1 /*
2 * Copyright (C) 2009, 2010, 2012, 2013 Apple Inc. All rights reserved.
3 * Copyright (C) 2012 Google Inc. All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 * 1. Redistributions of source code must retain the above copyright
9 * notice, this list of conditions and the following disclaimer.
10 * 2. Redistributions in binary form must reproduce the above copyright
11 * notice, this list of conditions and the following disclaimer in the
12 * documentation and/or other materials provided with the distribution.
13 *
14 * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY
15 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
16 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
17 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE INC. OR
18 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
19 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
20 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
21 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
22 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
23 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
24 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
25 */
26
27 #ifndef StringBuilder_h
28 #define StringBuilder_h
29
30 #include "wtf/WTFExport.h"
31 #include "wtf/text/AtomicString.h"
32 #include "wtf/text/WTFString.h"
33
34 namespace WTF {
35
36 class WTF_EXPORT StringBuilder {
37 // Disallow copying since it's expensive and we don't want code to do it by accident.
38 WTF_MAKE_NONCOPYABLE(StringBuilder);
39
40 public:
StringBuilder()41 StringBuilder()
42 : m_bufferCharacters8(0)
43 , m_length(0)
44 , m_is8Bit(true)
45 {
46 }
47
48 void append(const UChar*, unsigned);
49 void append(const LChar*, unsigned);
50
append(const char * characters,unsigned length)51 ALWAYS_INLINE void append(const char* characters, unsigned length) { append(reinterpret_cast<const LChar*>(characters), length); }
52
append(const String & string)53 void append(const String& string)
54 {
55 if (!string.length())
56 return;
57
58 // If we're appending to an empty string, and there is not a buffer (reserveCapacity has not been called)
59 // then just retain the string.
60 if (!m_length && !m_buffer) {
61 m_string = string;
62 m_length = string.length();
63 m_is8Bit = m_string.is8Bit();
64 return;
65 }
66
67 if (string.is8Bit())
68 append(string.characters8(), string.length());
69 else
70 append(string.characters16(), string.length());
71 }
72
append(const StringBuilder & other)73 void append(const StringBuilder& other)
74 {
75 if (!other.m_length)
76 return;
77
78 // If we're appending to an empty string, and there is not a buffer (reserveCapacity has not been called)
79 // then just retain the string.
80 if (!m_length && !m_buffer && !other.m_string.isNull()) {
81 m_string = other.m_string;
82 m_length = other.m_length;
83 return;
84 }
85
86 if (other.is8Bit())
87 append(other.characters8(), other.m_length);
88 else
89 append(other.characters16(), other.m_length);
90 }
91
append(const String & string,unsigned offset,unsigned length)92 void append(const String& string, unsigned offset, unsigned length)
93 {
94 if (!string.length())
95 return;
96
97 unsigned extent = offset + length;
98 if (extent < offset || extent > string.length())
99 return;
100
101 if (string.is8Bit())
102 append(string.characters8() + offset, length);
103 else
104 append(string.characters16() + offset, length);
105 }
106
append(const StringView & string)107 void append(const StringView& string)
108 {
109 if (!string.length())
110 return;
111
112 if (string.is8Bit())
113 append(string.characters8(), string.length());
114 else
115 append(string.characters16(), string.length());
116 }
117
append(const char * characters)118 void append(const char* characters)
119 {
120 if (characters)
121 append(characters, strlen(characters));
122 }
123
append(UChar c)124 void append(UChar c)
125 {
126 if (m_buffer && m_length < m_buffer->length() && m_string.isNull()) {
127 if (!m_is8Bit) {
