1 /* 2 * (C) 1999 Lars Knoll (knoll (at) kde.org) 3 * Copyright (C) 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2012, 2013 Apple Inc. All rights reserved. 4 * 5 * This library is free software; you can redistribute it and/or 6 * modify it under the terms of the GNU Library General Public 7 * License as published by the Free Software Foundation; either 8 * version 2 of the License, or (at your option) any later version. 9 * 10 * This library is distributed in the hope that it will be useful, 11 * but WITHOUT ANY WARRANTY; without even the implied warranty of 12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 13 * Library General Public License for more details. 14 * 15 * You should have received a copy of the GNU Library General Public License 16 * along with this library; see the file COPYING.LIB. If not, write to 17 * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, 18 * Boston, MA 02110-1301, USA. 19 * 20 */ 21 22 #ifndef WTFString_h 23 #define WTFString_h 24 25 // This file would be called String.h, but that conflicts with <string.h> 26 // on systems without case-sensitive file systems. 27 28 #include "wtf/HashTableDeletedValueType.h" 29 #include "wtf/WTFExport.h" 30 #include "wtf/text/ASCIIFastPath.h" 31 #include "wtf/text/StringImpl.h" 32 #include "wtf/text/StringView.h" 33 34 #ifdef __OBJC__ 35 #include <objc/objc.h> 36 #endif 37 38 namespace WTF { 39 40 class CString; 41 struct StringHash; 42 43 // Declarations of string operations 44 45 WTF_EXPORT int charactersToIntStrict(const LChar*, size_t, bool* ok = 0, int base = 10); 46 WTF_EXPORT int charactersToIntStrict(const UChar*, size_t, bool* ok = 0, int base = 10); 47 WTF_EXPORT unsigned charactersToUIntStrict(const LChar*, size_t, bool* ok = 0, int base = 10); 48 WTF_EXPORT unsigned charactersToUIntStrict(const UChar*, size_t, bool* ok = 0, int base = 10); 49 WTF_EXPORT int64_t charactersToInt64Strict(const LChar*, size_t, bool* ok = 0, int base = 10); 50 WTF_EXPORT int64_t charactersToInt64Strict(const UChar*, size_t, bool* ok = 0, int base = 10); 51 WTF_EXPORT uint64_t charactersToUInt64Strict(const LChar*, size_t, bool* ok = 0, int base = 10); 52 WTF_EXPORT uint64_t charactersToUInt64Strict(const UChar*, size_t, bool* ok = 0, int base = 10); 53 WTF_EXPORT intptr_t charactersToIntPtrStrict(const LChar*, size_t, bool* ok = 0, int base = 10); 54 WTF_EXPORT intptr_t charactersToIntPtrStrict(const UChar*, size_t, bool* ok = 0, int base = 10); 55 56 WTF_EXPORT int charactersToInt(const LChar*, size_t, bool* ok = 0); // ignores trailing garbage 57 WTF_EXPORT int charactersToInt(const UChar*, size_t, bool* ok = 0); // ignores trailing garbage 58 WTF_EXPORT unsigned charactersToUInt(const LChar*, size_t, bool* ok = 0); // ignores trailing garbage 59 WTF_EXPORT unsigned charactersToUInt(const UChar*, size_t, bool* ok = 0); // ignores trailing garbage 60 WTF_EXPORT int64_t charactersToInt64(const LChar*, size_t, bool* ok = 0); // ignores trailing garbage 61 WTF_EXPORT int64_t charactersToInt64(const UChar*, size_t, bool* ok = 0); // ignores trailing garbage 62 WTF_EXPORT uint64_t charactersToUInt64(const LChar*, size_t, bool* ok = 0); // ignores trailing garbage 63 WTF_EXPORT uint64_t charactersToUInt64(const UChar*, size_t, bool* ok = 0); // ignores trailing garbage 64 WTF_EXPORT intptr_t charactersToIntPtr(const LChar*, size_t, bool* ok = 0); // ignores trailing garbage 65 WTF_EXPORT intptr_t charactersToIntPtr(const UChar*, size_t, bool* ok = 0); // ignores trailing garbage 66 67 // FIXME: Like the strict functions above, these give false for "ok" when there is trailing garbage. 