1 /*
2 **********************************************************************
3 * Copyright (C) 1998-2012, 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 #include "unicode/ucasemap.h"
35
36 struct UConverter; // unicode/ucnv.h
37 class StringThreadTest;
38
39 #ifndef U_COMPARE_CODE_POINT_ORDER
40 /* see also ustring.h and unorm.h */
41 /**
42 * Option bit for u_strCaseCompare, u_strcasecmp, unorm_compare, etc:
43 * Compare strings in code point order instead of code unit order.
44 * @stable ICU 2.2
45 */
46 #define U_COMPARE_CODE_POINT_ORDER 0x8000
47 #endif
48
49 #ifndef USTRING_H
50 /**
51 * \ingroup ustring_ustrlen
52 */
53 U_STABLE int32_t U_EXPORT2
54 u_strlen(const UChar *s);
55 #endif
56
57 /**
58 * \def U_STRING_CASE_MAPPER_DEFINED
59 * @internal
60 */
61
62 #ifndef U_STRING_CASE_MAPPER_DEFINED
63 #define U_STRING_CASE_MAPPER_DEFINED
64
65 /**
66 * Internal string case mapping function type.
67 * @internal
68 */
69 typedef int32_t U_CALLCONV
70 UStringCaseMapper(const UCaseMap *csm,
71 UChar *dest, int32_t destCapacity,
72 const UChar *src, int32_t srcLength,
73 UErrorCode *pErrorCode);
74
75 #endif
76
77 U_NAMESPACE_BEGIN
78
79 class BreakIterator; // unicode/brkiter.h
80 class Locale; // unicode/locid.h
81 class StringCharacterIterator;
82 class UnicodeStringAppendable; // unicode/appendable.h
83
84 /* The <iostream> include has been moved to unicode/ustream.h */
85
86 /**
87 * Constant to be used in the UnicodeString(char *, int32_t, EInvariant) constructor
88 * which constructs a Unicode string from an invariant-character char * string.
89 * About invariant characters see utypes.h.
90 * This constructor has no runtime dependency on conversion code and is
91 * therefore recommended over ones taking a charset name string
92 * (where the empty string "" indicates invariant-character conversion).
93 *
94 * @stable ICU 3.2
95 */
96 #define US_INV icu::UnicodeString::kInvariant
97
98 /**
99 * Unicode String literals in C++.
100 * Dependent on the platform properties, different UnicodeString
101 * constructors should be used to create a UnicodeString object from
102 * a string literal.
103 * The macros are defined for maximum performance.
104 * They work only for strings that contain "invariant characters", i.e.,
105 * only latin letters, digits, and some punctuation.
106 * See utypes.h for details.
107 *
108 * The string parameter must be a C string literal.
109 * The length of the string, not including the terminating
110 * <code>NUL</code>, must be specified as a constant.
111 * The U_STRING_DECL macro should be invoked exactly once for one
112 * such string variable before it is used.
113 * @stable ICU 2.0
114 */
115 #if defined(U_DECLARE_UTF16)
116 # define UNICODE_STRING(cs, _length) icu::UnicodeString(TRUE, (const UChar *)U_DECLARE_UTF16(cs), _length)
117 #elif U_SIZEOF_WCHAR_T==U_SIZEOF_UCHAR && (U_CHARSET_FAMILY==U_ASCII_FAMILY || (U_SIZEOF_UCHAR == 2 && defined(U_WCHAR_IS_UTF16)))
118 # define UNICODE_STRING(cs, _length) icu::UnicodeString(TRUE, (const UChar *)L ## cs, _length)
119 #elif U_SIZEOF_UCHAR==1 && U_CHARSET_FAMILY==U_ASCII_FAMILY
120 # define UNICODE_STRING(cs, _length) icu::UnicodeString(TRUE, (const UChar *)cs, _length)
121 #else
122 # define UNICODE_STRING(cs, _length) icu::UnicodeString(cs, _length, US_INV)
123 #endif
124
125 /**
126 * Unicode String literals in C++.
127 * Dependent on the platform properties, different UnicodeString
128 * constructors should be used to create a UnicodeString object from
129 * a string literal.
130 * The macros are defined for improved performance.
131 * They work only for strings that contain "invariant characters", i.e.,
132 * only latin letters, digits, and some punctuation.
133 * See utypes.h for details.
134 *
135 * The string parameter must be a C string literal.
136 * @stable ICU 2.0
137 */
138 #define UNICODE_STRING_SIMPLE(cs) UNICODE_STRING(cs, -1)
139
140 /**
141 * \def UNISTR_FROM_CHAR_EXPLICIT
142 * This can be defined to be empty or "explicit".
143 * If explicit, then the UnicodeString(UChar) and UnicodeString(UChar32)
144 * constructors are marked as explicit, preventing their inadvertent use.
145 * @draft ICU 49
146 */
147 #ifndef UNISTR_FROM_CHAR_EXPLICIT
148 # if defined(U_COMBINED_IMPLEMENTATION) || defined(U_COMMON_IMPLEMENTATION) || defined(U_I18N_IMPLEMENTATION) || defined(U_IO_IMPLEMENTATION)
149 // Auto-"explicit" in ICU library code.
150 # define UNISTR_FROM_CHAR_EXPLICIT explicit
151 # else
152 // Empty by default for source code compatibility.
153 # define UNISTR_FROM_CHAR_EXPLICIT
154 # endif
155 #endif
156
157 /**
158 * \def UNISTR_FROM_STRING_EXPLICIT
159 * This can be defined to be empty or "explicit".
160 * If explicit, then the UnicodeString(const char *) and UnicodeString(const UChar *)
161 * constructors are marked as explicit, preventing their inadvertent use.
162 *
163 * In particular, this helps prevent accidentally depending on ICU conversion code
164 * by passing a string literal into an API with a const UnicodeString & parameter.
165 * @draft ICU 49
166 */
167 #ifndef UNISTR_FROM_STRING_EXPLICIT
168 # if defined(U_COMBINED_IMPLEMENTATION) || defined(U_COMMON_IMPLEMENTATION) || defined(U_I18N_IMPLEMENTATION) || defined(U_IO_IMPLEMENTATION)
169 // Auto-"explicit" in ICU library code.
170 # define UNISTR_FROM_STRING_EXPLICIT explicit
171 # else
172 // Empty by default for source code compatibility.
173 # define UNISTR_FROM_STRING_EXPLICIT
174 # endif
175 #endif
176
177 /**
178 * UnicodeString is a string class that stores Unicode characters directly and provides
179 * similar functionality as the Java String and StringBuffer classes.
180 * It is a concrete implementation of the abstract class Replaceable (for transliteration).
181 *
182 * The UnicodeString class is not suitable for subclassing.
183 *
184 * <p>For an overview of Unicode strings in C and C++ see the
185 * <a href="http://icu-project.org/userguide/strings.html">User Guide Strings chapter</a>.</p>
186 *
187 * <p>In ICU, a Unicode string consists of 16-bit Unicode <em>code units</em>.
188 * A Unicode character may be stored with either one code unit
189 * (the most common case) or with a matched pair of special code units
190 * ("surrogates"). The data type for code units is UChar.
191 * For single-character handling, a Unicode character code <em>point</em> is a value
192 * in the range 0..0x10ffff. ICU uses the UChar32 type for code points.</p>
193 *
194 * <p>Indexes and offsets into and lengths of strings always count code units, not code points.
195 * This is the same as with multi-byte char* strings in traditional string handling.
196 * Operations on partial strings typically do not test for code point boundaries.
197 * If necessary, the user needs to take care of such boundaries by testing for the code unit
198 * values or by using functions like
199 * UnicodeString::getChar32Start() and UnicodeString::getChar32Limit()
200 * (or, in C, the equivalent macros U16_SET_CP_START() and U16_SET_CP_LIMIT(), see utf.h).</p>
201 *
202 * UnicodeString methods are more lenient with regard to input parameter values
203 * than other ICU APIs. In particular:
204 * - If indexes are out of bounds for a UnicodeString object
205 * (<0 or >length()) then they are "pinned" to the nearest boundary.
206 * - If primitive string pointer values (e.g., const UChar * or char *)
207 * for input strings are NULL, then those input string parameters are treated
208 * as if they pointed to an empty string.
209 * However, this is <em>not</em> the case for char * parameters for charset names
210 * or other IDs.
211 * - Most UnicodeString methods do not take a UErrorCode parameter because
212 * there are usually very few opportunities for failure other than a shortage
213 * of memory, error codes in low-level C++ string methods would be inconvenient,
214 * and the error code as the last parameter (ICU convention) would prevent
215 * the use of default parameter values.
216 * Instead, such methods set the UnicodeString into a "bogus" state
217 * (see isBogus()) if an error occurs.
218 *
219 * In string comparisons, two UnicodeString objects that are both "bogus"
220 * compare equal (to be transitive and prevent endless loops in sorting),
221 * and a "bogus" string compares less than any non-"bogus" one.
222 *
223 * Const UnicodeString methods are thread-safe. Multiple threads can use
224 * const methods on the same UnicodeString object simultaneously,
225 * but non-const methods must not be called concurrently (in multiple threads)
226 * with any other (const or non-const) methods.
227 *
228 * Similarly, const UnicodeString & parameters are thread-safe.
229 * One object may be passed in as such a parameter concurrently in multiple threads.
230 * This includes the const UnicodeString & parameters for
231 * copy construction, assignment, and cloning.
232 *
233 * <p>UnicodeString uses several storage methods.
234 * String contents can be stored inside the UnicodeString object itself,
235 * in an allocated and shared buffer, or in an outside buffer that is "aliased".
236 * Most of this is done transparently, but careful aliasing in particular provides
237 * significant performance improvements.
238 * Also, the internal buffer is accessible via special functions.
239 * For details see the
240 * <a href="http://icu-project.org/userguide/strings.html">User Guide Strings chapter</a>.</p>
241 *
242 * @see utf.h
243 * @see CharacterIterator
244 * @stable ICU 2.0
245 */
246 class U_COMMON_API UnicodeString : public Replaceable
247 {
248 public:
249
250 /**
251 * Constant to be used in the UnicodeString(char *, int32_t, EInvariant) constructor
252 * which constructs a Unicode string from an invariant-character char * string.
253 * Use the macro US_INV instead of the full qualification for this value.
254 *
255 * @see US_INV
256 * @stable ICU 3.2
257 */
258 enum EInvariant {
259 /**
260 * @see EInvariant
261 * @stable ICU 3.2
262 */
263 kInvariant
264 };
265
266 //========================================
267 // Read-only operations
268 //========================================
269
270 /* Comparison - bitwise only - for international comparison use collation */
271
272 /**
273 * Equality operator. Performs only bitwise comparison.
274 * @param text The UnicodeString to compare to this one.
275 * @return TRUE if <TT>text</TT> contains the same characters as this one,
276 * FALSE otherwise.
277 * @stable ICU 2.0
278 */
279 inline UBool operator== (const UnicodeString& text) const;
280
281 /**
282 * Inequality operator. Performs only bitwise comparison.
283 * @param text The UnicodeString to compare to this one.
284 * @return FALSE if <TT>text</TT> contains the same characters as this one,
285 * TRUE otherwise.
286 * @stable ICU 2.0
287 */
288 inline UBool operator!= (const UnicodeString& text) const;
289
290 /**
291 * Greater than operator. Performs only bitwise comparison.
292 * @param text The UnicodeString to compare to this one.
293 * @return TRUE if the characters in this are bitwise
294 * greater than the characters in <code>text</code>, FALSE otherwise
295 * @stable ICU 2.0
296 */
297 inline UBool operator> (const UnicodeString& text) const;
298
299 /**
300 * Less than operator. Performs only bitwise comparison.
301 * @param text The UnicodeString to compare to this one.
302 * @return TRUE if the characters in this are bitwise
303 * less than the characters in <code>text</code>, FALSE otherwise
304 * @stable ICU 2.0
305 */
306 inline UBool operator< (const UnicodeString& text) const;
307
308 /**
309 * Greater than or equal operator. Performs only bitwise comparison.
310 * @param text The UnicodeString to compare to this one.
311 * @return TRUE if the characters in this are bitwise
312 * greater than or equal to the characters in <code>text</code>, FALSE otherwise
313 * @stable ICU 2.0
314 */
315 inline UBool operator>= (const UnicodeString& text) const;
316
317 /**
318 * Less than or equal operator. Performs only bitwise comparison.
319 * @param text The UnicodeString to compare to this one.
320 * @return TRUE if the characters in this are bitwise
321 * less than or equal to the characters in <code>text</code>, FALSE otherwise
322 * @stable ICU 2.0
323 */
324 inline UBool operator<= (const UnicodeString& text) const;
325
326 /**
327 * Compare the characters bitwise in this UnicodeString to
328 * the characters in <code>text</code>.
329 * @param text The UnicodeString to compare to this one.
330 * @return The result of bitwise character comparison: 0 if this
331 * contains the same characters as <code>text</code>, -1 if the characters in
332 * this are bitwise less than the characters in <code>text</code>, +1 if the
333 * characters in this are bitwise greater than the characters
334 * in <code>text</code>.
335 * @stable ICU 2.0
336 */
337 inline int8_t compare(const UnicodeString& text) const;
338
339 /**
340 * Compare the characters bitwise in the range
341 * [<TT>start</TT>, <TT>start + length</TT>) with the characters
342 * in <TT>text</TT>
343 * @param start the offset at which the compare operation begins
344 * @param length the number of characters of text to compare.
345 * @param text the other text to be compared against this string.
346 * @return The result of bitwise character comparison: 0 if this
347 * contains the same characters as <code>text</code>, -1 if the characters in
348 * this are bitwise less than the characters in <code>text</code>, +1 if the
349 * characters in this are bitwise greater than the characters
350 * in <code>text</code>.
351 * @stable ICU 2.0
352 */
353 inline int8_t compare(int32_t start,
354 int32_t length,
355 const UnicodeString& text) const;
356
357 /**
358 * Compare the characters bitwise in the range
359 * [<TT>start</TT>, <TT>start + length</TT>) with the characters
360 * in <TT>srcText</TT> in the range
361 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
362 * @param start the offset at which the compare operation begins
363 * @param length the number of characters in this to compare.
364 * @param srcText the text to be compared
365 * @param srcStart the offset into <TT>srcText</TT> to start comparison
366 * @param srcLength the number of characters in <TT>src</TT> to compare
367 * @return The result of bitwise character comparison: 0 if this
368 * contains the same characters as <code>srcText</code>, -1 if the characters in
369 * this are bitwise less than the characters in <code>srcText</code>, +1 if the
370 * characters in this are bitwise greater than the characters
371 * in <code>srcText</code>.
372 * @stable ICU 2.0
373 */
374 inline int8_t compare(int32_t start,
375 int32_t length,
376 const UnicodeString& srcText,
377 int32_t srcStart,
378 int32_t srcLength) const;
379
380 /**
381 * Compare the characters bitwise in this UnicodeString with the first
382 * <TT>srcLength</TT> characters in <TT>srcChars</TT>.
383 * @param srcChars The characters to compare to this UnicodeString.
384 * @param srcLength the number of characters in <TT>srcChars</TT> to compare
385 * @return The result of bitwise character comparison: 0 if this
386 * contains the same characters as <code>srcChars</code>, -1 if the characters in
387 * this are bitwise less than the characters in <code>srcChars</code>, +1 if the
388 * characters in this are bitwise greater than the characters
389 * in <code>srcChars</code>.
390 * @stable ICU 2.0
391 */
392 inline int8_t compare(const UChar *srcChars,
393 int32_t srcLength) const;
394
395 /**
396 * Compare the characters bitwise in the range
397 * [<TT>start</TT>, <TT>start + length</TT>) with the first
398 * <TT>length</TT> characters in <TT>srcChars</TT>
399 * @param start the offset at which the compare operation begins
400 * @param length the number of characters to compare.
401 * @param srcChars the characters to be compared
402 * @return The result of bitwise character comparison: 0 if this
403 * contains the same characters as <code>srcChars</code>, -1 if the characters in
404 * this are bitwise less than the characters in <code>srcChars</code>, +1 if the
405 * characters in this are bitwise greater than the characters
406 * in <code>srcChars</code>.
