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 template<bool isSpecialCharacter(UChar), typename CharacterType> 83 bool isAllSpecialCharacters(const CharacterType*, size_t); 84 85 // You can find documentation about this class in this doc: 86 // https://docs.google.com/document/d/1kOCUlJdh2WJMJGDf-WoEQhmnjKLaOYRbiHz5TiGJl14/edit?usp=sharing 87 class WTF_EXPORT String { 88 public: 89 // Construct a null string, distinguishable from an empty string. 90 String() { } 91 92 // Construct a string with UTF-16 data. 93 String(const UChar* characters, unsigned length); 94 95 // Construct a string by copying the contents of a vector. 96 // This method will never create a null string. Vectors with size() == 0 97 // will return the empty string. 98 // NOTE: This is different from String(vector.data(), vector.size()) 99 // which will sometimes return a null string when vector.data() is null 100 // which can only occur for vectors without inline capacity. 101 // See: https://bugs.webkit.org/show_bug.cgi?id=109792 102 template<size_t inlineCapacity> 103 explicit String(const Vector<UChar, inlineCapacity>&); 104 105 // Construct a string with UTF-16 data, from a null-terminated source. 106 String(const UChar*); 107 108 // Construct a string with latin1 data. 109 String(const LChar* characters, unsigned length); 110 String(const char* characters, unsigned length); 111 112 // Construct a string with latin1 data, from a null-terminated source. 113 String(const LChar* characters); 114 String(const char* characters); 115 116 // Construct a string referencing an existing StringImpl. 117 String(StringImpl* impl) : m_impl(impl) { } 118 String(PassRefPtr<StringImpl> impl) : m_impl(impl) { } 119 120 #if COMPILER_SUPPORTS(CXX_RVALUE_REFERENCES) 121 // We have to declare the copy constructor and copy assignment operator as well, otherwise 122 // they'll be implicitly deleted by adding the move constructor and move assignment operator. 123 String(const String& other) : m_impl(other.m_impl) { } 124 String(String&& other) : m_impl(other.m_impl.release()) { } 125 String& operator=(const String& other) { m_impl = other.m_impl; return *this; } 126 String& operator=(String&& other) { m_impl = other.m_impl.release(); return *this; } 127 #endif 128 129 // Inline the destructor. 130 ALWAYS_INLINE ~String() { } 131 132 void swap(String& o) { m_impl.swap(o.m_impl); } 133 134 template<typename CharType> 135 static String adopt(StringBuffer<CharType>& buffer) 136 { 137 if (!buffer.length()) 138 return StringImpl::empty(); 139 return String(buffer.release()); 140 } 141 142 bool isNull() const { return !m_impl; } 143 bool isEmpty() const { return !m_impl || !m_impl->length(); } 144 145 StringImpl* impl() const { return m_impl.get(); } 146 PassRefPtr<StringImpl> releaseImpl() { return m_impl.release(); } 147 148 unsigned length() const 149 { 150 if (!m_impl) 151 return 0; 152 return m_impl->length(); 153 } 154 155 const LChar* characters8() const 156 { 157 if (!m_impl) 158 return 0; 159 ASSERT(m_impl->is8Bit()); 160 return m_impl->characters8(); 161 } 162 163 const UChar* characters16() const 164 { 165 if (!m_impl) 166 return 0; 167 ASSERT(!m_impl->is8Bit()); 168 return m_impl->characters16(); 169 } 170 171 // Return characters8() or characters16() depending on CharacterType. 