68 // Like the non-strict functions above, these return the value when there is trailing garbage. 69 // It would be better if these were more consistent with the above functions instead. 70 WTF_EXPORT double charactersToDouble(const LChar*, size_t, bool* ok = 0); 71 WTF_EXPORT double charactersToDouble(const UChar*, size_t, bool* ok = 0); 72 WTF_EXPORT float charactersToFloat(const LChar*, size_t, bool* ok = 0); 73 WTF_EXPORT float charactersToFloat(const UChar*, size_t, bool* ok = 0); 74 WTF_EXPORT float charactersToFloat(const LChar*, size_t, size_t& parsedLength); 75 WTF_EXPORT float charactersToFloat(const UChar*, size_t, size_t& parsedLength); 76 77 enum TrailingZerosTruncatingPolicy { 78 KeepTrailingZeros, 79 TruncateTrailingZeros 80 }; 81 82 enum UTF8ConversionMode { 83 LenientUTF8Conversion, 84 StrictUTF8Conversion, 85 StrictUTF8ConversionReplacingUnpairedSurrogatesWithFFFD 86 }; 87 88 template<bool isSpecialCharacter(UChar), typename CharacterType> 89 bool isAllSpecialCharacters(const CharacterType*, size_t); 90 91 // You can find documentation about this class in this doc: 92 // https://docs.google.com/document/d/1kOCUlJdh2WJMJGDf-WoEQhmnjKLaOYRbiHz5TiGJl14/edit?usp=sharing 93 class WTF_EXPORT String { 94 public: 95 // Construct a null string, distinguishable from an empty string. 96 String() { } 97 98 // Construct a string with UTF-16 data. 99 String(const UChar* characters, unsigned length); 100 101 // Construct a string by copying the contents of a vector. 102 // This method will never create a null string. Vectors with size() == 0 103 // will return the empty string. 104 // NOTE: This is different from String(vector.data(), vector.size()) 105 // which will sometimes return a null string when vector.data() is null 106 // which can only occur for vectors without inline capacity. 107 // See: https://bugs.webkit.org/show_bug.cgi?id=109792 108 template<size_t inlineCapacity> 109 explicit String(const Vector<UChar, inlineCapacity>&); 110 111 // Construct a string with UTF-16 data, from a null-terminated source. 112 String(const UChar*); 113 114 // Construct a string with latin1 data. 115 String(const LChar* characters, unsigned length); 116 String(const char* characters, unsigned length); 117 118 // Construct a string with latin1 data, from a null-terminated source. 119 String(const LChar* characters); 120 String(const char* characters); 121 122 // Construct a string referencing an existing StringImpl. 123 String(StringImpl* impl) : m_impl(impl) { } 124 String(PassRefPtr<StringImpl> impl) : m_impl(impl) { } 125 126 void swap(String& o) { m_impl.swap(o.m_impl); } 127 128 template<typename CharType> 129 static String adopt(StringBuffer<CharType>& buffer) 130 { 131 if (!buffer.length()) 132 return StringImpl::empty(); 133 return String(buffer.release()); 134 } 135 136 bool isNull() const { return !m_impl; } 137 bool isEmpty() const { return !m_impl || !m_impl->length(); } 138 139 StringImpl* impl() const { return m_impl.get(); } 140 PassRefPtr<StringImpl> releaseImpl() { return m_impl.release(); } 141 142 unsigned length() const 143 { 144 if (!m_impl) 145 return 0; 146 return m_impl->length(); 147 } 148 149 const LChar* characters8() const 150 { 151 if (!m_impl) 152 return 0; 153 ASSERT(m_impl->is8Bit()); 154 return m_impl->characters8(); 155 } 156 157 const UChar* characters16() const 158 { 159 if (!m_impl) 160 return 0; 161 ASSERT(!m_impl->is8Bit()); 162 return m_impl->characters16(); 163 } 164 165 // Return characters8() or characters16() depending on CharacterType. 