407 * @stable ICU 2.0
408 */
409 inline int8_t compare(int32_t start,
410 int32_t length,
411 const UChar *srcChars) const;
412
413 /**
414 * Compare the characters bitwise in the range
415 * [<TT>start</TT>, <TT>start + length</TT>) with the characters
416 * in <TT>srcChars</TT> in the range
417 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
418 * @param start the offset at which the compare operation begins
419 * @param length the number of characters in this to compare
420 * @param srcChars the characters to be compared
421 * @param srcStart the offset into <TT>srcChars</TT> to start comparison
422 * @param srcLength the number of characters in <TT>srcChars</TT> to compare
423 * @return The result of bitwise character comparison: 0 if this
424 * contains the same characters as <code>srcChars</code>, -1 if the characters in
425 * this are bitwise less than the characters in <code>srcChars</code>, +1 if the
426 * characters in this are bitwise greater than the characters
427 * in <code>srcChars</code>.
428 * @stable ICU 2.0
429 */
430 inline int8_t compare(int32_t start,
431 int32_t length,
432 const UChar *srcChars,
433 int32_t srcStart,
434 int32_t srcLength) const;
435
436 /**
437 * Compare the characters bitwise in the range
438 * [<TT>start</TT>, <TT>limit</TT>) with the characters
439 * in <TT>srcText</TT> in the range
440 * [<TT>srcStart</TT>, <TT>srcLimit</TT>).
441 * @param start the offset at which the compare operation begins
442 * @param limit the offset immediately following the compare operation
443 * @param srcText the text to be compared
444 * @param srcStart the offset into <TT>srcText</TT> to start comparison
445 * @param srcLimit the offset into <TT>srcText</TT> to limit comparison
446 * @return The result of bitwise character comparison: 0 if this
447 * contains the same characters as <code>srcText</code>, -1 if the characters in
448 * this are bitwise less than the characters in <code>srcText</code>, +1 if the
449 * characters in this are bitwise greater than the characters
450 * in <code>srcText</code>.
451 * @stable ICU 2.0
452 */
453 inline int8_t compareBetween(int32_t start,
454 int32_t limit,
455 const UnicodeString& srcText,
456 int32_t srcStart,
457 int32_t srcLimit) const;
458
459 /**
460 * Compare two Unicode strings in code point order.
461 * The result may be different from the results of compare(), operator<, etc.
462 * if supplementary characters are present:
463 *
464 * In UTF-16, supplementary characters (with code points U+10000 and above) are
465 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
466 * which means that they compare as less than some other BMP characters like U+feff.
467 * This function compares Unicode strings in code point order.
468 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
469 *
470 * @param text Another string to compare this one to.
471 * @return a negative/zero/positive integer corresponding to whether
472 * this string is less than/equal to/greater than the second one
473 * in code point order
474 * @stable ICU 2.0
475 */
476 inline int8_t compareCodePointOrder(const UnicodeString& text) const;
477
478 /**
479 * Compare two Unicode strings in code point order.
480 * The result may be different from the results of compare(), operator<, etc.
481 * if supplementary characters are present:
482 *
483 * In UTF-16, supplementary characters (with code points U+10000 and above) are
484 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
485 * which means that they compare as less than some other BMP characters like U+feff.
486 * This function compares Unicode strings in code point order.
487 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
488 *
489 * @param start The start offset in this string at which the compare operation begins.
490 * @param length The number of code units from this string to compare.
491 * @param srcText Another string to compare this one to.
492 * @return a negative/zero/positive integer corresponding to whether
493 * this string is less than/equal to/greater than the second one
494 * in code point order
495 * @stable ICU 2.0
496 */
497 inline int8_t compareCodePointOrder(int32_t start,
498 int32_t length,
499 const UnicodeString& srcText) const;
500
501 /**
502 * Compare two Unicode strings in code point order.
503 * The result may be different from the results of compare(), operator<, etc.
504 * if supplementary characters are present:
505 *
506 * In UTF-16, supplementary characters (with code points U+10000 and above) are
507 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
508 * which means that they compare as less than some other BMP characters like U+feff.
509 * This function compares Unicode strings in code point order.
510 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
511 *
512 * @param start The start offset in this string at which the compare operation begins.
513 * @param length The number of code units from this string to compare.
514 * @param srcText Another string to compare this one to.
515 * @param srcStart The start offset in that string at which the compare operation begins.
516 * @param srcLength The number of code units from that string to compare.
517 * @return a negative/zero/positive integer corresponding to whether
518 * this string is less than/equal to/greater than the second one
519 * in code point order
520 * @stable ICU 2.0
521 */
522 inline int8_t compareCodePointOrder(int32_t start,
523 int32_t length,
524 const UnicodeString& srcText,
525 int32_t srcStart,
526 int32_t srcLength) const;
527
528 /**
529 * Compare two Unicode strings in code point order.
530 * The result may be different from the results of compare(), operator<, etc.
531 * if supplementary characters are present:
532 *
533 * In UTF-16, supplementary characters (with code points U+10000 and above) are
534 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
535 * which means that they compare as less than some other BMP characters like U+feff.
536 * This function compares Unicode strings in code point order.
537 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
538 *
539 * @param srcChars A pointer to another string to compare this one to.
540 * @param srcLength The number of code units from that string to compare.
541 * @return a negative/zero/positive integer corresponding to whether
542 * this string is less than/equal to/greater than the second one
543 * in code point order
544 * @stable ICU 2.0
545 */
546 inline int8_t compareCodePointOrder(const UChar *srcChars,
547 int32_t srcLength) const;
548
549 /**
550 * Compare two Unicode strings in code point order.
551 * The result may be different from the results of compare(), operator<, etc.
552 * if supplementary characters are present:
553 *
554 * In UTF-16, supplementary characters (with code points U+10000 and above) are
555 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
556 * which means that they compare as less than some other BMP characters like U+feff.
557 * This function compares Unicode strings in code point order.
558 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
559 *
560 * @param start The start offset in this string at which the compare operation begins.
561 * @param length The number of code units from this string to compare.
562 * @param srcChars A pointer to another string to compare this one to.
563 * @return a negative/zero/positive integer corresponding to whether
564 * this string is less than/equal to/greater than the second one
565 * in code point order
566 * @stable ICU 2.0
567 */
568 inline int8_t compareCodePointOrder(int32_t start,
569 int32_t length,
570 const UChar *srcChars) const;
571
572 /**
573 * Compare two Unicode strings in code point order.
574 * The result may be different from the results of compare(), operator<, etc.
575 * if supplementary characters are present:
576 *
577 * In UTF-16, supplementary characters (with code points U+10000 and above) are
578 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
579 * which means that they compare as less than some other BMP characters like U+feff.
580 * This function compares Unicode strings in code point order.
581 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
582 *
583 * @param start The start offset in this string at which the compare operation begins.
584 * @param length The number of code units from this string to compare.
585 * @param srcChars A pointer to another string to compare this one to.
586 * @param srcStart The start offset in that string at which the compare operation begins.
587 * @param srcLength The number of code units from that string to compare.
588 * @return a negative/zero/positive integer corresponding to whether
589 * this string is less than/equal to/greater than the second one
590 * in code point order
591 * @stable ICU 2.0
592 */
593 inline int8_t compareCodePointOrder(int32_t start,
594 int32_t length,
595 const UChar *srcChars,
596 int32_t srcStart,
597 int32_t srcLength) const;
598
599 /**
600 * Compare two Unicode strings in code point order.
601 * The result may be different from the results of compare(), operator<, etc.
602 * if supplementary characters are present:
603 *
604 * In UTF-16, supplementary characters (with code points U+10000 and above) are
605 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
606 * which means that they compare as less than some other BMP characters like U+feff.
607 * This function compares Unicode strings in code point order.
608 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
609 *
610 * @param start The start offset in this string at which the compare operation begins.
611 * @param limit The offset after the last code unit from this string to compare.
612 * @param srcText Another string to compare this one to.
613 * @param srcStart The start offset in that string at which the compare operation begins.
614 * @param srcLimit The offset after the last code unit from that string to compare.
615 * @return a negative/zero/positive integer corresponding to whether
616 * this string is less than/equal to/greater than the second one
617 * in code point order
618 * @stable ICU 2.0
619 */
620 inline int8_t compareCodePointOrderBetween(int32_t start,
621 int32_t limit,
622 const UnicodeString& srcText,
623 int32_t srcStart,
624 int32_t srcLimit) const;
625
626 /**
627 * Compare two strings case-insensitively using full case folding.
628 * This is equivalent to this->foldCase(options).compare(text.foldCase(options)).
629 *
630 * @param text Another string to compare this one to.
631 * @param options A bit set of options:
632 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
633 * Comparison in code unit order with default case folding.
634 *
635 * - U_COMPARE_CODE_POINT_ORDER
636 * Set to choose code point order instead of code unit order
637 * (see u_strCompare for details).
638 *
639 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
640 *
641 * @return A negative, zero, or positive integer indicating the comparison result.
642 * @stable ICU 2.0
643 */
644 inline int8_t caseCompare(const UnicodeString& text, uint32_t options) const;
645
646 /**
647 * Compare two strings case-insensitively using full case folding.
648 * This is equivalent to this->foldCase(options).compare(srcText.foldCase(options)).
649 *
650 * @param start The start offset in this string at which the compare operation begins.
651 * @param length The number of code units from this string to compare.
652 * @param srcText Another string to compare this one to.
653 * @param options A bit set of options:
654 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
655 * Comparison in code unit order with default case folding.
656 *
657 * - U_COMPARE_CODE_POINT_ORDER
658 * Set to choose code point order instead of code unit order
659 * (see u_strCompare for details).
660 *
661 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
662 *
663 * @return A negative, zero, or positive integer indicating the comparison result.
664 * @stable ICU 2.0
665 */
666 inline int8_t caseCompare(int32_t start,
667 int32_t length,
668 const UnicodeString& srcText,
669 uint32_t options) const;
670
671 /**
672 * Compare two strings case-insensitively using full case folding.
673 * This is equivalent to this->foldCase(options).compare(srcText.foldCase(options)).
674 *
675 * @param start The start offset in this string at which the compare operation begins.
676 * @param length The number of code units from this string to compare.
677 * @param srcText Another string to compare this one to.
678 * @param srcStart The start offset in that string at which the compare operation begins.
679 * @param srcLength The number of code units from that string to compare.
680 * @param options A bit set of options:
681 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
682 * Comparison in code unit order with default case folding.
683 *
684 * - U_COMPARE_CODE_POINT_ORDER
685 * Set to choose code point order instead of code unit order
686 * (see u_strCompare for details).
687 *
688 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
689 *
690 * @return A negative, zero, or positive integer indicating the comparison result.
691 * @stable ICU 2.0
692 */
693 inline int8_t caseCompare(int32_t start,
694 int32_t length,
695 const UnicodeString& srcText,
696 int32_t srcStart,
697 int32_t srcLength,
698 uint32_t options) const;
699
700 /**
701 * Compare two strings case-insensitively using full case folding.
702 * This is equivalent to this->foldCase(options).compare(srcChars.foldCase(options)).
703 *
704 * @param srcChars A pointer to another string to compare this one to.
705 * @param srcLength The number of code units from that string to compare.
706 * @param options A bit set of options:
707 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
708 * Comparison in code unit order with default case folding.
709 *
710 * - U_COMPARE_CODE_POINT_ORDER
711 * Set to choose code point order instead of code unit order
712 * (see u_strCompare for details).
713 *
714 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
715 *
716 * @return A negative, zero, or positive integer indicating the comparison result.
717 * @stable ICU 2.0
718 */
719 inline int8_t caseCompare(const UChar *srcChars,
720 int32_t srcLength,
721 uint32_t options) const;
722
723 /**
724 * Compare two strings case-insensitively using full case folding.
725 * This is equivalent to this->foldCase(options).compare(srcChars.foldCase(options)).
726 *
727 * @param start The start offset in this string at which the compare operation begins.
728 * @param length The number of code units from this string to compare.
729 * @param srcChars A pointer to another string to compare this one to.
730 * @param options A bit set of options:
731 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
732 * Comparison in code unit order with default case folding.
733 *
734 * - U_COMPARE_CODE_POINT_ORDER
735 * Set to choose code point order instead of code unit order
736 * (see u_strCompare for details).
737 *
738 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
739 *
740 * @return A negative, zero, or positive integer indicating the comparison result.
741 * @stable ICU 2.0
742 */
743 inline int8_t caseCompare(int32_t start,
744 int32_t length,
745 const UChar *srcChars,
746 uint32_t options) const;
747
748 /**
749 * Compare two strings case-insensitively using full case folding.
750 * This is equivalent to this->foldCase(options).compare(srcChars.foldCase(options)).
751 *
752 * @param start The start offset in this string at which the compare operation begins.
753 * @param length The number of code units from this string to compare.
754 * @param srcChars A pointer to another string to compare this one to.
755 * @param srcStart The start offset in that string at which the compare operation begins.
756 * @param srcLength The number of code units from that string to compare.
757 * @param options A bit set of options:
758 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
759 * Comparison in code unit order with default case folding.
760 *
761 * - U_COMPARE_CODE_POINT_ORDER
762 * Set to choose code point order instead of code unit order
763 * (see u_strCompare for details).
764 *
765 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
766 *
767 * @return A negative, zero, or positive integer indicating the comparison result.
768 * @stable ICU 2.0
769 */
770 inline int8_t caseCompare(int32_t start,
771 int32_t length,
772 const UChar *srcChars,
773 int32_t srcStart,
774 int32_t srcLength,
775 uint32_t options) const;
776
777 /**
778 * Compare two strings case-insensitively using full case folding.
779 * This is equivalent to this->foldCase(options).compareBetween(text.foldCase(options)).
780 *
781 * @param start The start offset in this string at which the compare operation begins.
782 * @param limit The offset after the last code unit from this string to compare.
783 * @param srcText Another string to compare this one to.
784 * @param srcStart The start offset in that string at which the compare operation begins.
785 * @param srcLimit The offset after the last code unit from that string to compare.
786 * @param options A bit set of options:
787 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
788 * Comparison in code unit order with default case folding.
789 *
790 * - U_COMPARE_CODE_POINT_ORDER
791 * Set to choose code point order instead of code unit order
792 * (see u_strCompare for details).
793 *
794 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
795 *
796 * @return A negative, zero, or positive integer indicating the comparison result.
797 * @stable ICU 2.0
798 */
799 inline int8_t caseCompareBetween(int32_t start,
800 int32_t limit,
801 const UnicodeString& srcText,
802 int32_t srcStart,
803 int32_t srcLimit,
804 uint32_t options) const;
805
806 /**
807 * Determine if this starts with the characters in <TT>text</TT>
808 * @param text The text to match.
809 * @return TRUE if this starts with the characters in <TT>text</TT>,
810 * FALSE otherwise
811 * @stable ICU 2.0
812 */
813 inline UBool startsWith(const UnicodeString& text) const;
814
815 /**
816 * Determine if this starts with the characters in <TT>srcText</TT>
817 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
818 * @param srcText The text to match.
819 * @param srcStart the offset into <TT>srcText</TT> to start matching
820 * @param srcLength the number of characters in <TT>srcText</TT> to match
821 * @return TRUE if this starts with the characters in <TT>text</TT>,
822 * FALSE otherwise
823 * @stable ICU 2.0
824 */
825 inline UBool startsWith(const UnicodeString& srcText,
826 int32_t srcStart,
827 int32_t srcLength) const;
828
829 /**
830 * Determine if this starts with the characters in <TT>srcChars</TT>
831 * @param srcChars The characters to match.
832 * @param srcLength the number of characters in <TT>srcChars</TT>
833 * @return TRUE if this starts with the characters in <TT>srcChars</TT>,
834 * FALSE otherwise
835 * @stable ICU 2.0
836 */
837 inline UBool startsWith(const UChar *srcChars,
838 int32_t srcLength) const;
839
840 /**
841 * Determine if this ends with the characters in <TT>srcChars</TT>
842 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
843 * @param srcChars The characters to match.
844 * @param srcStart the offset into <TT>srcText</TT> to start matching
845 * @param srcLength the number of characters in <TT>srcChars</TT> to match
846 * @return TRUE if this ends with the characters in <TT>srcChars</TT>, FALSE otherwise
847 * @stable ICU 2.0
848 */
849 inline UBool startsWith(const UChar *srcChars,
850 int32_t srcStart,
851 int32_t srcLength) const;
852
853 /**
854 * Determine if this ends with the characters in <TT>text</TT>
855 * @param text The text to match.
856 * @return TRUE if this ends with the characters in <TT>text</TT>,
857 * FALSE otherwise
858 * @stable ICU 2.0
859 */
860 inline UBool endsWith(const UnicodeString& text) const;
861
862 /**
863 * Determine if this ends with the characters in <TT>srcText</TT>
864 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
865 * @param srcText The text to match.
