OLD | NEW |
| (Empty) |
1 /* | |
2 ********************************************************************** | |
3 * Copyright (C) 1998-2010, International Business Machines | |
4 * Corporation and others. All Rights Reserved. | |
5 ********************************************************************** | |
6 * | |
7 * File unistr.h | |
8 * | |
9 * Modification History: | |
10 * | |
11 * Date Name Description | |
12 * 09/25/98 stephen Creation. | |
13 * 11/11/98 stephen Changed per 11/9 code review. | |
14 * 04/20/99 stephen Overhauled per 4/16 code review. | |
15 * 11/18/99 aliu Made to inherit from Replaceable. Added method | |
16 * handleReplaceBetween(); other methods unchanged. | |
17 * 06/25/01 grhoten Remove dependency on iostream. | |
18 ****************************************************************************** | |
19 */ | |
20 | |
21 #ifndef UNISTR_H | |
22 #define UNISTR_H | |
23 | |
24 /** | |
25 * \file | |
26 * \brief C++ API: Unicode String | |
27 */ | |
28 | |
29 #include "unicode/utypes.h" | |
30 #include "unicode/rep.h" | |
31 #include "unicode/std_string.h" | |
32 #include "unicode/stringpiece.h" | |
33 #include "unicode/bytestream.h" | |
34 | |
35 struct UConverter; // unicode/ucnv.h | |
36 class StringThreadTest; | |
37 | |
38 #ifndef U_COMPARE_CODE_POINT_ORDER | |
39 /* see also ustring.h and unorm.h */ | |
40 /** | |
41 * Option bit for u_strCaseCompare, u_strcasecmp, unorm_compare, etc: | |
42 * Compare strings in code point order instead of code unit order. | |
43 * @stable ICU 2.2 | |
44 */ | |
45 #define U_COMPARE_CODE_POINT_ORDER 0x8000 | |
46 #endif | |
47 | |
48 #ifndef USTRING_H | |
49 /** | |
50 * \ingroup ustring_ustrlen | |
51 */ | |
52 U_STABLE int32_t U_EXPORT2 | |
53 u_strlen(const UChar *s); | |
54 #endif | |
55 | |
56 U_NAMESPACE_BEGIN | |
57 | |
58 class Locale; // unicode/locid.h | |
59 class StringCharacterIterator; | |
60 class BreakIterator; // unicode/brkiter.h | |
61 | |
62 /* The <iostream> include has been moved to unicode/ustream.h */ | |
63 | |
64 /** | |
65 * Constant to be used in the UnicodeString(char *, int32_t, EInvariant) constru
ctor | |
66 * which constructs a Unicode string from an invariant-character char * string. | |
67 * About invariant characters see utypes.h. | |
68 * This constructor has no runtime dependency on conversion code and is | |
69 * therefore recommended over ones taking a charset name string | |
70 * (where the empty string "" indicates invariant-character conversion). | |
71 * | |
72 * @stable ICU 3.2 | |
73 */ | |
74 #define US_INV U_NAMESPACE_QUALIFIER UnicodeString::kInvariant | |
75 | |
76 /** | |
77 * Unicode String literals in C++. | |
78 * Dependent on the platform properties, different UnicodeString | |
79 * constructors should be used to create a UnicodeString object from | |
80 * a string literal. | |
81 * The macros are defined for maximum performance. | |
82 * They work only for strings that contain "invariant characters", i.e., | |
83 * only latin letters, digits, and some punctuation. | |
84 * See utypes.h for details. | |
85 * | |
86 * The string parameter must be a C string literal. | |
87 * The length of the string, not including the terminating | |
88 * <code>NUL</code>, must be specified as a constant. | |
89 * The U_STRING_DECL macro should be invoked exactly once for one | |
90 * such string variable before it is used. | |
91 * @stable ICU 2.0 | |
92 */ | |
93 #if defined(U_DECLARE_UTF16) | |
94 # define UNICODE_STRING(cs, _length) U_NAMESPACE_QUALIFIER UnicodeString(TRUE,
(const UChar *)U_DECLARE_UTF16(cs), _length) | |
95 #elif U_SIZEOF_WCHAR_T==U_SIZEOF_UCHAR && (U_CHARSET_FAMILY==U_ASCII_FAMILY || (
U_SIZEOF_UCHAR == 2 && defined(U_WCHAR_IS_UTF16))) | |
96 # define UNICODE_STRING(cs, _length) U_NAMESPACE_QUALIFIER UnicodeString(TRUE,
(const UChar *)L ## cs, _length) | |
97 #elif U_SIZEOF_UCHAR==1 && U_CHARSET_FAMILY==U_ASCII_FAMILY | |
98 # define UNICODE_STRING(cs, _length) U_NAMESPACE_QUALIFIER UnicodeString(TRUE,
(const UChar *)cs, _length) | |
99 #else | |
100 # define UNICODE_STRING(cs, _length) U_NAMESPACE_QUALIFIER UnicodeString(cs, _
length, US_INV) | |
101 #endif | |
102 | |
103 /** | |
104 * Unicode String literals in C++. | |
105 * Dependent on the platform properties, different UnicodeString | |
106 * constructors should be used to create a UnicodeString object from | |
107 * a string literal. | |
108 * The macros are defined for improved performance. | |
109 * They work only for strings that contain "invariant characters", i.e., | |
110 * only latin letters, digits, and some punctuation. | |
111 * See utypes.h for details. | |
112 * | |
113 * The string parameter must be a C string literal. | |
114 * @stable ICU 2.0 | |
115 */ | |
116 #define UNICODE_STRING_SIMPLE(cs) UNICODE_STRING(cs, -1) | |
117 | |
118 /** | |
119 * UnicodeString is a string class that stores Unicode characters directly and p
rovides | |
120 * similar functionality as the Java String and StringBuffer classes. | |
121 * It is a concrete implementation of the abstract class Replaceable (for transl
iteration). | |
122 * | |
123 * The UnicodeString class is not suitable for subclassing. | |
124 * | |
125 * <p>For an overview of Unicode strings in C and C++ see the | |
126 * <a href="http://icu-project.org/userguide/strings.html">User Guide Strings ch
apter</a>.</p> | |
127 * | |
128 * <p>In ICU, a Unicode string consists of 16-bit Unicode <em>code units</em>. | |
129 * A Unicode character may be stored with either one code unit | |
130 * (the most common case) or with a matched pair of special code units | |
131 * ("surrogates"). The data type for code units is UChar. | |
132 * For single-character handling, a Unicode character code <em>point</em> is a v
alue | |
133 * in the range 0..0x10ffff. ICU uses the UChar32 type for code points.</p> | |
134 * | |
135 * <p>Indexes and offsets into and lengths of strings always count code units, n
ot code points. | |
136 * This is the same as with multi-byte char* strings in traditional string handl
ing. | |
137 * Operations on partial strings typically do not test for code point boundaries
. | |
138 * If necessary, the user needs to take care of such boundaries by testing for t
he code unit | |
139 * values or by using functions like | |
140 * UnicodeString::getChar32Start() and UnicodeString::getChar32Limit() | |
141 * (or, in C, the equivalent macros U16_SET_CP_START() and U16_SET_CP_LIMIT(), s
ee utf.h).</p> | |
142 * | |
143 * UnicodeString methods are more lenient with regard to input parameter values | |
144 * than other ICU APIs. In particular: | |
145 * - If indexes are out of bounds for a UnicodeString object | |
146 * (<0 or >length()) then they are "pinned" to the nearest boundary. | |
147 * - If primitive string pointer values (e.g., const UChar * or char *) | |
148 * for input strings are NULL, then those input string parameters are treated | |
149 * as if they pointed to an empty string. | |
150 * However, this is <em>not</em> the case for char * parameters for charset na
mes | |
151 * or other IDs. | |
152 * - Most UnicodeString methods do not take a UErrorCode parameter because | |
153 * there are usually very few opportunities for failure other than a shortage | |
154 * of memory, error codes in low-level C++ string methods would be inconvenien
t, | |
155 * and the error code as the last parameter (ICU convention) would prevent | |
156 * the use of default parameter values. | |
157 * Instead, such methods set the UnicodeString into a "bogus" state | |
158 * (see isBogus()) if an error occurs. | |
159 * | |
160 * In string comparisons, two UnicodeString objects that are both "bogus" | |
161 * compare equal (to be transitive and prevent endless loops in sorting), | |
162 * and a "bogus" string compares less than any non-"bogus" one. | |
163 * | |
164 * Const UnicodeString methods are thread-safe. Multiple threads can use | |
165 * const methods on the same UnicodeString object simultaneously, | |
166 * but non-const methods must not be called concurrently (in multiple threads) | |
167 * with any other (const or non-const) methods. | |
168 * | |
169 * Similarly, const UnicodeString & parameters are thread-safe. | |
170 * One object may be passed in as such a parameter concurrently in multiple thre
ads. | |
171 * This includes the const UnicodeString & parameters for | |
172 * copy construction, assignment, and cloning. | |
173 * | |
174 * <p>UnicodeString uses several storage methods. | |
175 * String contents can be stored inside the UnicodeString object itself, | |
176 * in an allocated and shared buffer, or in an outside buffer that is "aliased". | |
177 * Most of this is done transparently, but careful aliasing in particular provid
es | |
178 * significant performance improvements. | |
179 * Also, the internal buffer is accessible via special functions. | |
180 * For details see the | |
181 * <a href="http://icu-project.org/userguide/strings.html">User Guide Strings ch
apter</a>.</p> | |
182 * | |
183 * @see utf.h | |
184 * @see CharacterIterator | |
185 * @stable ICU 2.0 | |
186 */ | |
187 class U_COMMON_API UnicodeString : public Replaceable | |
188 { | |
189 public: | |
190 | |
191 /** | |
192 * Constant to be used in the UnicodeString(char *, int32_t, EInvariant) const
ructor | |
193 * which constructs a Unicode string from an invariant-character char * string
. | |
194 * Use the macro US_INV instead of the full qualification for this value. | |
195 * | |
196 * @see US_INV | |
197 * @stable ICU 3.2 | |
198 */ | |
199 enum EInvariant { | |
200 /** | |
201 * @see EInvariant | |
202 * @stable ICU 3.2 | |
203 */ | |
204 kInvariant | |
205 }; | |
206 | |
207 //======================================== | |
208 // Read-only operations | |
209 //======================================== | |
210 | |
211 /* Comparison - bitwise only - for international comparison use collation */ | |
212 | |
213 /** | |
214 * Equality operator. Performs only bitwise comparison. | |
215 * @param text The UnicodeString to compare to this one. | |
216 * @return TRUE if <TT>text</TT> contains the same characters as this one, | |
217 * FALSE otherwise. | |
218 * @stable ICU 2.0 | |
219 */ | |
220 inline UBool operator== (const UnicodeString& text) const; | |
221 | |
222 /** | |
223 * Inequality operator. Performs only bitwise comparison. | |
224 * @param text The UnicodeString to compare to this one. | |
225 * @return FALSE if <TT>text</TT> contains the same characters as this one, | |
226 * TRUE otherwise. | |
227 * @stable ICU 2.0 | |
228 */ | |
229 inline UBool operator!= (const UnicodeString& text) const; | |
230 | |
231 /** | |
232 * Greater than operator. Performs only bitwise comparison. | |
233 * @param text The UnicodeString to compare to this one. | |
234 * @return TRUE if the characters in this are bitwise | |
235 * greater than the characters in <code>text</code>, FALSE otherwise | |
236 * @stable ICU 2.0 | |
237 */ | |
238 inline UBool operator> (const UnicodeString& text) const; | |
239 | |
240 /** | |
241 * Less than operator. Performs only bitwise comparison. | |
242 * @param text The UnicodeString to compare to this one. | |
243 * @return TRUE if the characters in this are bitwise | |
244 * less than the characters in <code>text</code>, FALSE otherwise | |
245 * @stable ICU 2.0 | |
246 */ | |
247 inline UBool operator< (const UnicodeString& text) const; | |
248 | |
249 /** | |
250 * Greater than or equal operator. Performs only bitwise comparison. | |
251 * @param text The UnicodeString to compare to this one. | |
252 * @return TRUE if the characters in this are bitwise | |
253 * greater than or equal to the characters in <code>text</code>, FALSE otherwi
se | |
254 * @stable ICU 2.0 | |
255 */ | |
256 inline UBool operator>= (const UnicodeString& text) const; | |
257 | |
258 /** | |
259 * Less than or equal operator. Performs only bitwise comparison. | |
260 * @param text The UnicodeString to compare to this one. | |
261 * @return TRUE if the characters in this are bitwise | |
262 * less than or equal to the characters in <code>text</code>, FALSE otherwise | |
263 * @stable ICU 2.0 | |
264 */ | |
265 inline UBool operator<= (const UnicodeString& text) const; | |
266 | |
267 /** | |
268 * Compare the characters bitwise in this UnicodeString to | |
269 * the characters in <code>text</code>. | |
270 * @param text The UnicodeString to compare to this one. | |
271 * @return The result of bitwise character comparison: 0 if this | |
272 * contains the same characters as <code>text</code>, -1 if the characters in | |
273 * this are bitwise less than the characters in <code>text</code>, +1 if the | |
274 * characters in this are bitwise greater than the characters | |
275 * in <code>text</code>. | |
276 * @stable ICU 2.0 | |
277 */ | |
278 inline int8_t compare(const UnicodeString& text) const; | |
279 | |
280 /** | |
281 * Compare the characters bitwise in the range | |
282 * [<TT>start</TT>, <TT>start + length</TT>) with the characters | |
283 * in <TT>text</TT> | |
284 * @param start the offset at which the compare operation begins | |
285 * @param length the number of characters of text to compare. | |
286 * @param text the other text to be compared against this string. | |
287 * @return The result of bitwise character comparison: 0 if this | |
288 * contains the same characters as <code>text</code>, -1 if the characters in | |
289 * this are bitwise less than the characters in <code>text</code>, +1 if the | |
290 * characters in this are bitwise greater than the characters | |
291 * in <code>text</code>. | |
292 * @stable ICU 2.0 | |
293 */ | |
294 inline int8_t compare(int32_t start, | |
295 int32_t length, | |
296 const UnicodeString& text) const; | |
297 | |
298 /** | |
299 * Compare the characters bitwise in the range | |
300 * [<TT>start</TT>, <TT>start + length</TT>) with the characters | |
301 * in <TT>srcText</TT> in the range | |
302 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). | |
303 * @param start the offset at which the compare operation begins | |
304 * @param length the number of characters in this to compare. | |
305 * @param srcText the text to be compared | |
306 * @param srcStart the offset into <TT>srcText</TT> to start comparison | |
307 * @param srcLength the number of characters in <TT>src</TT> to compare | |
308 * @return The result of bitwise character comparison: 0 if this | |
309 * contains the same characters as <code>srcText</code>, -1 if the characters
in | |
310 * this are bitwise less than the characters in <code>srcText</code>, +1 if th
e | |
311 * characters in this are bitwise greater than the characters | |
312 * in <code>srcText</code>. | |
313 * @stable ICU 2.0 | |
314 */ | |
315 inline int8_t compare(int32_t start, | |
316 int32_t length, | |
317 const UnicodeString& srcText, | |
318 int32_t srcStart, | |
319 int32_t srcLength) const; | |
320 | |
321 /** | |
322 * Compare the characters bitwise in this UnicodeString with the first | |
323 * <TT>srcLength</TT> characters in <TT>srcChars</TT>. | |
324 * @param srcChars The characters to compare to this UnicodeString. | |
325 * @param srcLength the number of characters in <TT>srcChars</TT> to compare | |
326 * @return The result of bitwise character comparison: 0 if this | |
327 * contains the same characters as <code>srcChars</code>, -1 if the characters
in | |
328 * this are bitwise less than the characters in <code>srcChars</code>, +1 if t
he | |
329 * characters in this are bitwise greater than the characters | |
330 * in <code>srcChars</code>. | |
331 * @stable ICU 2.0 | |
332 */ | |
333 inline int8_t compare(const UChar *srcChars, | |
334 int32_t srcLength) const; | |
335 | |
336 /** | |
337 * Compare the characters bitwise in the range | |
338 * [<TT>start</TT>, <TT>start + length</TT>) with the first | |
339 * <TT>length</TT> characters in <TT>srcChars</TT> | |
340 * @param start the offset at which the compare operation begins | |
341 * @param length the number of characters to compare. | |
342 * @param srcChars the characters to be compared | |
343 * @return The result of bitwise character comparison: 0 if this | |
344 * contains the same characters as <code>srcChars</code>, -1 if the characters
in | |
345 * this are bitwise less than the characters in <code>srcChars</code>, +1 if t
he | |
346 * characters in this are bitwise greater than the characters | |
347 * in <code>srcChars</code>. | |
348 * @stable ICU 2.0 | |
349 */ | |
350 inline int8_t compare(int32_t start, | |
351 int32_t length, | |
352 const UChar *srcChars) const; | |
353 | |
354 /** | |
355 * Compare the characters bitwise in the range | |
356 * [<TT>start</TT>, <TT>start + length</TT>) with the characters | |
357 * in <TT>srcChars</TT> in the range | |
358 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). | |
359 * @param start the offset at which the compare operation begins | |
360 * @param length the number of characters in this to compare | |
361 * @param srcChars the characters to be compared | |
362 * @param srcStart the offset into <TT>srcChars</TT> to start comparison | |
363 * @param srcLength the number of characters in <TT>srcChars</TT> to compare | |
364 * @return The result of bitwise character comparison: 0 if this | |
365 * contains the same characters as <code>srcChars</code>, -1 if the characters
in | |
366 * this are bitwise less than the characters in <code>srcChars</code>, +1 if t
he | |
367 * characters in this are bitwise greater than the characters | |
368 * in <code>srcChars</code>. | |
369 * @stable ICU 2.0 | |
370 */ | |
371 inline int8_t compare(int32_t start, | |
372 int32_t length, | |
373 const UChar *srcChars, | |
374 int32_t srcStart, | |
375 int32_t srcLength) const; | |
376 | |
377 /** | |
378 * Compare the characters bitwise in the range | |
379 * [<TT>start</TT>, <TT>limit</TT>) with the characters | |
380 * in <TT>srcText</TT> in the range | |
381 * [<TT>srcStart</TT>, <TT>srcLimit</TT>). | |
382 * @param start the offset at which the compare operation begins | |
383 * @param limit the offset immediately following the compare operation | |
384 * @param srcText the text to be compared | |
385 * @param srcStart the offset into <TT>srcText</TT> to start comparison | |
386 * @param srcLimit the offset into <TT>srcText</TT> to limit comparison | |
387 * @return The result of bitwise character comparison: 0 if this | |
388 * contains the same characters as <code>srcText</code>, -1 if the characters
in | |
389 * this are bitwise less than the characters in <code>srcText</code>, +1 if th
e | |
390 * characters in this are bitwise greater than the characters | |
391 * in <code>srcText</code>. | |
392 * @stable ICU 2.0 | |
393 */ | |
394 inline int8_t compareBetween(int32_t start, | |
395 int32_t limit, | |
396 const UnicodeString& srcText, | |
397 int32_t srcStart, | |
398 int32_t srcLimit) const; | |
399 | |
400 /** | |
401 * Compare two Unicode strings in code point order. | |
402 * The result may be different from the results of compare(), operator<, etc. | |
403 * if supplementary characters are present: | |
404 * | |
405 * In UTF-16, supplementary characters (with code points U+10000 and above) ar
e | |
406 * stored with pairs of surrogate code units. These have values from 0xd800 to
0xdfff, | |
407 * which means that they compare as less than some other BMP characters like U
+feff. | |
408 * This function compares Unicode strings in code point order. | |
409 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired su
rrogates), then the result is not defined. | |
410 * | |
411 * @param text Another string to compare this one to. | |
412 * @return a negative/zero/positive integer corresponding to whether | |
413 * this string is less than/equal to/greater than the second one | |
414 * in code point order | |
415 * @stable ICU 2.0 | |
416 */ | |
417 inline int8_t compareCodePointOrder(const UnicodeString& text) const; | |
418 | |
419 /** | |
420 * Compare two Unicode strings in code point order. | |
421 * The result may be different from the results of compare(), operator<, etc. | |
422 * if supplementary characters are present: | |
423 * | |
424 * In UTF-16, supplementary characters (with code points U+10000 and above) ar
e | |
425 * stored with pairs of surrogate code units. These have values from 0xd800 to
0xdfff, | |
426 * which means that they compare as less than some other BMP characters like U
+feff. | |
427 * This function compares Unicode strings in code point order. | |
428 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired su
rrogates), then the result is not defined. | |
429 * | |
430 * @param start The start offset in this string at which the compare operation
begins. | |
431 * @param length The number of code units from this string to compare. | |
432 * @param srcText Another string to compare this one to. | |
433 * @return a negative/zero/positive integer corresponding to whether | |
434 * this string is less than/equal to/greater than the second one | |
435 * in code point order | |
436 * @stable ICU 2.0 | |
437 */ | |
438 inline int8_t compareCodePointOrder(int32_t start, | |
439 int32_t length, | |
440 const UnicodeString& srcText) const; | |
441 | |
442 /** | |
443 * Compare two Unicode strings in code point order. | |
444 * The result may be different from the results of compare(), operator<, etc. | |
445 * if supplementary characters are present: | |
446 * | |
447 * In UTF-16, supplementary characters (with code points U+10000 and above) ar
e | |
448 * stored with pairs of surrogate code units. These have values from 0xd800 to
0xdfff, | |
449 * which means that they compare as less than some other BMP characters like U
+feff. | |
450 * This function compares Unicode strings in code point order. | |
451 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired su
rrogates), then the result is not defined. | |
452 * | |
453 * @param start The start offset in this string at which the compare operation
begins. | |
454 * @param length The number of code units from this string to compare. | |
455 * @param srcText Another string to compare this one to. | |
456 * @param srcStart The start offset in that string at which the compare operat
ion begins. | |
457 * @param srcLength The number of code units from that string to compare. | |
458 * @return a negative/zero/positive integer corresponding to whether | |
459 * this string is less than/equal to/greater than the second one | |
460 * in code point order | |
461 * @stable ICU 2.0 | |
462 */ | |
463 inline int8_t compareCodePointOrder(int32_t start, | |
464 int32_t length, | |
465 const UnicodeString& srcText, | |
466 int32_t srcStart, | |
467 int32_t srcLength) const; | |
468 | |
469 /** | |
470 * Compare two Unicode strings in code point order. | |
471 * The result may be different from the results of compare(), operator<, etc. | |
472 * if supplementary characters are present: | |
473 * | |
474 * In UTF-16, supplementary characters (with code points U+10000 and above) ar
e | |
475 * stored with pairs of surrogate code units. These have values from 0xd800 to
0xdfff, | |
476 * which means that they compare as less than some other BMP characters like U
+feff. | |