166 template <typename CharacterType> 167 inline const CharacterType* getCharacters() const; 168 169 bool is8Bit() const { return m_impl->is8Bit(); } 170 171 unsigned sizeInBytes() const 172 { 173 if (!m_impl) 174 return 0; 175 return m_impl->length() * (is8Bit() ? sizeof(LChar) : sizeof(UChar)); 176 } 177 178 CString ascii() const; 179 CString latin1() const; 180 CString utf8(UTF8ConversionMode = LenientUTF8Conversion) const; 181 182 UChar operator[](unsigned index) const 183 { 184 if (!m_impl || index >= m_impl->length()) 185 return 0; 186 return (*m_impl)[index]; 187 } 188 189 static String number(int); 190 static String number(unsigned); 191 static String number(long); 192 static String number(unsigned long); 193 static String number(long long); 194 static String number(unsigned long long); 195 196 static String number(double, unsigned precision = 6, TrailingZerosTruncatingPolicy = TruncateTrailingZeros); 197 198 // Number to String conversion following the ECMAScript definition. 199 static String numberToStringECMAScript(double); 200 static String numberToStringFixedWidth(double, unsigned decimalPlaces); 201 202 // Find a single character or string, also with match function & latin1 forms. 203 size_t find(UChar c, unsigned start = 0) const 204 { return m_impl ? m_impl->find(c, start) : kNotFound; } 205 206 size_t find(const String& str) const 207 { return m_impl ? m_impl->find(str.impl()) : kNotFound; } 208 size_t find(const String& str, unsigned start) const 209 { return m_impl ? m_impl->find(str.impl(), start) : kNotFound; } 210 211 size_t find(CharacterMatchFunctionPtr matchFunction, unsigned start = 0) const 212 { return m_impl ? m_impl->find(matchFunction, start) : kNotFound; } 213 size_t find(const LChar* str, unsigned start = 0) const 214 { return m_impl ? m_impl->find(str, start) : kNotFound; } 215 216 size_t findNextLineStart(unsigned start = 0) const 217 { return m_impl ? m_impl->findNextLineStart(start) : kNotFound; } 218 219 // Find the last instance of a single character or string. 220 size_t reverseFind(UChar c, unsigned start = UINT_MAX) const 221 { return m_impl ? m_impl->reverseFind(c, start) : kNotFound; } 222 size_t reverseFind(const String& str, unsigned start = UINT_MAX) const 223 { return m_impl ? m_impl->reverseFind(str.impl(), start) : kNotFound; } 224 225 // Case insensitive string matching. 226 size_t findIgnoringCase(const LChar* str, unsigned start = 0) const 227 { return m_impl ? m_impl->findIgnoringCase(str, start) : kNotFound; } 228 size_t findIgnoringCase(const String& str, unsigned start = 0) const 229 { return m_impl ? m_impl->findIgnoringCase(str.impl(), start) : kNotFound; } 230 size_t reverseFindIgnoringCase(const String& str, unsigned start = UINT_MAX) const 231 { return m_impl ? m_impl->reverseFindIgnoringCase(str.impl(), start) : kNotFound; } 232 233 // Wrappers for find & reverseFind adding dynamic sensitivity check. 234 size_t find(const LChar* str, unsigned start, bool caseSensitive) const 235 { return caseSensitive ? find(str, start) : findIgnoringCase(str, start); } 236 size_t find(const String& str, unsigned start, bool caseSensitive) const 237 { return caseSensitive ? find(str, start) : findIgnoringCase(str, start); } 238 size_t reverseFind(const String& str, unsigned start, bool caseSensitive) const 239 { return caseSensitive ? reverseFind(str, start) : reverseFindIgnoringCase(str, start); } 240 241 Vector<UChar> charactersWithNullTermination() const; 242 unsigned copyTo(UChar* buffer, unsigned pos, unsigned maxLength) const; 243 244 template<size_t inlineCapacity> 245 void appendTo(Vector<UChar, inlineCapacity>&, unsigned pos = 0, unsigned len = UINT_MAX) const; 246 247 template<typename BufferType> 248 void appendTo(BufferType&, unsigned pos = 0, unsigned len = UINT_MAX) const; 249 250 