866 * @param srcStart the offset into <TT>srcText</TT> to start matching
867 * @param srcLength the number of characters in <TT>srcText</TT> to match
868 * @return TRUE if this ends with the characters in <TT>text</TT>,
869 * FALSE otherwise
870 * @stable ICU 2.0
871 */
872 inline UBool endsWith(const UnicodeString& srcText,
873 int32_t srcStart,
874 int32_t srcLength) const;
875
876 /**
877 * Determine if this ends with the characters in <TT>srcChars</TT>
878 * @param srcChars The characters to match.
879 * @param srcLength the number of characters in <TT>srcChars</TT>
880 * @return TRUE if this ends with the characters in <TT>srcChars</TT>,
881 * FALSE otherwise
882 * @stable ICU 2.0
883 */
884 inline UBool endsWith(const UChar *srcChars,
885 int32_t srcLength) const;
886
887 /**
888 * Determine if this ends with the characters in <TT>srcChars</TT>
889 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
890 * @param srcChars The characters to match.
891 * @param srcStart the offset into <TT>srcText</TT> to start matching
892 * @param srcLength the number of characters in <TT>srcChars</TT> to match
893 * @return TRUE if this ends with the characters in <TT>srcChars</TT>,
894 * FALSE otherwise
895 * @stable ICU 2.0
896 */
897 inline UBool endsWith(const UChar *srcChars,
898 int32_t srcStart,
899 int32_t srcLength) const;
900
901
902 /* Searching - bitwise only */
903
904 /**
905 * Locate in this the first occurrence of the characters in <TT>text</TT>,
906 * using bitwise comparison.
907 * @param text The text to search for.
908 * @return The offset into this of the start of <TT>text</TT>,
909 * or -1 if not found.
910 * @stable ICU 2.0
911 */
912 inline int32_t indexOf(const UnicodeString& text) const;
913
914 /**
915 * Locate in this the first occurrence of the characters in <TT>text</TT>
916 * starting at offset <TT>start</TT>, using bitwise comparison.
917 * @param text The text to search for.
918 * @param start The offset at which searching will start.
919 * @return The offset into this of the start of <TT>text</TT>,
920 * or -1 if not found.
921 * @stable ICU 2.0
922 */
923 inline int32_t indexOf(const UnicodeString& text,
924 int32_t start) const;
925
926 /**
927 * Locate in this the first occurrence in the range
928 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
929 * in <TT>text</TT>, using bitwise comparison.
930 * @param text The text to search for.
931 * @param start The offset at which searching will start.
932 * @param length The number of characters to search
933 * @return The offset into this of the start of <TT>text</TT>,
934 * or -1 if not found.
935 * @stable ICU 2.0
936 */
937 inline int32_t indexOf(const UnicodeString& text,
938 int32_t start,
939 int32_t length) const;
940
941 /**
942 * Locate in this the first occurrence in the range
943 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
944 * in <TT>srcText</TT> in the range
945 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>),
946 * using bitwise comparison.
947 * @param srcText The text to search for.
948 * @param srcStart the offset into <TT>srcText</TT> at which
949 * to start matching
950 * @param srcLength the number of characters in <TT>srcText</TT> to match
951 * @param start the offset into this at which to start matching
952 * @param length the number of characters in this to search
953 * @return The offset into this of the start of <TT>text</TT>,
954 * or -1 if not found.
955 * @stable ICU 2.0
956 */
957 inline int32_t indexOf(const UnicodeString& srcText,
958 int32_t srcStart,
959 int32_t srcLength,
960 int32_t start,
961 int32_t length) const;
962
963 /**
964 * Locate in this the first occurrence of the characters in
965 * <TT>srcChars</TT>
966 * starting at offset <TT>start</TT>, using bitwise comparison.
967 * @param srcChars The text to search for.
968 * @param srcLength the number of characters in <TT>srcChars</TT> to match
969 * @param start the offset into this at which to start matching
970 * @return The offset into this of the start of <TT>text</TT>,
971 * or -1 if not found.
972 * @stable ICU 2.0
973 */
974 inline int32_t indexOf(const UChar *srcChars,
975 int32_t srcLength,
976 int32_t start) const;
977
978 /**
979 * Locate in this the first occurrence in the range
980 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
981 * in <TT>srcChars</TT>, using bitwise comparison.
982 * @param srcChars The text to search for.
983 * @param srcLength the number of characters in <TT>srcChars</TT>
984 * @param start The offset at which searching will start.
985 * @param length The number of characters to search
986 * @return The offset into this of the start of <TT>srcChars</TT>,
987 * or -1 if not found.
988 * @stable ICU 2.0
989 */
990 inline int32_t indexOf(const UChar *srcChars,
991 int32_t srcLength,
992 int32_t start,
993 int32_t length) const;
994
995 /**
996 * Locate in this the first occurrence in the range
997 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
998 * in <TT>srcChars</TT> in the range
999 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>),
1000 * using bitwise comparison.
1001 * @param srcChars The text to search for.
1002 * @param srcStart the offset into <TT>srcChars</TT> at which
1003 * to start matching
1004 * @param srcLength the number of characters in <TT>srcChars</TT> to match
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 the start of <TT>text</TT>,
1008 * or -1 if not found.
1009 * @stable ICU 2.0
1010 */
1011 int32_t indexOf(const UChar *srcChars,
1012 int32_t srcStart,
1013 int32_t srcLength,
1014 int32_t start,
1015 int32_t length) const;
1016
1017 /**
1018 * Locate in this the first occurrence of the BMP code point <code>c</code>,
1019 * using bitwise comparison.
1020 * @param c The code unit to search for.
1021 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1022 * @stable ICU 2.0
1023 */
1024 inline int32_t indexOf(UChar c) const;
1025
1026 /**
1027 * Locate in this the first occurrence of the code point <TT>c</TT>,
1028 * using bitwise comparison.
1029 *
1030 * @param c The code point to search for.
1031 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1032 * @stable ICU 2.0
1033 */
1034 inline int32_t indexOf(UChar32 c) const;
1035
1036 /**
1037 * Locate in this the first occurrence of the BMP code point <code>c</code>,
1038 * starting at offset <TT>start</TT>, using bitwise comparison.
1039 * @param c The code unit to search for.
1040 * @param start The offset at which searching will start.
1041 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1042 * @stable ICU 2.0
1043 */
1044 inline int32_t indexOf(UChar c,
1045 int32_t start) const;
1046
1047 /**
1048 * Locate in this the first occurrence of the code point <TT>c</TT>
1049 * starting at offset <TT>start</TT>, using bitwise comparison.
1050 *
1051 * @param c The code point to search for.
1052 * @param start The offset at which searching will start.
1053 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1054 * @stable ICU 2.0
1055 */
1056 inline int32_t indexOf(UChar32 c,
1057 int32_t start) const;
1058
1059 /**
1060 * Locate in this the first occurrence of the BMP code point <code>c</code>
1061 * in the range [<TT>start</TT>, <TT>start + length</TT>),
1062 * using bitwise comparison.
1063 * @param c The code unit to search for.
1064 * @param start the offset into this at which to start matching
1065 * @param length the number of characters in this to search
1066 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1067 * @stable ICU 2.0
1068 */
1069 inline int32_t indexOf(UChar c,
1070 int32_t start,
1071 int32_t length) const;
1072
1073 /**
1074 * Locate in this the first occurrence of the code point <TT>c</TT>
1075 * in the range [<TT>start</TT>, <TT>start + length</TT>),
1076 * using bitwise comparison.
1077 *
1078 * @param c The code point to search for.
1079 * @param start the offset into this at which to start matching
1080 * @param length the number of characters in this to search
1081 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1082 * @stable ICU 2.0
1083 */
1084 inline int32_t indexOf(UChar32 c,
1085 int32_t start,
1086 int32_t length) const;
1087
1088 /**
1089 * Locate in this the last occurrence of the characters in <TT>text</TT>,
1090 * using bitwise comparison.
1091 * @param text The text to search for.
1092 * @return The offset into this of the start of <TT>text</TT>,
1093 * or -1 if not found.
1094 * @stable ICU 2.0
1095 */
1096 inline int32_t lastIndexOf(const UnicodeString& text) const;
1097
1098 /**
1099 * Locate in this the last occurrence of the characters in <TT>text</TT>
1100 * starting at offset <TT>start</TT>, using bitwise comparison.
1101 * @param text The text to search for.
1102 * @param start The offset at which searching will start.
1103 * @return The offset into this of the start of <TT>text</TT>,
1104 * or -1 if not found.
1105 * @stable ICU 2.0
1106 */
1107 inline int32_t lastIndexOf(const UnicodeString& text,
1108 int32_t start) const;
1109
1110 /**
1111 * Locate in this the last occurrence in the range
1112 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
1113 * in <TT>text</TT>, using bitwise comparison.
1114 * @param text The text to search for.
1115 * @param start The offset at which searching will start.
1116 * @param length The number of characters to search
1117 * @return The offset into this of the start of <TT>text</TT>,
1118 * or -1 if not found.
1119 * @stable ICU 2.0
1120 */
1121 inline int32_t lastIndexOf(const UnicodeString& text,
1122 int32_t start,
1123 int32_t length) const;
1124
1125 /**
1126 * Locate in this the last occurrence in the range
1127 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
1128 * in <TT>srcText</TT> in the range
1129 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>),
1130 * using bitwise comparison.
1131 * @param srcText The text to search for.
1132 * @param srcStart the offset into <TT>srcText</TT> at which
1133 * to start matching
1134 * @param srcLength the number of characters in <TT>srcText</TT> to match
1135 * @param start the offset into this at which to start matching
1136 * @param length the number of characters in this to search
1137 * @return The offset into this of the start of <TT>text</TT>,
1138 * or -1 if not found.
1139 * @stable ICU 2.0
1140 */
1141 inline int32_t lastIndexOf(const UnicodeString& srcText,
1142 int32_t srcStart,
1143 int32_t srcLength,
1144 int32_t start,
1145 int32_t length) const;
1146
1147 /**
1148 * Locate in this the last occurrence of the characters in <TT>srcChars</TT>
1149 * starting at offset <TT>start</TT>, using bitwise comparison.
1150 * @param srcChars The text to search for.
1151 * @param srcLength the number of characters in <TT>srcChars</TT> to match
1152 * @param start the offset into this at which to start matching
1153 * @return The offset into this of the start of <TT>text</TT>,
1154 * or -1 if not found.
1155 * @stable ICU 2.0
1156 */
1157 inline int32_t lastIndexOf(const UChar *srcChars,
1158 int32_t srcLength,
1159 int32_t start) const;
1160
1161 /**
1162 * Locate in this the last occurrence in the range
1163 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
1164 * in <TT>srcChars</TT>, using bitwise comparison.
1165 * @param srcChars The text to search for.
1166 * @param srcLength the number of characters in <TT>srcChars</TT>
1167 * @param start The offset at which searching will start.
1168 * @param length The number of characters to search
1169 * @return The offset into this of the start of <TT>srcChars</TT>,
1170 * or -1 if not found.
1171 * @stable ICU 2.0
1172 */
1173 inline int32_t lastIndexOf(const UChar *srcChars,
1174 int32_t srcLength,
1175 int32_t start,
1176 int32_t length) const;
1177
1178 /**
1179 * Locate in this the last occurrence in the range
1180 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
1181 * in <TT>srcChars</TT> in the range
1182 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>),
1183 * using bitwise comparison.
1184 * @param srcChars The text to search for.
1185 * @param srcStart the offset into <TT>srcChars</TT> at which
1186 * to start matching
1187 * @param srcLength the number of characters in <TT>srcChars</TT> to match
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 the start of <TT>text</TT>,
1191 * or -1 if not found.
1192 * @stable ICU 2.0
1193 */
1194 int32_t lastIndexOf(const UChar *srcChars,
1195 int32_t srcStart,
1196 int32_t srcLength,
1197 int32_t start,
1198 int32_t length) const;
1199
1200 /**
1201 * Locate in this the last occurrence of the BMP code point <code>c</code>,
1202 * using bitwise comparison.
1203 * @param c The code unit to search for.
1204 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1205 * @stable ICU 2.0
1206 */
1207 inline int32_t lastIndexOf(UChar c) const;
1208
1209 /**
1210 * Locate in this the last occurrence of the code point <TT>c</TT>,
1211 * using bitwise comparison.
1212 *
1213 * @param c The code point to search for.
1214 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1215 * @stable ICU 2.0
1216 */
1217 inline int32_t lastIndexOf(UChar32 c) const;
1218
1219 /**
1220 * Locate in this the last occurrence of the BMP code point <code>c</code>
1221 * starting at offset <TT>start</TT>, using bitwise comparison.
1222 * @param c The code unit to search for.
1223 * @param start The offset at which searching will start.
1224 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1225 * @stable ICU 2.0
1226 */
1227 inline int32_t lastIndexOf(UChar c,
1228 int32_t start) const;
1229
1230 /**
1231 * Locate in this the last occurrence of the code point <TT>c</TT>
1232 * starting at offset <TT>start</TT>, using bitwise comparison.
1233 *
1234 * @param c The code point to search for.
1235 * @param start The offset at which searching will start.
1236 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1237 * @stable ICU 2.0
1238 */
1239 inline int32_t lastIndexOf(UChar32 c,
1240 int32_t start) const;
1241
1242 /**
1243 * Locate in this the last occurrence of the BMP code point <code>c</code>
1244 * in the range [<TT>start</TT>, <TT>start + length</TT>),
1245 * using bitwise comparison.
1246 * @param c The code unit to search for.
1247 * @param start the offset into this at which to start matching
1248 * @param length the number of characters in this to search
1249 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1250 * @stable ICU 2.0
1251 */
1252 inline int32_t lastIndexOf(UChar c,
1253 int32_t start,
1254 int32_t length) const;
1255
1256 /**
1257 * Locate in this the last occurrence of the code point <TT>c</TT>
1258 * in the range [<TT>start</TT>, <TT>start + length</TT>),
1259 * using bitwise comparison.
1260 *
1261 * @param c The code point to search for.
1262 * @param start the offset into this at which to start matching
1263 * @param length the number of characters in this to search
1264 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1265 * @stable ICU 2.0
1266 */
1267 inline int32_t lastIndexOf(UChar32 c,
1268 int32_t start,
1269 int32_t length) const;
1270
1271
1272 /* Character access */
1273
1274 /**
1275 * Return the code unit at offset <tt>offset</tt>.
1276 * If the offset is not valid (0..length()-1) then U+ffff is returned.
1277 * @param offset a valid offset into the text
1278 * @return the code unit at offset <tt>offset</tt>
1279 * or 0xffff if the offset is not valid for this string
1280 * @stable ICU 2.0
1281 */
1282 inline UChar charAt(int32_t offset) const;
1283
1284 /**
1285 * Return the code unit at offset <tt>offset</tt>.
1286 * If the offset is not valid (0..length()-1) then U+ffff is returned.
1287 * @param offset a valid offset into the text
1288 * @return the code unit at offset <tt>offset</tt>
1289 * @stable ICU 2.0
1290 */
1291 inline UChar operator[] (int32_t offset) const;
1292
1293 /**
1294 * Return the code point that contains the code unit
1295 * at offset <tt>offset</tt>.
1296 * If the offset is not valid (0..length()-1) then U+ffff is returned.
1297 * @param offset a valid offset into the text
1298 * that indicates the text offset of any of the code units
1299 * that will be assembled into a code point (21-bit value) and returned
1300 * @return the code point of text at <tt>offset</tt>
1301 * or 0xffff if the offset is not valid for this string
1302 * @stable ICU 2.0
1303 */
1304 UChar32 char32At(int32_t offset) const;
1305
1306 /**
1307 * Adjust a random-access offset so that
1308 * it points to the beginning of a Unicode character.
1309 * The offset that is passed in points to
1310 * any code unit of a code point,
1311 * while the returned offset will point to the first code unit
1312 * of the same code point.
1313 * In UTF-16, if the input offset points to a second surrogate
1314 * of a surrogate pair, then the returned offset will point
1315 * to the first surrogate.
1316 * @param offset a valid offset into one code point of the text
1317 * @return offset of the first code unit of the same code point
1318 * @see U16_SET_CP_START
1319 * @stable ICU 2.0
1320 */
1321 int32_t getChar32Start(int32_t offset) const;
1322
1323 /**
1324 * Adjust a random-access offset so that
1325 * it points behind a Unicode character.
