeSpellCheck(kTestCases[i].locale); | |
| 1318 size_t suggestion_length = 0; | |
| 1319 if (kTestCases[i].suggestion != NULL) | |
| 1320 suggestion_length = strlen(kTestCases[i].suggestion); | |
| 1321 | |
| 1322 // First check that the NOSUGGEST flag didn't mark this word as not being in | |
| 1323 // the dictionary. | |
| 1324 int misspelling_start = 0; | |
| 1325 int misspelling_length = 0; | |
| 1326 bool result = spell_check()->SpellCheckWord( | |
| 1327 base::ASCIIToUTF16(kTestCases[i].suggestion).c_str(), | |
| 1328 kNoOffset, | |
| 1329 static_cast<int>(suggestion_length), | |
| 1330 kNoTag, | |
| 1331 &misspelling_start, | |
| 1332 &misspelling_length, NULL); | |
| 1333 | |
| 1334 EXPECT_EQ(kTestCases[i].should_pass, result) << kTestCases[i].suggestion << | |
| 1335 " in " << kTestCases[i].locale; | |
| 1336 | |
| 1337 // Now verify that this test case does not show up as a suggestion. | |
| 1338 std::vector<base::string16> suggestions; | |
| 1339 size_t input_length = 0; | |
| 1340 if (kTestCases[i].input != NULL) | |
| 1341 input_length = strlen(kTestCases[i].input); | |
| 1342 result = spell_check()->SpellCheckWord( | |
| 1343 base::ASCIIToUTF16(kTestCases[i].input).c_str(), | |
| 1344 kNoOffset, | |
| 1345 static_cast<int>(input_length), | |
| 1346 kNoTag, | |
| 1347 &misspelling_start, | |
| 1348 &misspelling_length, | |
| 1349 &suggestions); | |
| 1350 // Input word should be a misspelling. | |
| 1351 EXPECT_FALSE(result) << kTestCases[i].input | |
| 1352 << " is not a misspelling in " | |
| 1353 << kTestCases[i].locale; | |
| 1354 // Check if the suggested words occur. | |
| 1355 for (int j = 0; j < static_cast<int>(suggestions.size()); j++) { | |
| 1356 for (size_t t = 0; t < test_cases_size; t++) { | |
| 1357 int compare_result = suggestions.at(j).compare( | |
| 1358 base::ASCIIToUTF16(kTestCases[t].suggestion)); | |
| 1359 EXPECT_FALSE(compare_result == 0) << kTestCases[t].suggestion << | |
| 1360 " in " << kTestCases[i].locale; | |
| 1361 } | |
| 1362 } | |
| 1363 } | |
| 1364 } | |
| 1365 | |
| 1366 // Check that the correct dictionary files are checked in. | |
| 1367 TEST_F(SpellCheckTest, DictionaryFiles) { | |
| 1368 std::vector<std::string> spellcheck_languages; | |
| 1369 spellcheck::SpellCheckLanguages(&spellcheck_languages); | |
| 1370 EXPECT_FALSE(spellcheck_languages.empty()); | |
| 1371 | |
| 1372 base::FilePath hunspell = GetHunspellDirectory(); | |
| 1373 for (size_t i = 0; i < spellcheck_languages.size(); ++i) { | |
| 1374 base::FilePath dict = | |
| 1375 spellcheck::GetVersionedFileName(spellcheck_languages[i], hunspell); | |
| 1376 EXPECT_TRUE(base::PathExists(dict)) << dict.value() << " not found"; | |
| 1377 } | |
| 1378 } | |
| 1379 | |
| 1380 // TODO(groby): Add a test for hunspell itself, when MAXWORDLEN is exceeded. | |
| 1381 TEST_F(SpellCheckTest, SpellingEngine_CheckSpelling) { | |
| 1382 static const struct { | |