477 * This function compares Unicode strings in code point order. | |
478 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired su
rrogates), then the result is not defined. | |
479 * | |
480 * @param srcChars A pointer to another string to compare this one to. | |
481 * @param srcLength The number of code units from that string to compare. | |
482 * @return a negative/zero/positive integer corresponding to whether | |
483 * this string is less than/equal to/greater than the second one | |
484 * in code point order | |
485 * @stable ICU 2.0 | |
486 */ | |
487 inline int8_t compareCodePointOrder(const UChar *srcChars, | |
488 int32_t srcLength) const; | |
489 | |
490 /** | |
491 * Compare two Unicode strings in code point order. | |
492 * The result may be different from the results of compare(), operator<, etc. | |
493 * if supplementary characters are present: | |
494 * | |
495 * In UTF-16, supplementary characters (with code points U+10000 and above) ar
e | |
496 * stored with pairs of surrogate code units. These have values from 0xd800 to
0xdfff, | |
497 * which means that they compare as less than some other BMP characters like U
+feff. | |
498 * This function compares Unicode strings in code point order. | |
499 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired su
rrogates), then the result is not defined. | |
500 * | |
501 * @param start The start offset in this string at which the compare operation
begins. | |
502 * @param length The number of code units from this string to compare. | |
503 * @param srcChars A pointer to another string to compare this one to. | |
504 * @return a negative/zero/positive integer corresponding to whether | |
505 * this string is less than/equal to/greater than the second one | |
506 * in code point order | |
507 * @stable ICU 2.0 | |
508 */ | |
509 inline int8_t compareCodePointOrder(int32_t start, | |
510 int32_t length, | |
511 const UChar *srcChars) const; | |
512 | |
513 /** | |
514 * Compare two Unicode strings in code point order. | |
515 * The result may be different from the results of compare(), operator<, etc. | |
516 * if supplementary characters are present: | |
517 * | |
518 * In UTF-16, supplementary characters (with code points U+10000 and above) ar
e | |
519 * stored with pairs of surrogate code units. These have values from 0xd800 to
0xdfff, | |
520 * which means that they compare as less than some other BMP characters like U
+feff. | |
521 * This function compares Unicode strings in code point order. | |
522 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired su
rrogates), then the result is not defined. | |
523 * | |
524 * @param start The start offset in this string at which the compare operation
begins. | |
525 * @param length The number of code units from this string to compare. | |
526 * @param srcChars A pointer to another string to compare this one to. | |
527 * @param srcStart The start offset in that string at which the compare operat
ion begins. | |
528 * @param srcLength The number of code units from that string to compare. | |
529 * @return a negative/zero/positive integer corresponding to whether | |
530 * this string is less than/equal to/greater than the second one | |
531 * in code point order | |
532 * @stable ICU 2.0 | |
533 */ | |
534 inline int8_t compareCodePointOrder(int32_t start, | |
535 int32_t length, | |
536 const UChar *srcChars, | |
537 int32_t srcStart, | |
538 int32_t srcLength) const; | |
539 | |
540 /** | |
541 * Compare two Unicode strings in code point order. | |
542 * The result may be different from the results of compare(), operator<, etc. | |
543 * if supplementary characters are present: | |
544 * | |
545 * In UTF-16, supplementary characters (with code points U+10000 and above) ar
e | |
546 * stored with pairs of surrogate code units. These have values from 0xd800 to
0xdfff, | |
547 * which means that they compare as less than some other BMP characters like U
+feff. | |
548 * This function compares Unicode strings in code point order. | |
549 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired su
rrogates), then the result is not defined. | |
550 * | |
551 * @param start The start offset in this string at which the compare operation
begins. | |
552 * @param limit The offset after the last code unit from this string to compar
e. | |
553 * @param srcText Another string to compare this one to. | |
554 * @param srcStart The start offset in that string at which the compare operat
ion begins. | |
555 * @param srcLimit The offset after the last code unit from that string to com
pare. | |
556 * @return a negative/zero/positive integer corresponding to whether | |
557 * this string is less than/equal to/greater than the second one | |
558 * in code point order | |
559 * @stable ICU 2.0 | |
560 */ | |
561 inline int8_t compareCodePointOrderBetween(int32_t start, | |
562 int32_t limit, | |
563 const UnicodeString& srcText, | |
564 int32_t srcStart, | |
565 int32_t srcLimit) const; | |
566 | |
567 /** | |
568 * Compare two strings case-insensitively using full case folding. | |
569 * This is equivalent to this->foldCase(options).compare(text.foldCase(options
)). | |
570 * | |
571 * @param text Another string to compare this one to. | |
572 * @param options A bit set of options: | |
573 * - U_FOLD_CASE_DEFAULT or 0 is used for default options: | |
574 * Comparison in code unit order with default case folding. | |
575 * | |
576 * - U_COMPARE_CODE_POINT_ORDER | |
577 * Set to choose code point order instead of code unit order | |
578 * (see u_strCompare for details). | |
579 * | |
580 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I | |
581 * | |
582 * @return A negative, zero, or positive integer indicating the comparison res
ult. | |
583 * @stable ICU 2.0 | |
584 */ | |
585 inline int8_t caseCompare(const UnicodeString& text, uint32_t options) const; | |
586 | |
587 /** | |
588 * Compare two strings case-insensitively using full case folding. | |
589 * This is equivalent to this->foldCase(options).compare(srcText.foldCase(opti
ons)). | |
590 * | |
591 * @param start The start offset in this string at which the compare operation
begins. | |
592 * @param length The number of code units from this string to compare. | |
593 * @param srcText Another string to compare this one to. | |
594 * @param options A bit set of options: | |
595 * - U_FOLD_CASE_DEFAULT or 0 is used for default options: | |
596 * Comparison in code unit order with default case folding. | |
597 * | |
598 * - U_COMPARE_CODE_POINT_ORDER | |
599 * Set to choose code point order instead of code unit order | |
600 * (see u_strCompare for details). | |
601 * | |
602 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I | |
603 * | |
604 * @return A negative, zero, or positive integer indicating the comparison res
ult. | |
605 * @stable ICU 2.0 | |
606 */ | |
607 inline int8_t caseCompare(int32_t start, | |
608 int32_t length, | |
609 const UnicodeString& srcText, | |
610 uint32_t options) const; | |
611 | |
612 /** | |
613 * Compare two strings case-insensitively using full case folding. | |
614 * This is equivalent to this->foldCase(options).compare(srcText.foldCase(opti
ons)). | |
615 * | |
616 * @param start The start offset in this string at which the compare operation
begins. | |
617 * @param length The number of code units from this string to compare. | |
618 * @param srcText Another string to compare this one to. | |
619 * @param srcStart The start offset in that string at which the compare operat
ion begins. | |
620 * @param srcLength The number of code units from that string to compare. | |
621 * @param options A bit set of options: | |
622 * - U_FOLD_CASE_DEFAULT or 0 is used for default options: | |
623 * Comparison in code unit order with default case folding. | |
624 * | |
625 * - U_COMPARE_CODE_POINT_ORDER | |
626 * Set to choose code point order instead of code unit order | |
627 * (see u_strCompare for details). | |
628 * | |
629 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I | |
630 * | |
631 * @return A negative, zero, or positive integer indicating the comparison res
ult. | |
632 * @stable ICU 2.0 | |
633 */ | |
634 inline int8_t caseCompare(int32_t start, | |
635 int32_t length, | |
636 const UnicodeString& srcText, | |
637 int32_t srcStart, | |
638 int32_t srcLength, | |
639 uint32_t options) const; | |
640 | |
641 /** | |
642 * Compare two strings case-insensitively using full case folding. | |
643 * This is equivalent to this->foldCase(options).compare(srcChars.foldCase(opt
ions)). | |
644 * | |
645 * @param srcChars A pointer to another string to compare this one to. | |
646 * @param srcLength The number of code units from that string to compare. | |
647 * @param options A bit set of options: | |
648 * - U_FOLD_CASE_DEFAULT or 0 is used for default options: | |
649 * Comparison in code unit order with default case folding. | |
650 * | |
651 * - U_COMPARE_CODE_POINT_ORDER | |
652 * Set to choose code point order instead of code unit order | |
653 * (see u_strCompare for details). | |
654 * | |
655 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I | |
656 * | |
657 * @return A negative, zero, or positive integer indicating the comparison res
ult. | |
658 * @stable ICU 2.0 | |
659 */ | |
660 inline int8_t caseCompare(const UChar *srcChars, | |
661 int32_t srcLength, | |
662 uint32_t options) const; | |
663 | |
664 /** | |
665 * Compare two strings case-insensitively using full case folding. | |
666 * This is equivalent to this->foldCase(options).compare(srcChars.foldCase(opt
ions)). | |
667 * | |
668 * @param start The start offset in this string at which the compare operation
begins. | |
669 * @param length The number of code units from this string to compare. | |
670 * @param srcChars A pointer to another string to compare this one to. | |
671 * @param options A bit set of options: | |
672 * - U_FOLD_CASE_DEFAULT or 0 is used for default options: | |
673 * Comparison in code unit order with default case folding. | |
674 * | |
675 * - U_COMPARE_CODE_POINT_ORDER | |
676 * Set to choose code point order instead of code unit order | |
677 * (see u_strCompare for details). | |
678 * | |
679 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I | |
680 * | |
681 * @return A negative, zero, or positive integer indicating the comparison res
ult. | |
682 * @stable ICU 2.0 | |
683 */ | |
684 inline int8_t caseCompare(int32_t start, | |
685 int32_t length, | |
686 const UChar *srcChars, | |
687 uint32_t options) const; | |
688 | |
689 /** | |
690 * Compare two strings case-insensitively using full case folding. | |
691 * This is equivalent to this->foldCase(options).compare(srcChars.foldCase(opt
ions)). | |
692 * | |
693 * @param start The start offset in this string at which the compare operation
begins. | |
694 * @param length The number of code units from this string to compare. | |
695 * @param srcChars A pointer to another string to compare this one to. | |
696 * @param srcStart The start offset in that string at which the compare operat
ion begins. | |
697 * @param srcLength The number of code units from that string to compare. | |
698 * @param options A bit set of options: | |
699 * - U_FOLD_CASE_DEFAULT or 0 is used for default options: | |
700 * Comparison in code unit order with default case folding. | |
701 * | |
702 * - U_COMPARE_CODE_POINT_ORDER | |
703 * Set to choose code point order instead of code unit order | |
704 * (see u_strCompare for details). | |
705 * | |
706 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I | |
707 * | |
708 * @return A negative, zero, or positive integer indicating the comparison res
ult. | |
709 * @stable ICU 2.0 | |
710 */ | |
711 inline int8_t caseCompare(int32_t start, | |
712 int32_t length, | |
713 const UChar *srcChars, | |
714 int32_t srcStart, | |
715 int32_t srcLength, | |
716 uint32_t options) const; | |
717 | |
718 /** | |
719 * Compare two strings case-insensitively using full case folding. | |
720 * This is equivalent to this->foldCase(options).compareBetween(text.foldCase(
options)). | |
721 * | |
722 * @param start The start offset in this string at which the compare operation
begins. | |
723 * @param limit The offset after the last code unit from this string to compar
e. | |
724 * @param srcText Another string to compare this one to. | |
725 * @param srcStart The start offset in that string at which the compare operat
ion begins. | |
726 * @param srcLimit The offset after the last code unit from that string to com
pare. | |
727 * @param options A bit set of options: | |
728 * - U_FOLD_CASE_DEFAULT or 0 is used for default options: | |
729 * Comparison in code unit order with default case folding. | |
730 * | |
731 * - U_COMPARE_CODE_POINT_ORDER | |
732 * Set to choose code point order instead of code unit order | |
733 * (see u_strCompare for details). | |
734 * | |
735 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I | |
736 * | |
737 * @return A negative, zero, or positive integer indicating the comparison res
ult. | |
738 * @stable ICU 2.0 | |
739 */ | |
740 inline int8_t caseCompareBetween(int32_t start, | |
741 int32_t limit, | |
742 const UnicodeString& srcText, | |
743 int32_t srcStart, | |
744 int32_t srcLimit, | |
745 uint32_t options) const; | |
746 | |
747 /** | |
748 * Determine if this starts with the characters in <TT>text</TT> | |
749 * @param text The text to match. | |
750 * @return TRUE if this starts with the characters in <TT>text</TT>, | |
751 * FALSE otherwise | |
752 * @stable ICU 2.0 | |
753 */ | |
754 inline UBool startsWith(const UnicodeString& text) const; | |
755 | |
756 /** | |
757 * Determine if this starts with the characters in <TT>srcText</TT> | |
758 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). | |
759 * @param srcText The text to match. | |
760 * @param srcStart the offset into <TT>srcText</TT> to start matching | |
761 * @param srcLength the number of characters in <TT>srcText</TT> to match | |
762 * @return TRUE if this starts with the characters in <TT>text</TT>, | |
763 * FALSE otherwise | |
764 * @stable ICU 2.0 | |
765 */ | |
766 inline UBool startsWith(const UnicodeString& srcText, | |
767 int32_t srcStart, | |
768 int32_t srcLength) const; | |
769 | |
770 /** | |
771 * Determine if this starts with the characters in <TT>srcChars</TT> | |
772 * @param srcChars The characters to match. | |
773 * @param srcLength the number of characters in <TT>srcChars</TT> | |
774 * @return TRUE if this starts with the characters in <TT>srcChars</TT>, | |
775 * FALSE otherwise | |
776 * @stable ICU 2.0 | |
777 */ | |
778 inline UBool startsWith(const UChar *srcChars, | |
779 int32_t srcLength) const; | |
780 | |
781 /** | |
782 * Determine if this ends with the characters in <TT>srcChars</TT> | |
783 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). | |
784 * @param srcChars The characters to match. | |
785 * @param srcStart the offset into <TT>srcText</TT> to start matching | |
786 * @param srcLength the number of characters in <TT>srcChars</TT> to match | |
787 * @return TRUE if this ends with the characters in <TT>srcChars</TT>, FALSE o
therwise | |
788 * @stable ICU 2.0 | |
789 */ | |
790 inline UBool startsWith(const UChar *srcChars, | |
791 int32_t srcStart, | |
792 int32_t srcLength) const; | |
793 | |
794 /** | |
795 * Determine if this ends with the characters in <TT>text</TT> | |
796 * @param text The text to match. | |
797 * @return TRUE if this ends with the characters in <TT>text</TT>, | |
798 * FALSE otherwise | |
799 * @stable ICU 2.0 | |
800 */ | |
801 inline UBool endsWith(const UnicodeString& text) const; | |
802 | |
803 /** | |
804 * Determine if this ends with the characters in <TT>srcText</TT> | |
805 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). | |
806 * @param srcText The text to match. | |
807 * @param srcStart the offset into <TT>srcText</TT> to start matching | |
808 * @param srcLength the number of characters in <TT>srcText</TT> to match | |
809 * @return TRUE if this ends with the characters in <TT>text</TT>, | |
810 * FALSE otherwise | |
811 * @stable ICU 2.0 | |
812 */ | |
813 inline UBool endsWith(const UnicodeString& srcText, | |
814 int32_t srcStart, | |
815 int32_t srcLength) const; | |
816 | |
817 /** | |
818 * Determine if this ends with the characters in <TT>srcChars</TT> | |
819 * @param srcChars The characters to match. | |
820 * @param srcLength the number of characters in <TT>srcChars</TT> | |
821 * @return TRUE if this ends with the characters in <TT>srcChars</TT>, | |
822 * FALSE otherwise | |
823 * @stable ICU 2.0 | |
824 */ | |
825 inline UBool endsWith(const UChar *srcChars, | |
826 int32_t srcLength) const; | |
827 | |
828 /** | |
829 * Determine if this ends with the characters in <TT>srcChars</TT> | |
830 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). | |
831 * @param srcChars The characters to match. | |
832 * @param srcStart the offset into <TT>srcText</TT> to start matching | |
833 * @param srcLength the number of characters in <TT>srcChars</TT> to match | |
834 * @return TRUE if this ends with the characters in <TT>srcChars</TT>, | |
835 * FALSE otherwise | |
836 * @stable ICU 2.0 | |
837 */ | |
838 inline UBool endsWith(const UChar *srcChars, | |
839 int32_t srcStart, | |
840 int32_t srcLength) const; | |
841 | |
842 | |
843 /* Searching - bitwise only */ | |
844 | |
845 /** | |
846 * Locate in this the first occurrence of the characters in <TT>text</TT>, | |
847 * using bitwise comparison. | |
848 * @param text The text to search for. | |
849 * @return The offset into this of the start of <TT>text</TT>, | |
850 * or -1 if not found. | |
851 * @stable ICU 2.0 | |
852 */ | |
853 inline int32_t indexOf(const UnicodeString& text) const; | |
854 | |
855 /** | |
856 * Locate in this the first occurrence of the characters in <TT>text</TT> | |
857 * starting at offset <TT>start</TT>, using bitwise comparison. | |
858 * @param text The text to search for. | |
859 * @param start The offset at which searching will start. | |
860 * @return The offset into this of the start of <TT>text</TT>, | |
861 * or -1 if not found. | |
862 * @stable ICU 2.0 | |
863 */ | |
864 inline int32_t indexOf(const UnicodeString& text, | |
865 int32_t start) const; | |
866 | |
867 /** | |
868 * Locate in this the first occurrence in the range | |
869 * [<TT>start</TT>, <TT>start + length</TT>) of the characters | |
870 * in <TT>text</TT>, using bitwise comparison. | |
871 * @param text The text to search for. | |
872 * @param start The offset at which searching will start. | |
873 * @param length The number of characters to search | |
874 * @return The offset into this of the start of <TT>text</TT>, | |
875 * or -1 if not found. | |
876 * @stable ICU 2.0 | |
877 */ | |
878 inline int32_t indexOf(const UnicodeString& text, | |
879 int32_t start, | |
880 int32_t length) const; | |
881 | |
882 /** | |
883 * Locate in this the first occurrence in the range | |
884 * [<TT>start</TT>, <TT>start + length</TT>) of the characters | |
885 * in <TT>srcText</TT> in the range | |
886 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>), | |
887 * using bitwise comparison. | |
888 * @param srcText The text to search for. | |
889 * @param srcStart the offset into <TT>srcText</TT> at which | |
890 * to start matching | |
891 * @param srcLength the number of characters in <TT>srcText</TT> to match | |
892 * @param start the offset into this at which to start matching | |
893 * @param length the number of characters in this to search | |
894 * @return The offset into this of the start of <TT>text</TT>, | |
895 * or -1 if not found. | |
896 * @stable ICU 2.0 | |
897 */ | |
898 inline int32_t indexOf(const UnicodeString& srcText, | |
899 int32_t srcStart, | |
900 int32_t srcLength, | |
901 int32_t start, | |
902 int32_t length) const; | |
903 | |
904 /** | |
905 * Locate in this the first occurrence of the characters in | |
906 * <TT>srcChars</TT> | |
907 * starting at offset <TT>start</TT>, using bitwise comparison. | |
908 * @param srcChars The text to search for. | |
909 * @param srcLength the number of characters in <TT>srcChars</TT> to match | |
910 * @param start the offset into this at which to start matching | |
911 * @return The offset into this of the start of <TT>text</TT>, | |
912 * or -1 if not found. | |
913 * @stable ICU 2.0 | |
914 */ | |
915 inline int32_t indexOf(const UChar *srcChars, | |
916 int32_t srcLength, | |
917 int32_t start) const; | |
918 | |
919 /** | |
920 * Locate in this the first occurrence in the range | |
921 * [<TT>start</TT>, <TT>start + length</TT>) of the characters | |
922 * in <TT>srcChars</TT>, using bitwise comparison. | |
923 * @param srcChars The text to search for. | |
924 * @param srcLength the number of characters in <TT>srcChars</TT> | |
925 * @param start The offset at which searching will start. | |
926 * @param length The number of characters to search | |
927 * @return The offset into this of the start of <TT>srcChars</TT>, | |
928 * or -1 if not found. | |
929 * @stable ICU 2.0 | |
930 */ | |
931 inline int32_t indexOf(const UChar *srcChars, | |
932 int32_t srcLength, | |
933 int32_t start, | |
934 int32_t length) const; | |
935 | |
936 /** | |
937 * Locate in this the first occurrence in the range | |
938 * [<TT>start</TT>, <TT>start + length</TT>) of the characters | |
939 * in <TT>srcChars</TT> in the range | |
940 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>), | |
941 * using bitwise comparison. | |
942 * @param srcChars The text to search for. | |
943 * @param srcStart the offset into <TT>srcChars</TT> at which | |
944 * to start matching | |
945 * @param srcLength the number of characters in <TT>srcChars</TT> to match | |
946 * @param start the offset into this at which to start matching | |
947 * @param length the number of characters in this to search | |
948 * @return The offset into this of the start of <TT>text</TT>, | |
949 * or -1 if not found. | |
950 * @stable ICU 2.0 | |
951 */ | |
952 int32_t indexOf(const UChar *srcChars, | |
953 int32_t srcStart, | |
954 int32_t srcLength, | |
955 int32_t start, | |
956 int32_t length) const; | |
957 | |
958 /** | |
959 * Locate in this the first occurrence of the BMP code point <code>c</code>, | |
960 * using bitwise comparison. | |
961 * @param c The code unit to search for. | |
962 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
963 * @stable ICU 2.0 | |
964 */ | |
965 inline int32_t indexOf(UChar c) const; | |
966 | |
967 /** | |
968 * Locate in this the first occurrence of the code point <TT>c</TT>, | |
969 * using bitwise comparison. | |
970 * | |
971 * @param c The code point to search for. | |
972 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
973 * @stable ICU 2.0 | |
974 */ | |
975 inline int32_t indexOf(UChar32 c) const; | |
976 | |
977 /** | |
978 * Locate in this the first occurrence of the BMP code point <code>c</code>, | |
979 * starting at offset <TT>start</TT>, using bitwise comparison. | |
980 * @param c The code unit to search for. | |
981 * @param start The offset at which searching will start. | |
982 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
983 * @stable ICU 2.0 | |
984 */ | |
985 inline int32_t indexOf(UChar c, | |