172 template <typename CharacterType> 173 inline const CharacterType* getCharacters() const; 174 175 bool is8Bit() const { return m_impl->is8Bit(); } 176 177 unsigned sizeInBytes() const 178 { 179 if (!m_impl) 180 return 0; 181 return m_impl->length() * (is8Bit() ? sizeof(LChar) : sizeof(UChar)); 182 } 183 184 CString ascii() const; 185 CString latin1() const; 186 187 typedef enum { 188 LenientConversion, 189 StrictConversion, 190 StrictConversionReplacingUnpairedSurrogatesWithFFFD, 191 } ConversionMode; 192 193 CString utf8(ConversionMode = LenientConversion) const; 194 195 UChar operator[](unsigned index) const 196 { 197 if (!m_impl || index >= m_impl->length()) 198 return 0; 199 return (*m_impl)[index]; 200 } 201 202 static String number(int); 203 static String number(unsigned); 204 static String number(long); 205 static String number(unsigned long); 206 static String number(long long); 207 static String number(unsigned long long); 208 209 static String number(double, unsigned precision = 6, TrailingZerosTruncatingPolicy = TruncateTrailingZeros); 210 211 // Number to String conversion following the ECMAScript definition. 212 static String numberToStringECMAScript(double); 213 static String numberToStringFixedWidth(double, unsigned decimalPlaces); 214 215 // Find a single character or string, also with match function & latin1 forms. 216 size_t find(UChar c, unsigned start = 0) const 217 { return m_impl ? m_impl->find(c, start) : kNotFound; } 218 219 size_t find(const String& str) const 220 { return m_impl ? m_impl->find(str.impl()) : kNotFound; } 221 size_t find(const String& str, unsigned start) const 222 { return m_impl ? m_impl->find(str.impl(), start) : kNotFound; } 223 224 size_t find(CharacterMatchFunctionPtr matchFunction, unsigned start = 0) const 225 { return m_impl ? m_impl->find(matchFunction, start) : kNotFound; } 226 size_t find(const LChar* str, unsigned start = 0) const 227 { return m_impl ? m_impl->find(str, start) : kNotFound; } 228 229 size_t findNextLineStart(unsigned start = 0) const 230 { return m_impl ? m_impl->findNextLineStart(start) : kNotFound; } 231 232 // Find the last instance of a single character or string. 233 size_t reverseFind(UChar c, unsigned start = UINT_MAX) const 234 { return m_impl ? m_impl->reverseFind(c, start) : kNotFound; } 235 size_t reverseFind(const String& str, unsigned start = UINT_MAX) const 236 { return m_impl ? m_impl->reverseFind(str.impl(), start) : kNotFound; } 237 238 // Case insensitive string matching. 239 size_t findIgnoringCase(const LChar* str, unsigned start = 0) const 240 { return m_impl ? m_impl->findIgnoringCase(str, start) : kNotFound; } 241 size_t findIgnoringCase(const String& str, unsigned start = 0) const 242 { return m_impl ? m_impl->findIgnoringCase(str.impl(), start) : kNotFound; } 243 size_t reverseFindIgnoringCase(const String& str, unsigned start = UINT_MAX) const 244 { return m_impl ? m_impl->reverseFindIgnoringCase(str.impl(), start) : kNotFound; } 245 246 // Wrappers for find & reverseFind adding dynamic sensitivity check. 247 size_t find(const LChar* str, unsigned start, bool caseSensitive) const 248 { return caseSensitive ? find(str, start) : findIgnoringCase(str, start); } 249 size_t find(const String& str, unsigned start, bool caseSensitive) const 250 { return caseSensitive ? find(str, start) : findIgnoringCase(str, start); } 251 size_t reverseFind(const String& str, unsigned start, bool caseSensitive) const 252 { return caseSensitive ? reverseFind(str, start) : reverseFindIgnoringCase(str, start); } 253 254 Vector<UChar> charactersWithNullTermination() const; 255 unsigned copyTo(UChar* buffer, unsigned pos, unsigned maxLength) const; 256 257 template<size_t inlineCapacity> 258 void appendTo(Vector<UChar, inlineCapacity>&, unsigned pos = 0, unsigned len = UINT_MAX) const; 259 260 template<typename BufferType> 261 void appendTo(BufferType&, unsigned pos = 0, unsigned len = UINT_MAX) const; 262 263 template<size_t inlineCapacity> 264 void