1326 * The offset that is passed in points behind
1327 * any code unit of a code point,
1328 * while the returned offset will point behind the last code unit
1329 * of the same code point.
1330 * In UTF-16, if the input offset points behind the first surrogate
1331 * (i.e., to the second surrogate)
1332 * of a surrogate pair, then the returned offset will point
1333 * behind the second surrogate (i.e., to the first surrogate).
1334 * @param offset a valid offset after any code unit of a code point of the text
1335 * @return offset of the first code unit after the same code point
1336 * @see U16_SET_CP_LIMIT
1337 * @stable ICU 2.0
1338 */
1339 int32_t getChar32Limit(int32_t offset) const;
1340
1341 /**
1342 * Move the code unit index along the string by delta code points.
1343 * Interpret the input index as a code unit-based offset into the string,
1344 * move the index forward or backward by delta code points, and
1345 * return the resulting index.
1346 * The input index should point to the first code unit of a code point,
1347 * if there is more than one.
1348 *
1349 * Both input and output indexes are code unit-based as for all
1350 * string indexes/offsets in ICU (and other libraries, like MBCS char*).
1351 * If delta<0 then the index is moved backward (toward the start of the string).
1352 * If delta>0 then the index is moved forward (toward the end of the string).
1353 *
1354 * This behaves like CharacterIterator::move32(delta, kCurrent).
1355 *
1356 * Behavior for out-of-bounds indexes:
1357 * <code>moveIndex32</code> pins the input index to 0..length(), i.e.,
1358 * if the input index<0 then it is pinned to 0;
1359 * if it is index>length() then it is pinned to length().
1360 * Afterwards, the index is moved by <code>delta</code> code points
1361 * forward or backward,
1362 * but no further backward than to 0 and no further forward than to length().
1363 * The resulting index return value will be in between 0 and length(), inclusively.
1364 *
1365 * Examples:
1366 * <pre>
1367 * // s has code points 'a' U+10000 'b' U+10ffff U+2029
1368 * UnicodeString s=UNICODE_STRING("a\\U00010000b\\U0010ffff\\u2029", 31).unescape();
1369 *
1370 * // initial index: position of U+10000
1371 * int32_t index=1;
1372 *
1373 * // the following examples will all result in index==4, position of U+10ffff
1374 *
1375 * // skip 2 code points from some position in the string
1376 * index=s.moveIndex32(index, 2); // skips U+10000 and 'b'
1377 *
1378 * // go to the 3rd code point from the start of s (0-based)
1379 * index=s.moveIndex32(0, 3); // skips 'a', U+10000, and 'b'
1380 *
1381 * // go to the next-to-last code point of s
1382 * index=s.moveIndex32(s.length(), -2); // backward-skips U+2029 and U+10ffff
1383 * </pre>
1384 *
1385 * @param index input code unit index
1386 * @param delta (signed) code point count to move the index forward or backward
1387 * in the string
1388 * @return the resulting code unit index
1389 * @stable ICU 2.0
1390 */
1391 int32_t moveIndex32(int32_t index, int32_t delta) const;
1392
1393 /* Substring extraction */
1394
1395 /**
1396 * Copy the characters in the range
1397 * [<tt>start</tt>, <tt>start + length</tt>) into the array <tt>dst</tt>,
1398 * beginning at <tt>dstStart</tt>.
1399 * If the string aliases to <code>dst</code> itself as an external buffer,
1400 * then extract() will not copy the contents.
1401 *
1402 * @param start offset of first character which will be copied into the array
1403 * @param length the number of characters to extract
1404 * @param dst array in which to copy characters. The length of <tt>dst</tt>
1405 * must be at least (<tt>dstStart + length</tt>).
1406 * @param dstStart the offset in <TT>dst</TT> where the first character
1407 * will be extracted
1408 * @stable ICU 2.0
1409 */
1410 inline void extract(int32_t start,
1411 int32_t length,
1412 UChar *dst,
1413 int32_t dstStart = 0) const;
1414
1415 /**
1416 * Copy the contents of the string into dest.
1417 * This is a convenience function that
1418 * checks if there is enough space in dest,
1419 * extracts the entire string if possible,
1420 * and NUL-terminates dest if possible.
1421 *
1422 * If the string fits into dest but cannot be NUL-terminated
1423 * (length()==destCapacity) then the error code is set to U_STRING_NOT_TERMINATED_WARNING.
1424 * If the string itself does not fit into dest
1425 * (length()>destCapacity) then the error code is set to U_BUFFER_OVERFLOW_ERROR.
1426 *
1427 * If the string aliases to <code>dest</code> itself as an external buffer,
1428 * then extract() will not copy the contents.
1429 *
1430 * @param dest Destination string buffer.
1431 * @param destCapacity Number of UChars available at dest.
1432 * @param errorCode ICU error code.
1433 * @return length()
1434 * @stable ICU 2.0
1435 */
1436 int32_t
1437 extract(UChar *dest, int32_t destCapacity,
1438 UErrorCode &errorCode) const;
1439
1440 /**
1441 * Copy the characters in the range
1442 * [<tt>start</tt>, <tt>start + length</tt>) into the UnicodeString
1443 * <tt>target</tt>.
1444 * @param start offset of first character which will be copied
1445 * @param length the number of characters to extract
1446 * @param target UnicodeString into which to copy characters.
1447 * @return A reference to <TT>target</TT>
1448 * @stable ICU 2.0
1449 */
1450 inline void extract(int32_t start,
1451 int32_t length,
1452 UnicodeString& target) const;
1453
1454 /**
1455 * Copy the characters in the range [<tt>start</tt>, <tt>limit</tt>)
1456 * into the array <tt>dst</tt>, beginning at <tt>dstStart</tt>.
1457 * @param start offset of first character which will be copied into the array
1458 * @param limit offset immediately following the last character to be copied
1459 * @param dst array in which to copy characters. The length of <tt>dst</tt>
1460 * must be at least (<tt>dstStart + (limit - start)</tt>).
1461 * @param dstStart the offset in <TT>dst</TT> where the first character
1462 * will be extracted
1463 * @stable ICU 2.0
1464 */
1465 inline void extractBetween(int32_t start,
1466 int32_t limit,
1467 UChar *dst,
1468 int32_t dstStart = 0) const;
1469
1470 /**
1471 * Copy the characters in the range [<tt>start</tt>, <tt>limit</tt>)
1472 * into the UnicodeString <tt>target</tt>. Replaceable API.
1473 * @param start offset of first character which will be copied
1474 * @param limit offset immediately following the last character to be copied
1475 * @param target UnicodeString into which to copy characters.
1476 * @return A reference to <TT>target</TT>
1477 * @stable ICU 2.0
1478 */
1479 virtual void extractBetween(int32_t start,
1480 int32_t limit,
1481 UnicodeString& target) const;
1482
1483 /**
1484 * Copy the characters in the range
1485 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters.
1486 * All characters must be invariant (see utypes.h).
1487 * Use US_INV as the last, signature-distinguishing parameter.
1488 *
1489 * This function does not write any more than <code>targetLength</code>
1490 * characters but returns the length of the entire output string
1491 * so that one can allocate a larger buffer and call the function again
1492 * if necessary.
1493 * The output string is NUL-terminated if possible.
1494 *
1495 * @param start offset of first character which will be copied
1496 * @param startLength the number of characters to extract
1497 * @param target the target buffer for extraction, can be NULL
1498 * if targetLength is 0
1499 * @param targetCapacity the length of the target buffer
1500 * @param inv Signature-distinguishing paramater, use US_INV.
1501 * @return the output string length, not including the terminating NUL
1502 * @stable ICU 3.2
1503 */
1504 int32_t extract(int32_t start,
1505 int32_t startLength,
1506 char *target,
1507 int32_t targetCapacity,
1508 enum EInvariant inv) const;
1509
1510 #if U_CHARSET_IS_UTF8 || !UCONFIG_NO_CONVERSION
1511
1512 /**
1513 * Copy the characters in the range
1514 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters
1515 * in the platform's default codepage.
1516 * This function does not write any more than <code>targetLength</code>
1517 * characters but returns the length of the entire output string
1518 * so that one can allocate a larger buffer and call the function again
1519 * if necessary.
1520 * The output string is NUL-terminated if possible.
1521 *
1522 * @param start offset of first character which will be copied
1523 * @param startLength the number of characters to extract
1524 * @param target the target buffer for extraction
1525 * @param targetLength the length of the target buffer
1526 * If <TT>target</TT> is NULL, then the number of bytes required for
1527 * <TT>target</TT> is returned.
1528 * @return the output string length, not including the terminating NUL
1529 * @stable ICU 2.0
1530 */
1531 int32_t extract(int32_t start,
1532 int32_t startLength,
1533 char *target,
1534 uint32_t targetLength) const;
1535
1536 #endif
1537
1538 #if !UCONFIG_NO_CONVERSION
1539
1540 /**
1541 * Copy the characters in the range
1542 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters
1543 * in a specified codepage.
1544 * The output string is NUL-terminated.
1545 *
1546 * Recommendation: For invariant-character strings use
1547 * extract(int32_t start, int32_t length, char *target, int32_t targetCapacity, enum EInvariant inv) const
1548 * because it avoids object code dependencies of UnicodeString on
1549 * the conversion code.
1550 *
1551 * @param start offset of first character which will be copied
1552 * @param startLength the number of characters to extract
1553 * @param target the target buffer for extraction
1554 * @param codepage the desired codepage for the characters. 0 has
1555 * the special meaning of the default codepage
1556 * If <code>codepage</code> is an empty string (<code>""</code>),
1557 * then a simple conversion is performed on the codepage-invariant
1558 * subset ("invariant characters") of the platform encoding. See utypes.h.
1559 * If <TT>target</TT> is NULL, then the number of bytes required for
1560 * <TT>target</TT> is returned. It is assumed that the target is big enough
1561 * to fit all of the characters.
1562 * @return the output string length, not including the terminating NUL
1563 * @stable ICU 2.0
1564 */
1565 inline int32_t extract(int32_t start,
1566 int32_t startLength,
1567 char *target,
1568 const char *codepage = 0) const;
1569
1570 /**
1571 * Copy the characters in the range
1572 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters
1573 * in a specified codepage.
1574 * This function does not write any more than <code>targetLength</code>
1575 * characters but returns the length of the entire output string
1576 * so that one can allocate a larger buffer and call the function again
1577 * if necessary.
1578 * The output string is NUL-terminated if possible.
1579 *
1580 * Recommendation: For invariant-character strings use
1581 * extract(int32_t start, int32_t length, char *target, int32_t targetCapacity, enum EInvariant inv) const
1582 * because it avoids object code dependencies of UnicodeString on
1583 * the conversion code.
1584 *
1585 * @param start offset of first character which will be copied
1586 * @param startLength the number of characters to extract
1587 * @param target the target buffer for extraction
1588 * @param targetLength the length of the target buffer
1589 * @param codepage the desired codepage for the characters. 0 has
1590 * the special meaning of the default codepage
1591 * If <code>codepage</code> is an empty string (<code>""</code>),
1592 * then a simple conversion is performed on the codepage-invariant
1593 * subset ("invariant characters") of the platform encoding. See utypes.h.
1594 * If <TT>target</TT> is NULL, then the number of bytes required for
1595 * <TT>target</TT> is returned.
1596 * @return the output string length, not including the terminating NUL
1597 * @stable ICU 2.0
1598 */
1599 int32_t extract(int32_t start,
1600 int32_t startLength,
1601 char *target,
1602 uint32_t targetLength,
1603 const char *codepage) const;
1604
1605 /**
1606 * Convert the UnicodeString into a codepage string using an existing UConverter.
1607 * The output string is NUL-terminated if possible.
1608 *
1609 * This function avoids the overhead of opening and closing a converter if
1610 * multiple strings are extracted.
1611 *
1612 * @param dest destination string buffer, can be NULL if destCapacity==0
1613 * @param destCapacity the number of chars available at dest
1614 * @param cnv the converter object to be used (ucnv_resetFromUnicode() will be called),
1615 * or NULL for the default converter
1616 * @param errorCode normal ICU error code
1617 * @return the length of the output string, not counting the terminating NUL;
1618 * if the length is greater than destCapacity, then the string will not fit
1619 * and a buffer of the indicated length would need to be passed in
1620 * @stable ICU 2.0
1621 */
1622 int32_t extract(char *dest, int32_t destCapacity,
1623 UConverter *cnv,
1624 UErrorCode &errorCode) const;
1625
1626 #endif
1627
1628 /**
1629 * Create a temporary substring for the specified range.
1630 * Unlike the substring constructor and setTo() functions,
1631 * the object returned here will be a read-only alias (using getBuffer())
1632 * rather than copying the text.
1633 * As a result, this substring operation is much faster but requires
1634 * that the original string not be modified or deleted during the lifetime
1635 * of the returned substring object.
1636 * @param start offset of the first character visible in the substring
1637 * @param length length of the substring
1638 * @return a read-only alias UnicodeString object for the substring
1639 * @stable ICU 4.4
1640 */
1641 UnicodeString tempSubString(int32_t start=0, int32_t length=INT32_MAX) const;
1642
1643 /**
1644 * Create a temporary substring for the specified range.
1645 * Same as tempSubString(start, length) except that the substring range
1646 * is specified as a (start, limit) pair (with an exclusive limit index)
1647 * rather than a (start, length) pair.
1648 * @param start offset of the first character visible in the substring
1649 * @param limit offset immediately following the last character visible in the substring
1650 * @return a read-only alias UnicodeString object for the substring
1651 * @stable ICU 4.4
1652 */
1653 inline UnicodeString tempSubStringBetween(int32_t start, int32_t limit=INT32_MAX) const;
1654
1655 /**
1656 * Convert the UnicodeString to UTF-8 and write the result
1657 * to a ByteSink. This is called by toUTF8String().
1658 * Unpaired surrogates are replaced with U+FFFD.
1659 * Calls u_strToUTF8WithSub().
1660 *
1661 * @param sink A ByteSink to which the UTF-8 version of the string is written.
1662 * sink.Flush() is called at the end.
1663 * @stable ICU 4.2
1664 * @see toUTF8String
1665 */
1666 void toUTF8(ByteSink &sink) const;
1667
1668 #if U_HAVE_STD_STRING
1669
1670 /**
1671 * Convert the UnicodeString to UTF-8 and append the result
1672 * to a standard string.
1673 * Unpaired surrogates are replaced with U+FFFD.
1674 * Calls toUTF8().
1675 *
1676 * @param result A standard string (or a compatible object)
1677 * to which the UTF-8 version of the string is appended.
1678 * @return The string object.
1679 * @stable ICU 4.2
1680 * @see toUTF8
1681 */
1682 template<typename StringClass>
toUTF8String(StringClass & result)1683 StringClass &toUTF8String(StringClass &result) const {
1684 StringByteSink<StringClass> sbs(&result);
1685 toUTF8(sbs);
1686 return result;
1687 }
1688
1689 #endif
1690
1691 /**
1692 * Convert the UnicodeString to UTF-32.
1693 * Unpaired surrogates are replaced with U+FFFD.
1694 * Calls u_strToUTF32WithSub().
1695 *
1696 * @param utf32 destination string buffer, can be NULL if capacity==0
1697 * @param capacity the number of UChar32s available at utf32
1698 * @param errorCode Standard ICU error code. Its input value must
1699 * pass the U_SUCCESS() test, or else the function returns
1700 * immediately. Check for U_FAILURE() on output or use with
1701 * function chaining. (See User Guide for details.)
1702 * @return The length of the UTF-32 string.
1703 * @see fromUTF32
1704 * @stable ICU 4.2
1705 */
1706 int32_t toUTF32(UChar32 *utf32, int32_t capacity, UErrorCode &errorCode) const;
1707
1708 /* Length operations */
1709
1710 /**
1711 * Return the length of the UnicodeString object.
1712 * The length is the number of UChar code units are in the UnicodeString.
1713 * If you want the number of code points, please use countChar32().
1714 * @return the length of the UnicodeString object
1715 * @see countChar32
1716 * @stable ICU 2.0
1717 */
1718 inline int32_t length(void) const;
1719
1720 /**
1721 * Count Unicode code points in the length UChar code units of the string.
1722 * A code point may occupy either one or two UChar code units.
1723 * Counting code points involves reading all code units.
1724 *
1725 * This functions is basically the inverse of moveIndex32().