986 int32_t start) const; | |
987 | |
988 /** | |
989 * Locate in this the first occurrence of the code point <TT>c</TT> | |
990 * starting at offset <TT>start</TT>, using bitwise comparison. | |
991 * | |
992 * @param c The code point to search for. | |
993 * @param start The offset at which searching will start. | |
994 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
995 * @stable ICU 2.0 | |
996 */ | |
997 inline int32_t indexOf(UChar32 c, | |
998 int32_t start) const; | |
999 | |
1000 /** | |
1001 * Locate in this the first occurrence of the BMP code point <code>c</code> | |
1002 * in the range [<TT>start</TT>, <TT>start + length</TT>), | |
1003 * using bitwise comparison. | |
1004 * @param c The code unit to search for. | |
1005 * @param start the offset into this at which to start matching | |
1006 * @param length the number of characters in this to search | |
1007 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
1008 * @stable ICU 2.0 | |
1009 */ | |
1010 inline int32_t indexOf(UChar c, | |
1011 int32_t start, | |
1012 int32_t length) const; | |
1013 | |
1014 /** | |
1015 * Locate in this the first occurrence of the code point <TT>c</TT> | |
1016 * in the range [<TT>start</TT>, <TT>start + length</TT>), | |
1017 * using bitwise comparison. | |
1018 * | |
1019 * @param c The code point to search for. | |
1020 * @param start the offset into this at which to start matching | |
1021 * @param length the number of characters in this to search | |
1022 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
1023 * @stable ICU 2.0 | |
1024 */ | |
1025 inline int32_t indexOf(UChar32 c, | |
1026 int32_t start, | |
1027 int32_t length) const; | |
1028 | |
1029 /** | |
1030 * Locate in this the last occurrence of the characters in <TT>text</TT>, | |
1031 * using bitwise comparison. | |
1032 * @param text The text to search for. | |
1033 * @return The offset into this of the start of <TT>text</TT>, | |
1034 * or -1 if not found. | |
1035 * @stable ICU 2.0 | |
1036 */ | |
1037 inline int32_t lastIndexOf(const UnicodeString& text) const; | |
1038 | |
1039 /** | |
1040 * Locate in this the last occurrence of the characters in <TT>text</TT> | |
1041 * starting at offset <TT>start</TT>, using bitwise comparison. | |
1042 * @param text The text to search for. | |
1043 * @param start The offset at which searching will start. | |
1044 * @return The offset into this of the start of <TT>text</TT>, | |
1045 * or -1 if not found. | |
1046 * @stable ICU 2.0 | |
1047 */ | |
1048 inline int32_t lastIndexOf(const UnicodeString& text, | |
1049 int32_t start) const; | |
1050 | |
1051 /** | |
1052 * Locate in this the last occurrence in the range | |
1053 * [<TT>start</TT>, <TT>start + length</TT>) of the characters | |
1054 * in <TT>text</TT>, using bitwise comparison. | |
1055 * @param text The text to search for. | |
1056 * @param start The offset at which searching will start. | |
1057 * @param length The number of characters to search | |
1058 * @return The offset into this of the start of <TT>text</TT>, | |
1059 * or -1 if not found. | |
1060 * @stable ICU 2.0 | |
1061 */ | |
1062 inline int32_t lastIndexOf(const UnicodeString& text, | |
1063 int32_t start, | |
1064 int32_t length) const; | |
1065 | |
1066 /** | |
1067 * Locate in this the last occurrence in the range | |
1068 * [<TT>start</TT>, <TT>start + length</TT>) of the characters | |
1069 * in <TT>srcText</TT> in the range | |
1070 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>), | |
1071 * using bitwise comparison. | |
1072 * @param srcText The text to search for. | |
1073 * @param srcStart the offset into <TT>srcText</TT> at which | |
1074 * to start matching | |
1075 * @param srcLength the number of characters in <TT>srcText</TT> to match | |
1076 * @param start the offset into this at which to start matching | |
1077 * @param length the number of characters in this to search | |
1078 * @return The offset into this of the start of <TT>text</TT>, | |
1079 * or -1 if not found. | |
1080 * @stable ICU 2.0 | |
1081 */ | |
1082 inline int32_t lastIndexOf(const UnicodeString& srcText, | |
1083 int32_t srcStart, | |
1084 int32_t srcLength, | |
1085 int32_t start, | |
1086 int32_t length) const; | |
1087 | |
1088 /** | |
1089 * Locate in this the last occurrence of the characters in <TT>srcChars</TT> | |
1090 * starting at offset <TT>start</TT>, using bitwise comparison. | |
1091 * @param srcChars The text to search for. | |
1092 * @param srcLength the number of characters in <TT>srcChars</TT> to match | |
1093 * @param start the offset into this at which to start matching | |
1094 * @return The offset into this of the start of <TT>text</TT>, | |
1095 * or -1 if not found. | |
1096 * @stable ICU 2.0 | |
1097 */ | |
1098 inline int32_t lastIndexOf(const UChar *srcChars, | |
1099 int32_t srcLength, | |
1100 int32_t start) const; | |
1101 | |
1102 /** | |
1103 * Locate in this the last occurrence in the range | |
1104 * [<TT>start</TT>, <TT>start + length</TT>) of the characters | |
1105 * in <TT>srcChars</TT>, using bitwise comparison. | |
1106 * @param srcChars The text to search for. | |
1107 * @param srcLength the number of characters in <TT>srcChars</TT> | |
1108 * @param start The offset at which searching will start. | |
1109 * @param length The number of characters to search | |
1110 * @return The offset into this of the start of <TT>srcChars</TT>, | |
1111 * or -1 if not found. | |
1112 * @stable ICU 2.0 | |
1113 */ | |
1114 inline int32_t lastIndexOf(const UChar *srcChars, | |
1115 int32_t srcLength, | |
1116 int32_t start, | |
1117 int32_t length) const; | |
1118 | |
1119 /** | |
1120 * Locate in this the last occurrence in the range | |
1121 * [<TT>start</TT>, <TT>start + length</TT>) of the characters | |
1122 * in <TT>srcChars</TT> in the range | |
1123 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>), | |
1124 * using bitwise comparison. | |
1125 * @param srcChars The text to search for. | |
1126 * @param srcStart the offset into <TT>srcChars</TT> at which | |
1127 * to start matching | |
1128 * @param srcLength the number of characters in <TT>srcChars</TT> to match | |
1129 * @param start the offset into this at which to start matching | |
1130 * @param length the number of characters in this to search | |
1131 * @return The offset into this of the start of <TT>text</TT>, | |
1132 * or -1 if not found. | |
1133 * @stable ICU 2.0 | |
1134 */ | |
1135 int32_t lastIndexOf(const UChar *srcChars, | |
1136 int32_t srcStart, | |
1137 int32_t srcLength, | |
1138 int32_t start, | |
1139 int32_t length) const; | |
1140 | |
1141 /** | |
1142 * Locate in this the last occurrence of the BMP code point <code>c</code>, | |
1143 * using bitwise comparison. | |
1144 * @param c The code unit to search for. | |
1145 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
1146 * @stable ICU 2.0 | |
1147 */ | |
1148 inline int32_t lastIndexOf(UChar c) const; | |
1149 | |
1150 /** | |
1151 * Locate in this the last occurrence of the code point <TT>c</TT>, | |
1152 * using bitwise comparison. | |
1153 * | |
1154 * @param c The code point to search for. | |
1155 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
1156 * @stable ICU 2.0 | |
1157 */ | |
1158 inline int32_t lastIndexOf(UChar32 c) const; | |
1159 | |
1160 /** | |
1161 * Locate in this the last occurrence of the BMP code point <code>c</code> | |
1162 * starting at offset <TT>start</TT>, using bitwise comparison. | |
1163 * @param c The code unit to search for. | |
1164 * @param start The offset at which searching will start. | |
1165 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
1166 * @stable ICU 2.0 | |
1167 */ | |
1168 inline int32_t lastIndexOf(UChar c, | |
1169 int32_t start) const; | |
1170 | |
1171 /** | |
1172 * Locate in this the last occurrence of the code point <TT>c</TT> | |
1173 * starting at offset <TT>start</TT>, using bitwise comparison. | |
1174 * | |
1175 * @param c The code point to search for. | |
1176 * @param start The offset at which searching will start. | |
1177 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
1178 * @stable ICU 2.0 | |
1179 */ | |
1180 inline int32_t lastIndexOf(UChar32 c, | |
1181 int32_t start) const; | |
1182 | |
1183 /** | |
1184 * Locate in this the last occurrence of the BMP code point <code>c</code> | |
1185 * in the range [<TT>start</TT>, <TT>start + length</TT>), | |
1186 * using bitwise comparison. | |
1187 * @param c The code unit to search for. | |
1188 * @param start the offset into this at which to start matching | |
1189 * @param length the number of characters in this to search | |
1190 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
1191 * @stable ICU 2.0 | |
1192 */ | |
1193 inline int32_t lastIndexOf(UChar c, | |
1194 int32_t start, | |
1195 int32_t length) const; | |
1196 | |
1197 /** | |
1198 * Locate in this the last occurrence of the code point <TT>c</TT> | |
1199 * in the range [<TT>start</TT>, <TT>start + length</TT>), | |
1200 * using bitwise comparison. | |
1201 * | |
1202 * @param c The code point to search for. | |
1203 * @param start the offset into this at which to start matching | |
1204 * @param length the number of characters in this to search | |
1205 * @return The offset into this of <TT>c</TT>, or -1 if not found. | |
1206 * @stable ICU 2.0 | |
1207 */ | |
1208 inline int32_t lastIndexOf(UChar32 c, | |
1209 int32_t start, | |
1210 int32_t length) const; | |
1211 | |
1212 | |
1213 /* Character access */ | |
1214 | |
1215 /** | |
1216 * Return the code unit at offset <tt>offset</tt>. | |
1217 * If the offset is not valid (0..length()-1) then U+ffff is returned. | |
1218 * @param offset a valid offset into the text | |
1219 * @return the code unit at offset <tt>offset</tt> | |
1220 * or 0xffff if the offset is not valid for this string | |
1221 * @stable ICU 2.0 | |
1222 */ | |
1223 inline UChar charAt(int32_t offset) const; | |
1224 | |
1225 /** | |
1226 * Return the code unit at offset <tt>offset</tt>. | |
1227 * If the offset is not valid (0..length()-1) then U+ffff is returned. | |
1228 * @param offset a valid offset into the text | |
1229 * @return the code unit at offset <tt>offset</tt> | |
1230 * @stable ICU 2.0 | |
1231 */ | |
1232 inline UChar operator[] (int32_t offset) const; | |
1233 | |
1234 /** | |
1235 * Return the code point that contains the code unit | |
1236 * at offset <tt>offset</tt>. | |
1237 * If the offset is not valid (0..length()-1) then U+ffff is returned. | |
1238 * @param offset a valid offset into the text | |
1239 * that indicates the text offset of any of the code units | |
1240 * that will be assembled into a code point (21-bit value) and returned | |
1241 * @return the code point of text at <tt>offset</tt> | |
1242 * or 0xffff if the offset is not valid for this string | |
1243 * @stable ICU 2.0 | |
1244 */ | |
1245 inline UChar32 char32At(int32_t offset) const; | |
1246 | |
1247 /** | |
1248 * Adjust a random-access offset so that | |
1249 * it points to the beginning of a Unicode character. | |
1250 * The offset that is passed in points to | |
1251 * any code unit of a code point, | |
1252 * while the returned offset will point to the first code unit | |
1253 * of the same code point. | |
1254 * In UTF-16, if the input offset points to a second surrogate | |
1255 * of a surrogate pair, then the returned offset will point | |
1256 * to the first surrogate. | |
1257 * @param offset a valid offset into one code point of the text | |
1258 * @return offset of the first code unit of the same code point | |
1259 * @see U16_SET_CP_START | |
1260 * @stable ICU 2.0 | |
1261 */ | |
1262 inline int32_t getChar32Start(int32_t offset) const; | |
1263 | |
1264 /** | |
1265 * Adjust a random-access offset so that | |
1266 * it points behind a Unicode character. | |
1267 * The offset that is passed in points behind | |
1268 * any code unit of a code point, | |
1269 * while the returned offset will point behind the last code unit | |
1270 * of the same code point. | |
1271 * In UTF-16, if the input offset points behind the first surrogate | |
1272 * (i.e., to the second surrogate) | |
1273 * of a surrogate pair, then the returned offset will point | |
1274 * behind the second surrogate (i.e., to the first surrogate). | |
1275 * @param offset a valid offset after any code unit of a code point of the tex
t | |
1276 * @return offset of the first code unit after the same code point | |
1277 * @see U16_SET_CP_LIMIT | |
1278 * @stable ICU 2.0 | |
1279 */ | |
1280 inline int32_t getChar32Limit(int32_t offset) const; | |
1281 | |
1282 /** | |
1283 * Move the code unit index along the string by delta code points. | |
1284 * Interpret the input index as a code unit-based offset into the string, | |
1285 * move the index forward or backward by delta code points, and | |
1286 * return the resulting index. | |
1287 * The input index should point to the first code unit of a code point, | |
1288 * if there is more than one. | |
1289 * | |
1290 * Both input and output indexes are code unit-based as for all | |
1291 * string indexes/offsets in ICU (and other libraries, like MBCS char*). | |
1292 * If delta<0 then the index is moved backward (toward the start of the string
). | |
1293 * If delta>0 then the index is moved forward (toward the end of the string). | |
1294 * | |
1295 * This behaves like CharacterIterator::move32(delta, kCurrent). | |
1296 * | |
1297 * Behavior for out-of-bounds indexes: | |
1298 * <code>moveIndex32</code> pins the input index to 0..length(), i.e., | |
1299 * if the input index<0 then it is pinned to 0; | |
1300 * if it is index>length() then it is pinned to length(). | |
1301 * Afterwards, the index is moved by <code>delta</code> code points | |
1302 * forward or backward, | |
1303 * but no further backward than to 0 and no further forward than to length(). | |
1304 * The resulting index return value will be in between 0 and length(), inclusi
vely. | |
1305 * | |
1306 * Examples: | |
1307 * <pre> | |
1308 * // s has code points 'a' U+10000 'b' U+10ffff U+2029 | |
1309 * UnicodeString s=UNICODE_STRING("a\\U00010000b\\U0010ffff\\u2029", 31).unesc
ape(); | |
1310 * | |
1311 * // initial index: position of U+10000 | |
1312 * int32_t index=1; | |
1313 * | |
1314 * // the following examples will all result in index==4, position of U+10ffff | |
1315 * | |
1316 * // skip 2 code points from some position in the string | |
1317 * index=s.moveIndex32(index, 2); // skips U+10000 and 'b' | |
1318 * | |
1319 * // go to the 3rd code point from the start of s (0-based) | |
1320 * index=s.moveIndex32(0, 3); // skips 'a', U+10000, and 'b' | |
1321 * | |
1322 * // go to the next-to-last code point of s | |
1323 * index=s.moveIndex32(s.length(), -2); // backward-skips U+2029 and U+10ffff | |
1324 * </pre> | |
1325 * | |
1326 * @param index input code unit index | |
1327 * @param delta (signed) code point count to move the index forward or backwar
d | |
1328 * in the string | |
1329 * @return the resulting code unit index | |
1330 * @stable ICU 2.0 | |
1331 */ | |
1332 int32_t moveIndex32(int32_t index, int32_t delta) const; | |
1333 | |
1334 /* Substring extraction */ | |
1335 | |
1336 /** | |
1337 * Copy the characters in the range | |
1338 * [<tt>start</tt>, <tt>start + length</tt>) into the array <tt>dst</tt>, | |
1339 * beginning at <tt>dstStart</tt>. | |
1340 * If the string aliases to <code>dst</code> itself as an external buffer, | |
1341 * then extract() will not copy the contents. | |
1342 * | |
1343 * @param start offset of first character which will be copied into the array | |
1344 * @param length the number of characters to extract | |
1345 * @param dst array in which to copy characters. The length of <tt>dst</tt> | |
1346 * must be at least (<tt>dstStart + length</tt>). | |
1347 * @param dstStart the offset in <TT>dst</TT> where the first character | |
1348 * will be extracted | |
1349 * @stable ICU 2.0 | |
1350 */ | |
1351 inline void extract(int32_t start, | |
1352 int32_t length, | |
1353 UChar *dst, | |
1354 int32_t dstStart = 0) const; | |
1355 | |
1356 /** | |
1357 * Copy the contents of the string into dest. | |
1358 * This is a convenience function that | |
1359 * checks if there is enough space in dest, | |
1360 * extracts the entire string if possible, | |
1361 * and NUL-terminates dest if possible. | |
1362 * | |
1363 * If the string fits into dest but cannot be NUL-terminated | |
1364 * (length()==destCapacity) then the error code is set to U_STRING_NOT_TERMINA
TED_WARNING. | |
1365 * If the string itself does not fit into dest | |
1366 * (length()>destCapacity) then the error code is set to U_BUFFER_OVERFLOW_ERR
OR. | |
1367 * | |
1368 * If the string aliases to <code>dest</code> itself as an external buffer, | |
1369 * then extract() will not copy the contents. | |
1370 * | |
1371 * @param dest Destination string buffer. | |
1372 * @param destCapacity Number of UChars available at dest. | |
1373 * @param errorCode ICU error code. | |
1374 * @return length() | |
1375 * @stable ICU 2.0 | |
1376 */ | |
1377 int32_t | |
1378 extract(UChar *dest, int32_t destCapacity, | |
1379 UErrorCode &errorCode) const; | |
1380 | |
1381 /** | |
1382 * Copy the characters in the range | |
1383 * [<tt>start</tt>, <tt>start + length</tt>) into the UnicodeString | |
1384 * <tt>target</tt>. | |
1385 * @param start offset of first character which will be copied | |
1386 * @param length the number of characters to extract | |
1387 * @param target UnicodeString into which to copy characters. | |
1388 * @return A reference to <TT>target</TT> | |
1389 * @stable ICU 2.0 | |
1390 */ | |
1391 inline void extract(int32_t start, | |
1392 int32_t length, | |
1393 UnicodeString& target) const; | |
1394 | |
1395 /** | |
1396 * Copy the characters in the range [<tt>start</tt>, <tt>limit</tt>) | |
1397 * into the array <tt>dst</tt>, beginning at <tt>dstStart</tt>. | |
1398 * @param start offset of first character which will be copied into the array | |
1399 * @param limit offset immediately following the last character to be copied | |
1400 * @param dst array in which to copy characters. The length of <tt>dst</tt> | |
1401 * must be at least (<tt>dstStart + (limit - start)</tt>). | |
1402 * @param dstStart the offset in <TT>dst</TT> where the first character | |
1403 * will be extracted | |
1404 * @stable ICU 2.0 | |
1405 */ | |
1406 inline void extractBetween(int32_t start, | |
1407 int32_t limit, | |
1408 UChar *dst, | |
1409 int32_t dstStart = 0) const; | |
1410 | |
1411 /** | |
1412 * Copy the characters in the range [<tt>start</tt>, <tt>limit</tt>) | |
1413 * into the UnicodeString <tt>target</tt>. Replaceable API. | |
1414 * @param start offset of first character which will be copied | |
1415 * @param limit offset immediately following the last character to be copied | |
1416 * @param target UnicodeString into which to copy characters. | |
1417 * @return A reference to <TT>target</TT> | |
1418 * @stable ICU 2.0 | |
1419 */ | |
1420 virtual void extractBetween(int32_t start, | |
1421 int32_t limit, | |
1422 UnicodeString& target) const; | |
1423 | |
1424 /** | |
1425 * Copy the characters in the range | |
1426 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters. | |
1427 * All characters must be invariant (see utypes.h). | |
1428 * Use US_INV as the last, signature-distinguishing parameter. | |
1429 * | |
1430 * This function does not write any more than <code>targetLength</code> | |
1431 * characters but returns the length of the entire output string | |
1432 * so that one can allocate a larger buffer and call the function again | |
1433 * if necessary. | |
1434 * The output string is NUL-terminated if possible. | |
1435 * | |
1436 * @param start offset of first character which will be copied | |
1437 * @param startLength the number of characters to extract | |
1438 * @param target the target buffer for extraction, can be NULL | |
1439 * if targetLength is 0 | |
1440 * @param targetCapacity the length of the target buffer | |
1441 * @param inv Signature-distinguishing paramater, use US_INV. | |
1442 * @return the output string length, not including the terminating NUL | |
1443 * @stable ICU 3.2 | |
1444 */ | |
1445 int32_t extract(int32_t start, | |
1446 int32_t startLength, | |
1447 char *target, | |
1448 int32_t targetCapacity, | |
1449 enum EInvariant inv) const; | |
1450 | |
1451 #if U_CHARSET_IS_UTF8 || !UCONFIG_NO_CONVERSION | |
1452 | |
1453 /** | |
1454 * Copy the characters in the range | |
1455 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters | |
1456 * in the platform's default codepage. | |
1457 * This function does not write any more than <code>targetLength</code> | |
1458 * characters but returns the length of the entire output string | |
1459 * so that one can allocate a larger buffer and call the function again | |
1460 * if necessary. | |
1461 * The output string is NUL-terminated if possible. | |
1462 * | |
1463 * @param start offset of first character which will be copied | |
1464 * @param startLength the number of characters to extract | |
1465 * @param target the target buffer for extraction | |
1466 * @param targetLength the length of the target buffer | |
1467 * If <TT>target</TT> is NULL, then the number of bytes required for | |
1468 * <TT>target</TT> is returned. | |
1469 * @return the output string length, not including the terminating NUL | |
1470 * @stable ICU 2.0 | |
1471 */ | |
1472 int32_t extract(int32_t start, | |
1473 int32_t startLength, | |
1474 char *target, | |
1475 uint32_t targetLength) const; | |
1476 | |
1477 #endif | |
1478 | |
1479 #if !UCONFIG_NO_CONVERSION | |
1480 | |
1481 /** | |
1482 * Copy the characters in the range | |
1483 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters | |
1484 * in a specified codepage. | |
1485 * The output string is NUL-terminated. | |
1486 * | |
1487 * Recommendation: For invariant-character strings use | |
1488 * extract(int32_t start, int32_t length, char *target, int32_t targetCapacity
, enum EInvariant inv) const | |
1489 * because it avoids object code dependencies of UnicodeString on | |
1490 * the conversion code. | |
1491 * | |
1492 * @param start offset of first character which will be copied | |
1493 * @param startLength the number of characters to extract | |
1494 * @param target the target buffer for extraction | |
1495 * @param codepage the desired codepage for the characters. 0 has | |
1496 * the special meaning of the default codepage | |
1497 * If <code>codepage</code> is an empty string (<code>""</code>), | |
1498 * then a simple conversion is performed on the codepage-invariant | |
1499 * subset ("invariant characters") of the platform encoding. See utypes.h. | |
1500 * If <TT>target</TT> is NULL, then the number of bytes required for | |
1501 * <TT>target</TT> is returned. It is assumed that the target is big enough | |
1502 * to fit all of the characters. | |
1503 * @return the output string length, not including the terminating NUL | |
1504 * @stable ICU 2.0 | |
1505 */ | |
1506 inline int32_t extract(int32_t start, | |
1507 int32_t startLength, | |
1508 char *target, | |
1509 const char *codepage = 0) const; | |
1510 | |
1511 /** | |