prependTo(Vector<UChar, inlineCapacity>&, unsigned pos = 0, unsigned len = UINT_MAX) const; 265 266 UChar32 characterStartingAt(unsigned) const; 267 268 bool contains(UChar c) const { return find(c) != kNotFound; } 269 bool contains(const LChar* str, bool caseSensitive = true) const { return find(str, 0, caseSensitive) != kNotFound; } 270 bool contains(const String& str, bool caseSensitive = true) const { return find(str, 0, caseSensitive) != kNotFound; } 271 272 bool startsWith(const String& s, bool caseSensitive = true) const 273 { return m_impl ? m_impl->startsWith(s.impl(), caseSensitive) : s.isEmpty(); } 274 bool startsWith(UChar character) const 275 { return m_impl ? m_impl->startsWith(character) : false; } 276 template<unsigned matchLength> 277 bool startsWith(const char (&prefix)[matchLength], bool caseSensitive = true) const 278 { return m_impl ? m_impl->startsWith<matchLength>(prefix, caseSensitive) : !matchLength; } 279 280 bool endsWith(const String& s, bool caseSensitive = true) const 281 { return m_impl ? m_impl->endsWith(s.impl(), caseSensitive) : s.isEmpty(); } 282 bool endsWith(UChar character) const 283 { return m_impl ? m_impl->endsWith(character) : false; } 284 template<unsigned matchLength> 285 bool endsWith(const char (&prefix)[matchLength], bool caseSensitive = true) const 286 { return m_impl ? m_impl->endsWith<matchLength>(prefix, caseSensitive) : !matchLength; } 287 288 void append(const String&); 289 void append(LChar); 290 void append(char c) { append(static_cast<LChar>(c)); }; 291 void append(UChar); 292 void append(const LChar*, unsigned length); 293 void append(const UChar*, unsigned length); 294 void insert(const String&, unsigned pos); 295 void insert(const LChar*, unsigned length, unsigned pos); 296 void insert(const UChar*, unsigned length, unsigned pos); 297 298 String& replace(UChar a, UChar b) { if (m_impl) m_impl = m_impl->replace(a, b); return *this; } 299 String& replace(UChar a, const String& b) { if (m_impl) m_impl = m_impl->replace(a, b.impl()); return *this; } 300 String& replace(const String& a, const String& b) { if (m_impl) m_impl = m_impl->replace(a.impl(), b.impl()); return *this; } 301 String& replace(unsigned index, unsigned len, const String& b) { if (m_impl) m_impl = m_impl->replace(index, len, b.impl()); return *this; } 302 303 template<unsigned charactersCount> 304 ALWAYS_INLINE String& replaceWithLiteral(UChar a, const char (&characters)[charactersCount]) 305 { 306 if (m_impl) 307 m_impl = m_impl->replace(a, characters, charactersCount - 1); 308 309 return *this; 310 } 311 312 void fill(UChar c) { if (m_impl) m_impl = m_impl->fill(c); } 313 314 void ensure16Bit(); 315 316 void truncate(unsigned len); 317 void remove(unsigned pos, int len = 1); 318 319 String substring(unsigned pos, unsigned len = UINT_MAX) const; 320 String left(unsigned len) const { return substring(0, len); } 321 String right(unsigned len) const { return substring(length() - len, len); } 322 323 StringView createView() const { return StringView(impl()); } 324 StringView createView(unsigned offset, unsigned length) const { return StringView(impl(), offset, length); } 325 326 // Returns a lowercase/uppercase version of the string 327 String lower() const; 328 String upper() const; 329 330 String lower(const AtomicString& localeIdentifier) const; 331 String upper(const AtomicString& localeIdentifier) const; 332 333 String stripWhiteSpace() const; 334 String stripWhiteSpace(IsWhiteSpaceFunctionPtr) const; 335 String simplifyWhiteSpace(StripBehavior stripBehavior = StripExtraWhiteSpace) const; 336 String simplifyWhiteSpace(IsWhiteSpaceFunctionPtr, StripBehavior stripBehavior = StripExtraWhiteSpace) const; 337 338 String removeCharacters(CharacterMatchFunctionPtr) const; 339 template<bool isSpecialCharacter(UChar)> bool isAllSpecialCharacters() const; 340 341 // Return the string with case folded for case insensitive comparison. 