| 1383 const char* word; | |
| 1384 bool expected_result; | |
| 1385 } kTestCases[] = { | |
| 1386 { "", true }, | |
| 1387 { "automatic", true }, | |
| 1388 { "hello", true }, | |
| 1389 { "forglobantic", false }, | |
| 1390 { "xfdssfsdfaasds", false }, | |
| 1391 { // 64 chars are the longest word to check - this should fail checking. | |
| 1392 "abcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijkl", | |
| 1393 false | |
| 1394 }, | |
| 1395 { // Any word longer than 64 chars should be exempt from checking. | |
| 1396 "reallylongwordthatabsolutelyexceedsthespecifiedcharacterlimitabit", | |
| 1397 true | |
| 1398 } | |
| 1399 }; | |
| 1400 | |
| 1401 // Initialization magic - call InitializeIfNeeded twice. The first one simply | |
| 1402 // flags internal state that a dictionary was requested. The second one will | |
| 1403 // take the passed-in file and initialize hunspell with it. (The file was | |
| 1404 // passed to hunspell in the ctor for the test fixture). | |
| 1405 // This needs to be done since we need to ensure the SpellingEngine objects | |
| 1406 // contained in the SpellcheckLanguages in |languages_| from the test fixture | |
| 1407 // does get initialized. | |
| 1408 // TODO(groby): Clean up this mess. | |
| 1409 InitializeIfNeeded(); | |
| 1410 ASSERT_FALSE(InitializeIfNeeded()); | |
| 1411 | |
| 1412 for (size_t i = 0; i < arraysize(kTestCases); ++i) { | |
| 1413 bool result = CheckSpelling(kTestCases[i].word, 0); | |
| 1414 EXPECT_EQ(kTestCases[i].expected_result, result) << | |
| 1415 "Failed test for " << kTestCases[i].word; | |
| 1416 } | |
| 1417 } | |
| 1418 | |
| 1419 // Chrome should not suggest "Othello" for "hellllo" or "identically" for | |
| 1420 // "accidently". | |
| 1421 TEST_F(SpellCheckTest, LogicalSuggestions) { | |
| 1422 static const struct { | |
| 1423 const char* misspelled; | |
| 1424 const char* suggestion; | |
| 1425 } kTestCases[] = { | |
| 1426 { "hellllo", "hello" }, | |
| 1427 { "accidently", "accidentally" } | |
| 1428 }; | |
| 1429 | |
| 1430 for (size_t i = 0; i < arraysize(kTestCases); ++i) { | |
| 1431 int misspelling_start = 0; | |
| 1432 int misspelling_length = 0; | |
| 1433 std::vector<base::string16> suggestions; | |
| 1434 EXPECT_FALSE(spell_check()->SpellCheckWord( | |
| 1435 base::ASCIIToUTF16(kTestCases[i].misspelled).c_str(), | |
| 1436 kNoOffset, | |
| 1437 strlen(kTestCases[i].misspelled), | |
| 1438 kNoTag, | |
| 1439 &misspelling_start, | |
| 1440 &misspelling_length, | |
| 1441 &suggestions)); | |
| 1442 EXPECT_GE(suggestions.size(), static_cast<size_t>(1)); | |
| 1443 if (suggestions.size() > 0) | |
| 1444 EXPECT_EQ(suggestions[0], base::ASCIIToUTF16(kTestCases[i].suggestion)); | |
| 1445 } | |
| 1446 } | |
| 1447 | |
| 1448 // Words with apostrophes should be valid contractions. | |
| 1449 TEST_F(SpellCheckTest, IsValidContraction) { | |
| 1450 static const char* kLanguages[] = { | |