template<size_t inlineCapacity> 251 void prependTo(Vector<UChar, inlineCapacity>&, unsigned pos = 0, unsigned len = UINT_MAX) const; 252 253 UChar32 characterStartingAt(unsigned) const; 254 255 bool contains(UChar c) const { return find(c) != kNotFound; } 256 bool contains(const LChar* str, bool caseSensitive = true) const { return find(str, 0, caseSensitive) != kNotFound; } 257 bool contains(const String& str, bool caseSensitive = true) const { return find(str, 0, caseSensitive) != kNotFound; } 258 259 bool startsWith(const String& s, bool caseSensitive = true) const 260 { return m_impl ? m_impl->startsWith(s.impl(), caseSensitive) : s.isEmpty(); } 261 bool startsWith(UChar character) const 262 { return m_impl ? m_impl->startsWith(character) : false; } 263 template<unsigned matchLength> 264 bool startsWith(const char (&prefix)[matchLength], bool caseSensitive = true) const 265 { return m_impl ? m_impl->startsWith<matchLength>(prefix, caseSensitive) : !matchLength; } 266 267 bool endsWith(const String& s, bool caseSensitive = true) const 268 { return m_impl ? m_impl->endsWith(s.impl(), caseSensitive) : s.isEmpty(); } 269 bool endsWith(UChar character) const 270 { return m_impl ? m_impl->endsWith(character) : false; } 271 template<unsigned matchLength> 272 bool endsWith(const char (&prefix)[matchLength], bool caseSensitive = true) const 273 { return m_impl ? m_impl->endsWith<matchLength>(prefix, caseSensitive) : !matchLength; } 274 275 void append(const String&); 276 void append(LChar); 277 void append(char c) { append(static_cast<LChar>(c)); } 278 void append(UChar); 279 void append(const LChar*, unsigned length); 280 void append(const char* charactersToAppend, unsigned length) { append(reinterpret_cast<const LChar*>(charactersToAppend), length); } 281 void append(const UChar*, unsigned length); 282 void insert(const String&, unsigned pos); 283 void insert(const LChar*, unsigned length, unsigned pos); 284 void insert(const UChar*, unsigned length, unsigned pos); 285 286 String& replace(UChar a, UChar b) { if (m_impl) m_impl = m_impl->replace(a, b); return *this; } 287 String& replace(UChar a, const String& b) { if (m_impl) m_impl = m_impl->replace(a, b.impl()); return *this; } 288 String& replace(const String& a, const String& b) { if (m_impl) m_impl = m_impl->replace(a.impl(), b.impl()); return *this; } 289 String& replace(unsigned index, unsigned len, const String& b) { if (m_impl) m_impl = m_impl->replace(index, len, b.impl()); return *this; } 290 291 template<unsigned charactersCount> 292 ALWAYS_INLINE String& replaceWithLiteral(UChar a, const char (&characters)[charactersCount]) 293 { 294 if (m_impl) 295 m_impl = m_impl->replace(a, characters, charactersCount - 1); 296 297 return *this; 298 } 299 300 void fill(UChar c) { if (m_impl) m_impl = m_impl->fill(c); } 301 302 void ensure16Bit(); 303 304 void truncate(unsigned len); 305 void remove(unsigned pos, int len = 1); 306 307 String substring(unsigned pos, unsigned len = UINT_MAX) const; 308 String left(unsigned len) const { return substring(0, len); } 309 String right(unsigned len) const { return substring(length() - len, len); } 310 311 StringView createView() const { return StringView(impl()); } 312 StringView createView(unsigned offset, unsigned length) const { return StringView(impl(), offset, length); } 313 314 // Returns a lowercase/uppercase version of the string 315 String lower() const; 316 String upper() const; 317 318 String lower(const AtomicString& localeIdentifier) const; 319 String upper(const AtomicString& localeIdentifier) const; 320 321 String stripWhiteSpace() const; 322 String stripWhiteSpace(IsWhiteSpaceFunctionPtr) const; 323 String simplifyWhiteSpace(StripBehavior stripBehavior = StripExtraWhiteSpace) const; 324 String simplifyWhiteSpace(IsWhiteSpaceFunctionPtr, StripBehavior stripBehavior = StripExtraWhiteSpace) const; 325 326 String removeCharacters(CharacterMatchFunctionPtr) const; 327 template<bool isSpecialCharacter(UChar)> bool isAllSpecialCharacters() const; 328 329 // Return the string with case folded for case insensitive comparison. 