342 String foldCase() const; 343 344 static String format(const char *, ...) WTF_ATTRIBUTE_PRINTF(1, 2); 345 346 // Returns an uninitialized string. The characters needs to be written 347 // into the buffer returned in data before the returned string is used. 348 // Failure to do this will have unpredictable results. 349 static String createUninitialized(unsigned length, UChar*& data) { return StringImpl::createUninitialized(length, data); } 350 static String createUninitialized(unsigned length, LChar*& data) { return StringImpl::createUninitialized(length, data); } 351 352 void split(const String& separator, bool allowEmptyEntries, Vector<String>& result) const; 353 void split(const String& separator, Vector<String>& result) const 354 { 355 split(separator, false, result); 356 } 357 void split(UChar separator, bool allowEmptyEntries, Vector<String>& result) const; 358 void split(UChar separator, Vector<String>& result) const 359 { 360 split(separator, false, result); 361 } 362 363 int toIntStrict(bool* ok = 0, int base = 10) const; 364 unsigned toUIntStrict(bool* ok = 0, int base = 10) const; 365 int64_t toInt64Strict(bool* ok = 0, int base = 10) const; 366 uint64_t toUInt64Strict(bool* ok = 0, int base = 10) const; 367 intptr_t toIntPtrStrict(bool* ok = 0, int base = 10) const; 368 369 int toInt(bool* ok = 0) const; 370 unsigned toUInt(bool* ok = 0) const; 371 int64_t toInt64(bool* ok = 0) const; 372 uint64_t toUInt64(bool* ok = 0) const; 373 intptr_t toIntPtr(bool* ok = 0) const; 374 375 // FIXME: Like the strict functions above, these give false for "ok" when there is trailing garbage. 376 // Like the non-strict functions above, these return the value when there is trailing garbage. 377 // It would be better if these were more consistent with the above functions instead. 378 double toDouble(bool* ok = 0) const; 379 float toFloat(bool* ok = 0) const; 380 381 bool percentage(int& percentage) const; 382 383 String isolatedCopy() const; 384 bool isSafeToSendToAnotherThread() const; 385 386 // Prevent Strings from being implicitly convertable to bool as it will be ambiguous on any platform that 387 // allows implicit conversion to another pointer type (e.g., Mac allows implicit conversion to NSString*). 388 typedef struct ImplicitConversionFromWTFStringToBoolDisallowedA* (String::*UnspecifiedBoolTypeA); 389 typedef struct ImplicitConversionFromWTFStringToBoolDisallowedB* (String::*UnspecifiedBoolTypeB); 390 operator UnspecifiedBoolTypeA() const; 391 operator UnspecifiedBoolTypeB() const; 392 393 #if USE(CF) 394 String(CFStringRef); 395 RetainPtr<CFStringRef> createCFString() const; 396 #endif 397 398 #ifdef __OBJC__ 399 String(NSString*); 400 401 // This conversion maps NULL to "", which loses the meaning of NULL, but we 402 // need this mapping because AppKit crashes when passed nil NSStrings. 403 operator NSString*() const { if (!m_impl) return @""; return *m_impl; } 404 #endif 405 406 static String make8BitFrom16BitSource(const UChar*, size_t); 407 template<size_t inlineCapacity> 408 static String make8BitFrom16BitSource(const Vector<UChar, inlineCapacity>& buffer) 409 { 410 return make8BitFrom16BitSource(buffer.data(), buffer.size()); 411 } 412 413 static String make16BitFrom8BitSource(const LChar*, size_t); 414 415 // String::fromUTF8 will return a null string if 416 // the input data contains invalid UTF-8 sequences. 