1512 * Copy the characters in the range | |
1513 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters | |
1514 * in a specified codepage. | |
1515 * This function does not write any more than <code>targetLength</code> | |
1516 * characters but returns the length of the entire output string | |
1517 * so that one can allocate a larger buffer and call the function again | |
1518 * if necessary. | |
1519 * The output string is NUL-terminated if possible. | |
1520 * | |
1521 * Recommendation: For invariant-character strings use | |
1522 * extract(int32_t start, int32_t length, char *target, int32_t targetCapacity
, enum EInvariant inv) const | |
1523 * because it avoids object code dependencies of UnicodeString on | |
1524 * the conversion code. | |
1525 * | |
1526 * @param start offset of first character which will be copied | |
1527 * @param startLength the number of characters to extract | |
1528 * @param target the target buffer for extraction | |
1529 * @param targetLength the length of the target buffer | |
1530 * @param codepage the desired codepage for the characters. 0 has | |
1531 * the special meaning of the default codepage | |
1532 * If <code>codepage</code> is an empty string (<code>""</code>), | |
1533 * then a simple conversion is performed on the codepage-invariant | |
1534 * subset ("invariant characters") of the platform encoding. See utypes.h. | |
1535 * If <TT>target</TT> is NULL, then the number of bytes required for | |
1536 * <TT>target</TT> is returned. | |
1537 * @return the output string length, not including the terminating NUL | |
1538 * @stable ICU 2.0 | |
1539 */ | |
1540 int32_t extract(int32_t start, | |
1541 int32_t startLength, | |
1542 char *target, | |
1543 uint32_t targetLength, | |
1544 const char *codepage) const; | |
1545 | |
1546 /** | |
1547 * Convert the UnicodeString into a codepage string using an existing UConvert
er. | |
1548 * The output string is NUL-terminated if possible. | |
1549 * | |
1550 * This function avoids the overhead of opening and closing a converter if | |
1551 * multiple strings are extracted. | |
1552 * | |
1553 * @param dest destination string buffer, can be NULL if destCapacity==0 | |
1554 * @param destCapacity the number of chars available at dest | |
1555 * @param cnv the converter object to be used (ucnv_resetFromUnicode() will be
called), | |
1556 * or NULL for the default converter | |
1557 * @param errorCode normal ICU error code | |
1558 * @return the length of the output string, not counting the terminating NUL; | |
1559 * if the length is greater than destCapacity, then the string will no
t fit | |
1560 * and a buffer of the indicated length would need to be passed in | |
1561 * @stable ICU 2.0 | |
1562 */ | |
1563 int32_t extract(char *dest, int32_t destCapacity, | |
1564 UConverter *cnv, | |
1565 UErrorCode &errorCode) const; | |
1566 | |
1567 #endif | |
1568 | |
1569 /** | |
1570 * Create a temporary substring for the specified range. | |
1571 * Unlike the substring constructor and setTo() functions, | |
1572 * the object returned here will be a read-only alias (using getBuffer()) | |
1573 * rather than copying the text. | |
1574 * As a result, this substring operation is much faster but requires | |
1575 * that the original string not be modified or deleted during the lifetime | |
1576 * of the returned substring object. | |
1577 * @param start offset of the first character visible in the substring | |
1578 * @param length length of the substring | |
1579 * @return a read-only alias UnicodeString object for the substring | |
1580 * @stable ICU 4.4 | |
1581 */ | |
1582 UnicodeString tempSubString(int32_t start=0, int32_t length=INT32_MAX) const; | |
1583 | |
1584 /** | |
1585 * Create a temporary substring for the specified range. | |
1586 * Same as tempSubString(start, length) except that the substring range | |
1587 * is specified as a (start, limit) pair (with an exclusive limit index) | |
1588 * rather than a (start, length) pair. | |
1589 * @param start offset of the first character visible in the substring | |
1590 * @param limit offset immediately following the last character visible in the
substring | |
1591 * @return a read-only alias UnicodeString object for the substring | |
1592 * @stable ICU 4.4 | |
1593 */ | |
1594 inline UnicodeString tempSubStringBetween(int32_t start, int32_t limit=INT32_M
AX) const; | |
1595 | |
1596 /** | |
1597 * Convert the UnicodeString to UTF-8 and write the result | |
1598 * to a ByteSink. This is called by toUTF8String(). | |
1599 * Unpaired surrogates are replaced with U+FFFD. | |
1600 * Calls u_strToUTF8WithSub(). | |
1601 * | |
1602 * @param sink A ByteSink to which the UTF-8 version of the string is written. | |
1603 * sink.Flush() is called at the end. | |
1604 * @stable ICU 4.2 | |
1605 * @see toUTF8String | |
1606 */ | |
1607 void toUTF8(ByteSink &sink) const; | |
1608 | |
1609 #if U_HAVE_STD_STRING | |
1610 | |
1611 /** | |
1612 * Convert the UnicodeString to UTF-8 and append the result | |
1613 * to a standard string. | |
1614 * Unpaired surrogates are replaced with U+FFFD. | |
1615 * Calls toUTF8(). | |
1616 * | |
1617 * @param result A standard string (or a compatible object) | |
1618 * to which the UTF-8 version of the string is appended. | |
1619 * @return The string object. | |
1620 * @stable ICU 4.2 | |
1621 * @see toUTF8 | |
1622 */ | |
1623 template<typename StringClass> | |
1624 StringClass &toUTF8String(StringClass &result) const { | |
1625 StringByteSink<StringClass> sbs(&result); | |
1626 toUTF8(sbs); | |
1627 return result; | |
1628 } | |
1629 | |
1630 #endif | |
1631 | |
1632 /** | |
1633 * Convert the UnicodeString to UTF-32. | |
1634 * Unpaired surrogates are replaced with U+FFFD. | |
1635 * Calls u_strToUTF32WithSub(). | |
1636 * | |
1637 * @param utf32 destination string buffer, can be NULL if capacity==0 | |
1638 * @param capacity the number of UChar32s available at utf32 | |
1639 * @param errorCode Standard ICU error code. Its input value must | |
1640 * pass the U_SUCCESS() test, or else the function returns | |
1641 * immediately. Check for U_FAILURE() on output or use with | |
1642 * function chaining. (See User Guide for details.) | |
1643 * @return The length of the UTF-32 string. | |
1644 * @see fromUTF32 | |
1645 * @stable ICU 4.2 | |
1646 */ | |
1647 int32_t toUTF32(UChar32 *utf32, int32_t capacity, UErrorCode &errorCode) const
; | |
1648 | |
1649 /* Length operations */ | |
1650 | |
1651 /** | |
1652 * Return the length of the UnicodeString object. | |
1653 * The length is the number of UChar code units are in the UnicodeString. | |
1654 * If you want the number of code points, please use countChar32(). | |
1655 * @return the length of the UnicodeString object | |
1656 * @see countChar32 | |
1657 * @stable ICU 2.0 | |
1658 */ | |
1659 inline int32_t length(void) const; | |
1660 | |
1661 /** | |
1662 * Count Unicode code points in the length UChar code units of the string. | |
1663 * A code point may occupy either one or two UChar code units. | |
1664 * Counting code points involves reading all code units. | |
1665 * | |
1666 * This functions is basically the inverse of moveIndex32(). | |
1667 * | |
1668 * @param start the index of the first code unit to check | |
1669 * @param length the number of UChar code units to check | |
1670 * @return the number of code points in the specified code units | |
1671 * @see length | |
1672 * @stable ICU 2.0 | |
1673 */ | |
1674 int32_t | |
1675 countChar32(int32_t start=0, int32_t length=INT32_MAX) const; | |
1676 | |
1677 /** | |
1678 * Check if the length UChar code units of the string | |
1679 * contain more Unicode code points than a certain number. | |
1680 * This is more efficient than counting all code points in this part of the st
ring | |
1681 * and comparing that number with a threshold. | |
1682 * This function may not need to scan the string at all if the length | |
1683 * falls within a certain range, and | |
1684 * never needs to count more than 'number+1' code points. | |
1685 * Logically equivalent to (countChar32(start, length)>number). | |
1686 * A Unicode code point may occupy either one or two UChar code units. | |
1687 * | |
1688 * @param start the index of the first code unit to check (0 for the entire st
ring) | |
1689 * @param length the number of UChar code units to check | |
1690 * (use INT32_MAX for the entire string; remember that start/len
gth | |
1691 * values are pinned) | |
1692 * @param number The number of code points in the (sub)string is compared agai
nst | |
1693 * the 'number' parameter. | |
1694 * @return Boolean value for whether the string contains more Unicode code poi
nts | |
1695 * than 'number'. Same as (u_countChar32(s, length)>number). | |
1696 * @see countChar32 | |
1697 * @see u_strHasMoreChar32Than | |
1698 * @stable ICU 2.4 | |
1699 */ | |
1700 UBool | |
1701 hasMoreChar32Than(int32_t start, int32_t length, int32_t number) const; | |
1702 | |
1703 /** | |
1704 * Determine if this string is empty. | |
1705 * @return TRUE if this string contains 0 characters, FALSE otherwise. | |
1706 * @stable ICU 2.0 | |
1707 */ | |
1708 inline UBool isEmpty(void) const; | |
1709 | |
1710 /** | |
1711 * Return the capacity of the internal buffer of the UnicodeString object. | |
1712 * This is useful together with the getBuffer functions. | |
1713 * See there for details. | |
1714 * | |
1715 * @return the number of UChars available in the internal buffer | |
1716 * @see getBuffer | |
1717 * @stable ICU 2.0 | |
1718 */ | |
1719 inline int32_t getCapacity(void) const; | |
1720 | |
1721 /* Other operations */ | |
1722 | |
1723 /** | |
1724 * Generate a hash code for this object. | |
1725 * @return The hash code of this UnicodeString. | |
1726 * @stable ICU 2.0 | |
1727 */ | |
1728 inline int32_t hashCode(void) const; | |
1729 | |
1730 /** | |
1731 * Determine if this object contains a valid string. | |
1732 * A bogus string has no value. It is different from an empty string, | |
1733 * although in both cases isEmpty() returns TRUE and length() returns 0. | |
1734 * setToBogus() and isBogus() can be used to indicate that no string value is
available. | |
1735 * For a bogus string, getBuffer() and getTerminatedBuffer() return NULL, and | |
1736 * length() returns 0. | |
1737 * | |
1738 * @return TRUE if the string is valid, FALSE otherwise | |
1739 * @see setToBogus() | |
1740 * @stable ICU 2.0 | |
1741 */ | |
1742 inline UBool isBogus(void) const; | |
1743 | |
1744 | |
1745 //======================================== | |
1746 // Write operations | |
1747 //======================================== | |
1748 | |
1749 /* Assignment operations */ | |
1750 | |
1751 /** | |
1752 * Assignment operator. Replace the characters in this UnicodeString | |
1753 * with the characters from <TT>srcText</TT>. | |
1754 * @param srcText The text containing the characters to replace | |
1755 * @return a reference to this | |
1756 * @stable ICU 2.0 | |
1757 */ | |
1758 UnicodeString &operator=(const UnicodeString &srcText); | |
1759 | |
1760 /** | |
1761 * Almost the same as the assignment operator. | |
1762 * Replace the characters in this UnicodeString | |
1763 * with the characters from <code>srcText</code>. | |
1764 * | |
1765 * This function works the same for all strings except for ones that | |
1766 * are readonly aliases. | |
1767 * Starting with ICU 2.4, the assignment operator and the copy constructor | |
1768 * allocate a new buffer and copy the buffer contents even for readonly aliase
s. | |
1769 * This function implements the old, more efficient but less safe behavior | |
1770 * of making this string also a readonly alias to the same buffer. | |
1771 * The fastCopyFrom function must be used only if it is known that the lifetim
e of | |
1772 * this UnicodeString is at least as long as the lifetime of the aliased buffe
r | |
1773 * including its contents, for example for strings from resource bundles | |
1774 * or aliases to string contents. | |
1775 * | |
1776 * @param src The text containing the characters to replace. | |
1777 * @return a reference to this | |
1778 * @stable ICU 2.4 | |
1779 */ | |
1780 UnicodeString &fastCopyFrom(const UnicodeString &src); | |
1781 | |
1782 /** | |
1783 * Assignment operator. Replace the characters in this UnicodeString | |
1784 * with the code unit <TT>ch</TT>. | |
1785 * @param ch the code unit to replace | |
1786 * @return a reference to this | |
1787 * @stable ICU 2.0 | |
1788 */ | |
1789 inline UnicodeString& operator= (UChar ch); | |
1790 | |
1791 /** | |
1792 * Assignment operator. Replace the characters in this UnicodeString | |
1793 * with the code point <TT>ch</TT>. | |
1794 * @param ch the code point to replace | |
1795 * @return a reference to this | |
1796 * @stable ICU 2.0 | |
1797 */ | |
1798 inline UnicodeString& operator= (UChar32 ch); | |
1799 | |
1800 /** | |
1801 * Set the text in the UnicodeString object to the characters | |
1802 * in <TT>srcText</TT> in the range | |
1803 * [<TT>srcStart</TT>, <TT>srcText.length()</TT>). | |
1804 * <TT>srcText</TT> is not modified. | |
1805 * @param srcText the source for the new characters | |
1806 * @param srcStart the offset into <TT>srcText</TT> where new characters | |
1807 * will be obtained | |
1808 * @return a reference to this | |
1809 * @stable ICU 2.2 | |
1810 */ | |
1811 inline UnicodeString& setTo(const UnicodeString& srcText, | |
1812 int32_t srcStart); | |
1813 | |
1814 /** | |
1815 * Set the text in the UnicodeString object to the characters | |
1816 * in <TT>srcText</TT> in the range | |
1817 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). | |
1818 * <TT>srcText</TT> is not modified. | |
1819 * @param srcText the source for the new characters | |
1820 * @param srcStart the offset into <TT>srcText</TT> where new characters | |
1821 * will be obtained | |
1822 * @param srcLength the number of characters in <TT>srcText</TT> in the | |
1823 * replace string. | |
1824 * @return a reference to this | |
1825 * @stable ICU 2.0 | |
1826 */ | |
1827 inline UnicodeString& setTo(const UnicodeString& srcText, | |
1828 int32_t srcStart, | |
1829 int32_t srcLength); | |
1830 | |
1831 /** | |
1832 * Set the text in the UnicodeString object to the characters in | |
1833 * <TT>srcText</TT>. | |
1834 * <TT>srcText</TT> is not modified. | |
1835 * @param srcText the source for the new characters | |
1836 * @return a reference to this | |
1837 * @stable ICU 2.0 | |
1838 */ | |
1839 inline UnicodeString& setTo(const UnicodeString& srcText); | |
1840 | |
1841 /** | |
1842 * Set the characters in the UnicodeString object to the characters | |
1843 * in <TT>srcChars</TT>. <TT>srcChars</TT> is not modified. | |
1844 * @param srcChars the source for the new characters | |
1845 * @param srcLength the number of Unicode characters in srcChars. | |
1846 * @return a reference to this | |
1847 * @stable ICU 2.0 | |
1848 */ | |
1849 inline UnicodeString& setTo(const UChar *srcChars, | |
1850 int32_t srcLength); | |
1851 | |
1852 /** | |
1853 * Set the characters in the UnicodeString object to the code unit | |
1854 * <TT>srcChar</TT>. | |
1855 * @param srcChar the code unit which becomes the UnicodeString's character | |
1856 * content | |
1857 * @return a reference to this | |
1858 * @stable ICU 2.0 | |
1859 */ | |
1860 UnicodeString& setTo(UChar srcChar); | |
1861 | |
1862 /** | |
1863 * Set the characters in the UnicodeString object to the code point | |
1864 * <TT>srcChar</TT>. | |
1865 * @param srcChar the code point which becomes the UnicodeString's character | |
1866 * content | |
1867 * @return a reference to this | |
1868 * @stable ICU 2.0 | |
1869 */ | |
1870 UnicodeString& setTo(UChar32 srcChar); | |
1871 | |
1872 /** | |
1873 * Aliasing setTo() function, analogous to the readonly-aliasing UChar* constr
uctor. | |
1874 * The text will be used for the UnicodeString object, but | |
1875 * it will not be released when the UnicodeString is destroyed. | |
1876 * This has copy-on-write semantics: | |
1877 * When the string is modified, then the buffer is first copied into | |
1878 * newly allocated memory. | |
1879 * The aliased buffer is never modified. | |
1880 * In an assignment to another UnicodeString, the text will be aliased again, | |
1881 * so that both strings then alias the same readonly-text. | |
1882 * | |
1883 * @param isTerminated specifies if <code>text</code> is <code>NUL</code>-term
inated. | |
1884 * This must be true if <code>textLength==-1</code>. | |
1885 * @param text The characters to alias for the UnicodeString. | |
1886 * @param textLength The number of Unicode characters in <code>text</code> to
alias. | |
1887 * If -1, then this constructor will determine the length | |
1888 * by calling <code>u_strlen()</code>. | |
1889 * @return a reference to this | |
1890 * @stable ICU 2.0 | |
1891 */ | |
1892 UnicodeString &setTo(UBool isTerminated, | |
1893 const UChar *text, | |
1894 int32_t textLength); | |
1895 | |
1896 /** | |
1897 * Aliasing setTo() function, analogous to the writable-aliasing UChar* constr
uctor. | |
1898 * The text will be used for the UnicodeString object, but | |
1899 * it will not be released when the UnicodeString is destroyed. | |
1900 * This has write-through semantics: | |
1901 * For as long as the capacity of the buffer is sufficient, write operations | |
1902 * will directly affect the buffer. When more capacity is necessary, then | |
1903 * a new buffer will be allocated and the contents copied as with regularly | |
1904 * constructed strings. | |
1905 * In an assignment to another UnicodeString, the buffer will be copied. | |
1906 * The extract(UChar *dst) function detects whether the dst pointer is the sam
e | |
1907 * as the string buffer itself and will in this case not copy the contents. | |
1908 * | |
1909 * @param buffer The characters to alias for the UnicodeString. | |
1910 * @param buffLength The number of Unicode characters in <code>buffer</code> t
o alias. | |
1911 * @param buffCapacity The size of <code>buffer</code> in UChars. | |
1912 * @return a reference to this | |
1913 * @stable ICU 2.0 | |
1914 */ | |
1915 UnicodeString &setTo(UChar *buffer, | |
1916 int32_t buffLength, | |
1917 int32_t buffCapacity); | |
1918 | |
1919 /** | |
1920 * Make this UnicodeString object invalid. | |
1921 * The string will test TRUE with isBogus(). | |
1922 * | |
1923 * A bogus string has no value. It is different from an empty string. | |
1924 * It can be used to indicate that no string value is available. | |
1925 * getBuffer() and getTerminatedBuffer() return NULL, and | |
1926 * length() returns 0. | |
1927 * | |
1928 * This utility function is used throughout the UnicodeString | |
1929 * implementation to indicate that a UnicodeString operation failed, | |
1930 * and may be used in other functions, | |
1931 * especially but not exclusively when such functions do not | |
1932 * take a UErrorCode for simplicity. | |
1933 * | |
1934 * The following methods, and no others, will clear a string object's bogus fl
ag: | |
1935 * - remove() | |
1936 * - remove(0, INT32_MAX) | |
1937 * - truncate(0) | |
1938 * - operator=() (assignment operator) | |
1939 * - setTo(...) | |
1940 * | |
1941 * The simplest ways to turn a bogus string into an empty one | |
1942 * is to use the remove() function. | |
1943 * Examples for other functions that are equivalent to "set to empty string": | |
1944 * \code | |
1945 * if(s.isBogus()) { | |
1946 * s.remove(); // set to an empty string (remove all), or | |
1947 * s.remove(0, INT32_MAX); // set to an empty string (remove all), or | |
1948 * s.truncate(0); // set to an empty string (complete truncation), or | |
1949 * s=UnicodeString(); // assign an empty string, or | |
1950 * s.setTo((UChar32)-1); // set to a pseudo code point that is out of range,
or | |
1951 * static const UChar nul=0; | |
1952 * s.setTo(&nul, 0); // set to an empty C Unicode string | |
1953 * } | |
1954 * \endcode | |
1955 * | |
1956 * @see isBogus() | |
1957 * @stable ICU 2.0 | |
1958 */ | |
1959 void setToBogus(); | |
1960 | |
1961 /** | |
1962 * Set the character at the specified offset to the specified character. | |
1963 * @param offset A valid offset into the text of the character to set | |
1964 * @param ch The new character | |
1965 * @return A reference to this | |
1966 * @stable ICU 2.0 | |
1967 */ | |
1968 UnicodeString& setCharAt(int32_t offset, | |
1969 UChar ch); | |
1970 | |
1971 | |
1972 /* Append operations */ | |
1973 | |
1974 /** | |
1975 * Append operator. Append the code unit <TT>ch</TT> to the UnicodeString | |
1976 * object. | |
1977 * @param ch the code unit to be appended | |
1978 * @return a reference to this | |
1979 * @stable ICU 2.0 | |
1980 */ | |
1981 inline UnicodeString& operator+= (UChar ch); | |
1982 | |
1983 /** | |
1984 * Append operator. Append the code point <TT>ch</TT> to the UnicodeString | |
1985 * object. | |
1986 * @param ch the code point to be appended | |
1987 * @return a reference to this | |
1988 * @stable ICU 2.0 | |
1989 */ | |
1990 inline UnicodeString& operator+= (UChar32 ch); | |
1991 | |
1992 /** | |
1993 * Append operator. Append the characters in <TT>srcText</TT> to the | |
1994 * UnicodeString object at offset <TT>start</TT>. <TT>srcText</TT> is | |
1995 * not modified. | |
1996 * @param srcText the source for the new characters | |
1997 * @return a reference to this | |
1998 * @stable ICU 2.0 | |
1999 */ | |
2000 inline UnicodeString& operator+= (const UnicodeString& srcText); | |
2001 | |
2002 /** | |
2003 * Append the characters | |
2004 * in <TT>srcText</TT> in the range | |
2005 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) to the | |
2006 * UnicodeString object at offset <TT>start</TT>. <TT>srcText</TT> | |
2007 * is not modified. | |
2008 * @param srcText the source for the new characters | |
2009 * @param srcStart the offset into <TT>srcText</TT> where new characters | |
2010 * will be obtained | |
2011 * @param srcLength the number of characters in <TT>srcText</TT> in | |
2012 * the append string | |
2013 * @return a reference to this | |
2014 * @stable ICU 2.0 | |
2015 */ | |
2016 inline UnicodeString& append(const UnicodeString& srcText, | |
2017 int32_t srcStart, | |
2018 int32_t srcLength); | |
2019 | |
2020 /** | |
2021 * Append the characters in <TT>srcText</TT> to the UnicodeString object at | |
2022 * offset <TT>start</TT>. <TT>srcText</TT> is not modified. | |
2023 * @param srcText the source for the new characters | |
2024 * @return a reference to this | |
2025 * @stable ICU 2.0 | |
2026 */ | |
2027 inline UnicodeString& append(const UnicodeString& srcText); | |
2028 | |
2029 /** | |
2030 * Append the characters in <TT>srcChars</TT> in the range | |
2031 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) to the UnicodeString | |
2032 * object at offset | |
2033 * <TT>start</TT>. <TT>srcChars</TT> is not modified. | |
2034 * @param srcChars the source for the new characters | |
2035 * @param srcStart the offset into <TT>srcChars</TT> where new characters | |
2036 * will be obtained | |
2037 * @param srcLength the number of characters in <TT>srcChars</TT> in | |
2038 * the append string | |
2039 * @return a reference to this | |
2040 * @stable ICU 2.0 | |