330 String foldCase() const; 331 332 static String format(const char *, ...) WTF_ATTRIBUTE_PRINTF(1, 2); 333 334 // Returns an uninitialized string. The characters needs to be written 335 // into the buffer returned in data before the returned string is used. 336 // Failure to do this will have unpredictable results. 337 static String createUninitialized(unsigned length, UChar*& data) { return StringImpl::createUninitialized(length, data); } 338 static String createUninitialized(unsigned length, LChar*& data) { return StringImpl::createUninitialized(length, data); } 339 340 void split(const String& separator, bool allowEmptyEntries, Vector<String>& result) const; 341 void split(const String& separator, Vector<String>& result) const 342 { 343 split(separator, false, result); 344 } 345 void split(UChar separator, bool allowEmptyEntries, Vector<String>& result) const; 346 void split(UChar separator, Vector<String>& result) const 347 { 348 split(separator, false, result); 349 } 350 351 int toIntStrict(bool* ok = 0, int base = 10) const; 352 unsigned toUIntStrict(bool* ok = 0, int base = 10) const; 353 int64_t toInt64Strict(bool* ok = 0, int base = 10) const; 354 uint64_t toUInt64Strict(bool* ok = 0, int base = 10) const; 355 intptr_t toIntPtrStrict(bool* ok = 0, int base = 10) const; 356 357 int toInt(bool* ok = 0) const; 358 unsigned toUInt(bool* ok = 0) const; 359 int64_t toInt64(bool* ok = 0) const; 360 uint64_t toUInt64(bool* ok = 0) const; 361 intptr_t toIntPtr(bool* ok = 0) const; 362 363 // FIXME: Like the strict functions above, these give false for "ok" when there is trailing garbage. 364 // Like the non-strict functions above, these return the value when there is trailing garbage. 365 // It would be better if these were more consistent with the above functions instead. 366 double toDouble(bool* ok = 0) const; 367 float toFloat(bool* ok = 0) const; 368 369 bool percentage(int& percentage) const; 370 371 String isolatedCopy() const; 372 bool isSafeToSendToAnotherThread() const; 373 374 #if USE(CF) 375 String(CFStringRef); 376 RetainPtr<CFStringRef> createCFString() const; 377 #endif 378 379 #ifdef __OBJC__ 380 String(NSString*); 381 382 // This conversion maps NULL to "", which loses the meaning of NULL, but we 383 // need this mapping because AppKit crashes when passed nil NSStrings. 384 operator NSString*() const { if (!m_impl) return @""; return *m_impl; } 385 #endif 386 387 static String make8BitFrom16BitSource(const UChar*, size_t); 388 template<size_t inlineCapacity> 389 static String make8BitFrom16BitSource(const Vector<UChar, inlineCapacity>& buffer) 390 { 391 return make8BitFrom16BitSource(buffer.data(), buffer.size()); 392 } 393 394 static String make16BitFrom8BitSource(const LChar*, size_t); 395 396 // String::fromUTF8 will return a null string if 397 // the input data contains invalid UTF-8 sequences. 398 static String fromUTF8(const LChar*, size_t); 399 static String fromUTF8(const LChar*); 400 static String fromUTF8(const char* s, size_t length) { return fromUTF8(reinterpret_cast<const LChar*>(s), length); }; 401 static String fromUTF8(const char* s) { return fromUTF8(reinterpret_cast<const LChar*>(s)); }; 402 static String fromUTF8(const CString&); 403 404 // Tries to convert the passed in string to UTF-8, but will fall back to Latin-1 if the string is not valid UTF-8. 