417 static String fromUTF8(const LChar*, size_t); 418 static String fromUTF8(const LChar*); 419 static String fromUTF8(const char* s, size_t length) { return fromUTF8(reinterpret_cast<const LChar*>(s), length); }; 420 static String fromUTF8(const char* s) { return fromUTF8(reinterpret_cast<const LChar*>(s)); }; 421 static String fromUTF8(const CString&); 422 423 // 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. 424 static String fromUTF8WithLatin1Fallback(const LChar*, size_t); 425 static String fromUTF8WithLatin1Fallback(const char* s, size_t length) { return fromUTF8WithLatin1Fallback(reinterpret_cast<const LChar*>(s), length); }; 426 427 bool containsOnlyASCII() const; 428 bool containsOnlyLatin1() const; 429 bool containsOnlyWhitespace() const { return !m_impl || m_impl->containsOnlyWhitespace(); } 430 431 // Hash table deleted values, which are only constructed and never copied or destroyed. 432 String(WTF::HashTableDeletedValueType) : m_impl(WTF::HashTableDeletedValue) { } 433 bool isHashTableDeletedValue() const { return m_impl.isHashTableDeletedValue(); } 434 435 #ifndef NDEBUG 436 void show() const; 437 #endif 438 439 // Workaround for a compiler bug. Use operator[] instead. 440 UChar characterAt(unsigned index) const 441 { 442 if (!m_impl || index >= m_impl->length()) 443 return 0; 444 return (*m_impl)[index]; 445 } 446 447 private: 448 template <typename CharacterType> 449 void removeInternal(const CharacterType*, unsigned, int); 450 451 template <typename CharacterType> 452 void appendInternal(CharacterType); 453 454 RefPtr<StringImpl> m_impl; 455 }; 456 457 inline bool operator==(const String& a, const String& b) { return equal(a.impl(), b.impl()); } 458 inline bool operator==(const String& a, const LChar* b) { return equal(a.impl(), b); } 459 inline bool operator==(const String& a, const char* b) { return equal(a.impl(), reinterpret_cast<const LChar*>(b)); } 460 inline bool operator==(const LChar* a, const String& b) { return equal(a, b.impl()); } 461 inline bool operator==(const char* a, const String& b) { return equal(reinterpret_cast<const LChar*>(a), b.impl()); } 462 template<size_t inlineCapacity> 463 inline bool operator==(const Vector<char, inlineCapacity>& a, const String& b) { return equal(b.impl(), a.data(), a.size()); } 464 template<size_t inlineCapacity> 465 inline bool operator==(const String& a, const Vector<char, inlineCapacity>& b) { return b == a; } 466 467 468 inline bool operator!=(const String& a, const String& b) { return !equal(a.impl(), b.impl()); } 469 inline bool operator!=(const String& a, const LChar* b) { return !equal(a.impl(), b); } 470 inline bool operator!=(const String& a, const char* b) { return !equal(a.impl(), reinterpret_cast<const LChar*>(b)); } 471 inline bool operator!=(const LChar* a, const String& b) { return !equal(a, b.impl()); } 472 inline bool operator!=(const char* a, const String& b) { return !equal(reinterpret_cast<const LChar*>(a), b.impl()); } 473 template<size_t inlineCapacity> 474 inline bool operator!=(const Vector<char, inlineCapacity>& a, const String& b) { return !(a == b); } 475 template<size_t inlineCapacity> 476 inline bool operator!