2041 */ | |
2042 inline UnicodeString& append(const UChar *srcChars, | |
2043 int32_t srcStart, | |
2044 int32_t srcLength); | |
2045 | |
2046 /** | |
2047 * Append the characters in <TT>srcChars</TT> to the UnicodeString object | |
2048 * at offset <TT>start</TT>. <TT>srcChars</TT> is not modified. | |
2049 * @param srcChars the source for the new characters | |
2050 * @param srcLength the number of Unicode characters in <TT>srcChars</TT> | |
2051 * @return a reference to this | |
2052 * @stable ICU 2.0 | |
2053 */ | |
2054 inline UnicodeString& append(const UChar *srcChars, | |
2055 int32_t srcLength); | |
2056 | |
2057 /** | |
2058 * Append the code unit <TT>srcChar</TT> to the UnicodeString object. | |
2059 * @param srcChar the code unit to append | |
2060 * @return a reference to this | |
2061 * @stable ICU 2.0 | |
2062 */ | |
2063 inline UnicodeString& append(UChar srcChar); | |
2064 | |
2065 /** | |
2066 * Append the code point <TT>srcChar</TT> to the UnicodeString object. | |
2067 * @param srcChar the code point to append | |
2068 * @return a reference to this | |
2069 * @stable ICU 2.0 | |
2070 */ | |
2071 inline UnicodeString& append(UChar32 srcChar); | |
2072 | |
2073 | |
2074 /* Insert operations */ | |
2075 | |
2076 /** | |
2077 * Insert the characters in <TT>srcText</TT> in the range | |
2078 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) into the UnicodeString | |
2079 * object at offset <TT>start</TT>. <TT>srcText</TT> is not modified. | |
2080 * @param start the offset where the insertion begins | |
2081 * @param srcText the source for the new characters | |
2082 * @param srcStart the offset into <TT>srcText</TT> where new characters | |
2083 * will be obtained | |
2084 * @param srcLength the number of characters in <TT>srcText</TT> in | |
2085 * the insert string | |
2086 * @return a reference to this | |
2087 * @stable ICU 2.0 | |
2088 */ | |
2089 inline UnicodeString& insert(int32_t start, | |
2090 const UnicodeString& srcText, | |
2091 int32_t srcStart, | |
2092 int32_t srcLength); | |
2093 | |
2094 /** | |
2095 * Insert the characters in <TT>srcText</TT> into the UnicodeString object | |
2096 * at offset <TT>start</TT>. <TT>srcText</TT> is not modified. | |
2097 * @param start the offset where the insertion begins | |
2098 * @param srcText the source for the new characters | |
2099 * @return a reference to this | |
2100 * @stable ICU 2.0 | |
2101 */ | |
2102 inline UnicodeString& insert(int32_t start, | |
2103 const UnicodeString& srcText); | |
2104 | |
2105 /** | |
2106 * Insert the characters in <TT>srcChars</TT> in the range | |
2107 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) into the UnicodeString | |
2108 * object at offset <TT>start</TT>. <TT>srcChars</TT> is not modified. | |
2109 * @param start the offset at which the insertion begins | |
2110 * @param srcChars the source for the new characters | |
2111 * @param srcStart the offset into <TT>srcChars</TT> where new characters | |
2112 * will be obtained | |
2113 * @param srcLength the number of characters in <TT>srcChars</TT> | |
2114 * in the insert string | |
2115 * @return a reference to this | |
2116 * @stable ICU 2.0 | |
2117 */ | |
2118 inline UnicodeString& insert(int32_t start, | |
2119 const UChar *srcChars, | |
2120 int32_t srcStart, | |
2121 int32_t srcLength); | |
2122 | |
2123 /** | |
2124 * Insert the characters in <TT>srcChars</TT> into the UnicodeString object | |
2125 * at offset <TT>start</TT>. <TT>srcChars</TT> is not modified. | |
2126 * @param start the offset where the insertion begins | |
2127 * @param srcChars the source for the new characters | |
2128 * @param srcLength the number of Unicode characters in srcChars. | |
2129 * @return a reference to this | |
2130 * @stable ICU 2.0 | |
2131 */ | |
2132 inline UnicodeString& insert(int32_t start, | |
2133 const UChar *srcChars, | |
2134 int32_t srcLength); | |
2135 | |
2136 /** | |
2137 * Insert the code unit <TT>srcChar</TT> into the UnicodeString object at | |
2138 * offset <TT>start</TT>. | |
2139 * @param start the offset at which the insertion occurs | |
2140 * @param srcChar the code unit to insert | |
2141 * @return a reference to this | |
2142 * @stable ICU 2.0 | |
2143 */ | |
2144 inline UnicodeString& insert(int32_t start, | |
2145 UChar srcChar); | |
2146 | |
2147 /** | |
2148 * Insert the code point <TT>srcChar</TT> into the UnicodeString object at | |
2149 * offset <TT>start</TT>. | |
2150 * @param start the offset at which the insertion occurs | |
2151 * @param srcChar the code point to insert | |
2152 * @return a reference to this | |
2153 * @stable ICU 2.0 | |
2154 */ | |
2155 inline UnicodeString& insert(int32_t start, | |
2156 UChar32 srcChar); | |
2157 | |
2158 | |
2159 /* Replace operations */ | |
2160 | |
2161 /** | |
2162 * Replace the characters in the range | |
2163 * [<TT>start</TT>, <TT>start + length</TT>) with the characters in | |
2164 * <TT>srcText</TT> in the range | |
2165 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). | |
2166 * <TT>srcText</TT> is not modified. | |
2167 * @param start the offset at which the replace operation begins | |
2168 * @param length the number of characters to replace. The character at | |
2169 * <TT>start + length</TT> is not modified. | |
2170 * @param srcText the source for the new characters | |
2171 * @param srcStart the offset into <TT>srcText</TT> where new characters | |
2172 * will be obtained | |
2173 * @param srcLength the number of characters in <TT>srcText</TT> in | |
2174 * the replace string | |
2175 * @return a reference to this | |
2176 * @stable ICU 2.0 | |
2177 */ | |
2178 UnicodeString& replace(int32_t start, | |
2179 int32_t length, | |
2180 const UnicodeString& srcText, | |
2181 int32_t srcStart, | |
2182 int32_t srcLength); | |
2183 | |
2184 /** | |
2185 * Replace the characters in the range | |
2186 * [<TT>start</TT>, <TT>start + length</TT>) | |
2187 * with the characters in <TT>srcText</TT>. <TT>srcText</TT> is | |
2188 * not modified. | |
2189 * @param start the offset at which the replace operation begins | |
2190 * @param length the number of characters to replace. The character at | |
2191 * <TT>start + length</TT> is not modified. | |
2192 * @param srcText the source for the new characters | |
2193 * @return a reference to this | |
2194 * @stable ICU 2.0 | |
2195 */ | |
2196 UnicodeString& replace(int32_t start, | |
2197 int32_t length, | |
2198 const UnicodeString& srcText); | |
2199 | |
2200 /** | |
2201 * Replace the characters in the range | |
2202 * [<TT>start</TT>, <TT>start + length</TT>) with the characters in | |
2203 * <TT>srcChars</TT> in the range | |
2204 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). <TT>srcChars</TT> | |
2205 * is not modified. | |
2206 * @param start the offset at which the replace operation begins | |
2207 * @param length the number of characters to replace. The character at | |
2208 * <TT>start + length</TT> is not modified. | |
2209 * @param srcChars the source for the new characters | |
2210 * @param srcStart the offset into <TT>srcChars</TT> where new characters | |
2211 * will be obtained | |
2212 * @param srcLength the number of characters in <TT>srcChars</TT> | |
2213 * in the replace string | |
2214 * @return a reference to this | |
2215 * @stable ICU 2.0 | |
2216 */ | |
2217 UnicodeString& replace(int32_t start, | |
2218 int32_t length, | |
2219 const UChar *srcChars, | |
2220 int32_t srcStart, | |
2221 int32_t srcLength); | |
2222 | |
2223 /** | |
2224 * Replace the characters in the range | |
2225 * [<TT>start</TT>, <TT>start + length</TT>) with the characters in | |
2226 * <TT>srcChars</TT>. <TT>srcChars</TT> is not modified. | |
2227 * @param start the offset at which the replace operation begins | |
2228 * @param length number of characters to replace. The character at | |
2229 * <TT>start + length</TT> is not modified. | |
2230 * @param srcChars the source for the new characters | |
2231 * @param srcLength the number of Unicode characters in srcChars | |
2232 * @return a reference to this | |
2233 * @stable ICU 2.0 | |
2234 */ | |
2235 inline UnicodeString& replace(int32_t start, | |
2236 int32_t length, | |
2237 const UChar *srcChars, | |
2238 int32_t srcLength); | |
2239 | |
2240 /** | |
2241 * Replace the characters in the range | |
2242 * [<TT>start</TT>, <TT>start + length</TT>) with the code unit | |
2243 * <TT>srcChar</TT>. | |
2244 * @param start the offset at which the replace operation begins | |
2245 * @param length the number of characters to replace. The character at | |
2246 * <TT>start + length</TT> is not modified. | |
2247 * @param srcChar the new code unit | |
2248 * @return a reference to this | |
2249 * @stable ICU 2.0 | |
2250 */ | |
2251 inline UnicodeString& replace(int32_t start, | |
2252 int32_t length, | |
2253 UChar srcChar); | |
2254 | |
2255 /** | |
2256 * Replace the characters in the range | |
2257 * [<TT>start</TT>, <TT>start + length</TT>) with the code point | |
2258 * <TT>srcChar</TT>. | |
2259 * @param start the offset at which the replace operation begins | |
2260 * @param length the number of characters to replace. The character at | |
2261 * <TT>start + length</TT> is not modified. | |
2262 * @param srcChar the new code point | |
2263 * @return a reference to this | |
2264 * @stable ICU 2.0 | |
2265 */ | |
2266 inline UnicodeString& replace(int32_t start, | |
2267 int32_t length, | |
2268 UChar32 srcChar); | |
2269 | |
2270 /** | |
2271 * Replace the characters in the range [<TT>start</TT>, <TT>limit</TT>) | |
2272 * with the characters in <TT>srcText</TT>. <TT>srcText</TT> is not modified. | |
2273 * @param start the offset at which the replace operation begins | |
2274 * @param limit the offset immediately following the replace range | |
2275 * @param srcText the source for the new characters | |
2276 * @return a reference to this | |
2277 * @stable ICU 2.0 | |
2278 */ | |
2279 inline UnicodeString& replaceBetween(int32_t start, | |
2280 int32_t limit, | |
2281 const UnicodeString& srcText); | |
2282 | |
2283 /** | |
2284 * Replace the characters in the range [<TT>start</TT>, <TT>limit</TT>) | |
2285 * with the characters in <TT>srcText</TT> in the range | |
2286 * [<TT>srcStart</TT>, <TT>srcLimit</TT>). <TT>srcText</TT> is not modified. | |
2287 * @param start the offset at which the replace operation begins | |
2288 * @param limit the offset immediately following the replace range | |
2289 * @param srcText the source for the new characters | |
2290 * @param srcStart the offset into <TT>srcChars</TT> where new characters | |
2291 * will be obtained | |
2292 * @param srcLimit the offset immediately following the range to copy | |
2293 * in <TT>srcText</TT> | |
2294 * @return a reference to this | |
2295 * @stable ICU 2.0 | |
2296 */ | |
2297 inline UnicodeString& replaceBetween(int32_t start, | |
2298 int32_t limit, | |
2299 const UnicodeString& srcText, | |
2300 int32_t srcStart, | |
2301 int32_t srcLimit); | |
2302 | |
2303 /** | |
2304 * Replace a substring of this object with the given text. | |
2305 * @param start the beginning index, inclusive; <code>0 <= start | |
2306 * <= limit</code>. | |
2307 * @param limit the ending index, exclusive; <code>start <= limit | |
2308 * <= length()</code>. | |
2309 * @param text the text to replace characters <code>start</code> | |
2310 * to <code>limit - 1</code> | |
2311 * @stable ICU 2.0 | |
2312 */ | |
2313 virtual void handleReplaceBetween(int32_t start, | |
2314 int32_t limit, | |
2315 const UnicodeString& text); | |
2316 | |
2317 /** | |
2318 * Replaceable API | |
2319 * @return TRUE if it has MetaData | |
2320 * @stable ICU 2.4 | |
2321 */ | |
2322 virtual UBool hasMetaData() const; | |
2323 | |
2324 /** | |
2325 * Copy a substring of this object, retaining attribute (out-of-band) | |
2326 * information. This method is used to duplicate or reorder substrings. | |
2327 * The destination index must not overlap the source range. | |
2328 * | |
2329 * @param start the beginning index, inclusive; <code>0 <= start <= | |
2330 * limit</code>. | |
2331 * @param limit the ending index, exclusive; <code>start <= limit <= | |
2332 * length()</code>. | |
2333 * @param dest the destination index. The characters from | |
2334 * <code>start..limit-1</code> will be copied to <code>dest</code>. | |
2335 * Implementations of this method may assume that <code>dest <= start || | |
2336 * dest >= limit</code>. | |
2337 * @stable ICU 2.0 | |
2338 */ | |
2339 virtual void copy(int32_t start, int32_t limit, int32_t dest); | |
2340 | |
2341 /* Search and replace operations */ | |
2342 | |
2343 /** | |
2344 * Replace all occurrences of characters in oldText with the characters | |
2345 * in newText | |
2346 * @param oldText the text containing the search text | |
2347 * @param newText the text containing the replacement text | |
2348 * @return a reference to this | |
2349 * @stable ICU 2.0 | |
2350 */ | |
2351 inline UnicodeString& findAndReplace(const UnicodeString& oldText, | |
2352 const UnicodeString& newText); | |
2353 | |
2354 /** | |
2355 * Replace all occurrences of characters in oldText with characters | |
2356 * in newText | |
2357 * in the range [<TT>start</TT>, <TT>start + length</TT>). | |
2358 * @param start the start of the range in which replace will performed | |
2359 * @param length the length of the range in which replace will be performed | |
2360 * @param oldText the text containing the search text | |
2361 * @param newText the text containing the replacement text | |
2362 * @return a reference to this | |
2363 * @stable ICU 2.0 | |
2364 */ | |
2365 inline UnicodeString& findAndReplace(int32_t start, | |
2366 int32_t length, | |
2367 const UnicodeString& oldText, | |
2368 const UnicodeString& newText); | |
2369 | |
2370 /** | |
2371 * Replace all occurrences of characters in oldText in the range | |
2372 * [<TT>oldStart</TT>, <TT>oldStart + oldLength</TT>) with the characters | |
2373 * in newText in the range | |
2374 * [<TT>newStart</TT>, <TT>newStart + newLength</TT>) | |
2375 * in the range [<TT>start</TT>, <TT>start + length</TT>). | |
2376 * @param start the start of the range in which replace will performed | |
2377 * @param length the length of the range in which replace will be performed | |
2378 * @param oldText the text containing the search text | |
2379 * @param oldStart the start of the search range in <TT>oldText</TT> | |
2380 * @param oldLength the length of the search range in <TT>oldText</TT> | |
2381 * @param newText the text containing the replacement text | |
2382 * @param newStart the start of the replacement range in <TT>newText</TT> | |
2383 * @param newLength the length of the replacement range in <TT>newText</TT> | |
2384 * @return a reference to this | |
2385 * @stable ICU 2.0 | |
2386 */ | |
2387 UnicodeString& findAndReplace(int32_t start, | |
2388 int32_t length, | |
2389 const UnicodeString& oldText, | |
2390 int32_t oldStart, | |
2391 int32_t oldLength, | |
2392 const UnicodeString& newText, | |
2393 int32_t newStart, | |
2394 int32_t newLength); | |
2395 | |
2396 | |
2397 /* Remove operations */ | |
2398 | |
2399 /** | |
2400 * Remove all characters from the UnicodeString object. | |
2401 * @return a reference to this | |
2402 * @stable ICU 2.0 | |
2403 */ | |
2404 inline UnicodeString& remove(void); | |
2405 | |
2406 /** | |
2407 * Remove the characters in the range | |
2408 * [<TT>start</TT>, <TT>start + length</TT>) from the UnicodeString object. | |
2409 * @param start the offset of the first character to remove | |
2410 * @param length the number of characters to remove | |
2411 * @return a reference to this | |
2412 * @stable ICU 2.0 | |
2413 */ | |
2414 inline UnicodeString& remove(int32_t start, | |
2415 int32_t length = (int32_t)INT32_MAX); | |
2416 | |
2417 /** | |
2418 * Remove the characters in the range | |
2419 * [<TT>start</TT>, <TT>limit</TT>) from the UnicodeString object. | |
2420 * @param start the offset of the first character to remove | |
2421 * @param limit the offset immediately following the range to remove | |
2422 * @return a reference to this | |
2423 * @stable ICU 2.0 | |
2424 */ | |
2425 inline UnicodeString& removeBetween(int32_t start, | |
2426 int32_t limit = (int32_t)INT32_MAX); | |
2427 | |
2428 /** | |
2429 * Retain only the characters in the range | |
2430 * [<code>start</code>, <code>limit</code>) from the UnicodeString object. | |
2431 * Removes characters before <code>start</code> and at and after <code>limit</
code>. | |
2432 * @param start the offset of the first character to retain | |
2433 * @param limit the offset immediately following the range to retain | |
2434 * @return a reference to this | |
2435 * @stable ICU 4.4 | |
2436 */ | |
2437 inline UnicodeString &retainBetween(int32_t start, int32_t limit = INT32_MAX); | |
2438 | |
2439 /* Length operations */ | |
2440 | |
2441 /** | |
2442 * Pad the start of this UnicodeString with the character <TT>padChar</TT>. | |
2443 * If the length of this UnicodeString is less than targetLength, | |
2444 * length() - targetLength copies of padChar will be added to the | |
2445 * beginning of this UnicodeString. | |
2446 * @param targetLength the desired length of the string | |
2447 * @param padChar the character to use for padding. Defaults to | |
2448 * space (U+0020) | |
2449 * @return TRUE if the text was padded, FALSE otherwise. | |
2450 * @stable ICU 2.0 | |
2451 */ | |
2452 UBool padLeading(int32_t targetLength, | |
2453 UChar padChar = 0x0020); | |
2454 | |
2455 /** | |
2456 * Pad the end of this UnicodeString with the character <TT>padChar</TT>. | |
2457 * If the length of this UnicodeString is less than targetLength, | |
2458 * length() - targetLength copies of padChar will be added to the | |
2459 * end of this UnicodeString. | |
2460 * @param targetLength the desired length of the string | |
2461 * @param padChar the character to use for padding. Defaults to | |
2462 * space (U+0020) | |
2463 * @return TRUE if the text was padded, FALSE otherwise. | |
2464 * @stable ICU 2.0 | |
2465 */ | |
2466 UBool padTrailing(int32_t targetLength, | |
2467 UChar padChar = 0x0020); | |
2468 | |
2469 /** | |
2470 * Truncate this UnicodeString to the <TT>targetLength</TT>. | |
2471 * @param targetLength the desired length of this UnicodeString. | |
2472 * @return TRUE if the text was truncated, FALSE otherwise | |
2473 * @stable ICU 2.0 | |
2474 */ | |
2475 inline UBool truncate(int32_t targetLength); | |
2476 | |
2477 /** | |
2478 * Trims leading and trailing whitespace from this UnicodeString. | |
2479 * @return a reference to this | |
2480 * @stable ICU 2.0 | |
2481 */ | |
2482 UnicodeString& trim(void); | |
2483 | |
2484 | |
2485 /* Miscellaneous operations */ | |
2486 | |
2487 /** | |
2488 * Reverse this UnicodeString in place. | |
2489 * @return a reference to this | |
2490 * @stable ICU 2.0 | |
2491 */ | |
2492 inline UnicodeString& reverse(void); | |
2493 | |
2494 /** | |
2495 * Reverse the range [<TT>start</TT>, <TT>start + length</TT>) in | |
2496 * this UnicodeString. | |
2497 * @param start the start of the range to reverse | |
2498 * @param length the number of characters to to reverse | |
2499 * @return a reference to this | |
2500 * @stable ICU 2.0 | |
2501 */ | |
2502 inline UnicodeString& reverse(int32_t start, | |
2503 int32_t length); | |
2504 | |
2505 /** | |
2506 * Convert the characters in this to UPPER CASE following the conventions of | |
2507 * the default locale. | |
2508 * @return A reference to this. | |
2509 * @stable ICU 2.0 | |
2510 */ | |
2511 UnicodeString& toUpper(void); | |
2512 | |
2513 /** | |
2514 * Convert the characters in this to UPPER CASE following the conventions of | |
2515 * a specific locale. | |
2516 * @param locale The locale containing the conventions to use. | |
2517 * @return A reference to this. | |
2518 * @stable ICU 2.0 | |
2519 */ | |
2520 UnicodeString& toUpper(const Locale& locale); | |
2521 | |
2522 /** | |
2523 * Convert the characters in this to lower case following the conventions of | |
2524 * the default locale. | |
2525 * @return A reference to this. | |
2526 * @stable ICU 2.0 | |
2527 */ | |
2528 UnicodeString& toLower(void); | |
2529 | |
2530 /** | |
2531 * Convert the characters in this to lower case following the conventions of | |
2532 * a specific locale. | |
2533 * @param locale The locale containing the conventions to use. | |
2534 * @return A reference to this. | |
2535 * @stable ICU 2.0 | |
2536 */ | |
2537 UnicodeString& toLower(const Locale& locale); | |
2538 | |
2539 #if !UCONFIG_NO_BREAK_ITERATION | |
2540 | |
2541 /** | |
2542 * Titlecase this string, convenience function using the default locale. | |
2543 * | |
2544 * Casing is locale-dependent and context-sensitive. | |
2545 * Titlecasing uses a break iterator to find the first characters of words | |
2546 * that are to be titlecased. It titlecases those characters and lowercases | |
2547 * all others. | |
2548 * | |
2549 * The titlecase break iterator can be provided to customize for arbitrary | |
2550 * styles, using rules and dictionaries beyond the standard iterators. | |
2551 * It may be more efficient to always provide an iterator to avoid | |
2552 * opening and closing one for each string. | |
2553 * The standard titlecase iterator for the root locale implements the | |
2554 * algorithm of Unicode TR 21. | |
2555 * | |
2556 * This function uses only the setText(), first() and next() methods of the | |
2557 * provided break iterator. | |
2558 * | |
2559 * @param titleIter A break iterator to find the first characters of words | |
2560 * that are to be titlecased. | |
2561 * If none is provided (0), then a standard titlecase | |
2562 * break iterator is opened. | |
2563 * Otherwise the provided iterator is set to the string's tex
t. | |
2564 * @return A reference to this. | |
2565 * @stable ICU 2.1 | |
2566 */ | |
2567 UnicodeString &toTitle(BreakIterator *titleIter); | |
2568 | |
2569 /** | |
2570 * Titlecase this string. | |
2571 * | |
2572 * Casing is locale-dependent and context-sensitive. | |
2573 * Titlecasing uses a break iterator to find the first characters of words | |
2574 * that are to be titlecased. It titlecases those characters and lowercases | |
2575 * all others. | |
2576 * | |
2577 * The titlecase break iterator can be provided to customize for arbitrary | |
2578 * styles, using rules and dictionaries beyond the standard iterators. | |
2579 * It may be more efficient to always provide an iterator to avoid | |
2580 * opening and closing one for each string. | |
2581 * The standard titlecase iterator for the root locale implements the | |
2582 * algorithm of Unicode TR 21. | |
2583 * | |
2584 * This function uses only the setText(), first() and next() methods of the | |
2585 * provided break iterator. | |
2586 * | |
2587 * @param titleIter A break iterator to find the first characters of words | |
2588 * that are to be titlecased. | |
2589 * If none is provided (0), then a standard titlecase | |
2590 * break iterator is opened. | |
2591 * Otherwise the provided iterator is set to the string's tex
t. | |
2592 * @param locale The locale to consider. | |
2593 * @return A reference to this. | |
2594 * @stable ICU 2.1 | |