1726 *
1727 * @param start the index of the first code unit to check
1728 * @param length the number of UChar code units to check
1729 * @return the number of code points in the specified code units
1730 * @see length
1731 * @stable ICU 2.0
1732 */
1733 int32_t
1734 countChar32(int32_t start=0, int32_t length=INT32_MAX) const;
1735
1736 /**
1737 * Check if the length UChar code units of the string
1738 * contain more Unicode code points than a certain number.
1739 * This is more efficient than counting all code points in this part of the string
1740 * and comparing that number with a threshold.
1741 * This function may not need to scan the string at all if the length
1742 * falls within a certain range, and
1743 * never needs to count more than 'number+1' code points.
1744 * Logically equivalent to (countChar32(start, length)>number).
1745 * A Unicode code point may occupy either one or two UChar code units.
1746 *
1747 * @param start the index of the first code unit to check (0 for the entire string)
1748 * @param length the number of UChar code units to check
1749 * (use INT32_MAX for the entire string; remember that start/length
1750 * values are pinned)
1751 * @param number The number of code points in the (sub)string is compared against
1752 * the 'number' parameter.
1753 * @return Boolean value for whether the string contains more Unicode code points
1754 * than 'number'. Same as (u_countChar32(s, length)>number).
1755 * @see countChar32
1756 * @see u_strHasMoreChar32Than
1757 * @stable ICU 2.4
1758 */
1759 UBool
1760 hasMoreChar32Than(int32_t start, int32_t length, int32_t number) const;
1761
1762 /**
1763 * Determine if this string is empty.
1764 * @return TRUE if this string contains 0 characters, FALSE otherwise.
1765 * @stable ICU 2.0
1766 */
1767 inline UBool isEmpty(void) const;
1768
1769 /**
1770 * Return the capacity of the internal buffer of the UnicodeString object.
1771 * This is useful together with the getBuffer functions.
1772 * See there for details.
1773 *
1774 * @return the number of UChars available in the internal buffer
1775 * @see getBuffer
1776 * @stable ICU 2.0
1777 */
1778 inline int32_t getCapacity(void) const;
1779
1780 /* Other operations */
1781
1782 /**
1783 * Generate a hash code for this object.
1784 * @return The hash code of this UnicodeString.
1785 * @stable ICU 2.0
1786 */
1787 inline int32_t hashCode(void) const;
1788
1789 /**
1790 * Determine if this object contains a valid string.
1791 * A bogus string has no value. It is different from an empty string,
1792 * although in both cases isEmpty() returns TRUE and length() returns 0.
1793 * setToBogus() and isBogus() can be used to indicate that no string value is available.
1794 * For a bogus string, getBuffer() and getTerminatedBuffer() return NULL, and
1795 * length() returns 0.
1796 *
1797 * @return TRUE if the string is valid, FALSE otherwise
1798 * @see setToBogus()
1799 * @stable ICU 2.0
1800 */
1801 inline UBool isBogus(void) const;
1802
1803
1804 //========================================
1805 // Write operations
1806 //========================================
1807
1808 /* Assignment operations */
1809
1810 /**
1811 * Assignment operator. Replace the characters in this UnicodeString
1812 * with the characters from <TT>srcText</TT>.
1813 * @param srcText The text containing the characters to replace
1814 * @return a reference to this
1815 * @stable ICU 2.0
1816 */
1817 UnicodeString &operator=(const UnicodeString &srcText);
1818
1819 /**
1820 * Almost the same as the assignment operator.
1821 * Replace the characters in this UnicodeString
1822 * with the characters from <code>srcText</code>.
1823 *
1824 * This function works the same as the assignment operator
1825 * for all strings except for ones that are readonly aliases.
1826 *
1827 * Starting with ICU 2.4, the assignment operator and the copy constructor
1828 * allocate a new buffer and copy the buffer contents even for readonly aliases.
1829 * This function implements the old, more efficient but less safe behavior
1830 * of making this string also a readonly alias to the same buffer.
1831 *
1832 * The fastCopyFrom function must be used only if it is known that the lifetime of
1833 * this UnicodeString does not exceed the lifetime of the aliased buffer
1834 * including its contents, for example for strings from resource bundles
1835 * or aliases to string constants.
1836 *
1837 * @param src The text containing the characters to replace.
1838 * @return a reference to this
1839 * @stable ICU 2.4
1840 */
1841 UnicodeString &fastCopyFrom(const UnicodeString &src);
1842
1843 /**
1844 * Assignment operator. Replace the characters in this UnicodeString
1845 * with the code unit <TT>ch</TT>.
1846 * @param ch the code unit to replace
1847 * @return a reference to this
1848 * @stable ICU 2.0
1849 */
1850 inline UnicodeString& operator= (UChar ch);
1851
1852 /**
1853 * Assignment operator. Replace the characters in this UnicodeString
1854 * with the code point <TT>ch</TT>.
1855 * @param ch the code point to replace
1856 * @return a reference to this
1857 * @stable ICU 2.0
1858 */
1859 inline UnicodeString& operator= (UChar32 ch);
1860
1861 /**
1862 * Set the text in the UnicodeString object to the characters
1863 * in <TT>srcText</TT> in the range
1864 * [<TT>srcStart</TT>, <TT>srcText.length()</TT>).
1865 * <TT>srcText</TT> is not modified.
1866 * @param srcText the source for the new characters
1867 * @param srcStart the offset into <TT>srcText</TT> where new characters
1868 * will be obtained
1869 * @return a reference to this
1870 * @stable ICU 2.2
1871 */
1872 inline UnicodeString& setTo(const UnicodeString& srcText,
1873 int32_t srcStart);
1874
1875 /**
1876 * Set the text in the UnicodeString object to the characters
1877 * in <TT>srcText</TT> in the range
1878 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
1879 * <TT>srcText</TT> is not modified.
1880 * @param srcText the source for the new characters
1881 * @param srcStart the offset into <TT>srcText</TT> where new characters
1882 * will be obtained
1883 * @param srcLength the number of characters in <TT>srcText</TT> in the
1884 * replace string.
1885 * @return a reference to this
1886 * @stable ICU 2.0
1887 */
1888 inline UnicodeString& setTo(const UnicodeString& srcText,
1889 int32_t srcStart,
1890 int32_t srcLength);
1891
1892 /**
1893 * Set the text in the UnicodeString object to the characters in
1894 * <TT>srcText</TT>.
1895 * <TT>srcText</TT> is not modified.
1896 * @param srcText the source for the new characters
1897 * @return a reference to this
1898 * @stable ICU 2.0
1899 */
1900 inline UnicodeString& setTo(const UnicodeString& srcText);
1901
1902 /**
1903 * Set the characters in the UnicodeString object to the characters
1904 * in <TT>srcChars</TT>. <TT>srcChars</TT> is not modified.
1905 * @param srcChars the source for the new characters
1906 * @param srcLength the number of Unicode characters in srcChars.
1907 * @return a reference to this
1908 * @stable ICU 2.0
1909 */
1910 inline UnicodeString& setTo(const UChar *srcChars,
1911 int32_t srcLength);
1912
1913 /**
1914 * Set the characters in the UnicodeString object to the code unit
1915 * <TT>srcChar</TT>.
1916 * @param srcChar the code unit which becomes the UnicodeString's character
1917 * content
1918 * @return a reference to this
1919 * @stable ICU 2.0
1920 */
1921 UnicodeString& setTo(UChar srcChar);
1922
1923 /**
1924 * Set the characters in the UnicodeString object to the code point
1925 * <TT>srcChar</TT>.
1926 * @param srcChar the code point which becomes the UnicodeString's character
1927 * content
1928 * @return a reference to this
1929 * @stable ICU 2.0
1930 */
1931 UnicodeString& setTo(UChar32 srcChar);
1932
1933 /**
1934 * Aliasing setTo() function, analogous to the readonly-aliasing UChar* constructor.
1935 * The text will be used for the UnicodeString object, but
1936 * it will not be released when the UnicodeString is destroyed.
1937 * This has copy-on-write semantics:
1938 * When the string is modified, then the buffer is first copied into
1939 * newly allocated memory.
1940 * The aliased buffer is never modified.
1941 *
1942 * In an assignment to another UnicodeString, when using the copy constructor
1943 * or the assignment operator, the text will be copied.
1944 * When using fastCopyFrom(), the text will be aliased again,
1945 * so that both strings then alias the same readonly-text.
1946 *
1947 * @param isTerminated specifies if <code>text</code> is <code>NUL</code>-terminated.
1948 * This must be true if <code>textLength==-1</code>.
1949 * @param text The characters to alias for the UnicodeString.
1950 * @param textLength The number of Unicode characters in <code>text</code> to alias.
1951 * If -1, then this constructor will determine the length
1952 * by calling <code>u_strlen()</code>.
1953 * @return a reference to this
1954 * @stable ICU 2.0
1955 */
1956 UnicodeString &setTo(UBool isTerminated,
1957 const UChar *text,
1958 int32_t textLength);
1959
1960 /**
1961 * Aliasing setTo() function, analogous to the writable-aliasing UChar* constructor.
1962 * The text will be used for the UnicodeString object, but
1963 * it will not be released when the UnicodeString is destroyed.
1964 * This has write-through semantics:
1965 * For as long as the capacity of the buffer is sufficient, write operations
1966 * will directly affect the buffer. When more capacity is necessary, then
1967 * a new buffer will be allocated and the contents copied as with regularly
1968 * constructed strings.
1969 * In an assignment to another UnicodeString, the buffer will be copied.
1970 * The extract(UChar *dst) function detects whether the dst pointer is the same
1971 * as the string buffer itself and will in this case not copy the contents.
1972 *
1973 * @param buffer The characters to alias for the UnicodeString.
1974 * @param buffLength The number of Unicode characters in <code>buffer</code> to alias.
1975 * @param buffCapacity The size of <code>buffer</code> in UChars.
1976 * @return a reference to this
1977 * @stable ICU 2.0
1978 */
1979 UnicodeString &setTo(UChar *buffer,
1980 int32_t buffLength,
1981 int32_t buffCapacity);
1982
1983 /**
1984 * Make this UnicodeString object invalid.
1985 * The string will test TRUE with isBogus().
1986 *
1987 * A bogus string has no value. It is different from an empty string.
1988 * It can be used to indicate that no string value is available.
1989 * getBuffer() and getTerminatedBuffer() return NULL, and
1990 * length() returns 0.
1991 *
1992 * This utility function is used throughout the UnicodeString
1993 * implementation to indicate that a UnicodeString operation failed,
1994 * and may be used in other functions,
1995 * especially but not exclusively when such functions do not
1996 * take a UErrorCode for simplicity.
1997 *
1998 * The following methods, and no others, will clear a string object's bogus flag:
1999 * - remove()
2000 * - remove(0, INT32_MAX)
2001 * - truncate(0)
2002 * - operator=() (assignment operator)
2003 * - setTo(...)
2004 *
2005 * The simplest ways to turn a bogus string into an empty one
2006 * is to use the remove() function.
2007 * Examples for other functions that are equivalent to "set to empty string":
2008 * \code
2009 * if(s.isBogus()) {
2010 * s.remove(); // set to an empty string (remove all), or
2011 * s.remove(0, INT32_MAX); // set to an empty string (remove all), or
2012 * s.truncate(0); // set to an empty string (complete truncation), or
2013 * s=UnicodeString(); // assign an empty string, or
2014 * s.setTo((UChar32)-1); // set to a pseudo code point that is out of range, or
2015 * static const UChar nul=0;
2016 * s.setTo(&nul, 0); // set to an empty C Unicode string
2017 * }
2018 * \endcode
2019 *
2020 * @see isBogus()
2021 * @stable ICU 2.0
2022 */
2023 void setToBogus();
2024
2025 /**
2026 * Set the character at the specified offset to the specified character.
2027 * @param offset A valid offset into the text of the character to set
2028 * @param ch The new character
2029 * @return A reference to this
2030 * @stable ICU 2.0
2031 */
2032 UnicodeString& setCharAt(int32_t offset,
2033 UChar ch);
2034
2035
2036 /* Append operations */
2037
2038 /**
2039 * Append operator. Append the code unit <TT>ch</TT> to the UnicodeString
2040 * object.
2041 * @param ch the code unit to be appended
2042 * @return a reference to this
2043 * @stable ICU 2.0
2044 */
2045 inline UnicodeString& operator+= (UChar ch);
2046
2047 /**
2048 * Append operator. Append the code point <TT>ch</TT> to the UnicodeString
2049 * object.
2050 * @param ch the code point to be appended
2051 * @return a reference to this
2052 * @stable ICU 2.0
2053 */
2054 inline UnicodeString& operator+= (UChar32 ch);
2055
2056 /**
2057 * Append operator. Append the characters in <TT>srcText</TT> to the
2058 * UnicodeString object. <TT>srcText</TT> is not modified.
2059 * @param srcText the source for the new characters
2060 * @return a reference to this
2061 * @stable ICU 2.0
2062 */
2063 inline UnicodeString& operator+= (const UnicodeString& srcText);
2064
2065 /**
2066 * Append the characters
2067 * in <TT>srcText</TT> in the range
2068 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) to the
2069 * UnicodeString object at offset <TT>start</TT>. <TT>srcText</TT>
2070 * is not modified.
2071 * @param srcText the source for the new characters
2072 * @param srcStart the offset into <TT>srcText</TT> where new characters
2073 * will be obtained
2074 * @param srcLength the number of characters in <TT>srcText</TT> in
2075 * the append string
2076 * @return a reference to this
2077 * @stable ICU 2.0
2078 */
2079 inline UnicodeString& append(const UnicodeString& srcText,
2080 int32_t srcStart,
2081 int32_t srcLength);
2082
2083 /**
2084 * Append the characters in <TT>srcText</TT> to the UnicodeString object.
2085 * <TT>srcText</TT> is not modified.
2086 * @param srcText the source for the new characters
2087 * @return a reference to this
2088 * @stable ICU 2.0
2089 */
2090 inline UnicodeString& append(const UnicodeString& srcText);
2091
2092 /**
2093 * Append the characters in <TT>srcChars</TT> in the range
2094 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) to the UnicodeString
2095 * object at offset
2096 * <TT>start</TT>. <TT>srcChars</TT> is not modified.
2097 * @param srcChars the source for the new characters
2098 * @param srcStart the offset into <TT>srcChars</TT> where new characters
2099 * will be obtained
2100 * @param srcLength the number of characters in <TT>srcChars</TT> in
2101 * the append string; can be -1 if <TT>srcChars</TT> is NUL-terminated
2102 * @return a reference to this
2103 * @stable ICU 2.0
2104 */
2105 inline UnicodeString& append(const UChar *srcChars,
2106 int32_t srcStart,
2107 int32_t srcLength);
2108
2109 /**
2110 * Append the characters in <TT>srcChars</TT> to the UnicodeString object
2111 * at offset <TT>start</TT>. <TT>srcChars</TT> is not modified.
2112 * @param srcChars the source for the new characters
2113 * @param srcLength the number of Unicode characters in <TT>srcChars</TT>;
2114 * can be -1 if <TT>srcChars</TT> is NUL-terminated
2115 * @return a reference to this
2116 * @stable ICU 2.0
2117 */
2118 inline UnicodeString& append(const UChar *srcChars,
2119 int32_t srcLength);
2120
2121 /**
2122 * Append the code unit <TT>srcChar</TT> to the UnicodeString object.
2123 * @param srcChar the code unit to append
2124 * @return a reference to this
2125 * @stable ICU 2.0
2126 */
2127 inline UnicodeString& append(UChar srcChar);
2128
2129 /**
2130 * Append the code point <TT>srcChar</TT> to the UnicodeString object.
2131 * @param srcChar the code point to append
2132 * @return a reference to this
2133 * @stable ICU 2.0
2134 */
2135 UnicodeString& append(UChar32 srcChar);
2136
2137
2138 /* Insert operations */
2139
2140 /**
2141 * Insert the characters in <TT>srcText</TT> in the range
2142 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) into the UnicodeString
2143 * object at offset <TT>start</TT>. <TT>srcText</TT> is not modified.
2144 * @param start the offset where the insertion begins
2145 * @param srcText the source for the new characters
2146 * @param srcStart the offset into <TT>srcText</TT> where new characters
2147 * will be obtained
2148 * @param srcLength the number of characters in <TT>srcText</TT> in
2149 * the insert string
2150 * @return a reference to this
2151 * @stable ICU 2.0
2152 */
2153 inline UnicodeString& insert(int32_t start,
2154 const UnicodeString& srcText,
2155 int32_t srcStart,
2156 int32_t srcLength);
2157
2158 /**
2159 * Insert the characters in <TT>srcText</TT> into the UnicodeString object
2160 * at offset <TT>start</TT>. <TT>srcText</TT> is not modified.