=(const String& a, const Vector<char, inlineCapacity>& b) { return b != a; } 477 478 inline bool equalIgnoringCase(const String& a, const String& b) { return equalIgnoringCase(a.impl(), b.impl()); } 479 inline bool equalIgnoringCase(const String& a, const LChar* b) { return equalIgnoringCase(a.impl(), b); } 480 inline bool equalIgnoringCase(const String& a, const char* b) { return equalIgnoringCase(a.impl(), reinterpret_cast<const LChar*>(b)); } 481 inline bool equalIgnoringCase(const LChar* a, const String& b) { return equalIgnoringCase(a, b.impl()); } 482 inline bool equalIgnoringCase(const char* a, const String& b) { return equalIgnoringCase(reinterpret_cast<const LChar*>(a), b.impl()); } 483 484 inline bool equalPossiblyIgnoringCase(const String& a, const String& b, bool ignoreCase) 485 { 486 return ignoreCase ? equalIgnoringCase(a, b) : (a == b); 487 } 488 489 inline bool equalIgnoringNullity(const String& a, const String& b) { return equalIgnoringNullity(a.impl(), b.impl()); } 490 491 template<size_t inlineCapacity> 492 inline bool equalIgnoringNullity(const Vector<UChar, inlineCapacity>& a, const String& b) { return equalIgnoringNullity(a, b.impl()); } 493 494 inline bool operator!(const String& str) { return str.isNull(); } 495 496 inline void swap(String& a, String& b) { a.swap(b); } 497 498 // Definitions of string operations 499 500 template<size_t inlineCapacity> 501 String::String(const Vector<UChar, inlineCapacity>& vector) 502 : m_impl(vector.size() ? StringImpl::create(vector.data(), vector.size()) : StringImpl::empty()) 503 { 504 } 505 506 template<> 507 inline const LChar* String::getCharacters<LChar>() const 508 { 509 ASSERT(is8Bit()); 510 return characters8(); 511 } 512 513 template<> 514 inline const UChar* String::getCharacters<UChar>() const 515 { 516 ASSERT(!is8Bit()); 517 return characters16(); 518 } 519 520 inline bool String::containsOnlyLatin1() const 521 { 522 if (isEmpty()) 523 return true; 524 525 if (is8Bit()) 526 return true; 527 528 const UChar* characters = characters16(); 529 UChar ored = 0; 530 for (size_t i = 0; i < m_impl->length(); ++i) 531 ored |= characters[i]; 532 return !(ored & 0xFF00); 533 } 534 535 536 #ifdef __OBJC__ 537 // This is for situations in WebKit where the long standing behavior has been 538 // "nil if empty", so we try to maintain longstanding behavior for the sake of 539 // entrenched clients 540 inline NSString* nsStringNilIfEmpty(const String& str) { return str.isEmpty() ? nil : (NSString*)str; } 541 #endif 542 543 inline bool String::containsOnlyASCII() const 544 { 545 if (isEmpty()) 546 return true; 547 548 if (is8Bit()) 549 return charactersAreAllASCII(characters8(), m_impl->length()); 550 551 return charactersAreAllASCII(characters16(), m_impl->length()); 552 } 553 554 WTF_EXPORT int codePointCompare(const String&, const String&); 555 556 inline bool codePointCompareLessThan(const String& a, const String& b) 557 { 558 return codePointCompare(a.impl(), b.impl()) < 0; 559 } 560 561 template<size_t inlineCapacity> 562 inline void append(Vector<UChar, inlineCapacity>& vector, const String& string) 563 { 564 unsigned length = string.length(); 565 if (!length) 566 return; 567 if (string.is8Bit()) { 568 const LChar* characters8 = string.characters8(); 569 vector.reserveCapacity(vector.size() + length); 570 for (size_t i = 0; i < length; ++i) 571 vector.uncheckedAppend(characters8[i]); 572 } else { 573 vector.append(string.characters16(), length); 574 } 575 } 576 577 template<typename CharacterType> 578 inline void appendNumber(Vector<CharacterType>& vector, unsigned char number) 579 { 580 int numberLength = number > 99 ? 3 : (number > 9 ? 