2595 */ | |
2596 UnicodeString &toTitle(BreakIterator *titleIter, const Locale &locale); | |
2597 | |
2598 /** | |
2599 * Titlecase this string, with options. | |
2600 * | |
2601 * Casing is locale-dependent and context-sensitive. | |
2602 * Titlecasing uses a break iterator to find the first characters of words | |
2603 * that are to be titlecased. It titlecases those characters and lowercases | |
2604 * all others. (This can be modified with options.) | |
2605 * | |
2606 * The titlecase break iterator can be provided to customize for arbitrary | |
2607 * styles, using rules and dictionaries beyond the standard iterators. | |
2608 * It may be more efficient to always provide an iterator to avoid | |
2609 * opening and closing one for each string. | |
2610 * The standard titlecase iterator for the root locale implements the | |
2611 * algorithm of Unicode TR 21. | |
2612 * | |
2613 * This function uses only the setText(), first() and next() methods of the | |
2614 * provided break iterator. | |
2615 * | |
2616 * @param titleIter A break iterator to find the first characters of words | |
2617 * that are to be titlecased. | |
2618 * If none is provided (0), then a standard titlecase | |
2619 * break iterator is opened. | |
2620 * Otherwise the provided iterator is set to the string's tex
t. | |
2621 * @param locale The locale to consider. | |
2622 * @param options Options bit set, see ucasemap_open(). | |
2623 * @return A reference to this. | |
2624 * @see U_TITLECASE_NO_LOWERCASE | |
2625 * @see U_TITLECASE_NO_BREAK_ADJUSTMENT | |
2626 * @see ucasemap_open | |
2627 * @stable ICU 3.8 | |
2628 */ | |
2629 UnicodeString &toTitle(BreakIterator *titleIter, const Locale &locale, uint32_
t options); | |
2630 | |
2631 #endif | |
2632 | |
2633 /** | |
2634 * Case-fold the characters in this string. | |
2635 * Case-folding is locale-independent and not context-sensitive, | |
2636 * but there is an option for whether to include or exclude mappings for dotte
d I | |
2637 * and dotless i that are marked with 'I' in CaseFolding.txt. | |
2638 * The result may be longer or shorter than the original. | |
2639 * | |
2640 * @param options Either U_FOLD_CASE_DEFAULT or U_FOLD_CASE_EXCLUDE_SPECIAL_I | |
2641 * @return A reference to this. | |
2642 * @stable ICU 2.0 | |
2643 */ | |
2644 UnicodeString &foldCase(uint32_t options=0 /*U_FOLD_CASE_DEFAULT*/); | |
2645 | |
2646 //======================================== | |
2647 // Access to the internal buffer | |
2648 //======================================== | |
2649 | |
2650 /** | |
2651 * Get a read/write pointer to the internal buffer. | |
2652 * The buffer is guaranteed to be large enough for at least minCapacity UChars
, | |
2653 * writable, and is still owned by the UnicodeString object. | |
2654 * Calls to getBuffer(minCapacity) must not be nested, and | |
2655 * must be matched with calls to releaseBuffer(newLength). | |
2656 * If the string buffer was read-only or shared, | |
2657 * then it will be reallocated and copied. | |
2658 * | |
2659 * An attempted nested call will return 0, and will not further modify the | |
2660 * state of the UnicodeString object. | |
2661 * It also returns 0 if the string is bogus. | |
2662 * | |
2663 * The actual capacity of the string buffer may be larger than minCapacity. | |
2664 * getCapacity() returns the actual capacity. | |
2665 * For many operations, the full capacity should be used to avoid reallocation
s. | |
2666 * | |
2667 * While the buffer is "open" between getBuffer(minCapacity) | |
2668 * and releaseBuffer(newLength), the following applies: | |
2669 * - The string length is set to 0. | |
2670 * - Any read API call on the UnicodeString object will behave like on a 0-len
gth string. | |
2671 * - Any write API call on the UnicodeString object is disallowed and will hav
e no effect. | |
2672 * - You can read from and write to the returned buffer. | |
2673 * - The previous string contents will still be in the buffer; | |
2674 * if you want to use it, then you need to call length() before getBuffer(mi
nCapacity). | |
2675 * If the length() was greater than minCapacity, then any contents after min
Capacity | |
2676 * may be lost. | |
2677 * The buffer contents is not NUL-terminated by getBuffer(). | |
2678 * If length()<getCapacity() then you can terminate it by writing a NUL | |
2679 * at index length(). | |
2680 * - You must call releaseBuffer(newLength) before and in order to | |
2681 * return to normal UnicodeString operation. | |
2682 * | |
2683 * @param minCapacity the minimum number of UChars that are to be available | |
2684 * in the buffer, starting at the returned pointer; | |
2685 * default to the current string capacity if minCapacity==-1 | |
2686 * @return a writable pointer to the internal string buffer, | |
2687 * or 0 if an error occurs (nested calls, out of memory) | |
2688 * | |
2689 * @see releaseBuffer | |
2690 * @see getTerminatedBuffer() | |
2691 * @stable ICU 2.0 | |
2692 */ | |
2693 UChar *getBuffer(int32_t minCapacity); | |
2694 | |
2695 /** | |
2696 * Release a read/write buffer on a UnicodeString object with an | |
2697 * "open" getBuffer(minCapacity). | |
2698 * This function must be called in a matched pair with getBuffer(minCapacity). | |
2699 * releaseBuffer(newLength) must be called if and only if a getBuffer(minCapac
ity) is "open". | |
2700 * | |
2701 * It will set the string length to newLength, at most to the current capacity
. | |
2702 * If newLength==-1 then it will set the length according to the | |
2703 * first NUL in the buffer, or to the capacity if there is no NUL. | |
2704 * | |
2705 * After calling releaseBuffer(newLength) the UnicodeString is back to normal
operation. | |
2706 * | |
2707 * @param newLength the new length of the UnicodeString object; | |
2708 * defaults to the current capacity if newLength is greater than that; | |
2709 * if newLength==-1, it defaults to u_strlen(buffer) but not more than | |
2710 * the current capacity of the string | |
2711 * | |
2712 * @see getBuffer(int32_t minCapacity) | |
2713 * @stable ICU 2.0 | |
2714 */ | |
2715 void releaseBuffer(int32_t newLength=-1); | |
2716 | |
2717 /** | |
2718 * Get a read-only pointer to the internal buffer. | |
2719 * This can be called at any time on a valid UnicodeString. | |
2720 * | |
2721 * It returns 0 if the string is bogus, or | |
2722 * during an "open" getBuffer(minCapacity). | |
2723 * | |
2724 * It can be called as many times as desired. | |
2725 * The pointer that it returns will remain valid until the UnicodeString objec
t is modified, | |
2726 * at which time the pointer is semantically invalidated and must not be used
any more. | |
2727 * | |
2728 * The capacity of the buffer can be determined with getCapacity(). | |
2729 * The part after length() may or may not be initialized and valid, | |
2730 * depending on the history of the UnicodeString object. | |
2731 * | |
2732 * The buffer contents is (probably) not NUL-terminated. | |
2733 * You can check if it is with | |
2734 * <code>(s.length()<s.getCapacity() && buffer[s.length()]==0)</code>. | |
2735 * (See getTerminatedBuffer().) | |
2736 * | |
2737 * The buffer may reside in read-only memory. Its contents must not | |
2738 * be modified. | |
2739 * | |
2740 * @return a read-only pointer to the internal string buffer, | |
2741 * or 0 if the string is empty or bogus | |
2742 * | |
2743 * @see getBuffer(int32_t minCapacity) | |
2744 * @see getTerminatedBuffer() | |
2745 * @stable ICU 2.0 | |
2746 */ | |
2747 inline const UChar *getBuffer() const; | |
2748 | |
2749 /** | |
2750 * Get a read-only pointer to the internal buffer, | |
2751 * making sure that it is NUL-terminated. | |
2752 * This can be called at any time on a valid UnicodeString. | |
2753 * | |
2754 * It returns 0 if the string is bogus, or | |
2755 * during an "open" getBuffer(minCapacity), or if the buffer cannot | |
2756 * be NUL-terminated (because memory allocation failed). | |
2757 * | |
2758 * It can be called as many times as desired. | |
2759 * The pointer that it returns will remain valid until the UnicodeString objec
t is modified, | |
2760 * at which time the pointer is semantically invalidated and must not be used
any more. | |
2761 * | |
2762 * The capacity of the buffer can be determined with getCapacity(). | |
2763 * The part after length()+1 may or may not be initialized and valid, | |
2764 * depending on the history of the UnicodeString object. | |
2765 * | |
2766 * The buffer contents is guaranteed to be NUL-terminated. | |
2767 * getTerminatedBuffer() may reallocate the buffer if a terminating NUL | |
2768 * is written. | |
2769 * For this reason, this function is not const, unlike getBuffer(). | |
2770 * Note that a UnicodeString may also contain NUL characters as part of its co
ntents. | |
2771 * | |
2772 * The buffer may reside in read-only memory. Its contents must not | |
2773 * be modified. | |
2774 * | |
2775 * @return a read-only pointer to the internal string buffer, | |
2776 * or 0 if the string is empty or bogus | |
2777 * | |
2778 * @see getBuffer(int32_t minCapacity) | |
2779 * @see getBuffer() | |
2780 * @stable ICU 2.2 | |
2781 */ | |
2782 inline const UChar *getTerminatedBuffer(); | |
2783 | |
2784 //======================================== | |
2785 // Constructors | |
2786 //======================================== | |
2787 | |
2788 /** Construct an empty UnicodeString. | |
2789 * @stable ICU 2.0 | |
2790 */ | |
2791 UnicodeString(); | |
2792 | |
2793 /** | |
2794 * Construct a UnicodeString with capacity to hold <TT>capacity</TT> UChars | |
2795 * @param capacity the number of UChars this UnicodeString should hold | |
2796 * before a resize is necessary; if count is greater than 0 and count | |
2797 * code points c take up more space than capacity, then capacity is adjusted | |
2798 * accordingly. | |
2799 * @param c is used to initially fill the string | |
2800 * @param count specifies how many code points c are to be written in the | |
2801 * string | |
2802 * @stable ICU 2.0 | |
2803 */ | |
2804 UnicodeString(int32_t capacity, UChar32 c, int32_t count); | |
2805 | |
2806 /** | |
2807 * Single UChar (code unit) constructor. | |
2808 * @param ch the character to place in the UnicodeString | |
2809 * @stable ICU 2.0 | |
2810 */ | |
2811 UnicodeString(UChar ch); | |
2812 | |
2813 /** | |
2814 * Single UChar32 (code point) constructor. | |
2815 * @param ch the character to place in the UnicodeString | |
2816 * @stable ICU 2.0 | |
2817 */ | |
2818 UnicodeString(UChar32 ch); | |
2819 | |
2820 /** | |
2821 * UChar* constructor. | |
2822 * @param text The characters to place in the UnicodeString. <TT>text</TT> | |
2823 * must be NULL (U+0000) terminated. | |
2824 * @stable ICU 2.0 | |
2825 */ | |
2826 UnicodeString(const UChar *text); | |
2827 | |
2828 /** | |
2829 * UChar* constructor. | |
2830 * @param text The characters to place in the UnicodeString. | |
2831 * @param textLength The number of Unicode characters in <TT>text</TT> | |
2832 * to copy. | |
2833 * @stable ICU 2.0 | |
2834 */ | |
2835 UnicodeString(const UChar *text, | |
2836 int32_t textLength); | |
2837 | |
2838 /** | |
2839 * Readonly-aliasing UChar* constructor. | |
2840 * The text will be used for the UnicodeString object, but | |
2841 * it will not be released when the UnicodeString is destroyed. | |
2842 * This has copy-on-write semantics: | |
2843 * When the string is modified, then the buffer is first copied into | |
2844 * newly allocated memory. | |
2845 * The aliased buffer is never modified. | |
2846 * In an assignment to another UnicodeString, the text will be aliased again, | |
2847 * so that both strings then alias the same readonly-text. | |
2848 * | |
2849 * @param isTerminated specifies if <code>text</code> is <code>NUL</code>-term
inated. | |
2850 * This must be true if <code>textLength==-1</code>. | |
2851 * @param text The characters to alias for the UnicodeString. | |
2852 * @param textLength The number of Unicode characters in <code>text</code> to
alias. | |
2853 * If -1, then this constructor will determine the length | |
2854 * by calling <code>u_strlen()</code>. | |
2855 * @stable ICU 2.0 | |
2856 */ | |
2857 UnicodeString(UBool isTerminated, | |
2858 const UChar *text, | |
2859 int32_t textLength); | |
2860 | |
2861 /** | |
2862 * Writable-aliasing UChar* constructor. | |
2863 * The text will be used for the UnicodeString object, but | |
2864 * it will not be released when the UnicodeString is destroyed. | |
2865 * This has write-through semantics: | |
2866 * For as long as the capacity of the buffer is sufficient, write operations | |
2867 * will directly affect the buffer. When more capacity is necessary, then | |
2868 * a new buffer will be allocated and the contents copied as with regularly | |
2869 * constructed strings. | |
2870 * In an assignment to another UnicodeString, the buffer will be copied. | |
2871 * The extract(UChar *dst) function detects whether the dst pointer is the sam
e | |
2872 * as the string buffer itself and will in this case not copy the contents. | |
2873 * | |
2874 * @param buffer The characters to alias for the UnicodeString. | |
2875 * @param buffLength The number of Unicode characters in <code>buffer</code> t
o alias. | |
2876 * @param buffCapacity The size of <code>buffer</code> in UChars. | |
2877 * @stable ICU 2.0 | |
2878 */ | |
2879 UnicodeString(UChar *buffer, int32_t buffLength, int32_t buffCapacity); | |
2880 | |
2881 #if U_CHARSET_IS_UTF8 || !UCONFIG_NO_CONVERSION | |
2882 | |
2883 /** | |
2884 * char* constructor. | |
2885 * @param codepageData an array of bytes, null-terminated, | |
2886 * in the platform's default codepage. | |
2887 * @stable ICU 2.0 | |
2888 */ | |
2889 UnicodeString(const char *codepageData); | |
2890 | |
2891 /** | |
2892 * char* constructor. | |
2893 * @param codepageData an array of bytes in the platform's default codepage. | |
2894 * @param dataLength The number of bytes in <TT>codepageData</TT>. | |
2895 * @stable ICU 2.0 | |
2896 */ | |
2897 UnicodeString(const char *codepageData, int32_t dataLength); | |
2898 | |
2899 #endif | |
2900 | |
2901 #if !UCONFIG_NO_CONVERSION | |
2902 | |
2903 /** | |
2904 * char* constructor. | |
2905 * @param codepageData an array of bytes, null-terminated | |
2906 * @param codepage the encoding of <TT>codepageData</TT>. The special | |
2907 * value 0 for <TT>codepage</TT> indicates that the text is in the | |
2908 * platform's default codepage. | |
2909 * | |
2910 * If <code>codepage</code> is an empty string (<code>""</code>), | |
2911 * then a simple conversion is performed on the codepage-invariant | |
2912 * subset ("invariant characters") of the platform encoding. See utypes.h. | |
2913 * Recommendation: For invariant-character strings use the constructor | |
2914 * UnicodeString(const char *src, int32_t length, enum EInvariant inv) | |
2915 * because it avoids object code dependencies of UnicodeString on | |
2916 * the conversion code. | |
2917 * | |
2918 * @stable ICU 2.0 | |
2919 */ | |
2920 UnicodeString(const char *codepageData, const char *codepage); | |
2921 | |
2922 /** | |
2923 * char* constructor. | |
2924 * @param codepageData an array of bytes. | |
2925 * @param dataLength The number of bytes in <TT>codepageData</TT>. | |
2926 * @param codepage the encoding of <TT>codepageData</TT>. The special | |
2927 * value 0 for <TT>codepage</TT> indicates that the text is in the | |
2928 * platform's default codepage. | |
2929 * If <code>codepage</code> is an empty string (<code>""</code>), | |
2930 * then a simple conversion is performed on the codepage-invariant | |
2931 * subset ("invariant characters") of the platform encoding. See utypes.h. | |
2932 * Recommendation: For invariant-character strings use the constructor | |
2933 * UnicodeString(const char *src, int32_t length, enum EInvariant inv) | |
2934 * because it avoids object code dependencies of UnicodeString on | |
2935 * the conversion code. | |
2936 * | |
2937 * @stable ICU 2.0 | |
2938 */ | |
2939 UnicodeString(const char *codepageData, int32_t dataLength, const char *codepa
ge); | |
2940 | |
2941 /** | |
2942 * char * / UConverter constructor. | |
2943 * This constructor uses an existing UConverter object to | |
2944 * convert the codepage string to Unicode and construct a UnicodeString | |
2945 * from that. | |
2946 * | |
2947 * The converter is reset at first. | |
2948 * If the error code indicates a failure before this constructor is called, | |
2949 * or if an error occurs during conversion or construction, | |
2950 * then the string will be bogus. | |
2951 * | |
2952 * This function avoids the overhead of opening and closing a converter if | |
2953 * multiple strings are constructed. | |
2954 * | |
2955 * @param src input codepage string | |
2956 * @param srcLength length of the input string, can be -1 for NUL-terminated s
trings | |
2957 * @param cnv converter object (ucnv_resetToUnicode() will be called), | |
2958 * can be NULL for the default converter | |
2959 * @param errorCode normal ICU error code | |
2960 * @stable ICU 2.0 | |
2961 */ | |
2962 UnicodeString( | |
2963 const char *src, int32_t srcLength, | |
2964 UConverter *cnv, | |
2965 UErrorCode &errorCode); | |
2966 | |
2967 #endif | |
2968 | |
2969 /** | |
2970 * Constructs a Unicode string from an invariant-character char * string. | |
2971 * About invariant characters see utypes.h. | |
2972 * This constructor has no runtime dependency on conversion code and is | |
2973 * therefore recommended over ones taking a charset name string | |
2974 * (where the empty string "" indicates invariant-character conversion). | |
2975 * | |
2976 * Use the macro US_INV as the third, signature-distinguishing parameter. | |
2977 * | |
2978 * For example: | |
2979 * \code | |
2980 * void fn(const char *s) { | |
2981 * UnicodeString ustr(s, -1, US_INV); | |
2982 * // use ustr ... | |
2983 * } | |
2984 * \endcode | |
2985 * | |
2986 * @param src String using only invariant characters. | |
2987 * @param length Length of src, or -1 if NUL-terminated. | |
2988 * @param inv Signature-distinguishing paramater, use US_INV. | |
2989 * | |
2990 * @see US_INV | |
2991 * @stable ICU 3.2 | |
2992 */ | |
2993 UnicodeString(const char *src, int32_t length, enum EInvariant inv); | |
2994 | |
2995 | |
2996 /** | |
2997 * Copy constructor. | |
2998 * @param that The UnicodeString object to copy. | |
2999 * @stable ICU 2.0 | |
3000 */ | |
3001 UnicodeString(const UnicodeString& that); | |
3002 | |
3003 /** | |
3004 * 'Substring' constructor from tail of source string. | |
3005 * @param src The UnicodeString object to copy. | |
3006 * @param srcStart The offset into <tt>src</tt> at which to start copying. | |
3007 * @stable ICU 2.2 | |
3008 */ | |
3009 UnicodeString(const UnicodeString& src, int32_t srcStart); | |
3010 | |
3011 /** | |
3012 * 'Substring' constructor from subrange of source string. | |
3013 * @param src The UnicodeString object to copy. | |
3014 * @param srcStart The offset into <tt>src</tt> at which to start copying. | |
3015 * @param srcLength The number of characters from <tt>src</tt> to copy. | |
3016 * @stable ICU 2.2 | |
3017 */ | |
3018 UnicodeString(const UnicodeString& src, int32_t srcStart, int32_t srcLength); | |
3019 | |
3020 /** | |
3021 * Clone this object, an instance of a subclass of Replaceable. | |
3022 * Clones can be used concurrently in multiple threads. | |
3023 * If a subclass does not implement clone(), or if an error occurs, | |
3024 * then NULL is returned. | |
3025 * The clone functions in all subclasses return a pointer to a Replaceable | |
3026 * because some compilers do not support covariant (same-as-this) | |
3027 * return types; cast to the appropriate subclass if necessary. | |
3028 * The caller must delete the clone. | |
3029 * | |
3030 * @return a clone of this object | |
3031 * | |
3032 * @see Replaceable::clone | |
3033 * @see getDynamicClassID | |
3034 * @stable ICU 2.6 | |
3035 */ | |
3036 virtual Replaceable *clone() const; | |
3037 | |
3038 /** Destructor. | |
3039 * @stable ICU 2.0 | |
3040 */ | |
3041 virtual ~UnicodeString(); | |
3042 | |
3043 /** | |
3044 * Create a UnicodeString from a UTF-8 string. | |
3045 * Illegal input is replaced with U+FFFD. Otherwise, errors result in a bogus
string. | |
3046 * Calls u_strFromUTF8WithSub(). | |
3047 * | |
3048 * @param utf8 UTF-8 input string. | |
3049 * Note that a StringPiece can be implicitly constructed | |
3050 * from a std::string or a NUL-terminated const char * string. | |
3051 * @return A UnicodeString with equivalent UTF-16 contents. | |
3052 * @see toUTF8 | |
3053 * @see toUTF8String | |
3054 * @stable ICU 4.2 | |
3055 */ | |
3056 static UnicodeString fromUTF8(const StringPiece &utf8); | |
3057 | |
3058 /** | |
3059 * Create a UnicodeString from a UTF-32 string. | |
3060 * Illegal input is replaced with U+FFFD. Otherwise, errors result in a bogus
string. | |
3061 * Calls u_strFromUTF32WithSub(). | |
3062 * | |
3063 * @param utf32 UTF-32 input string. Must not be NULL. | |
3064 * @param length Length of the input string, or -1 if NUL-terminated. | |
3065 * @return A UnicodeString with equivalent UTF-16 contents. | |
3066 * @see toUTF32 | |
3067 * @stable ICU 4.2 | |
3068 */ | |
3069 static UnicodeString fromUTF32(const UChar32 *utf32, int32_t length); | |
3070 | |
3071 /* Miscellaneous operations */ | |
3072 | |
3073 /** | |
3074 * Unescape a string of characters and return a string containing | |
3075 * the result. The following escape sequences are recognized: | |
3076 * | |
3077 * \\uhhhh 4 hex digits; h in [0-9A-Fa-f] | |
3078 * \\Uhhhhhhhh 8 hex digits | |
3079 * \\xhh 1-2 hex digits | |
3080 * \\ooo 1-3 octal digits; o in [0-7] | |
3081 * \\cX control-X; X is masked with 0x1F | |
3082 * | |
3083 * as well as the standard ANSI C escapes: | |
3084 * | |
3085 * \\a => U+0007, \\b => U+0008, \\t => U+0009, \\n => U+000A, | |
3086 * \\v => U+000B, \\f => U+000C, \\r => U+000D, \\e => U+001B, | |
3087 * \\" => U+0022, \\' => U+0027, \\? => U+003F, \\\\ => U+005C | |
3088 * | |
3089 * Anything else following a backslash is generically escaped. For | |
3090 * example, "[a\\-z]" returns "[a-z]". | |
3091 * | |
3092 * If an escape sequence is ill-formed, this method returns an empty | |
3093 * string. An example of an ill-formed sequence is "\\u" followed by | |
3094 * fewer than 4 hex digits. | |
3095 * | |
3096 * This function is similar to u_unescape() but not identical to it. | |
3097 * The latter takes a source char*, so it does escape recognition | |
3098 * and also invariant conversion. | |
3099 * | |
3100 * @return a string with backslash escapes interpreted, or an | |
3101 * empty string on error. | |
3102 * @see UnicodeString#unescapeAt() | |
3103 * @see u_unescape() | |
3104 * @see u_unescapeAt() | |
3105 * @stable ICU 2.0 | |
3106 */ | |
3107 UnicodeString unescape() const; | |
3108 | |
3109 /** | |
3110 * Unescape a single escape sequence and return the represented | |
3111 * character. See unescape() for a listing of the recognized escape | |
3112 * sequences. The character at offset-1 is assumed (without | |
3113 * checking) to be a backslash. If the escape sequence is | |
3114 * ill-formed, or the offset is out of range, (UChar32)0xFFFFFFFF is | |