| 1451 "en-AU", | |
| 1452 "en-CA", | |
| 1453 "en-GB", | |
| 1454 "en-US", | |
| 1455 }; | |
| 1456 | |
| 1457 static const wchar_t* kWords[] = { | |
| 1458 L"in'n'out", | |
| 1459 L"in" TYPOGRAPHICAL_APOSTROPHE L"n" TYPOGRAPHICAL_APOSTROPHE L"out", | |
| 1460 }; | |
| 1461 | |
| 1462 for (size_t i = 0; i < arraysize(kLanguages); ++i) { | |
| 1463 ReinitializeSpellCheck(kLanguages[i]); | |
| 1464 for (size_t j = 0; j < arraysize(kWords); ++j) | |
| 1465 EXPECT_TRUE(IsValidContraction(base::WideToUTF16(kWords[j]), 0)); | |
| 1466 } | |
| 1467 } | |
| 1468 | |
| 1469 TEST_F(SpellCheckTest, FillSuggestions_OneLanguageNoSuggestions) { | |
| 1470 std::vector<std::vector<base::string16>> suggestions_list; | |
| 1471 std::vector<base::string16> suggestion_results; | |
| 1472 | |
| 1473 suggestions_list.resize(1); | |
| 1474 | |
| 1475 FillSuggestions(suggestions_list, &suggestion_results); | |
| 1476 EXPECT_TRUE(suggestion_results.empty()); | |
| 1477 } | |
| 1478 | |
| 1479 TEST_F(SpellCheckTest, FillSuggestions_OneLanguageFewSuggestions) { | |
| 1480 std::vector<std::vector<base::string16>> suggestions_list; | |
| 1481 std::vector<base::string16> suggestion_results; | |
| 1482 | |
| 1483 suggestions_list.resize(1); | |
| 1484 suggestions_list[0].push_back(base::ASCIIToUTF16("foo")); | |
| 1485 | |
| 1486 FillSuggestions(suggestions_list, &suggestion_results); | |
| 1487 ASSERT_EQ(1U, suggestion_results.size()); | |
| 1488 EXPECT_EQ(base::ASCIIToUTF16("foo"), suggestion_results[0]); | |
| 1489 } | |
| 1490 | |
| 1491 TEST_F(SpellCheckTest, FillSuggestions_OneLanguageManySuggestions) { | |
| 1492 std::vector<std::vector<base::string16>> suggestions_list; | |
| 1493 std::vector<base::string16> suggestion_results; | |
| 1494 | |
| 1495 suggestions_list.resize(1); | |
| 1496 for (int i = 0; i < spellcheck::kMaxSuggestions + 2; ++i) | |
| 1497 suggestions_list[0].push_back(base::ASCIIToUTF16(base::IntToString(i))); | |
| 1498 | |
| 1499 FillSuggestions(suggestions_list, &suggestion_results); | |
| 1500 ASSERT_EQ(static_cast<size_t>(spellcheck::kMaxSuggestions), | |
| 1501 suggestion_results.size()); | |
| 1502 for (int i = 0; i < spellcheck::kMaxSuggestions; ++i) | |
| 1503 EXPECT_EQ(base::ASCIIToUTF16(base::IntToString(i)), suggestion_results[i]); | |
| 1504 } | |
| 1505 | |
| 1506 TEST_F(SpellCheckTest, FillSuggestions_RemoveDuplicates) { | |
| 1507 std::vector<std::vector<base::string16>> suggestions_list; | |
| 1508 std::vector<base::string16> suggestion_results; | |
| 1509 | |
| 1510 suggestions_list.resize(2); | |
| 1511 for (size_t i = 0; i < 2; ++i) { | |
| 1512 suggestions_list[i].push_back(base::ASCIIToUTF16("foo")); | |
| 1513 suggestions_list[i].push_back(base::ASCIIToUTF16("bar")); | |
| 1514 suggestions_list[i].push_back(base::ASCIIToUTF16("baz")); | |
| 1515 } | |
| 1516 | |
| 1517 FillSuggestions(suggestions_list, &suggestion_results); | |