2161 * @param start the offset where the insertion begins
2162 * @param srcText the source for the new characters
2163 * @return a reference to this
2164 * @stable ICU 2.0
2165 */
2166 inline UnicodeString& insert(int32_t start,
2167 const UnicodeString& srcText);
2168
2169 /**
2170 * Insert the characters in <TT>srcChars</TT> in the range
2171 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) into the UnicodeString
2172 * object at offset <TT>start</TT>. <TT>srcChars</TT> is not modified.
2173 * @param start the offset at which the insertion begins
2174 * @param srcChars the source for the new characters
2175 * @param srcStart the offset into <TT>srcChars</TT> where new characters
2176 * will be obtained
2177 * @param srcLength the number of characters in <TT>srcChars</TT>
2178 * in the insert string
2179 * @return a reference to this
2180 * @stable ICU 2.0
2181 */
2182 inline UnicodeString& insert(int32_t start,
2183 const UChar *srcChars,
2184 int32_t srcStart,
2185 int32_t srcLength);
2186
2187 /**
2188 * Insert the characters in <TT>srcChars</TT> into the UnicodeString object
2189 * at offset <TT>start</TT>. <TT>srcChars</TT> is not modified.
2190 * @param start the offset where the insertion begins
2191 * @param srcChars the source for the new characters
2192 * @param srcLength the number of Unicode characters in srcChars.
2193 * @return a reference to this
2194 * @stable ICU 2.0
2195 */
2196 inline UnicodeString& insert(int32_t start,
2197 const UChar *srcChars,
2198 int32_t srcLength);
2199
2200 /**
2201 * Insert the code unit <TT>srcChar</TT> into the UnicodeString object at
2202 * offset <TT>start</TT>.
2203 * @param start the offset at which the insertion occurs
2204 * @param srcChar the code unit to insert
2205 * @return a reference to this
2206 * @stable ICU 2.0
2207 */
2208 inline UnicodeString& insert(int32_t start,
2209 UChar srcChar);
2210
2211 /**
2212 * Insert the code point <TT>srcChar</TT> into the UnicodeString object at
2213 * offset <TT>start</TT>.
2214 * @param start the offset at which the insertion occurs
2215 * @param srcChar the code point to insert
2216 * @return a reference to this
2217 * @stable ICU 2.0
2218 */
2219 inline UnicodeString& insert(int32_t start,
2220 UChar32 srcChar);
2221
2222
2223 /* Replace operations */
2224
2225 /**
2226 * Replace the characters in the range
2227 * [<TT>start</TT>, <TT>start + length</TT>) with the characters in
2228 * <TT>srcText</TT> in the range
2229 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
2230 * <TT>srcText</TT> is not modified.
2231 * @param start the offset at which the replace operation begins
2232 * @param length the number of characters to replace. The character at
2233 * <TT>start + length</TT> is not modified.
2234 * @param srcText the source for the new characters
2235 * @param srcStart the offset into <TT>srcText</TT> where new characters
2236 * will be obtained
2237 * @param srcLength the number of characters in <TT>srcText</TT> in
2238 * the replace string
2239 * @return a reference to this
2240 * @stable ICU 2.0
2241 */
2242 UnicodeString& replace(int32_t start,
2243 int32_t length,
2244 const UnicodeString& srcText,
2245 int32_t srcStart,
2246 int32_t srcLength);
2247
2248 /**
2249 * Replace the characters in the range
2250 * [<TT>start</TT>, <TT>start + length</TT>)
2251 * with the characters in <TT>srcText</TT>. <TT>srcText</TT> is
2252 * not modified.
2253 * @param start the offset at which the replace operation begins
2254 * @param length the number of characters to replace. The character at
2255 * <TT>start + length</TT> is not modified.
2256 * @param srcText the source for the new characters
2257 * @return a reference to this
2258 * @stable ICU 2.0
2259 */
2260 UnicodeString& replace(int32_t start,
2261 int32_t length,
2262 const UnicodeString& srcText);
2263
2264 /**
2265 * Replace the characters in the range
2266 * [<TT>start</TT>, <TT>start + length</TT>) with the characters in
2267 * <TT>srcChars</TT> in the range
2268 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). <TT>srcChars</TT>
2269 * is not modified.
2270 * @param start the offset at which the replace operation begins
2271 * @param length the number of characters to replace. The character at
2272 * <TT>start + length</TT> is not modified.
2273 * @param srcChars the source for the new characters
2274 * @param srcStart the offset into <TT>srcChars</TT> where new characters
2275 * will be obtained
2276 * @param srcLength the number of characters in <TT>srcChars</TT>
2277 * in the replace string
2278 * @return a reference to this
2279 * @stable ICU 2.0
2280 */
2281 UnicodeString& replace(int32_t start,
2282 int32_t length,
2283 const UChar *srcChars,
2284 int32_t srcStart,
2285 int32_t srcLength);
2286
2287 /**
2288 * Replace the characters in the range
2289 * [<TT>start</TT>, <TT>start + length</TT>) with the characters in
2290 * <TT>srcChars</TT>. <TT>srcChars</TT> is not modified.
2291 * @param start the offset at which the replace operation begins
2292 * @param length number of characters to replace. The character at
2293 * <TT>start + length</TT> is not modified.
2294 * @param srcChars the source for the new characters
2295 * @param srcLength the number of Unicode characters in srcChars
2296 * @return a reference to this
2297 * @stable ICU 2.0
2298 */
2299 inline UnicodeString& replace(int32_t start,
2300 int32_t length,
2301 const UChar *srcChars,
2302 int32_t srcLength);
2303
2304 /**
2305 * Replace the characters in the range
2306 * [<TT>start</TT>, <TT>start + length</TT>) with the code unit
2307 * <TT>srcChar</TT>.
2308 * @param start the offset at which the replace operation begins
2309 * @param length the number of characters to replace. The character at
2310 * <TT>start + length</TT> is not modified.
2311 * @param srcChar the new code unit
2312 * @return a reference to this
2313 * @stable ICU 2.0
2314 */
2315 inline UnicodeString& replace(int32_t start,
2316 int32_t length,
2317 UChar srcChar);
2318
2319 /**
2320 * Replace the characters in the range
2321 * [<TT>start</TT>, <TT>start + length</TT>) with the code point
2322 * <TT>srcChar</TT>.
2323 * @param start the offset at which the replace operation begins
2324 * @param length the number of characters to replace. The character at
2325 * <TT>start + length</TT> is not modified.
2326 * @param srcChar the new code point
2327 * @return a reference to this
2328 * @stable ICU 2.0
2329 */
2330 UnicodeString& replace(int32_t start, int32_t length, UChar32 srcChar);
2331
2332 /**
2333 * Replace the characters in the range [<TT>start</TT>, <TT>limit</TT>)
2334 * with the characters in <TT>srcText</TT>. <TT>srcText</TT> is not modified.
2335 * @param start the offset at which the replace operation begins
2336 * @param limit the offset immediately following the replace range
2337 * @param srcText the source for the new characters
2338 * @return a reference to this
2339 * @stable ICU 2.0
2340 */
2341 inline UnicodeString& replaceBetween(int32_t start,
2342 int32_t limit,
2343 const UnicodeString& srcText);
2344
2345 /**
2346 * Replace the characters in the range [<TT>start</TT>, <TT>limit</TT>)
2347 * with the characters in <TT>srcText</TT> in the range
2348 * [<TT>srcStart</TT>, <TT>srcLimit</TT>). <TT>srcText</TT> is not modified.
2349 * @param start the offset at which the replace operation begins
2350 * @param limit the offset immediately following the replace range
2351 * @param srcText the source for the new characters
2352 * @param srcStart the offset into <TT>srcChars</TT> where new characters
2353 * will be obtained
2354 * @param srcLimit the offset immediately following the range to copy
2355 * in <TT>srcText</TT>
2356 * @return a reference to this
2357 * @stable ICU 2.0
2358 */
2359 inline UnicodeString& replaceBetween(int32_t start,
2360 int32_t limit,
2361 const UnicodeString& srcText,
2362 int32_t srcStart,
2363 int32_t srcLimit);
2364
2365 /**
2366 * Replace a substring of this object with the given text.
2367 * @param start the beginning index, inclusive; <code>0 <= start
2368 * <= limit</code>.
2369 * @param limit the ending index, exclusive; <code>start <= limit
2370 * <= length()</code>.
2371 * @param text the text to replace characters <code>start</code>
2372 * to <code>limit - 1</code>
2373 * @stable ICU 2.0
2374 */
2375 virtual void handleReplaceBetween(int32_t start,
2376 int32_t limit,
2377 const UnicodeString& text);
2378
2379 /**
2380 * Replaceable API
2381 * @return TRUE if it has MetaData
2382 * @stable ICU 2.4
2383 */
2384 virtual UBool hasMetaData() const;
2385
2386 /**
2387 * Copy a substring of this object, retaining attribute (out-of-band)
2388 * information. This method is used to duplicate or reorder substrings.
2389 * The destination index must not overlap the source range.
2390 *
2391 * @param start the beginning index, inclusive; <code>0 <= start <=
2392 * limit</code>.
2393 * @param limit the ending index, exclusive; <code>start <= limit <=
2394 * length()</code>.
2395 * @param dest the destination index. The characters from
2396 * <code>start..limit-1</code> will be copied to <code>dest</code>.
2397 * Implementations of this method may assume that <code>dest <= start ||
2398 * dest >= limit</code>.
2399 * @stable ICU 2.0
2400 */
2401 virtual void copy(int32_t start, int32_t limit, int32_t dest);
2402
2403 /* Search and replace operations */
2404
2405 /**
2406 * Replace all occurrences of characters in oldText with the characters
2407 * in newText
2408 * @param oldText the text containing the search text
2409 * @param newText the text containing the replacement text
2410 * @return a reference to this
2411 * @stable ICU 2.0
2412 */
2413 inline UnicodeString& findAndReplace(const UnicodeString& oldText,
2414 const UnicodeString& newText);
2415
2416 /**
2417 * Replace all occurrences of characters in oldText with characters
2418 * in newText
2419 * in the range [<TT>start</TT>, <TT>start + length</TT>).
2420 * @param start the start of the range in which replace will performed
2421 * @param length the length of the range in which replace will be performed
2422 * @param oldText the text containing the search text
2423 * @param newText the text containing the replacement text
2424 * @return a reference to this
2425 * @stable ICU 2.0
2426 */
2427 inline UnicodeString& findAndReplace(int32_t start,
2428 int32_t length,
2429 const UnicodeString& oldText,
2430 const UnicodeString& newText);
2431
2432 /**
2433 * Replace all occurrences of characters in oldText in the range
2434 * [<TT>oldStart</TT>, <TT>oldStart + oldLength</TT>) with the characters
2435 * in newText in the range
2436 * [<TT>newStart</TT>, <TT>newStart + newLength</TT>)
2437 * in the range [<TT>start</TT>, <TT>start + length</TT>).
2438 * @param start the start of the range in which replace will performed
2439 * @param length the length of the range in which replace will be performed
2440 * @param oldText the text containing the search text
2441 * @param oldStart the start of the search range in <TT>oldText</TT>
2442 * @param oldLength the length of the search range in <TT>oldText</TT>
2443 * @param newText the text containing the replacement text
2444 * @param newStart the start of the replacement range in <TT>newText</TT>
2445 * @param newLength the length of the replacement range in <TT>newText</TT>
2446 * @return a reference to this
2447 * @stable ICU 2.0
2448 */
2449 UnicodeString& findAndReplace(int32_t start,
2450 int32_t length,
2451 const UnicodeString& oldText,
2452 int32_t oldStart,
2453 int32_t oldLength,
2454 const UnicodeString& newText,
2455 int32_t newStart,
2456 int32_t newLength);
2457
2458
2459 /* Remove operations */
2460
2461 /**
2462 * Remove all characters from the UnicodeString object.
2463 * @return a reference to this
2464 * @stable ICU 2.0
2465 */
2466 inline UnicodeString& remove(void);
2467
2468 /**
2469 * Remove the characters in the range
2470 * [<TT>start</TT>, <TT>start + length</TT>) from the UnicodeString object.
2471 * @param start the offset of the first character to remove
2472 * @param length the number of characters to remove
2473 * @return a reference to this
2474 * @stable ICU 2.0
2475 */
2476 inline UnicodeString& remove(int32_t start,
2477 int32_t length = (int32_t)INT32_MAX);
2478
2479 /**
2480 * Remove the characters in the range
2481 * [<TT>start</TT>, <TT>limit</TT>) from the UnicodeString object.
2482 * @param start the offset of the first character to remove
2483 * @param limit the offset immediately following the range to remove
2484 * @return a reference to this
2485 * @stable ICU 2.0
2486 */
2487 inline UnicodeString& removeBetween(int32_t start,
2488 int32_t limit = (int32_t)INT32_MAX);
2489
2490 /**
2491 * Retain only the characters in the range
2492 * [<code>start</code>, <code>limit</code>) from the UnicodeString object.
2493 * Removes characters before <code>start</code> and at and after <code>limit</code>.
2494 * @param start the offset of the first character to retain
2495 * @param limit the offset immediately following the range to retain
2496 * @return a reference to this
2497 * @stable ICU 4.4
2498 */
2499 inline UnicodeString &retainBetween(int32_t start, int32_t limit = INT32_MAX);
2500
2501 /* Length operations */
2502
2503 /**
2504 * Pad the start of this UnicodeString with the character <TT>padChar</TT>.
2505 * If the length of this UnicodeString is less than targetLength,
2506 * length() - targetLength copies of padChar will be added to the
2507 * beginning of this UnicodeString.
2508 * @param targetLength the desired length of the string
2509 * @param padChar the character to use for padding. Defaults to
2510 * space (U+0020)
2511 * @return TRUE if the text was padded, FALSE otherwise.
2512 * @stable ICU 2.0
2513 */
2514 UBool padLeading(int32_t targetLength,
2515 UChar padChar = 0x0020);
2516
2517 /**
2518 * Pad the end of this UnicodeString with the character <TT>padChar</TT>.
2519 * If the length of this UnicodeString is less than targetLength,
2520 * length() - targetLength copies of padChar will be added to the
2521 * end of this UnicodeString.
2522 * @param targetLength the desired length of the string
2523 * @param padChar the character to use for padding. Defaults to
2524 * space (U+0020)
2525 * @return TRUE if the text was padded, FALSE otherwise.
2526 * @stable ICU 2.0
2527 */
2528 UBool padTrailing(int32_t targetLength,
2529 UChar padChar = 0x0020);
2530
2531 /**
2532 * Truncate this UnicodeString to the <TT>targetLength</TT>.
2533 * @param targetLength the desired length of this UnicodeString.
2534 * @return TRUE if the text was truncated, FALSE otherwise
2535 * @stable ICU 2.0
2536 */
2537 inline UBool truncate(int32_t targetLength);
2538
2539 /**
2540 * Trims leading and trailing whitespace from this UnicodeString.
2541 * @return a reference to this
2542 * @stable ICU 2.0
2543 */
2544 UnicodeString& trim(void);
2545
2546
2547 /* Miscellaneous operations */
2548
2549 /**
2550 * Reverse this UnicodeString in place.
2551 * @return a reference to this
2552 * @stable ICU 2.0
2553 */
2554 inline UnicodeString& reverse(void);
2555
2556 /**
2557 * Reverse the range [<TT>start</TT>, <TT>start + length</TT>) in
2558 * this UnicodeString.
2559 * @param start the start of the range to reverse
2560 * @param length the number of characters to to reverse
2561 * @return a reference to this
2562 * @stable ICU 2.0
2563 */
2564 inline UnicodeString& reverse(int32_t start,
2565 int32_t length);
2566
2567 /**
2568 * Convert the characters in this to UPPER CASE following the conventions of
2569 * the default locale.
2570 * @return A reference to this.
2571 * @stable ICU 2.0
2572 */
2573 UnicodeString& toUpper(void);
2574
2575 /**
2576 * Convert the characters in this to UPPER CASE following the conventions of
2577 * a specific locale.
2578 * @param locale The locale containing the conventions to use.
2579 * @return A reference to this.
2580 * @stable ICU 2.0
2581 */
2582 UnicodeString& toUpper(const Locale& locale);
2583
2584 /**
2585 * Convert the characters in this to lower case following the conventions of
2586 * the default locale.