3115 * returned. | |
3116 * | |
3117 * @param offset an input output parameter. On input, it is the | |
3118 * offset into this string where the escape sequence is located, | |
3119 * after the initial backslash. On output, it is advanced after the | |
3120 * last character parsed. On error, it is not advanced at all. | |
3121 * @return the character represented by the escape sequence at | |
3122 * offset, or (UChar32)0xFFFFFFFF on error. | |
3123 * @see UnicodeString#unescape() | |
3124 * @see u_unescape() | |
3125 * @see u_unescapeAt() | |
3126 * @stable ICU 2.0 | |
3127 */ | |
3128 UChar32 unescapeAt(int32_t &offset) const; | |
3129 | |
3130 /** | |
3131 * ICU "poor man's RTTI", returns a UClassID for this class. | |
3132 * | |
3133 * @stable ICU 2.2 | |
3134 */ | |
3135 static UClassID U_EXPORT2 getStaticClassID(); | |
3136 | |
3137 /** | |
3138 * ICU "poor man's RTTI", returns a UClassID for the actual class. | |
3139 * | |
3140 * @stable ICU 2.2 | |
3141 */ | |
3142 virtual UClassID getDynamicClassID() const; | |
3143 | |
3144 //======================================== | |
3145 // Implementation methods | |
3146 //======================================== | |
3147 | |
3148 protected: | |
3149 /** | |
3150 * Implement Replaceable::getLength() (see jitterbug 1027). | |
3151 * @stable ICU 2.4 | |
3152 */ | |
3153 virtual int32_t getLength() const; | |
3154 | |
3155 /** | |
3156 * The change in Replaceable to use virtual getCharAt() allows | |
3157 * UnicodeString::charAt() to be inline again (see jitterbug 709). | |
3158 * @stable ICU 2.4 | |
3159 */ | |
3160 virtual UChar getCharAt(int32_t offset) const; | |
3161 | |
3162 /** | |
3163 * The change in Replaceable to use virtual getChar32At() allows | |
3164 * UnicodeString::char32At() to be inline again (see jitterbug 709). | |
3165 * @stable ICU 2.4 | |
3166 */ | |
3167 virtual UChar32 getChar32At(int32_t offset) const; | |
3168 | |
3169 private: | |
3170 // For char* constructors. Could be made public. | |
3171 UnicodeString &setToUTF8(const StringPiece &utf8); | |
3172 // For extract(char*). | |
3173 // We could make a toUTF8(target, capacity, errorCode) public but not | |
3174 // this version: New API will be cleaner if we make callers create substrings | |
3175 // rather than having start+length on every method, | |
3176 // and it should take a UErrorCode&. | |
3177 int32_t | |
3178 toUTF8(int32_t start, int32_t len, | |
3179 char *target, int32_t capacity) const; | |
3180 | |
3181 | |
3182 inline int8_t | |
3183 doCompare(int32_t start, | |
3184 int32_t length, | |
3185 const UnicodeString& srcText, | |
3186 int32_t srcStart, | |
3187 int32_t srcLength) const; | |
3188 | |
3189 int8_t doCompare(int32_t start, | |
3190 int32_t length, | |
3191 const UChar *srcChars, | |
3192 int32_t srcStart, | |
3193 int32_t srcLength) const; | |
3194 | |
3195 inline int8_t | |
3196 doCompareCodePointOrder(int32_t start, | |
3197 int32_t length, | |
3198 const UnicodeString& srcText, | |
3199 int32_t srcStart, | |
3200 int32_t srcLength) const; | |
3201 | |
3202 int8_t doCompareCodePointOrder(int32_t start, | |
3203 int32_t length, | |
3204 const UChar *srcChars, | |
3205 int32_t srcStart, | |
3206 int32_t srcLength) const; | |
3207 | |
3208 inline int8_t | |
3209 doCaseCompare(int32_t start, | |
3210 int32_t length, | |
3211 const UnicodeString &srcText, | |
3212 int32_t srcStart, | |
3213 int32_t srcLength, | |
3214 uint32_t options) const; | |
3215 | |
3216 int8_t | |
3217 doCaseCompare(int32_t start, | |
3218 int32_t length, | |
3219 const UChar *srcChars, | |
3220 int32_t srcStart, | |
3221 int32_t srcLength, | |
3222 uint32_t options) const; | |
3223 | |
3224 int32_t doIndexOf(UChar c, | |
3225 int32_t start, | |
3226 int32_t length) const; | |
3227 | |
3228 int32_t doIndexOf(UChar32 c, | |
3229 int32_t start, | |
3230 int32_t length) const; | |
3231 | |
3232 int32_t doLastIndexOf(UChar c, | |
3233 int32_t start, | |
3234 int32_t length) const; | |
3235 | |
3236 int32_t doLastIndexOf(UChar32 c, | |
3237 int32_t start, | |
3238 int32_t length) const; | |
3239 | |
3240 void doExtract(int32_t start, | |
3241 int32_t length, | |
3242 UChar *dst, | |
3243 int32_t dstStart) const; | |
3244 | |
3245 inline void doExtract(int32_t start, | |
3246 int32_t length, | |
3247 UnicodeString& target) const; | |
3248 | |
3249 inline UChar doCharAt(int32_t offset) const; | |
3250 | |
3251 UnicodeString& doReplace(int32_t start, | |
3252 int32_t length, | |
3253 const UnicodeString& srcText, | |
3254 int32_t srcStart, | |
3255 int32_t srcLength); | |
3256 | |
3257 UnicodeString& doReplace(int32_t start, | |
3258 int32_t length, | |
3259 const UChar *srcChars, | |
3260 int32_t srcStart, | |
3261 int32_t srcLength); | |
3262 | |
3263 UnicodeString& doReverse(int32_t start, | |
3264 int32_t length); | |
3265 | |
3266 // calculate hash code | |
3267 int32_t doHashCode(void) const; | |
3268 | |
3269 // get pointer to start of array | |
3270 // these do not check for kOpenGetBuffer, unlike the public getBuffer() functi
on | |
3271 inline UChar* getArrayStart(void); | |
3272 inline const UChar* getArrayStart(void) const; | |
3273 | |
3274 // A UnicodeString object (not necessarily its current buffer) | |
3275 // is writable unless it isBogus() or it has an "open" getBuffer(minCapacity). | |
3276 inline UBool isWritable() const; | |
3277 | |
3278 // Is the current buffer writable? | |
3279 inline UBool isBufferWritable() const; | |
3280 | |
3281 // None of the following does releaseArray(). | |
3282 inline void setLength(int32_t len); // sets only fShortLength and fLeng
th | |
3283 inline void setToEmpty(); // sets fFlags=kShortString | |
3284 inline void setArray(UChar *array, int32_t len, int32_t capacity); // does not
set fFlags | |
3285 | |
3286 // allocate the array; result may be fStackBuffer | |
3287 // sets refCount to 1 if appropriate | |
3288 // sets fArray, fCapacity, and fFlags | |
3289 // returns boolean for success or failure | |
3290 UBool allocate(int32_t capacity); | |
3291 | |
3292 // release the array if owned | |
3293 void releaseArray(void); | |
3294 | |
3295 // turn a bogus string into an empty one | |
3296 void unBogus(); | |
3297 | |
3298 // implements assigment operator, copy constructor, and fastCopyFrom() | |
3299 UnicodeString ©From(const UnicodeString &src, UBool fastCopy=FALSE); | |
3300 | |
3301 // Pin start and limit to acceptable values. | |
3302 inline void pinIndex(int32_t& start) const; | |
3303 inline void pinIndices(int32_t& start, | |
3304 int32_t& length) const; | |
3305 | |
3306 #if !UCONFIG_NO_CONVERSION | |
3307 | |
3308 /* Internal extract() using UConverter. */ | |
3309 int32_t doExtract(int32_t start, int32_t length, | |
3310 char *dest, int32_t destCapacity, | |
3311 UConverter *cnv, | |
3312 UErrorCode &errorCode) const; | |
3313 | |
3314 /* | |
3315 * Real constructor for converting from codepage data. | |
3316 * It assumes that it is called with !fRefCounted. | |
3317 * | |
3318 * If <code>codepage==0</code>, then the default converter | |
3319 * is used for the platform encoding. | |
3320 * If <code>codepage</code> is an empty string (<code>""</code>), | |
3321 * then a simple conversion is performed on the codepage-invariant | |
3322 * subset ("invariant characters") of the platform encoding. See utypes.h. | |
3323 */ | |
3324 void doCodepageCreate(const char *codepageData, | |
3325 int32_t dataLength, | |
3326 const char *codepage); | |
3327 | |
3328 /* | |
3329 * Worker function for creating a UnicodeString from | |
3330 * a codepage string using a UConverter. | |
3331 */ | |
3332 void | |
3333 doCodepageCreate(const char *codepageData, | |
3334 int32_t dataLength, | |
3335 UConverter *converter, | |
3336 UErrorCode &status); | |
3337 | |
3338 #endif | |
3339 | |
3340 /* | |
3341 * This function is called when write access to the array | |
3342 * is necessary. | |
3343 * | |
3344 * We need to make a copy of the array if | |
3345 * the buffer is read-only, or | |
3346 * the buffer is refCounted (shared), and refCount>1, or | |
3347 * the buffer is too small. | |
3348 * | |
3349 * Return FALSE if memory could not be allocated. | |
3350 */ | |
3351 UBool cloneArrayIfNeeded(int32_t newCapacity = -1, | |
3352 int32_t growCapacity = -1, | |
3353 UBool doCopyArray = TRUE, | |
3354 int32_t **pBufferToDelete = 0, | |
3355 UBool forceClone = FALSE); | |
3356 | |
3357 // common function for case mappings | |
3358 UnicodeString & | |
3359 caseMap(BreakIterator *titleIter, | |
3360 const char *locale, | |
3361 uint32_t options, | |
3362 int32_t toWhichCase); | |
3363 | |
3364 // ref counting | |
3365 void addRef(void); | |
3366 int32_t removeRef(void); | |
3367 int32_t refCount(void) const; | |
3368 | |
3369 // constants | |
3370 enum { | |
3371 // Set the stack buffer size so that sizeof(UnicodeString) is a multiple of
sizeof(pointer): | |
3372 // 32-bit pointers: 4+1+1+13*2 = 32 bytes | |
3373 // 64-bit pointers: 8+1+1+15*2 = 40 bytes | |
3374 US_STACKBUF_SIZE= sizeof(void *)==4 ? 13 : 15, // Size of stack buffer for s
mall strings | |
3375 kInvalidUChar=0xffff, // invalid UChar index | |
3376 kGrowSize=128, // grow size for this buffer | |
3377 kInvalidHashCode=0, // invalid hash code | |
3378 kEmptyHashCode=1, // hash code for empty string | |
3379 | |
3380 // bit flag values for fFlags | |
3381 kIsBogus=1, // this string is bogus, i.e., not valid or NULL | |
3382 kUsingStackBuffer=2,// fArray==fStackBuffer | |
3383 kRefCounted=4, // there is a refCount field before the characters in fA
rray | |
3384 kBufferIsReadonly=8,// do not write to this buffer | |
3385 kOpenGetBuffer=16, // getBuffer(minCapacity) was called (is "open"), | |
3386 // and releaseBuffer(newLength) must be called | |
3387 | |
3388 // combined values for convenience | |
3389 kShortString=kUsingStackBuffer, | |
3390 kLongString=kRefCounted, | |
3391 kReadonlyAlias=kBufferIsReadonly, | |
3392 kWritableAlias=0 | |
3393 }; | |
3394 | |
3395 friend class StringThreadTest; | |
3396 | |
3397 union StackBufferOrFields; // forward declaration necessary before frie
nd declaration | |
3398 friend union StackBufferOrFields; // make US_STACKBUF_SIZE visible inside fUni
on | |
3399 | |
3400 /* | |
3401 * The following are all the class fields that are stored | |
3402 * in each UnicodeString object. | |
3403 * Note that UnicodeString has virtual functions, | |
3404 * therefore there is an implicit vtable pointer | |
3405 * as the first real field. | |
3406 * The fields should be aligned such that no padding is | |
3407 * necessary, mostly by having larger types first. | |
3408 * On 32-bit machines, the size should be 32 bytes, | |
3409 * on 64-bit machines (8-byte pointers), it should be 40 bytes. | |
3410 */ | |
3411 // (implicit) *vtable; | |
3412 int8_t fShortLength; // 0..127: length <0: real length is in fUnion.fFie
lds.fLength | |
3413 uint8_t fFlags; // bit flags: see constants above | |
3414 union StackBufferOrFields { | |
3415 // fStackBuffer is used iff (fFlags&kUsingStackBuffer) | |
3416 // else fFields is used | |
3417 UChar fStackBuffer [US_STACKBUF_SIZE]; // buffer for small strings | |
3418 struct { | |
3419 uint16_t fPadding; // align the following field at 8B (32b pointers) or
12B (64b) | |
3420 int32_t fLength; // number of characters in fArray if >127; else unde
fined | |
3421 UChar *fArray; // the Unicode data (aligned at 12B (32b pointers) o
r 16B (64b)) | |
3422 int32_t fCapacity; // sizeof fArray | |
3423 } fFields; | |
3424 } fUnion; | |
3425 }; | |
3426 | |
3427 /** | |
3428 * Create a new UnicodeString with the concatenation of two others. | |
3429 * | |
3430 * @param s1 The first string to be copied to the new one. | |
3431 * @param s2 The second string to be copied to the new one, after s1. | |
3432 * @return UnicodeString(s1).append(s2) | |
3433 * @stable ICU 2.8 | |
3434 */ | |
3435 U_COMMON_API UnicodeString U_EXPORT2 | |
3436 operator+ (const UnicodeString &s1, const UnicodeString &s2); | |
3437 | |
3438 //======================================== | |
3439 // Inline members | |
3440 //======================================== | |
3441 | |
3442 //======================================== | |
3443 // Privates | |
3444 //======================================== | |
3445 | |
3446 inline void | |
3447 UnicodeString::pinIndex(int32_t& start) const | |
3448 { | |
3449 // pin index | |
3450 if(start < 0) { | |
3451 start = 0; | |
3452 } else if(start > length()) { | |
3453 start = length(); | |
3454 } | |
3455 } | |
3456 | |
3457 inline void | |
3458 UnicodeString::pinIndices(int32_t& start, | |
3459 int32_t& _length) const | |
3460 { | |
3461 // pin indices | |
3462 int32_t len = length(); | |
3463 if(start < 0) { | |
3464 start = 0; | |
3465 } else if(start > len) { | |
3466 start = len; | |
3467 } | |
3468 if(_length < 0) { | |
3469 _length = 0; | |
3470 } else if(_length > (len - start)) { | |
3471 _length = (len - start); | |
3472 } | |
3473 } | |
3474 | |
3475 inline UChar* | |
3476 UnicodeString::getArrayStart() | |
3477 { return (fFlags&kUsingStackBuffer) ? fUnion.fStackBuffer : fUnion.fFields.fArra
y; } | |
3478 | |
3479 inline const UChar* | |
3480 UnicodeString::getArrayStart() const | |
3481 { return (fFlags&kUsingStackBuffer) ? fUnion.fStackBuffer : fUnion.fFields.fArra
y; } | |
3482 | |
3483 //======================================== | |
3484 // Read-only implementation methods | |
3485 //======================================== | |
3486 inline int32_t | |
3487 UnicodeString::length() const | |
3488 { return fShortLength>=0 ? fShortLength : fUnion.fFields.fLength; } | |
3489 | |
3490 inline int32_t | |
3491 UnicodeString::getCapacity() const | |
3492 { return (fFlags&kUsingStackBuffer) ? US_STACKBUF_SIZE : fUnion.fFields.fCapacit
y; } | |
3493 | |
3494 inline int32_t | |
3495 UnicodeString::hashCode() const | |
3496 { return doHashCode(); } | |
3497 | |
3498 inline UBool | |
3499 UnicodeString::isBogus() const | |
3500 { return (UBool)(fFlags & kIsBogus); } | |
3501 | |
3502 inline UBool | |
3503 UnicodeString::isWritable() const | |
3504 { return (UBool)!(fFlags&(kOpenGetBuffer|kIsBogus)); } | |
3505 | |
3506 inline UBool | |
3507 UnicodeString::isBufferWritable() const | |
3508 { | |
3509 return (UBool)( | |
3510 !(fFlags&(kOpenGetBuffer|kIsBogus|kBufferIsReadonly)) && | |
3511 (!(fFlags&kRefCounted) || refCount()==1)); | |
3512 } | |
3513 | |
3514 inline const UChar * | |
3515 UnicodeString::getBuffer() const { | |
3516 if(fFlags&(kIsBogus|kOpenGetBuffer)) { | |
3517 return 0; | |
3518 } else if(fFlags&kUsingStackBuffer) { | |
3519 return fUnion.fStackBuffer; | |
3520 } else { | |
3521 return fUnion.fFields.fArray; | |
3522 } | |
3523 } | |
3524 | |
3525 //======================================== | |
3526 // Read-only alias methods | |
3527 //======================================== | |
3528 inline int8_t | |
3529 UnicodeString::doCompare(int32_t start, | |
3530 int32_t thisLength, | |
3531 const UnicodeString& srcText, | |
3532 int32_t srcStart, | |
3533 int32_t srcLength) const | |
3534 { | |
3535 if(srcText.isBogus()) { | |
3536 return (int8_t)!isBogus(); // 0 if both are bogus, 1 otherwise | |
3537 } else { | |
3538 srcText.pinIndices(srcStart, srcLength); | |
3539 return doCompare(start, thisLength, srcText.getArrayStart(), srcStart, srcLe
ngth); | |
3540 } | |
3541 } | |
3542 | |
3543 inline UBool | |
3544 UnicodeString::operator== (const UnicodeString& text) const | |
3545 { | |
3546 if(isBogus()) { | |
3547 return text.isBogus(); | |
3548 } else { | |
3549 int32_t len = length(), textLength = text.length(); | |
3550 return | |
3551 !text.isBogus() && | |
3552 len == textLength && | |
3553 doCompare(0, len, text, 0, textLength) == 0; | |
3554 } | |
3555 } | |
3556 | |
3557 inline UBool | |
3558 UnicodeString::operator!= (const UnicodeString& text) const | |
3559 { return (! operator==(text)); } | |
3560 | |
3561 inline UBool | |
3562 UnicodeString::operator> (const UnicodeString& text) const | |
3563 { return doCompare(0, length(), text, 0, text.length()) == 1; } | |
3564 | |
3565 inline UBool | |
3566 UnicodeString::operator< (const UnicodeString& text) const | |
3567 { return doCompare(0, length(), text, 0, text.length()) == -1; } | |
3568 | |
3569 inline UBool | |
3570 UnicodeString::operator>= (const UnicodeString& text) const | |
3571 { return doCompare(0, length(), text, 0, text.length()) != -1; } | |
3572 | |
3573 inline UBool | |
3574 UnicodeString::operator<= (const UnicodeString& text) const | |
3575 { return doCompare(0, length(), text, 0, text.length()) != 1; } | |
3576 | |
3577 inline int8_t | |
3578 UnicodeString::compare(const UnicodeString& text) const | |
3579 { return doCompare(0, length(), text, 0, text.length()); } | |
3580 | |
3581 inline int8_t | |
3582 UnicodeString::compare(int32_t start, | |
3583 int32_t _length, | |
3584 const UnicodeString& srcText) const | |
3585 { return doCompare(start, _length, srcText, 0, srcText.length()); } | |
3586 | |
3587 inline int8_t | |
3588 UnicodeString::compare(const UChar *srcChars, | |
3589 int32_t srcLength) const | |
3590 { return doCompare(0, length(), srcChars, 0, srcLength); } | |
3591 | |
3592 inline int8_t | |
3593 UnicodeString::compare(int32_t start, | |
3594 int32_t _length, | |
3595 const UnicodeString& srcText, | |
3596 int32_t srcStart, | |
3597 int32_t srcLength) const | |
3598 { return doCompare(start, _length, srcText, srcStart, srcLength); } | |
3599 | |
3600 inline int8_t | |
3601 UnicodeString::compare(int32_t start, | |
3602 int32_t _length, | |
3603 const UChar *srcChars) const | |
3604 { return doCompare(start, _length, srcChars, 0, _length); } | |
3605 | |
3606 inline int8_t | |
3607 UnicodeString::compare(int32_t start, | |
3608 int32_t _length, | |
3609 const UChar *srcChars, | |
3610 int32_t srcStart, | |
3611 int32_t srcLength) const | |
3612 { return doCompare(start, _length, srcChars, srcStart, srcLength); } | |
3613 | |
3614 inline int8_t | |
3615 UnicodeString::compareBetween(int32_t start, | |
3616 int32_t limit, | |
3617 const UnicodeString& srcText, | |
3618 int32_t srcStart, | |
3619 int32_t srcLimit) const | |
3620 { return doCompare(start, limit - start, | |
3621 srcText, srcStart, srcLimit - srcStart); } | |
3622 | |
3623 inline int8_t | |
3624 UnicodeString::doCompareCodePointOrder(int32_t start, | |
3625 int32_t thisLength, | |
3626 const UnicodeString& srcText, | |
3627 int32_t srcStart, | |
3628 int32_t srcLength) const | |
3629 { | |
3630 if(srcText.isBogus()) { | |
3631 return (int8_t)!isBogus(); // 0 if both are bogus, 1 otherwise | |
3632 } else { | |
3633 srcText.pinIndices(srcStart, srcLength); | |
3634 return doCompareCodePointOrder(start, thisLength, srcText.getArrayStart(), s
rcStart, srcLength); | |
3635 } | |
3636 } | |
3637 | |
3638 inline int8_t | |
3639 UnicodeString::compareCodePointOrder(const UnicodeString& text) const | |
3640 { return doCompareCodePointOrder(0, length(), text, 0, text.length()); } | |
3641 | |
3642 inline int8_t | |
3643 UnicodeString::compareCodePointOrder(int32_t start, | |
3644 int32_t _length, | |
3645 const UnicodeString& srcText) const | |
3646 { return doCompareCodePointOrder(start, _length, srcText, 0, srcText.length());
} | |
3647 | |
3648 inline int8_t | |
3649 UnicodeString::compareCodePointOrder(const UChar *srcChars, | |
3650 int32_t srcLength) const | |
3651 { return doCompareCodePointOrder(0, length(), srcChars, 0, srcLength); } | |
3652 | |
3653 inline int8_t | |
3654 UnicodeString::compareCodePointOrder(int32_t start, | |
3655 int32_t _length, | |
3656 const UnicodeString& srcText, | |
3657 int32_t srcStart, | |
3658 int32_t srcLength) const | |
3659 { return doCompareCodePointOrder(start, _length, srcText, srcStart, srcLength);
} | |
3660 | |
3661 inline int8_t | |
3662 UnicodeString::compareCodePointOrder(int32_t start, | |
3663 int32_t _length, | |
3664 const UChar *srcChars) const | |
3665 { return doCompareCodePointOrder(start, _length, srcChars, 0, _length); } | |
3666 | |
3667 inline int8_t | |
3668 UnicodeString::compareCodePointOrder(int32_t start, | |
3669 int32_t _length, | |
3670 const UChar *srcChars, | |
3671 int32_t srcStart, | |
3672 int32_t srcLength) const | |
3673 { return doCompareCodePointOrder(start, _length, srcChars, srcStart, srcLength);
} | |
3674 | |
3675 inline int8_t | |
3676 UnicodeString::compareCodePointOrderBetween(int32_t start, | |
3677 int32_t limit, | |
3678 const UnicodeString& srcText, | |
3679 int32_t srcStart, | |
3680 int32_t srcLimit) const | |
3681 { return doCompareCodePointOrder(start, limit - start, | |
3682 srcText, srcStart, srcLimit - srcStart); } | |
3683 | |
3684 inline int8_t | |
3685 UnicodeString::doCaseCompare(int32_t start, | |
3686 int32_t thisLength, | |
3687 const UnicodeString &srcText, | |
3688 int32_t srcStart, | |
3689 int32_t srcLength, | |
3690 uint32_t options) const | |
3691 { | |
3692 if(srcText.isBogus()) { | |
3693 return (int8_t)!isBogus(); // 0 if both are bogus, 1 otherwise | |
3694 } else { | |
3695 srcText.pinIndices(srcStart, srcLength); | |
3696 return doCaseCompare(start, thisLength, srcText.getArrayStart(), srcStart, s
rcLength, options); | |
3697 } | |
3698 } | |
3699 | |
3700 inline int8_t | |
3701 UnicodeString::caseCompare(const UnicodeString &text, uint32_t options) const { | |
3702 return doCaseCompare(0, length(), text, 0, text.length(), options); | |
3703 } | |
3704 | |
3705 inline int8_t | |
3706 UnicodeString::caseCompare(int32_t start, | |
3707 int32_t _length, | |
3708 const UnicodeString &srcText, | |
3709 uint32_t options) const { | |
3710 return doCaseCompare(start, _length, srcText, 0, srcText.length(), options); | |
3711 } | |
3712 | |
3713 inline int8_t | |
3714 UnicodeString::caseCompare(const UChar *srcChars, | |
3715 int32_t srcLength, | |
3716 uint32_t options) const { | |
3717 return doCaseCompare(0, length(), srcChars, 0, srcLength, options); | |
3718 } | |
3719 | |
3720 inline int8_t | |
3721 UnicodeString::caseCompare(int32_t start, | |
3722 int32_t _length, | |
3723 const UnicodeString &srcText, | |
3724 int32_t srcStart, | |
3725 int32_t srcLength, | |
3726 uint32_t options) const { | |
3727 return doCaseCompare(start, _length, srcText, srcStart, srcLength, options); | |
3728 } | |
3729 | |
3730 inline int8_t | |
3731 UnicodeString::caseCompare(int32_t start, | |
3732 int32_t _length, | |
3733 const UChar *srcChars, | |
3734 uint32_t options) const { | |
3735 return doCaseCompare(start, _length, srcChars, 0, _length, options); | |
3736 } | |
3737 | |
3738 inline int8_t | |
3739 UnicodeString::caseCompare(int32_t start, | |
3740 int32_t _length, | |
3741 const UChar *srcChars, | |
3742 int32_t srcStart, | |
3743 int32_t srcLength, | |
3744 uint32_t options) const { | |
3745 return doCaseCompare(start, _length, srcChars, srcStart, srcLength, options); | |
3746 } | |
3747 | |
3748 inline int8_t | |
3749 UnicodeString::caseCompareBetween(int32_t start, | |
3750 int32_t limit, | |
3751 const UnicodeString &srcText, | |
3752 int32_t srcStart, | |
3753 int32_t srcLimit, | |
3754 uint32_t options) const { | |
3755 return doCaseCompare(start, limit - start, srcText, srcStart, srcLimit - srcSt
art, options); | |
3756 } | |
3757 | |