405 static String fromUTF8WithLatin1Fallback(const LChar*, size_t); 406 static String fromUTF8WithLatin1Fallback(const char* s, size_t length) { return fromUTF8WithLatin1Fallback(reinterpret_cast<const LChar*>(s), length); }; 407 408 bool containsOnlyASCII() const; 409 bool containsOnlyLatin1() const; 410 bool containsOnlyWhitespace() const { return !m_impl || m_impl->containsOnlyWhitespace(); } 411 412 // Hash table deleted values, which are only constructed and never copied or destroyed. 413 String(WTF::HashTableDeletedValueType) : m_impl(WTF::HashTableDeletedValue) { } 414 bool isHashTableDeletedValue() const { return m_impl.isHashTableDeletedValue(); } 415 416 #ifndef NDEBUG 417 void show() const; 418 #endif 419 420 // Workaround for a compiler bug. Use operator[] instead. 421 UChar characterAt(unsigned index) const 422 { 423 if (!m_impl || index >= m_impl->length()) 424 return 0; 425 return (*m_impl)[index]; 426 } 427 428 private: 429 typedef struct ImplicitConversionFromWTFStringToBoolDisallowed* (String::*UnspecifiedBoolType); 430 operator UnspecifiedBoolType() const; 431 432 template <typename CharacterType> 433 void removeInternal(const CharacterType*, unsigned, int); 434 435 template <typename CharacterType> 436 void appendInternal(CharacterType); 437 438 RefPtr<StringImpl> m_impl; 439 }; 440 441 inline bool operator==(const String& a, const String& b) { return equal(a.impl(), b.impl()); } 442 inline bool operator==(const String& a, const LChar* b) { return equal(a.impl(), b); } 443 inline bool operator==(const String& a, const char* b) { return equal(a.impl(), reinterpret_cast<const LChar*>(b)); } 444 inline bool operator==(const LChar* a, const String& b) { return equal(a, b.impl()); } 445 inline bool operator==(const char* a, const String& b) { return equal(reinterpret_cast<const LChar*>(a), b.impl()); } 446 template<size_t inlineCapacity> 447 inline bool operator==(const Vector<char, inlineCapacity>& a, const String& b) { return equal(b.impl(), a.data(), a.size()); } 448 template<size_t inlineCapacity> 449 inline bool operator==(const String& a, const Vector<char, inlineCapacity>& b) { return b == a; } 450 451 452 inline bool operator!=(const String& a, const String& b) { return !equal(a.impl(), b.impl()); } 453 inline bool operator!=(const String& a, const LChar* b) { return !equal(a.impl(), b); } 454 inline bool operator!=(const String& a, const char* b) { return !equal(a.impl(), reinterpret_cast<const LChar*>(b)); } 455 inline bool operator!=(const LChar* a, const String& b) { return !equal(a, b.impl()); } 456 inline bool operator!=(const char* a, const String& b) { return !equal(reinterpret_cast<const LChar*>(a), b.impl()); } 457 template<size_t inlineCapacity> 458 inline bool operator!=(const Vector<char, inlineCapacity>& a, const String& b) { return !(a == b); } 459 template<size_t inlineCapacity> 460 inline bool operator!=(const String& a, const Vector<char, inlineCapacity>& b) { return b != a; } 461 462 inline bool equalIgnoringCase(const String& a, const String& b) { return equalIgnoringCase(a.impl(), b.impl()); } 463 inline bool equalIgnoringCase(const String& a, const LChar* b) { return equalIgnoringCase(a.impl(), b); } 464 inline bool equalIgnoringCase(const String& a, const char* b) { return equalIgnoringCase(a.impl(), reinterpret_cast<const LChar*>(b)); } 465 inline bool equalIgnoringCase(const LChar* a, const String& b) { return equalIgnoringCase(a, b.impl()); } 466 inline bool equalIgnoringCase(const char* a, const String& b) { return