2587 * @return A reference to this.
2588 * @stable ICU 2.0
2589 */
2590 UnicodeString& toLower(void);
2591
2592 /**
2593 * Convert the characters in this to lower case following the conventions of
2594 * a specific locale.
2595 * @param locale The locale containing the conventions to use.
2596 * @return A reference to this.
2597 * @stable ICU 2.0
2598 */
2599 UnicodeString& toLower(const Locale& locale);
2600
2601 #if !UCONFIG_NO_BREAK_ITERATION
2602
2603 /**
2604 * Titlecase this string, convenience function using the default locale.
2605 *
2606 * Casing is locale-dependent and context-sensitive.
2607 * Titlecasing uses a break iterator to find the first characters of words
2608 * that are to be titlecased. It titlecases those characters and lowercases
2609 * all others.
2610 *
2611 * The titlecase break iterator can be provided to customize for arbitrary
2612 * styles, using rules and dictionaries beyond the standard iterators.
2613 * It may be more efficient to always provide an iterator to avoid
2614 * opening and closing one for each string.
2615 * The standard titlecase iterator for the root locale implements the
2616 * algorithm of Unicode TR 21.
2617 *
2618 * This function uses only the setText(), first() and next() methods of the
2619 * provided break iterator.
2620 *
2621 * @param titleIter A break iterator to find the first characters of words
2622 * that are to be titlecased.
2623 * If none is provided (0), then a standard titlecase
2624 * break iterator is opened.
2625 * Otherwise the provided iterator is set to the string's text.
2626 * @return A reference to this.
2627 * @stable ICU 2.1
2628 */
2629 UnicodeString &toTitle(BreakIterator *titleIter);
2630
2631 /**
2632 * Titlecase this string.
2633 *
2634 * Casing is locale-dependent and context-sensitive.
2635 * Titlecasing uses a break iterator to find the first characters of words
2636 * that are to be titlecased. It titlecases those characters and lowercases
2637 * all others.
2638 *
2639 * The titlecase break iterator can be provided to customize for arbitrary
2640 * styles, using rules and dictionaries beyond the standard iterators.
2641 * It may be more efficient to always provide an iterator to avoid
2642 * opening and closing one for each string.
2643 * The standard titlecase iterator for the root locale implements the
2644 * algorithm of Unicode TR 21.
2645 *
2646 * This function uses only the setText(), first() and next() methods of the
2647 * provided break iterator.
2648 *
2649 * @param titleIter A break iterator to find the first characters of words
2650 * that are to be titlecased.
2651 * If none is provided (0), then a standard titlecase
2652 * break iterator is opened.
2653 * Otherwise the provided iterator is set to the string's text.
2654 * @param locale The locale to consider.
2655 * @return A reference to this.
2656 * @stable ICU 2.1
2657 */
2658 UnicodeString &toTitle(BreakIterator *titleIter, const Locale &locale);
2659
2660 /**
2661 * Titlecase this string, with options.
2662 *
2663 * Casing is locale-dependent and context-sensitive.
2664 * Titlecasing uses a break iterator to find the first characters of words
2665 * that are to be titlecased. It titlecases those characters and lowercases
2666 * all others. (This can be modified with options.)
2667 *
2668 * The titlecase break iterator can be provided to customize for arbitrary
2669 * styles, using rules and dictionaries beyond the standard iterators.
2670 * It may be more efficient to always provide an iterator to avoid
2671 * opening and closing one for each string.
2672 * The standard titlecase iterator for the root locale implements the
2673 * algorithm of Unicode TR 21.
2674 *
2675 * This function uses only the setText(), first() and next() methods of the
2676 * provided break iterator.
2677 *
2678 * @param titleIter A break iterator to find the first characters of words
2679 * that are to be titlecased.
2680 * If none is provided (0), then a standard titlecase
2681 * break iterator is opened.
2682 * Otherwise the provided iterator is set to the string's text.
2683 * @param locale The locale to consider.
2684 * @param options Options bit set, see ucasemap_open().
2685 * @return A reference to this.
2686 * @see U_TITLECASE_NO_LOWERCASE
2687 * @see U_TITLECASE_NO_BREAK_ADJUSTMENT
2688 * @see ucasemap_open
2689 * @stable ICU 3.8
2690 */
2691 UnicodeString &toTitle(BreakIterator *titleIter, const Locale &locale, uint32_t options);
2692
2693 #endif
2694
2695 /**
2696 * Case-folds the characters in this string.
2697 *
2698 * Case-folding is locale-independent and not context-sensitive,
2699 * but there is an option for whether to include or exclude mappings for dotted I
2700 * and dotless i that are marked with 'T' in CaseFolding.txt.
2701 *
2702 * The result may be longer or shorter than the original.
2703 *
2704 * @param options Either U_FOLD_CASE_DEFAULT or U_FOLD_CASE_EXCLUDE_SPECIAL_I
2705 * @return A reference to this.
2706 * @stable ICU 2.0
2707 */
2708 UnicodeString &foldCase(uint32_t options=0 /*U_FOLD_CASE_DEFAULT*/);
2709
2710 //========================================
2711 // Access to the internal buffer
2712 //========================================
2713
2714 /**
2715 * Get a read/write pointer to the internal buffer.
2716 * The buffer is guaranteed to be large enough for at least minCapacity UChars,
2717 * writable, and is still owned by the UnicodeString object.
2718 * Calls to getBuffer(minCapacity) must not be nested, and
2719 * must be matched with calls to releaseBuffer(newLength).
2720 * If the string buffer was read-only or shared,
2721 * then it will be reallocated and copied.
2722 *
2723 * An attempted nested call will return 0, and will not further modify the
2724 * state of the UnicodeString object.
2725 * It also returns 0 if the string is bogus.
2726 *
2727 * The actual capacity of the string buffer may be larger than minCapacity.
2728 * getCapacity() returns the actual capacity.
2729 * For many operations, the full capacity should be used to avoid reallocations.
2730 *
2731 * While the buffer is "open" between getBuffer(minCapacity)
2732 * and releaseBuffer(newLength), the following applies:
2733 * - The string length is set to 0.
2734 * - Any read API call on the UnicodeString object will behave like on a 0-length string.
2735 * - Any write API call on the UnicodeString object is disallowed and will have no effect.
2736 * - You can read from and write to the returned buffer.
2737 * - The previous string contents will still be in the buffer;
2738 * if you want to use it, then you need to call length() before getBuffer(minCapacity).
2739 * If the length() was greater than minCapacity, then any contents after minCapacity
2740 * may be lost.
2741 * The buffer contents is not NUL-terminated by getBuffer().
2742 * If length()<getCapacity() then you can terminate it by writing a NUL
2743 * at index length().
2744 * - You must call releaseBuffer(newLength) before and in order to
2745 * return to normal UnicodeString operation.
2746 *
2747 * @param minCapacity the minimum number of UChars that are to be available
2748 * in the buffer, starting at the returned pointer;
2749 * default to the current string capacity if minCapacity==-1
2750 * @return a writable pointer to the internal string buffer,
2751 * or 0 if an error occurs (nested calls, out of memory)
2752 *
2753 * @see releaseBuffer
2754 * @see getTerminatedBuffer()
2755 * @stable ICU 2.0
2756 */
2757 UChar *getBuffer(int32_t minCapacity);
2758
2759 /**
2760 * Release a read/write buffer on a UnicodeString object with an
2761 * "open" getBuffer(minCapacity).
2762 * This function must be called in a matched pair with getBuffer(minCapacity).
2763 * releaseBuffer(newLength) must be called if and only if a getBuffer(minCapacity) is "open".
2764 *
2765 * It will set the string length to newLength, at most to the current capacity.
2766 * If newLength==-1 then it will set the length according to the
2767 * first NUL in the buffer, or to the capacity if there is no NUL.
2768 *
2769 * After calling releaseBuffer(newLength) the UnicodeString is back to normal operation.
2770 *
2771 * @param newLength the new length of the UnicodeString object;
2772 * defaults to the current capacity if newLength is greater than that;
2773 * if newLength==-1, it defaults to u_strlen(buffer) but not more than
2774 * the current capacity of the string
2775 *
2776 * @see getBuffer(int32_t minCapacity)
2777 * @stable ICU 2.0
2778 */
2779 void releaseBuffer(int32_t newLength=-1);
2780
2781 /**
2782 * Get a read-only pointer to the internal buffer.
2783 * This can be called at any time on a valid UnicodeString.
2784 *
2785 * It returns 0 if the string is bogus, or
2786 * during an "open" getBuffer(minCapacity).
2787 *
2788 * It can be called as many times as desired.
2789 * The pointer that it returns will remain valid until the UnicodeString object is modified,
2790 * at which time the pointer is semantically invalidated and must not be used any more.
2791 *
2792 * The capacity of the buffer can be determined with getCapacity().
2793 * The part after length() may or may not be initialized and valid,
2794 * depending on the history of the UnicodeString object.
2795 *
2796 * The buffer contents is (probably) not NUL-terminated.
2797 * You can check if it is with
2798 * <code>(s.length()<s.getCapacity() && buffer[s.length()]==0)</code>.
2799 * (See getTerminatedBuffer().)
2800 *
2801 * The buffer may reside in read-only memory. Its contents must not
2802 * be modified.
2803 *
2804 * @return a read-only pointer to the internal string buffer,
2805 * or 0 if the string is empty or bogus
2806 *
2807 * @see getBuffer(int32_t minCapacity)
2808 * @see getTerminatedBuffer()
2809 * @stable ICU 2.0
2810 */
2811 inline const UChar *getBuffer() const;
2812
2813 /**
2814 * Get a read-only pointer to the internal buffer,
2815 * making sure that it is NUL-terminated.
2816 * This can be called at any time on a valid UnicodeString.
2817 *
2818 * It returns 0 if the string is bogus, or
2819 * during an "open" getBuffer(minCapacity), or if the buffer cannot
2820 * be NUL-terminated (because memory allocation failed).
2821 *
2822 * It can be called as many times as desired.
2823 * The pointer that it returns will remain valid until the UnicodeString object is modified,
2824 * at which time the pointer is semantically invalidated and must not be used any more.
2825 *
2826 * The capacity of the buffer can be determined with getCapacity().
2827 * The part after length()+1 may or may not be initialized and valid,
2828 * depending on the history of the UnicodeString object.
2829 *
2830 * The buffer contents is guaranteed to be NUL-terminated.
2831 * getTerminatedBuffer() may reallocate the buffer if a terminating NUL
2832 * is written.
2833 * For this reason, this function is not const, unlike getBuffer().
2834 * Note that a UnicodeString may also contain NUL characters as part of its contents.
2835 *
2836 * The buffer may reside in read-only memory. Its contents must not
2837 * be modified.
2838 *
2839 * @return a read-only pointer to the internal string buffer,
2840 * or 0 if the string is empty or bogus
2841 *
2842 * @see getBuffer(int32_t minCapacity)
2843 * @see getBuffer()
2844 * @stable ICU 2.2
2845 */
2846 inline const UChar *getTerminatedBuffer();
2847
2848 //========================================
2849 // Constructors
2850 //========================================
2851
2852 /** Construct an empty UnicodeString.
2853 * @stable ICU 2.0
2854 */
2855 UnicodeString();
2856
2857 /**
2858 * Construct a UnicodeString with capacity to hold <TT>capacity</TT> UChars
2859 * @param capacity the number of UChars this UnicodeString should hold
2860 * before a resize is necessary; if count is greater than 0 and count
2861 * code points c take up more space than capacity, then capacity is adjusted
2862 * accordingly.
2863 * @param c is used to initially fill the string
2864 * @param count specifies how many code points c are to be written in the
2865 * string
2866 * @stable ICU 2.0
2867 */
2868 UnicodeString(int32_t capacity, UChar32 c, int32_t count);
2869
2870 /**
2871 * Single UChar (code unit) constructor.
2872 *
2873 * It is recommended to mark this constructor "explicit" by
2874 * <code>-DUNISTR_FROM_CHAR_EXPLICIT=explicit</code>
2875 * on the compiler command line or similar.
2876 * @param ch the character to place in the UnicodeString
2877 * @stable ICU 2.0
2878 */
2879 UNISTR_FROM_CHAR_EXPLICIT UnicodeString(UChar ch);
2880
2881 /**
2882 * Single UChar32 (code point) constructor.
2883 *
2884 * It is recommended to mark this constructor "explicit" by
2885 * <code>-DUNISTR_FROM_CHAR_EXPLICIT=explicit</code>
2886 * on the compiler command line or similar.
2887 * @param ch the character to place in the UnicodeString
2888 * @stable ICU 2.0
2889 */
2890 UNISTR_FROM_CHAR_EXPLICIT UnicodeString(UChar32 ch);
2891
2892 /**
2893 * UChar* constructor.
2894 *
2895 * It is recommended to mark this constructor "explicit" by
2896 * <code>-DUNISTR_FROM_STRING_EXPLICIT=explicit</code>
2897 * on the compiler command line or similar.
2898 * @param text The characters to place in the UnicodeString. <TT>text</TT>
2899 * must be NULL (U+0000) terminated.
2900 * @stable ICU 2.0
2901 */
2902 UNISTR_FROM_STRING_EXPLICIT UnicodeString(const UChar *text);
2903
2904 /**
2905 * UChar* constructor.
2906 * @param text The characters to place in the UnicodeString.
2907 * @param textLength The number of Unicode characters in <TT>text</TT>
2908 * to copy.
2909 * @stable ICU 2.0
2910 */
2911 UnicodeString(const UChar *text,
2912 int32_t textLength);
2913
2914 /**
2915 * Readonly-aliasing UChar* constructor.
2916 * The text will be used for the UnicodeString object, but
2917 * it will not be released when the UnicodeString is destroyed.
2918 * This has copy-on-write semantics:
2919 * When the string is modified, then the buffer is first copied into
2920 * newly allocated memory.
2921 * The aliased buffer is never modified.
2922 *
2923 * In an assignment to another UnicodeString, when using the copy constructor
2924 * or the assignment operator, the text will be copied.
2925 * When using fastCopyFrom(), the text will be aliased again,
2926 * so that both strings then alias the same readonly-text.
2927 *
2928 * @param isTerminated specifies if <code>text</code> is <code>NUL</code>-terminated.
2929 * This must be true if <code>textLength==-1</code>.
2930 * @param text The characters to alias for the UnicodeString.
2931 * @param textLength The number of Unicode characters in <code>text</code> to alias.
2932 * If -1, then this constructor will determine the length
2933 * by calling <code>u_strlen()</code>.
2934 * @stable ICU 2.0
2935 */
2936 UnicodeString(UBool isTerminated,
2937 const UChar *text,
2938 int32_t textLength);
2939
2940 /**
2941 * Writable-aliasing UChar* constructor.
2942 * The text will be used for the UnicodeString object, but
2943 * it will not be released when the UnicodeString is destroyed.
2944 * This has write-through semantics:
2945 * For as long as the capacity of the buffer is sufficient, write operations
2946 * will directly affect the buffer. When more capacity is necessary, then
2947 * a new buffer will be allocated and the contents copied as with regularly
2948 * constructed strings.
2949 * In an assignment to another UnicodeString, the buffer will be copied.
2950 * The extract(UChar *dst) function detects whether the dst pointer is the same
2951 * as the string buffer itself and will in this case not copy the contents.
2952 *
2953 * @param buffer The characters to alias for the UnicodeString.
2954 * @param buffLength The number of Unicode characters in <code>buffer</code> to alias.
2955 * @param buffCapacity The size of <code>buffer</code> in UChars.
2956 * @stable ICU 2.0
2957 */
2958 UnicodeString(UChar *buffer, int32_t buffLength, int32_t buffCapacity);
2959
2960 #if U_CHARSET_IS_UTF8 || !UCONFIG_NO_CONVERSION
2961
2962 /**
2963 * char* constructor.
2964 * Uses the default converter (and thus depends on the ICU conversion code)
2965 * unless U_CHARSET_IS_UTF8 is set to 1.
2966 *
2967 * For ASCII (really "invariant character") strings it is more efficient to use
2968 * the constructor that takes a US_INV (for its enum EInvariant).
2969 * For ASCII (invariant-character) string literals, see UNICODE_STRING and
2970 * UNICODE_STRING_SIMPLE.
2971 *
2972 * It is recommended to mark this constructor "explicit" by
2973 * <code>-DUNISTR_FROM_STRING_EXPLICIT=explicit</code>
2974 * on the compiler command line or similar.