3758 inline int32_t | |
3759 UnicodeString::indexOf(const UnicodeString& srcText, | |
3760 int32_t srcStart, | |
3761 int32_t srcLength, | |
3762 int32_t start, | |
3763 int32_t _length) const | |
3764 { | |
3765 if(!srcText.isBogus()) { | |
3766 srcText.pinIndices(srcStart, srcLength); | |
3767 if(srcLength > 0) { | |
3768 return indexOf(srcText.getArrayStart(), srcStart, srcLength, start, _lengt
h); | |
3769 } | |
3770 } | |
3771 return -1; | |
3772 } | |
3773 | |
3774 inline int32_t | |
3775 UnicodeString::indexOf(const UnicodeString& text) const | |
3776 { return indexOf(text, 0, text.length(), 0, length()); } | |
3777 | |
3778 inline int32_t | |
3779 UnicodeString::indexOf(const UnicodeString& text, | |
3780 int32_t start) const { | |
3781 pinIndex(start); | |
3782 return indexOf(text, 0, text.length(), start, length() - start); | |
3783 } | |
3784 | |
3785 inline int32_t | |
3786 UnicodeString::indexOf(const UnicodeString& text, | |
3787 int32_t start, | |
3788 int32_t _length) const | |
3789 { return indexOf(text, 0, text.length(), start, _length); } | |
3790 | |
3791 inline int32_t | |
3792 UnicodeString::indexOf(const UChar *srcChars, | |
3793 int32_t srcLength, | |
3794 int32_t start) const { | |
3795 pinIndex(start); | |
3796 return indexOf(srcChars, 0, srcLength, start, length() - start); | |
3797 } | |
3798 | |
3799 inline int32_t | |
3800 UnicodeString::indexOf(const UChar *srcChars, | |
3801 int32_t srcLength, | |
3802 int32_t start, | |
3803 int32_t _length) const | |
3804 { return indexOf(srcChars, 0, srcLength, start, _length); } | |
3805 | |
3806 inline int32_t | |
3807 UnicodeString::indexOf(UChar c, | |
3808 int32_t start, | |
3809 int32_t _length) const | |
3810 { return doIndexOf(c, start, _length); } | |
3811 | |
3812 inline int32_t | |
3813 UnicodeString::indexOf(UChar32 c, | |
3814 int32_t start, | |
3815 int32_t _length) const | |
3816 { return doIndexOf(c, start, _length); } | |
3817 | |
3818 inline int32_t | |
3819 UnicodeString::indexOf(UChar c) const | |
3820 { return doIndexOf(c, 0, length()); } | |
3821 | |
3822 inline int32_t | |
3823 UnicodeString::indexOf(UChar32 c) const | |
3824 { return indexOf(c, 0, length()); } | |
3825 | |
3826 inline int32_t | |
3827 UnicodeString::indexOf(UChar c, | |
3828 int32_t start) const { | |
3829 pinIndex(start); | |
3830 return doIndexOf(c, start, length() - start); | |
3831 } | |
3832 | |
3833 inline int32_t | |
3834 UnicodeString::indexOf(UChar32 c, | |
3835 int32_t start) const { | |
3836 pinIndex(start); | |
3837 return indexOf(c, start, length() - start); | |
3838 } | |
3839 | |
3840 inline int32_t | |
3841 UnicodeString::lastIndexOf(const UChar *srcChars, | |
3842 int32_t srcLength, | |
3843 int32_t start, | |
3844 int32_t _length) const | |
3845 { return lastIndexOf(srcChars, 0, srcLength, start, _length); } | |
3846 | |
3847 inline int32_t | |
3848 UnicodeString::lastIndexOf(const UChar *srcChars, | |
3849 int32_t srcLength, | |
3850 int32_t start) const { | |
3851 pinIndex(start); | |
3852 return lastIndexOf(srcChars, 0, srcLength, start, length() - start); | |
3853 } | |
3854 | |
3855 inline int32_t | |
3856 UnicodeString::lastIndexOf(const UnicodeString& srcText, | |
3857 int32_t srcStart, | |
3858 int32_t srcLength, | |
3859 int32_t start, | |
3860 int32_t _length) const | |
3861 { | |
3862 if(!srcText.isBogus()) { | |
3863 srcText.pinIndices(srcStart, srcLength); | |
3864 if(srcLength > 0) { | |
3865 return lastIndexOf(srcText.getArrayStart(), srcStart, srcLength, start, _l
ength); | |
3866 } | |
3867 } | |
3868 return -1; | |
3869 } | |
3870 | |
3871 inline int32_t | |
3872 UnicodeString::lastIndexOf(const UnicodeString& text, | |
3873 int32_t start, | |
3874 int32_t _length) const | |
3875 { return lastIndexOf(text, 0, text.length(), start, _length); } | |
3876 | |
3877 inline int32_t | |
3878 UnicodeString::lastIndexOf(const UnicodeString& text, | |
3879 int32_t start) const { | |
3880 pinIndex(start); | |
3881 return lastIndexOf(text, 0, text.length(), start, length() - start); | |
3882 } | |
3883 | |
3884 inline int32_t | |
3885 UnicodeString::lastIndexOf(const UnicodeString& text) const | |
3886 { return lastIndexOf(text, 0, text.length(), 0, length()); } | |
3887 | |
3888 inline int32_t | |
3889 UnicodeString::lastIndexOf(UChar c, | |
3890 int32_t start, | |
3891 int32_t _length) const | |
3892 { return doLastIndexOf(c, start, _length); } | |
3893 | |
3894 inline int32_t | |
3895 UnicodeString::lastIndexOf(UChar32 c, | |
3896 int32_t start, | |
3897 int32_t _length) const { | |
3898 return doLastIndexOf(c, start, _length); | |
3899 } | |
3900 | |
3901 inline int32_t | |
3902 UnicodeString::lastIndexOf(UChar c) const | |
3903 { return doLastIndexOf(c, 0, length()); } | |
3904 | |
3905 inline int32_t | |
3906 UnicodeString::lastIndexOf(UChar32 c) const { | |
3907 return lastIndexOf(c, 0, length()); | |
3908 } | |
3909 | |
3910 inline int32_t | |
3911 UnicodeString::lastIndexOf(UChar c, | |
3912 int32_t start) const { | |
3913 pinIndex(start); | |
3914 return doLastIndexOf(c, start, length() - start); | |
3915 } | |
3916 | |
3917 inline int32_t | |
3918 UnicodeString::lastIndexOf(UChar32 c, | |
3919 int32_t start) const { | |
3920 pinIndex(start); | |
3921 return lastIndexOf(c, start, length() - start); | |
3922 } | |
3923 | |
3924 inline UBool | |
3925 UnicodeString::startsWith(const UnicodeString& text) const | |
3926 { return compare(0, text.length(), text, 0, text.length()) == 0; } | |
3927 | |
3928 inline UBool | |
3929 UnicodeString::startsWith(const UnicodeString& srcText, | |
3930 int32_t srcStart, | |
3931 int32_t srcLength) const | |
3932 { return doCompare(0, srcLength, srcText, srcStart, srcLength) == 0; } | |
3933 | |
3934 inline UBool | |
3935 UnicodeString::startsWith(const UChar *srcChars, | |
3936 int32_t srcLength) const | |
3937 { return doCompare(0, srcLength, srcChars, 0, srcLength) == 0; } | |
3938 | |
3939 inline UBool | |
3940 UnicodeString::startsWith(const UChar *srcChars, | |
3941 int32_t srcStart, | |
3942 int32_t srcLength) const | |
3943 { return doCompare(0, srcLength, srcChars, srcStart, srcLength) == 0;} | |
3944 | |
3945 inline UBool | |
3946 UnicodeString::endsWith(const UnicodeString& text) const | |
3947 { return doCompare(length() - text.length(), text.length(), | |
3948 text, 0, text.length()) == 0; } | |
3949 | |
3950 inline UBool | |
3951 UnicodeString::endsWith(const UnicodeString& srcText, | |
3952 int32_t srcStart, | |
3953 int32_t srcLength) const { | |
3954 srcText.pinIndices(srcStart, srcLength); | |
3955 return doCompare(length() - srcLength, srcLength, | |
3956 srcText, srcStart, srcLength) == 0; | |
3957 } | |
3958 | |
3959 inline UBool | |
3960 UnicodeString::endsWith(const UChar *srcChars, | |
3961 int32_t srcLength) const { | |
3962 if(srcLength < 0) { | |
3963 srcLength = u_strlen(srcChars); | |
3964 } | |
3965 return doCompare(length() - srcLength, srcLength, | |
3966 srcChars, 0, srcLength) == 0; | |
3967 } | |
3968 | |
3969 inline UBool | |
3970 UnicodeString::endsWith(const UChar *srcChars, | |
3971 int32_t srcStart, | |
3972 int32_t srcLength) const { | |
3973 if(srcLength < 0) { | |
3974 srcLength = u_strlen(srcChars + srcStart); | |
3975 } | |
3976 return doCompare(length() - srcLength, srcLength, | |
3977 srcChars, srcStart, srcLength) == 0; | |
3978 } | |
3979 | |
3980 //======================================== | |
3981 // replace | |
3982 //======================================== | |
3983 inline UnicodeString& | |
3984 UnicodeString::replace(int32_t start, | |
3985 int32_t _length, | |
3986 const UnicodeString& srcText) | |
3987 { return doReplace(start, _length, srcText, 0, srcText.length()); } | |
3988 | |
3989 inline UnicodeString& | |
3990 UnicodeString::replace(int32_t start, | |
3991 int32_t _length, | |
3992 const UnicodeString& srcText, | |
3993 int32_t srcStart, | |
3994 int32_t srcLength) | |
3995 { return doReplace(start, _length, srcText, srcStart, srcLength); } | |
3996 | |
3997 inline UnicodeString& | |
3998 UnicodeString::replace(int32_t start, | |
3999 int32_t _length, | |
4000 const UChar *srcChars, | |
4001 int32_t srcLength) | |
4002 { return doReplace(start, _length, srcChars, 0, srcLength); } | |
4003 | |
4004 inline UnicodeString& | |
4005 UnicodeString::replace(int32_t start, | |
4006 int32_t _length, | |
4007 const UChar *srcChars, | |
4008 int32_t srcStart, | |
4009 int32_t srcLength) | |
4010 { return doReplace(start, _length, srcChars, srcStart, srcLength); } | |
4011 | |
4012 inline UnicodeString& | |
4013 UnicodeString::replace(int32_t start, | |
4014 int32_t _length, | |
4015 UChar srcChar) | |
4016 { return doReplace(start, _length, &srcChar, 0, 1); } | |
4017 | |
4018 inline UnicodeString& | |
4019 UnicodeString::replace(int32_t start, | |
4020 int32_t _length, | |
4021 UChar32 srcChar) { | |
4022 UChar buffer[U16_MAX_LENGTH]; | |
4023 int32_t count = 0; | |
4024 UBool isError = FALSE; | |
4025 U16_APPEND(buffer, count, U16_MAX_LENGTH, srcChar, isError); | |
4026 (void) isError; | |
4027 return doReplace(start, _length, buffer, 0, count); | |
4028 } | |
4029 | |
4030 inline UnicodeString& | |
4031 UnicodeString::replaceBetween(int32_t start, | |
4032 int32_t limit, | |
4033 const UnicodeString& srcText) | |
4034 { return doReplace(start, limit - start, srcText, 0, srcText.length()); } | |
4035 | |
4036 inline UnicodeString& | |
4037 UnicodeString::replaceBetween(int32_t start, | |
4038 int32_t limit, | |
4039 const UnicodeString& srcText, | |
4040 int32_t srcStart, | |
4041 int32_t srcLimit) | |
4042 { return doReplace(start, limit - start, srcText, srcStart, srcLimit - srcStart)
; } | |
4043 | |
4044 inline UnicodeString& | |
4045 UnicodeString::findAndReplace(const UnicodeString& oldText, | |
4046 const UnicodeString& newText) | |
4047 { return findAndReplace(0, length(), oldText, 0, oldText.length(), | |
4048 newText, 0, newText.length()); } | |
4049 | |
4050 inline UnicodeString& | |
4051 UnicodeString::findAndReplace(int32_t start, | |
4052 int32_t _length, | |
4053 const UnicodeString& oldText, | |
4054 const UnicodeString& newText) | |
4055 { return findAndReplace(start, _length, oldText, 0, oldText.length(), | |
4056 newText, 0, newText.length()); } | |
4057 | |
4058 // ============================ | |
4059 // extract | |
4060 // ============================ | |
4061 inline void | |
4062 UnicodeString::doExtract(int32_t start, | |
4063 int32_t _length, | |
4064 UnicodeString& target) const | |
4065 { target.replace(0, target.length(), *this, start, _length); } | |
4066 | |
4067 inline void | |
4068 UnicodeString::extract(int32_t start, | |
4069 int32_t _length, | |
4070 UChar *target, | |
4071 int32_t targetStart) const | |
4072 { doExtract(start, _length, target, targetStart); } | |
4073 | |
4074 inline void | |
4075 UnicodeString::extract(int32_t start, | |
4076 int32_t _length, | |
4077 UnicodeString& target) const | |
4078 { doExtract(start, _length, target); } | |
4079 | |
4080 #if !UCONFIG_NO_CONVERSION | |
4081 | |
4082 inline int32_t | |
4083 UnicodeString::extract(int32_t start, | |
4084 int32_t _length, | |
4085 char *dst, | |
4086 const char *codepage) const | |
4087 | |
4088 { | |
4089 // This dstSize value will be checked explicitly | |
4090 #if defined(__GNUC__) | |
4091 // Ticket #7039: Clip length to the maximum valid length to the end of address
able memory given the starting address | |
4092 // This is only an issue when using GCC and certain optimizations are turned o
n. | |
4093 return extract(start, _length, dst, dst!=0 ? ((dst >= (char*)((size_t)-1) - UI
NT32_MAX) ? static_cast<unsigned int>((((char*)UINT32_MAX) - dst)) : UINT32_MAX)
: 0, codepage); | |
4094 #else | |
4095 return extract(start, _length, dst, dst!=0 ? 0xffffffff : 0, codepage); | |
4096 #endif | |
4097 } | |
4098 | |
4099 #endif | |
4100 | |
4101 inline void | |
4102 UnicodeString::extractBetween(int32_t start, | |
4103 int32_t limit, | |
4104 UChar *dst, | |
4105 int32_t dstStart) const { | |
4106 pinIndex(start); | |
4107 pinIndex(limit); | |
4108 doExtract(start, limit - start, dst, dstStart); | |
4109 } | |
4110 | |
4111 inline UnicodeString | |
4112 UnicodeString::tempSubStringBetween(int32_t start, int32_t limit) const { | |
4113 return tempSubString(start, limit - start); | |
4114 } | |
4115 | |
4116 inline UChar | |
4117 UnicodeString::doCharAt(int32_t offset) const | |
4118 { | |
4119 if((uint32_t)offset < (uint32_t)length()) { | |
4120 return getArrayStart()[offset]; | |
4121 } else { | |
4122 return kInvalidUChar; | |
4123 } | |
4124 } | |
4125 | |
4126 inline UChar | |
4127 UnicodeString::charAt(int32_t offset) const | |
4128 { return doCharAt(offset); } | |
4129 | |
4130 inline UChar | |
4131 UnicodeString::operator[] (int32_t offset) const | |
4132 { return doCharAt(offset); } | |
4133 | |
4134 inline UChar32 | |
4135 UnicodeString::char32At(int32_t offset) const | |
4136 { | |
4137 int32_t len = length(); | |
4138 if((uint32_t)offset < (uint32_t)len) { | |
4139 const UChar *array = getArrayStart(); | |
4140 UChar32 c; | |
4141 U16_GET(array, 0, offset, len, c); | |
4142 return c; | |
4143 } else { | |
4144 return kInvalidUChar; | |
4145 } | |
4146 } | |
4147 | |
4148 inline int32_t | |
4149 UnicodeString::getChar32Start(int32_t offset) const { | |
4150 if((uint32_t)offset < (uint32_t)length()) { | |
4151 const UChar *array = getArrayStart(); | |
4152 U16_SET_CP_START(array, 0, offset); | |
4153 return offset; | |
4154 } else { | |
4155 return 0; | |
4156 } | |
4157 } | |
4158 | |
4159 inline int32_t | |
4160 UnicodeString::getChar32Limit(int32_t offset) const { | |
4161 int32_t len = length(); | |
4162 if((uint32_t)offset < (uint32_t)len) { | |
4163 const UChar *array = getArrayStart(); | |
4164 U16_SET_CP_LIMIT(array, 0, offset, len); | |
4165 return offset; | |
4166 } else { | |
4167 return len; | |
4168 } | |
4169 } | |
4170 | |
4171 inline UBool | |
4172 UnicodeString::isEmpty() const { | |
4173 return fShortLength == 0; | |
4174 } | |
4175 | |
4176 //======================================== | |
4177 // Write implementation methods | |
4178 //======================================== | |
4179 inline void | |
4180 UnicodeString::setLength(int32_t len) { | |
4181 if(len <= 127) { | |
4182 fShortLength = (int8_t)len; | |
4183 } else { | |
4184 fShortLength = (int8_t)-1; | |
4185 fUnion.fFields.fLength = len; | |
4186 } | |
4187 } | |
4188 | |
4189 inline void | |
4190 UnicodeString::setToEmpty() { | |
4191 fShortLength = 0; | |
4192 fFlags = kShortString; | |
4193 } | |
4194 | |
4195 inline void | |
4196 UnicodeString::setArray(UChar *array, int32_t len, int32_t capacity) { | |
4197 setLength(len); | |
4198 fUnion.fFields.fArray = array; | |
4199 fUnion.fFields.fCapacity = capacity; | |
4200 } | |
4201 | |
4202 inline const UChar * | |
4203 UnicodeString::getTerminatedBuffer() { | |
4204 if(!isWritable()) { | |
4205 return 0; | |
4206 } else { | |
4207 UChar *array = getArrayStart(); | |
4208 int32_t len = length(); | |
4209 if(len < getCapacity() && ((fFlags&kRefCounted) == 0 || refCount() == 1)) { | |
4210 /* | |
4211 * kRefCounted: Do not write the NUL if the buffer is shared. | |
4212 * That is mostly safe, except when the length of one copy was modified | |
4213 * without copy-on-write, e.g., via truncate(newLength) or remove(void). | |
4214 * Then the NUL would be written into the middle of another copy's string. | |
4215 */ | |
4216 if(!(fFlags&kBufferIsReadonly)) { | |
4217 /* | |
4218 * We must not write to a readonly buffer, but it is known to be | |
4219 * NUL-terminated if len<capacity. | |
4220 * A shared, allocated buffer (refCount()>1) must not have its contents | |
4221 * modified, but the NUL at [len] is beyond the string contents, | |
4222 * and multiple string objects and threads writing the same NUL into the | |
4223 * same location is harmless. | |
4224 * In all other cases, the buffer is fully writable and it is anyway saf
e | |
4225 * to write the NUL. | |
4226 * | |
4227 * Note: An earlier version of this code tested whether there is a NUL | |
4228 * at [len] already, but, while safe, it generated lots of warnings from | |
4229 * tools like valgrind and Purify. | |
4230 */ | |
4231 array[len] = 0; | |
4232 } | |
4233 return array; | |
4234 } else if(cloneArrayIfNeeded(len+1)) { | |
4235 array = getArrayStart(); | |
4236 array[len] = 0; | |
4237 return array; | |
4238 } else { | |
4239 return 0; | |
4240 } | |
4241 } | |
4242 } | |
4243 | |
4244 inline UnicodeString& | |
4245 UnicodeString::operator= (UChar ch) | |
4246 { return doReplace(0, length(), &ch, 0, 1); } | |
4247 | |
4248 inline UnicodeString& | |
4249 UnicodeString::operator= (UChar32 ch) | |
4250 { return replace(0, length(), ch); } | |
4251 | |
4252 inline UnicodeString& | |
4253 UnicodeString::setTo(const UnicodeString& srcText, | |
4254 int32_t srcStart, | |
4255 int32_t srcLength) | |
4256 { | |
4257 unBogus(); | |
4258 return doReplace(0, length(), srcText, srcStart, srcLength); | |
4259 } | |
4260 | |
4261 inline UnicodeString& | |
4262 UnicodeString::setTo(const UnicodeString& srcText, | |
4263 int32_t srcStart) | |
4264 { | |
4265 unBogus(); | |
4266 srcText.pinIndex(srcStart); | |
4267 return doReplace(0, length(), srcText, srcStart, srcText.length() - srcStart); | |
4268 } | |
4269 | |
4270 inline UnicodeString& | |
4271 UnicodeString::setTo(const UnicodeString& srcText) | |
4272 { | |
4273 unBogus(); | |
4274 return doReplace(0, length(), srcText, 0, srcText.length()); | |
4275 } | |
4276 | |
4277 inline UnicodeString& | |
4278 UnicodeString::setTo(const UChar *srcChars, | |
4279 int32_t srcLength) | |
4280 { | |
4281 unBogus(); | |
4282 return doReplace(0, length(), srcChars, 0, srcLength); | |
4283 } | |
4284 | |
4285 inline UnicodeString& | |
4286 UnicodeString::setTo(UChar srcChar) | |
4287 { | |
4288 unBogus(); | |
4289 return doReplace(0, length(), &srcChar, 0, 1); | |
4290 } | |
4291 | |
4292 inline UnicodeString& | |
4293 UnicodeString::setTo(UChar32 srcChar) | |
4294 { | |
4295 unBogus(); | |
4296 return replace(0, length(), srcChar); | |
4297 } | |
4298 | |
4299 inline UnicodeString& | |
4300 UnicodeString::append(const UnicodeString& srcText, | |
4301 int32_t srcStart, | |
4302 int32_t srcLength) | |
4303 { return doReplace(length(), 0, srcText, srcStart, srcLength); } | |
4304 | |
4305 inline UnicodeString& | |
4306 UnicodeString::append(const UnicodeString& srcText) | |
4307 { return doReplace(length(), 0, srcText, 0, srcText.length()); } | |
4308 | |
4309 inline UnicodeString& | |
4310 UnicodeString::append(const UChar *srcChars, | |
4311 int32_t srcStart, | |
4312 int32_t srcLength) | |
4313 { return doReplace(length(), 0, srcChars, srcStart, srcLength); } | |
4314 | |
4315 inline UnicodeString& | |
4316 UnicodeString::append(const UChar *srcChars, | |
4317 int32_t srcLength) | |
4318 { return doReplace(length(), 0, srcChars, 0, srcLength); } | |
4319 | |
4320 inline UnicodeString& | |
4321 UnicodeString::append(UChar srcChar) | |
4322 { return doReplace(length(), 0, &srcChar, 0, 1); } | |
4323 | |
4324 inline UnicodeString& | |
4325 UnicodeString::append(UChar32 srcChar) { | |
4326 UChar buffer[U16_MAX_LENGTH]; | |
4327 int32_t _length = 0; | |
4328 UBool isError = FALSE; | |
4329 U16_APPEND(buffer, _length, U16_MAX_LENGTH, srcChar, isError); | |
4330 (void) isError; | |
4331 return doReplace(length(), 0, buffer, 0, _length); | |
4332 } | |
4333 | |
4334 inline UnicodeString& | |
4335 UnicodeString::operator+= (UChar ch) | |
4336 { return doReplace(length(), 0, &ch, 0, 1); } | |
4337 | |
4338 inline UnicodeString& | |
4339 UnicodeString::operator+= (UChar32 ch) { | |
4340 return append(ch); | |
4341 } | |
4342 | |
4343 inline UnicodeString& | |
4344 UnicodeString::operator+= (const UnicodeString& srcText) | |
4345 { return doReplace(length(), 0, srcText, 0, srcText.length()); } | |
4346 | |
4347 inline UnicodeString& | |
4348 UnicodeString::insert(int32_t start, | |
4349 const UnicodeString& srcText, | |
4350 int32_t srcStart, | |
4351 int32_t srcLength) | |
4352 { return doReplace(start, 0, srcText, srcStart, srcLength); } | |
4353 | |
4354 inline UnicodeString& | |
4355 UnicodeString::insert(int32_t start, | |
4356 const UnicodeString& srcText) | |
4357 { return doReplace(start, 0, srcText, 0, srcText.length()); } | |
4358 | |
4359 inline UnicodeString& | |
4360 UnicodeString::insert(int32_t start, | |
4361 const UChar *srcChars, | |
4362 int32_t srcStart, | |
4363 int32_t srcLength) | |
4364 { return doReplace(start, 0, srcChars, srcStart, srcLength); } | |
4365 | |
4366 inline UnicodeString& | |
4367 UnicodeString::insert(int32_t start, | |
4368 const UChar *srcChars, | |
4369 int32_t srcLength) | |
4370 { return doReplace(start, 0, srcChars, 0, srcLength); } | |
4371 | |
4372 inline UnicodeString& | |
4373 UnicodeString::insert(int32_t start, | |
4374 UChar srcChar) | |
4375 { return doReplace(start, 0, &srcChar, 0, 1); } | |
4376 | |
4377 inline UnicodeString& | |
4378 UnicodeString::insert(int32_t start, | |
4379 UChar32 srcChar) | |
4380 { return replace(start, 0, srcChar); } | |
4381 | |
4382 | |
4383 inline UnicodeString& | |
4384 UnicodeString::remove() | |
4385 { | |
4386 // remove() of a bogus string makes the string empty and non-bogus | |
4387 // we also un-alias a read-only alias to deal with NUL-termination | |
4388 // issues with getTerminatedBuffer() | |
4389 if(fFlags & (kIsBogus|kBufferIsReadonly)) { | |
4390 setToEmpty(); | |
4391 } else { | |
4392 fShortLength = 0; | |
4393 } | |
4394 return *this; | |
4395 } | |
4396 | |
4397 inline UnicodeString& | |
4398 UnicodeString::remove(int32_t start, | |
4399 int32_t _length) | |
4400 { | |
4401 if(start <= 0 && _length == INT32_MAX) { | |
4402 // remove(guaranteed everything) of a bogus string makes the string empt
y and non-bogus | |
4403 return remove(); | |
4404 } | |
4405 return doReplace(start, _length, NULL, 0, 0); | |
4406 } | |
4407 | |
4408 inline UnicodeString& | |
4409 UnicodeString::removeBetween(int32_t start, | |
4410 int32_t limit) | |
4411 { return doReplace(start, limit - start, NULL, 0, 0); } | |
4412 | |
4413 inline UnicodeString & | |
4414 UnicodeString::retainBetween(int32_t start, int32_t limit) { | |
4415 truncate(limit); | |
4416 return doReplace(0, start, NULL, 0, 0); | |
4417 } | |
4418 | |
4419 inline UBool | |
4420 UnicodeString::truncate(int32_t targetLength) | |
4421 { | |
4422 if(isBogus() && targetLength == 0) { | |
4423 // truncate(0) of a bogus string makes the string empty and non-bogus | |
4424 unBogus(); | |
4425 return FALSE; | |
4426 } else if((uint32_t)targetLength < (uint32_t)length()) { | |
4427 setLength(targetLength); | |
4428 if(fFlags&kBufferIsReadonly) { | |
4429 fUnion.fFields.fCapacity = targetLength; // not NUL-terminated any more | |
4430 } | |
4431 return TRUE; | |
4432 } else { | |
4433 return FALSE; | |
4434 } | |
4435 } | |
4436 | |
4437 inline UnicodeString& | |
4438 UnicodeString::reverse() | |
4439 { return doReverse(0, length()); } | |
4440 | |
4441 inline UnicodeString& | |
4442 UnicodeString::reverse(int32_t start, | |
4443 int32_t _length) | |
4444 { return doReverse(start, _length); } | |
4445 | |
4446 U_NAMESPACE_END | |
4447 | |
4448 #endif | |
OLD | NEW |