128 m_bufferCharacters16[m_length++] = c;
129 return;
130 }
131
132 if (!(c & ~0xff)) {
133 m_bufferCharacters8[m_length++] = static_cast<LChar>(c);
134 return;
135 }
136 }
137 append(&c, 1);
138 }
139
append(LChar c)140 void append(LChar c)
141 {
142 if (m_buffer && m_length < m_buffer->length() && m_string.isNull()) {
143 if (m_is8Bit)
144 m_bufferCharacters8[m_length++] = c;
145 else
146 m_bufferCharacters16[m_length++] = c;
147 } else
148 append(&c, 1);
149 }
150
append(char c)151 void append(char c)
152 {
153 append(static_cast<LChar>(c));
154 }
155
append(UChar32 c)156 void append(UChar32 c)
157 {
158 if (U_IS_BMP(c)) {
159 append(static_cast<UChar>(c));
160 return;
161 }
162 append(U16_LEAD(c));
163 append(U16_TRAIL(c));
164 }
165
166 template<unsigned charactersCount>
appendLiteral(const char (& characters)[charactersCount])167 ALWAYS_INLINE void appendLiteral(const char (&characters)[charactersCount]) { append(characters, charactersCount - 1); }
168
169 void appendNumber(int);
170 void appendNumber(unsigned);
171 void appendNumber(long);
172 void appendNumber(unsigned long);
173 void appendNumber(long long);
174 void appendNumber(unsigned long long);
175
toString()176 String toString()
177 {
178 shrinkToFit();
179 if (m_string.isNull())
180 reifyString();
181 return m_string;
182 }
183
substring(unsigned position,unsigned length)184 String substring(unsigned position, unsigned length) const
185 {
186 if (!m_length)
187 return emptyString();
188 if (!m_string.isNull())
189 return m_string.substring(position, length);
190 return reifySubstring(position, length);
191 }
192
toAtomicString()193 AtomicString toAtomicString() const
194 {
195 if (!m_length)
196 return emptyAtom;
197
198 // If the buffer is sufficiently over-allocated, make a new AtomicString from a copy so its buffer is not so large.
199 if (canShrink()) {
200 if (is8Bit())
201 return AtomicString(characters8(), length());
202 return AtomicString(characters16(), length());
203 }
204
205 if (!m_string.isNull())
206 return AtomicString(m_string);
207
208 ASSERT(m_buffer);
209 return AtomicString(m_buffer.get(), 0, m_length);
210 }
211
length()212 unsigned length() const
213 {
214 return m_length;
215 }
216
isEmpty()217 bool isEmpty() const { return !m_length; }
218
219 void reserveCapacity(unsigned newCapacity);
220
capacity()221 unsigned capacity() const
222 {
223 return m_buffer ? m_buffer->length() : m_length;
224 }
225
226 void resize(unsigned newSize);
227
228 bool canShrink() const;
229
230 void shrinkToFit();
231
232 UChar operator[](unsigned i) const
233 {
234 ASSERT_WITH_SECURITY_IMPLICATION(i < m_length);
235 if (m_is8Bit)
236 return characters8()[i];
237 return characters16()[i];
238 }
239
characters8()240 const LChar* characters8() const
241 {
242 ASSERT(m_is8Bit);
243 if (!m_length)
244 return 0;
245 if (!m_string.isNull())
246 return m_string.characters8();
247 ASSERT(m_buffer);
248 return m_buffer->characters8();
249 }
250
characters16()251 const UChar* characters16() const
252 {
253 ASSERT(!m_is8Bit);
254 if (!m_length)
255 return 0;
256 if (!m_string.isNull())
257 return m_string.characters16();
258 ASSERT(m_buffer);
259 return m_buffer->characters16();
260 }
261
is8Bit()262 bool is8Bit() const { return m_is8Bit; }
263
clear()264 void clear()
265 {
266 m_length = 0;
267 m_string = String();
268 m_buffer = 0;
269 m_bufferCharacters8 = 0;
270 m_is8Bit = true;
271 }
272
swap(StringBuilder & stringBuilder)273 void swap(StringBuilder& stringBuilder)
274 {
275 std::swap(m_length, stringBuilder.m_length);
276 m_string.swap(stringBuilder.m_string);
277 m_buffer.swap(stringBuilder.m_buffer);
278 std::swap(m_is8Bit, stringBuilder.m_is8Bit);
279 std::swap(m_bufferCharacters8, stringBuilder.m_bufferCharacters8);
280 }
281
282 private:
283 void allocateBuffer(const LChar* currentCharacters, unsigned requiredLength);
284 void allocateBuffer(const UChar* currentCharacters, unsigned requiredLength);
285 void allocateBufferUpConvert(const LChar* currentCharacters, unsigned requiredLength);