2 : 1); 581 size_t vectorSize = vector.size(); 582 vector.grow(vectorSize + numberLength); 583 584 switch (numberLength) { 585 case 3: 586 vector[vectorSize + 2] = number % 10 + '0'; 587 number /= 10; 588 589 case 2: 590 vector[vectorSize + 1] = number % 10 + '0'; 591 number /= 10; 592 593 case 1: 594 vector[vectorSize] = number % 10 + '0'; 595 } 596 } 597 598 template<bool isSpecialCharacter(UChar), typename CharacterType> 599 inline bool isAllSpecialCharacters(const CharacterType* characters, size_t length) 600 { 601 for (size_t i = 0; i < length; ++i) { 602 if (!isSpecialCharacter(characters[i])) 603 return false; 604 } 605 return true; 606 } 607 608 template<bool isSpecialCharacter(UChar)> 609 inline bool String::isAllSpecialCharacters() const 610 { 611 size_t len = length(); 612 613 if (!len) 614 return true; 615 616 if (is8Bit()) 617 return WTF::isAllSpecialCharacters<isSpecialCharacter, LChar>(characters8(), len); 618 return WTF::isAllSpecialCharacters<isSpecialCharacter, UChar>(characters16(), len); 619 } 620 621 template<size_t inlineCapacity> 622 inline void String::appendTo(Vector<UChar, inlineCapacity>& result, unsigned pos, unsigned len) const 623 { 624 unsigned numberOfCharactersToCopy = std::min(len, length() - pos); 625 if (numberOfCharactersToCopy <= 0) 626 return; 627 result.reserveCapacity(result.size() + numberOfCharactersToCopy); 628 if (is8Bit()) { 629 const LChar* characters8 = m_impl->characters8(); 630 for (size_t i = 0; i < numberOfCharactersToCopy; ++i) 631 result.uncheckedAppend(characters8[pos + i]); 632 } else { 633 const UChar* characters16 = m_impl->characters16(); 634 result.append(characters16 + pos, numberOfCharactersToCopy); 635 } 636 } 637 638 template<typename BufferType> 639 inline void String::appendTo(BufferType& result, unsigned pos, unsigned len) const 640 { 641 unsigned numberOfCharactersToCopy = std::min(len, length() - pos); 642 if (numberOfCharactersToCopy <= 0) 643 return; 644 if (is8Bit()) 645 result.append(m_impl->characters8() + pos, numberOfCharactersToCopy); 646 else 647 result.append(m_impl->characters16() + pos, numberOfCharactersToCopy); 648 } 649 650 template<size_t inlineCapacity> 651 inline void String::prependTo(Vector<UChar, inlineCapacity>& result, unsigned pos, unsigned len) const 652 { 653 unsigned numberOfCharactersToCopy = std::min(len, length() - pos); 654 if (numberOfCharactersToCopy <= 0) 655 return; 656 if (is8Bit()) { 657 size_t oldSize = result.size(); 658 result.resize(oldSize + numberOfCharactersToCopy); 659 memmove(result.data() + numberOfCharactersToCopy, result.data(), oldSize * sizeof(UChar)); 660 StringImpl::copyChars(result.data(), m_impl->characters8() + pos, numberOfCharactersToCopy); 661 } else { 662 result.prepend(m_impl->characters16() + pos, numberOfCharactersToCopy); 663 } 664 } 665 666 // StringHash is the default hash for String 667 template<typename T> struct DefaultHash; 668 template<> struct DefaultHash<String> { 669 typedef StringHash Hash; 670 }; 671 672 template <> struct VectorTraits<String> : SimpleClassVectorTraits { 673 static const bool canCompareWithMemcmp = false; 674 }; 675 676 // Shared global empty string. 677 WTF_EXPORT const String& emptyString(); 678 679 } 680 681 using WTF::CString; 682 using WTF::KeepTrailingZeros; 683 using WTF::String; 684 using WTF::emptyString; 685 using WTF::append; 686 using WTF::appendNumber; 687 using WTF::charactersAreAllASCII; 688 using WTF::charactersToIntStrict; 689 using WTF::charactersToUIntStrict; 690 using WTF::charactersToInt64Strict; 691 using WTF::charactersToUInt64Strict; 692 using WTF::charactersToIntPtrStrict; 693 using WTF::charactersToInt; 694 using WTF::charactersToUInt; 695 using WTF::charactersToInt64; 696 using WTF::charactersToUInt64; 697 using WTF::charactersToIntPtr; 698 using WTF::charactersToDouble; 699 using WTF::charactersToFloat; 700 using WTF::equal; 701 using WTF::equalIgnoringCase; 702 using WTF::find; 703 using WTF::isAllSpecialCharacters; 704 using WTF::isSpaceOrNewline; 705 using WTF::reverseFind; 706 707 #include "wtf/text/AtomicString.h" 708 #endif 709