equalIgnoringCase(reinterpret_cast<const LChar*>(a), b.impl()); } 467 468 inline bool equalPossiblyIgnoringCase(const String& a, const String& b, bool ignoreCase) 469 { 470 return ignoreCase ? equalIgnoringCase(a, b) : (a == b); 471 } 472 473 inline bool equalIgnoringNullity(const String& a, const String& b) { return equalIgnoringNullity(a.impl(), b.impl()); } 474 475 template<size_t inlineCapacity> 476 inline bool equalIgnoringNullity(const Vector<UChar, inlineCapacity>& a, const String& b) { return equalIgnoringNullity(a, b.impl()); } 477 478 inline bool operator!(const String& str) { return str.isNull(); } 479 480 inline void swap(String& a, String& b) { a.swap(b); } 481 482 // Definitions of string operations 483 484 template<size_t inlineCapacity> 485 String::String(const Vector<UChar, inlineCapacity>& vector) 486 : m_impl(vector.size() ? StringImpl::create(vector.data(), vector.size()) : StringImpl::empty()) 487 { 488 } 489 490 template<> 491 inline const LChar* String::getCharacters<LChar>() const 492 { 493 ASSERT(is8Bit()); 494 return characters8(); 495 } 496 497 template<> 498 inline const UChar* String::getCharacters<UChar>() const 499 { 500 ASSERT(!is8Bit()); 501 return characters16(); 502 } 503 504 inline bool String::containsOnlyLatin1() const 505 { 506 if (isEmpty()) 507 return true; 508 509 if (is8Bit()) 510 return true; 511 512 const UChar* characters = characters16(); 513 UChar ored = 0; 514 for (size_t i = 0; i < m_impl->length(); ++i) 515 ored |= characters[i]; 516 return !(ored & 0xFF00); 517 } 518 519 520 #ifdef __OBJC__ 521 // This is for situations in WebKit where the long standing behavior has been 522 // "nil if empty", so we try to maintain longstanding behavior for the sake of 523 // entrenched clients 524 inline NSString* nsStringNilIfEmpty(const String& str) { return str.isEmpty() ? nil : (NSString*)str; } 525 #endif 526 527 inline bool String::containsOnlyASCII() const 528 { 529 if (isEmpty()) 530 return true; 531 532 if (is8Bit()) 533 return charactersAreAllASCII(characters8(), m_impl->length()); 534 535 return charactersAreAllASCII(characters16(), m_impl->length()); 536 } 537 538 WTF_EXPORT int codePointCompare(const String&, const String&); 539 540 inline bool codePointCompareLessThan(const String& a, const String& b) 541 { 542 return codePointCompare(a.impl(), b.impl()) < 0; 543 } 544 545 template<size_t inlineCapacity> 546 inline void append(Vector<UChar, inlineCapacity>& vector, const String& string) 547 { 548 unsigned length = string.length(); 549 if (!length) 550 return; 551 if (string.is8Bit()) { 552 const LChar* characters8 = string.characters8(); 553 vector.reserveCapacity(vector.size() + length); 554 for (size_t i = 0; i < length; ++i) 555 vector.uncheckedAppend(characters8[i]); 556 } else { 557 vector.append(string.characters16(), length); 558 } 559 } 560 561 template<typename CharacterType> 562 inline void appendNumber(Vector<CharacterType>& vector, unsigned char number) 563 { 564 int numberLength = number > 99 ? 3 : (number > 9 ? 2 : 1); 565 size_t vectorSize = vector.size(); 566 vector.grow(vectorSize + numberLength); 567 568 switch (numberLength) { 569 case 3: 570 vector[vectorSize + 2] = number % 10 + '0'; 571 number /= 10; 572 573 case 2: 574 vector[vectorSize + 1] = number % 10 + '0'; 575 number /= 10; 576 577 case 1: 578 vector[vectorSize] = number % 10 + '0'; 579 } 580 } 581 582 template<bool isSpecialCharacter(UChar), typename CharacterType> 583 inline bool isAllSpecialCharacters(const CharacterType* characters, size_t length) 584 { 585 for (size_t i = 0; i < length; ++i) { 586 if (!isSpecialCharacter(characters[i])) 587 return false; 588 } 589 return true; 590 } 591 592 template<bool isSpecialCharacter(UChar)> 593 inline bool String::isAllSpecialCharacters() const 594 { 595 size_t len = length(); 596 597 if (!len) 598 return true; 599 600 if (is8Bit()) 601 return WTF::isAllSpecialCharacters<isSpecialCharacter, LChar>(characters8(), len); 602 return WTF::isAllSpecialCharacters<isSpecialCharacter, UChar>(characters16(), len); 603 } 604 605 template<size_t inlineCapacity> 606 inline void String::appendTo(Vector<UChar, inlineCapacity>& result, unsigned pos, unsigned len) const 607 { 608 unsigned numberOfCharactersToCopy = std::min(len, length() - pos); 609 if (!numberOfCharactersToCopy) 610 return; 611 result.reserveCapacity(result.size() + numberOfCharactersToCopy); 612 if (is8Bit()) { 613 const LChar* characters8 = m_impl->characters8(); 614 for (size_t i = 0; i < numberOfCharactersToCopy; ++i) 615 result.uncheckedAppend(characters8[pos + i]); 616 } else { 617 const UChar* characters16 = m_impl->characters16(); 618 result.append(characters16 + pos, numberOfCharactersToCopy); 619 } 620 } 621 622 template<typename BufferType> 623 inline void String::appendTo(BufferType& result, unsigned pos, unsigned len) const 624 { 625 unsigned numberOfCharactersToCopy = std::min(len, length() - pos); 626 if (!numberOfCharactersToCopy) 627 return; 628 if (is8Bit()) 629 result.append(m_impl->characters8() + pos, numberOfCharactersToCopy); 630 else 631 result.append(m_impl->characters16() + pos, numberOfCharactersToCopy); 632 } 633 634 template<size_t inlineCapacity> 635 inline void String::prependTo(Vector<UChar, inlineCapacity>& result, unsigned pos, unsigned len) const 636 { 637 unsigned numberOfCharactersToCopy = std::min(len, length() - pos); 638 if (!numberOfCharactersToCopy) 639 return; 640 if (is8Bit()) { 641 size_t oldSize = result.size(); 642 result.resize(oldSize + numberOfCharactersToCopy); 643 memmove(result.data() + numberOfCharactersToCopy, result.data(), oldSize * sizeof(UChar)); 644 StringImpl::copyChars(result.data(), m_impl->characters8() + pos, numberOfCharactersToCopy); 645 } else { 646 result.prepend(m_impl->characters16() + pos, numberOfCharactersToCopy); 647 } 648 } 649 650 // StringHash is the default hash for String 651 template<typename T> struct DefaultHash; 652 template<> struct DefaultHash<String> { 653 typedef StringHash Hash; 654 }; 655 656 // Shared global empty string. 657 WTF_EXPORT const String& emptyString(); 658 WTF_EXPORT const String& emptyString16Bit(); 659 WTF_EXPORT extern const String& xmlnsWithColon; 660 661 } // namespace WTF 662 663 WTF_ALLOW_MOVE_AND_INIT_WITH_MEM_FUNCTIONS(String); 664 665 using WTF::CString; 666 using WTF::KeepTrailingZeros; 667 using WTF::StrictUTF8Conversion; 668 using WTF::StrictUTF8ConversionReplacingUnpairedSurrogatesWithFFFD; 669 using WTF::String; 670 using WTF::emptyString; 671 using WTF::emptyString16Bit; 672 using WTF::append; 673 using WTF::appendNumber; 674 using WTF::charactersAreAllASCII; 675 using WTF::charactersToIntStrict; 676 using WTF::charactersToUIntStrict; 677 using WTF::charactersToInt64Strict; 678 using WTF::charactersToUInt64Strict; 679 using WTF::charactersToIntPtrStrict; 680 using WTF::charactersToInt; 681 using WTF::charactersToUInt; 682 using WTF::charactersToInt64; 683 using WTF::charactersToUInt64; 684 using WTF::charactersToIntPtr; 685 using WTF::charactersToDouble; 686 using WTF::charactersToFloat; 687 using WTF::equal; 688 using WTF::equalIgnoringCase; 689 using WTF::find; 690 using WTF::isAllSpecialCharacters; 691 using WTF::isSpaceOrNewline; 692 using WTF::reverseFind; 693 694 #include "wtf/text/AtomicString.h" 695 #endif // WTFString_h 696