286 template <typename CharType>
287 void reallocateBuffer(unsigned requiredLength);
288 template <typename CharType>
289 ALWAYS_INLINE CharType* appendUninitialized(unsigned length);
290 template <typename CharType>
291 CharType* appendUninitializedSlow(unsigned length);
292 template <typename CharType>
293 ALWAYS_INLINE CharType * getBufferCharacters();
294 void reifyString();
295 String reifySubstring(unsigned position, unsigned length) const;
296
297 String m_string; // Pointers first: crbug.com/232031
298 RefPtr<StringImpl> m_buffer;
299 union {
300 LChar* m_bufferCharacters8;
301 UChar* m_bufferCharacters16;
302 };
303 unsigned m_length;
304 bool m_is8Bit;
305 };
306
307 template <>
308 ALWAYS_INLINE LChar* StringBuilder::getBufferCharacters<LChar>()
309 {
310 ASSERT(m_is8Bit);
311 return m_bufferCharacters8;
312 }
313
314 template <>
315 ALWAYS_INLINE UChar* StringBuilder::getBufferCharacters<UChar>()
316 {
317 ASSERT(!m_is8Bit);
318 return m_bufferCharacters16;
319 }
320
321 template <typename CharType>
equal(const StringBuilder & s,const CharType * buffer,unsigned length)322 bool equal(const StringBuilder& s, const CharType* buffer, unsigned length)
323 {
324 if (s.length() != length)
325 return false;
326
327 if (s.is8Bit())
328 return equal(s.characters8(), buffer, length);
329
330 return equal(s.characters16(), buffer, length);
331 }
332
333 template<typename CharType>
equalIgnoringCase(const StringBuilder & s,const CharType * buffer,unsigned length)334 bool equalIgnoringCase(const StringBuilder& s, const CharType* buffer, unsigned length)
335 {
336 if (s.length() != length)
337 return false;
338
339 if (s.is8Bit())
340 return equalIgnoringCase(s.characters8(), buffer, length);
341
342 return equalIgnoringCase(s.characters16(), buffer, length);
343 }
344
equalIgnoringCase(const StringBuilder & s,const char * string)345 inline bool equalIgnoringCase(const StringBuilder& s, const char* string)
346 {
347 return equalIgnoringCase(s, reinterpret_cast<const LChar*>(string), strlen(string));
348 }
349
350 template <typename StringType>
equal(const StringBuilder & a,const StringType & b)351 bool equal(const StringBuilder& a, const StringType& b)
352 {
353 if (a.length() != b.length())
354 return false;
355
356 if (!a.length())
357 return true;
358
359 if (a.is8Bit()) {
360 if (b.is8Bit())
361 return equal(a.characters8(), b.characters8(), a.length());
362 return equal(a.characters8(), b.characters16(), a.length());
363 }
364
365 if (b.is8Bit())
366 return equal(a.characters16(), b.characters8(), a.length());
367 return equal(a.characters16(), b.characters16(), a.length());
368 }
369
370 template <typename StringType>
equalIgnoringCase(const StringBuilder & a,const StringType & b)371 bool equalIgnoringCase(const StringBuilder& a, const StringType& b)
372 {
373 if (a.length() != b.length())
374 return false;
375
376 if (!a.length())
377 return true;
378
379 if (a.is8Bit()) {
380 if (b.is8Bit())
381 return equalIgnoringCase(a.characters8(), b.characters8(), a.length());
382 return equalIgnoringCase(a.characters8(), b.characters16(), a.length());
383 }
384
385 if (b.is8Bit())
386 return equalIgnoringCase(a.characters16(), b.characters8(), a.length());
387 return equalIgnoringCase(a.characters16(), b.characters16(), a.length());
388 }
389
390 inline bool operator==(const StringBuilder& a, const StringBuilder& b) { return equal(a, b); }
391 inline bool operator!=(const StringBuilder& a, const StringBuilder& b) { return !equal(a, b); }
392 inline bool operator==(const StringBuilder& a, const String& b) { return equal(a, b); }
393 inline bool operator!=(const StringBuilder& a, const String& b) { return !equal(a, b); }
394 inline bool operator==(const String& a, const StringBuilder& b) { return equal(b, a); }
395 inline bool operator!=(const String& a, const StringBuilder& b) { return !equal(b, a); }
396
397 } // namespace WTF
398
399 using WTF::StringBuilder;
400
401 #endif // StringBuilder_h
402