| 1518 ASSERT_EQ(3U, suggestion_results.size()); | |
| 1519 EXPECT_EQ(base::ASCIIToUTF16("foo"), suggestion_results[0]); | |
| 1520 EXPECT_EQ(base::ASCIIToUTF16("bar"), suggestion_results[1]); | |
| 1521 EXPECT_EQ(base::ASCIIToUTF16("baz"), suggestion_results[2]); | |
| 1522 } | |
| 1523 | |
| 1524 TEST_F(SpellCheckTest, FillSuggestions_TwoLanguages) { | |
| 1525 std::vector<std::vector<base::string16>> suggestions_list; | |
| 1526 std::vector<base::string16> suggestion_results; | |
| 1527 | |
| 1528 suggestions_list.resize(2); | |
| 1529 for (size_t i = 0; i < 2; ++i) { | |
| 1530 std::string prefix = base::IntToString(i); | |
| 1531 suggestions_list[i].push_back(base::ASCIIToUTF16(prefix + "foo")); | |
| 1532 suggestions_list[i].push_back(base::ASCIIToUTF16(prefix + "bar")); | |
| 1533 suggestions_list[i].push_back(base::ASCIIToUTF16(prefix + "baz")); | |
| 1534 } | |
| 1535 | |
| 1536 // Yes, this test assumes kMaxSuggestions is 5. If it isn't, the test needs | |
| 1537 // to be updated accordingly. | |
| 1538 ASSERT_EQ(5, spellcheck::kMaxSuggestions); | |
| 1539 FillSuggestions(suggestions_list, &suggestion_results); | |
| 1540 ASSERT_EQ(5U, suggestion_results.size()); | |
| 1541 EXPECT_EQ(base::ASCIIToUTF16("0foo"), suggestion_results[0]); | |
| 1542 EXPECT_EQ(base::ASCIIToUTF16("1foo"), suggestion_results[1]); | |
| 1543 EXPECT_EQ(base::ASCIIToUTF16("0bar"), suggestion_results[2]); | |
| 1544 EXPECT_EQ(base::ASCIIToUTF16("1bar"), suggestion_results[3]); | |
| 1545 EXPECT_EQ(base::ASCIIToUTF16("0baz"), suggestion_results[4]); | |
| 1546 } | |
| 1547 | |
| 1548 TEST_F(SpellCheckTest, FillSuggestions_ThreeLanguages) { | |
| 1549 std::vector<std::vector<base::string16>> suggestions_list; | |
| 1550 std::vector<base::string16> suggestion_results; | |
| 1551 | |
| 1552 suggestions_list.resize(3); | |
| 1553 for (size_t i = 0; i < 3; ++i) { | |
| 1554 std::string prefix = base::IntToString(i); | |
| 1555 suggestions_list[i].push_back(base::ASCIIToUTF16(prefix + "foo")); | |
| 1556 suggestions_list[i].push_back(base::ASCIIToUTF16(prefix + "bar")); | |
| 1557 suggestions_list[i].push_back(base::ASCIIToUTF16(prefix + "baz")); | |
| 1558 } | |
| 1559 | |
| 1560 // Yes, this test assumes kMaxSuggestions is 5. If it isn't, the test needs | |
| 1561 // to be updated accordingly. | |
| 1562 ASSERT_EQ(5, spellcheck::kMaxSuggestions); | |
| 1563 FillSuggestions(suggestions_list, &suggestion_results); | |
| 1564 ASSERT_EQ(5U, suggestion_results.size()); | |
| 1565 EXPECT_EQ(base::ASCIIToUTF16("0foo"), suggestion_results[0]); | |
| 1566 EXPECT_EQ(base::ASCIIToUTF16("1foo"), suggestion_results[1]); | |
| 1567 EXPECT_EQ(base::ASCIIToUTF16("2foo"), suggestion_results[2]); | |
| 1568 EXPECT_EQ(base::ASCIIToUTF16("0bar"), suggestion_results[3]); | |
| 1569 EXPECT_EQ(base::ASCIIToUTF16("1bar"), suggestion_results[4]); | |
| 1570 } | |
| OLD | NEW |