2975 * @param codepageData an array of bytes, null-terminated,
2976 * in the platform's default codepage.
2977 * @stable ICU 2.0
2978 * @see UNICODE_STRING
2979 * @see UNICODE_STRING_SIMPLE
2980 */
2981 UNISTR_FROM_STRING_EXPLICIT UnicodeString(const char *codepageData);
2982
2983 /**
2984 * char* constructor.
2985 * Uses the default converter (and thus depends on the ICU conversion code)
2986 * unless U_CHARSET_IS_UTF8 is set to 1.
2987 * @param codepageData an array of bytes in the platform's default codepage.
2988 * @param dataLength The number of bytes in <TT>codepageData</TT>.
2989 * @stable ICU 2.0
2990 */
2991 UnicodeString(const char *codepageData, int32_t dataLength);
2992
2993 #endif
2994
2995 #if !UCONFIG_NO_CONVERSION
2996
2997 /**
2998 * char* constructor.
2999 * @param codepageData an array of bytes, null-terminated
3000 * @param codepage the encoding of <TT>codepageData</TT>. The special
3001 * value 0 for <TT>codepage</TT> indicates that the text is in the
3002 * platform's default codepage.
3003 *
3004 * If <code>codepage</code> is an empty string (<code>""</code>),
3005 * then a simple conversion is performed on the codepage-invariant
3006 * subset ("invariant characters") of the platform encoding. See utypes.h.
3007 * Recommendation: For invariant-character strings use the constructor
3008 * UnicodeString(const char *src, int32_t length, enum EInvariant inv)
3009 * because it avoids object code dependencies of UnicodeString on
3010 * the conversion code.
3011 *
3012 * @stable ICU 2.0
3013 */
3014 UnicodeString(const char *codepageData, const char *codepage);
3015
3016 /**
3017 * char* constructor.
3018 * @param codepageData an array of bytes.
3019 * @param dataLength The number of bytes in <TT>codepageData</TT>.
3020 * @param codepage the encoding of <TT>codepageData</TT>. The special
3021 * value 0 for <TT>codepage</TT> indicates that the text is in the
3022 * platform's default codepage.
3023 * If <code>codepage</code> is an empty string (<code>""</code>),
3024 * then a simple conversion is performed on the codepage-invariant
3025 * subset ("invariant characters") of the platform encoding. See utypes.h.
3026 * Recommendation: For invariant-character strings use the constructor
3027 * UnicodeString(const char *src, int32_t length, enum EInvariant inv)
3028 * because it avoids object code dependencies of UnicodeString on
3029 * the conversion code.
3030 *
3031 * @stable ICU 2.0
3032 */
3033 UnicodeString(const char *codepageData, int32_t dataLength, const char *codepage);
3034
3035 /**
3036 * char * / UConverter constructor.
3037 * This constructor uses an existing UConverter object to
3038 * convert the codepage string to Unicode and construct a UnicodeString
3039 * from that.
3040 *
3041 * The converter is reset at first.
3042 * If the error code indicates a failure before this constructor is called,
3043 * or if an error occurs during conversion or construction,
3044 * then the string will be bogus.
3045 *
3046 * This function avoids the overhead of opening and closing a converter if
3047 * multiple strings are constructed.
3048 *
3049 * @param src input codepage string
3050 * @param srcLength length of the input string, can be -1 for NUL-terminated strings
3051 * @param cnv converter object (ucnv_resetToUnicode() will be called),
3052 * can be NULL for the default converter
3053 * @param errorCode normal ICU error code
3054 * @stable ICU 2.0
3055 */
3056 UnicodeString(
3057 const char *src, int32_t srcLength,
3058 UConverter *cnv,
3059 UErrorCode &errorCode);
3060
3061 #endif
3062
3063 /**
3064 * Constructs a Unicode string from an invariant-character char * string.
3065 * About invariant characters see utypes.h.
3066 * This constructor has no runtime dependency on conversion code and is
3067 * therefore recommended over ones taking a charset name string
3068 * (where the empty string "" indicates invariant-character conversion).
3069 *
3070 * Use the macro US_INV as the third, signature-distinguishing parameter.
3071 *
3072 * For example:
3073 * \code
3074 * void fn(const char *s) {
3075 * UnicodeString ustr(s, -1, US_INV);
3076 * // use ustr ...
3077 * }
3078 * \endcode
3079 *
3080 * @param src String using only invariant characters.
3081 * @param length Length of src, or -1 if NUL-terminated.
3082 * @param inv Signature-distinguishing paramater, use US_INV.
3083 *
3084 * @see US_INV
3085 * @stable ICU 3.2
3086 */
3087 UnicodeString(const char *src, int32_t length, enum EInvariant inv);
3088
3089
3090 /**
3091 * Copy constructor.
3092 * @param that The UnicodeString object to copy.
3093 * @stable ICU 2.0
3094 */
3095 UnicodeString(const UnicodeString& that);
3096
3097 /**
3098 * 'Substring' constructor from tail of source string.
3099 * @param src The UnicodeString object to copy.
3100 * @param srcStart The offset into <tt>src</tt> at which to start copying.
3101 * @stable ICU 2.2
3102 */
3103 UnicodeString(const UnicodeString& src, int32_t srcStart);
3104
3105 /**
3106 * 'Substring' constructor from subrange of source string.
3107 * @param src The UnicodeString object to copy.
3108 * @param srcStart The offset into <tt>src</tt> at which to start copying.
3109 * @param srcLength The number of characters from <tt>src</tt> to copy.
3110 * @stable ICU 2.2
3111 */
3112 UnicodeString(const UnicodeString& src, int32_t srcStart, int32_t srcLength);
3113
3114 /**
3115 * Clone this object, an instance of a subclass of Replaceable.
3116 * Clones can be used concurrently in multiple threads.
3117 * If a subclass does not implement clone(), or if an error occurs,
3118 * then NULL is returned.
3119 * The clone functions in all subclasses return a pointer to a Replaceable
3120 * because some compilers do not support covariant (same-as-this)
3121 * return types; cast to the appropriate subclass if necessary.
3122 * The caller must delete the clone.
3123 *
3124 * @return a clone of this object
3125 *
3126 * @see Replaceable::clone
3127 * @see getDynamicClassID
3128 * @stable ICU 2.6
3129 */
3130 virtual Replaceable *clone() const;
3131
3132 /** Destructor.
3133 * @stable ICU 2.0
3134 */
3135 virtual ~UnicodeString();
3136
3137 /**
3138 * Create a UnicodeString from a UTF-8 string.
3139 * Illegal input is replaced with U+FFFD. Otherwise, errors result in a bogus string.
3140 * Calls u_strFromUTF8WithSub().
3141 *
3142 * @param utf8 UTF-8 input string.
3143 * Note that a StringPiece can be implicitly constructed
3144 * from a std::string or a NUL-terminated const char * string.
3145 * @return A UnicodeString with equivalent UTF-16 contents.
3146 * @see toUTF8
3147 * @see toUTF8String
3148 * @stable ICU 4.2
3149 */
3150 static UnicodeString fromUTF8(const StringPiece &utf8);
3151
3152 /**
3153 * Create a UnicodeString from a UTF-32 string.
3154 * Illegal input is replaced with U+FFFD. Otherwise, errors result in a bogus string.
3155 * Calls u_strFromUTF32WithSub().
3156 *
3157 * @param utf32 UTF-32 input string. Must not be NULL.
3158 * @param length Length of the input string, or -1 if NUL-terminated.
3159 * @return A UnicodeString with equivalent UTF-16 contents.
3160 * @see toUTF32
3161 * @stable ICU 4.2
3162 */
3163 static UnicodeString fromUTF32(const UChar32 *utf32, int32_t length);
3164
3165 /* Miscellaneous operations */
3166
3167 /**
3168 * Unescape a string of characters and return a string containing
3169 * the result. The following escape sequences are recognized:
3170 *
3171 * \\uhhhh 4 hex digits; h in [0-9A-Fa-f]
3172 * \\Uhhhhhhhh 8 hex digits
3173 * \\xhh 1-2 hex digits
3174 * \\ooo 1-3 octal digits; o in [0-7]
3175 * \\cX control-X; X is masked with 0x1F
3176 *
3177 * as well as the standard ANSI C escapes:
3178 *
3179 * \\a => U+0007, \\b => U+0008, \\t => U+0009, \\n => U+000A,
3180 * \\v => U+000B, \\f => U+000C, \\r => U+000D, \\e => U+001B,
3181 * \\" => U+0022, \\' => U+0027, \\? => U+003F, \\\\ => U+005C
3182 *
3183 * Anything else following a backslash is generically escaped. For
3184 * example, "[a\\-z]" returns "[a-z]".
3185 *
3186 * If an escape sequence is ill-formed, this method returns an empty
3187 * string. An example of an ill-formed sequence is "\\u" followed by
3188 * fewer than 4 hex digits.
3189 *
3190 * This function is similar to u_unescape() but not identical to it.
3191 * The latter takes a source char*, so it does escape recognition
3192 * and also invariant conversion.
3193 *
3194 * @return a string with backslash escapes interpreted, or an
3195 * empty string on error.
3196 * @see UnicodeString#unescapeAt()
3197 * @see u_unescape()
3198 * @see u_unescapeAt()
3199 * @stable ICU 2.0
3200 */
3201 UnicodeString unescape() const;
3202
3203 /**
3204 * Unescape a single escape sequence and return the represented
3205 * character. See unescape() for a listing of the recognized escape
3206 * sequences. The character at offset-1 is assumed (without
3207 * checking) to be a backslash. If the escape sequence is
3208 * ill-formed, or the offset is out of range, (UChar32)0xFFFFFFFF is
3209 * returned.
3210 *
3211 * @param offset an input output parameter. On input, it is the
3212 * offset into this string where the escape sequence is located,
3213 * after the initial backslash. On output, it is advanced after the
3214 * last character parsed. On error, it is not advanced at all.
3215 * @return the character represented by the escape sequence at
3216 * offset, or (UChar32)0xFFFFFFFF on error.
3217 * @see UnicodeString#unescape()
3218 * @see u_unescape()
3219 * @see u_unescapeAt()
3220 * @stable ICU 2.0
3221 */
3222 UChar32 unescapeAt(int32_t &offset) const;
3223
3224 /**
3225 * ICU "poor man's RTTI", returns a UClassID for this class.
3226 *
3227 * @stable ICU 2.2
3228 */
3229 static UClassID U_EXPORT2 getStaticClassID();
3230
3231 /**
3232 * ICU "poor man's RTTI", returns a UClassID for the actual class.
3233 *
3234 * @stable ICU 2.2
3235 */
3236 virtual UClassID getDynamicClassID() const;
3237
3238 //========================================
3239 // Implementation methods
3240 //========================================
3241
3242 protected:
3243 /**
3244 * Implement Replaceable::getLength() (see jitterbug 1027).
3245 * @stable ICU 2.4
3246 */
3247 virtual int32_t getLength() const;
3248
3249 /**
3250 * The change in Replaceable to use virtual getCharAt() allows
3251 * UnicodeString::charAt() to be inline again (see jitterbug 709).
3252 * @stable ICU 2.4
3253 */
3254 virtual UChar getCharAt(int32_t offset) const;
3255
3256 /**
3257 * The change in Replaceable to use virtual getChar32At() allows
3258 * UnicodeString::char32At() to be inline again (see jitterbug 709).
3259 * @stable ICU 2.4
3260 */
3261 virtual UChar32 getChar32At(int32_t offset) const;
3262
3263 private:
3264 // For char* constructors. Could be made public.
3265 UnicodeString &setToUTF8(const StringPiece &utf8);
3266 // For extract(char*).
3267 // We could make a toUTF8(target, capacity, errorCode) public but not
3268 // this version: New API will be cleaner if we make callers create substrings
3269 // rather than having start+length on every method,
3270 // and it should take a UErrorCode&.
3271 int32_t
3272 toUTF8(int32_t start, int32_t len,
3273 char *target, int32_t capacity) const;
3274
3275 /**
3276 * Internal string contents comparison, called by operator==.
3277 * Requires: this & text not bogus and have same lengths.
3278 */
3279 UBool doEquals(const UnicodeString &text, int32_t len) const;
3280
3281 inline int8_t
3282 doCompare(int32_t start,
3283 int32_t length,
3284 const UnicodeString& srcText,
3285 int32_t srcStart,
3286 int32_t srcLength) const;
3287
3288 int8_t doCompare(int32_t start,
3289 int32_t length,
3290 const UChar *srcChars,
3291 int32_t srcStart,
3292 int32_t srcLength) const;
3293
3294 inline int8_t
3295 doCompareCodePointOrder(int32_t start,
3296 int32_t length,
3297 const UnicodeString& srcText,
3298 int32_t srcStart,
3299 int32_t srcLength) const;
3300
3301 int8_t doCompareCodePointOrder(int32_t start,
3302 int32_t length,
3303 const UChar *srcChars,
3304 int32_t srcStart,
3305 int32_t srcLength) const;
3306
3307 inline int8_t
3308 doCaseCompare(int32_t start,
3309 int32_t length,
3310 const UnicodeString &srcText,
3311 int32_t srcStart,
3312 int32_t srcLength,
3313 uint32_t options) const;
3314
3315 int8_t
3316 doCaseCompare(int32_t start,
3317 int32_t length,
3318 const UChar *srcChars,
3319 int32_t srcStart,
3320 int32_t srcLength,
3321 uint32_t options) const;
3322
3323 int32_t doIndexOf(UChar c,
3324 int32_t start,
3325 int32_t length) const;
3326
3327 int32_t doIndexOf(UChar32 c,
3328 int32_t start,
3329 int32_t length) const;
3330
3331 int32_t doLastIndexOf(UChar c,
3332 int32_t start,
3333 int32_t length) const;
3334
3335 int32_t doLastIndexOf(UChar32 c,
3336 int32_t start,
3337 int32_t length) const;
3338
3339 void doExtract(int32_t start,
3340 int32_t length,
3341 UChar *dst,
3342 int32_t dstStart) const;
3343
3344 inline void doExtract(int32_t start,
3345 int32_t length,
3346 UnicodeString& target) const;
3347
3348 inline UChar doCharAt(int32_t offset) const;
3349
3350 UnicodeString& doReplace(int32_t start,
3351 int32_t length,
3352 const UnicodeString& srcText,
3353 int32_t srcStart,
3354 int32_t srcLength);
3355
3356 UnicodeString& doReplace(int32_t start,
3357 int32_t length,
3358 const UChar *srcChars,
3359 int32_t srcStart,
3360 int32_t srcLength);
3361
3362 UnicodeString& doReverse(int32_t start,
3363 int32_t length);
3364
3365 // calculate hash code
3366 int32_t doHashCode(void) const;
3367
3368 // get pointer to start of array
3369 // these do not check for kOpenGetBuffer, unlike the public getBuffer() function
3370 inline UChar* getArrayStart(void);
3371 inline const UChar* getArrayStart(void) const;
3372
3373 // A UnicodeString object (not necessarily its current buffer)
3374 // is writable unless it isBogus() or it has an "open" getBuffer(minCapacity).
3375 inline UBool isWritable() const;
3376
3377 // Is the current buffer writable?
3378 inline UBool isBufferWritable() const;
3379
3380 // None of the following does releaseArray().
3381 inline void setLength(int32_t len); // sets only fShortLength and fLength
3382 inline void setToEmpty(); // sets fFlags=kShortString
3383 inline void setArray(UChar *array, int32_t len, int32_t capacity); // does not set fFlags
3384
3385 // allocate the array; result may be fStackBuffer
3386 // sets refCount to 1 if appropriate
3387 // sets fArray, fCapacity, and fFlags
3388 // returns boolean for success or failure
3389 UBool allocate(int32_t capacity);
3390
3391 // release the array if owned
3392 void releaseArray(void);
3393
3394 // turn a bogus string into an empty one
3395 void unBogus();
3396
3397 // implements assigment operator, copy constructor, and fastCopyFrom()
3398 UnicodeString ©From(const UnicodeString &src, UBool fastCopy=FALSE);
3399
3400 // Pin start and limit to acceptable values.
3401 inline void pinIndex(int32_t& start) const;
3402 inline void pinIndices(int32_t& start,
3403 int32_t& length) const;
3404
3405 #if !UCONFIG_NO_CONVERSION
3406
3407 /* Internal extract() using UConverter. */
3408 int32_t doExtract(int32_t start, int32_t length,
3409 char *dest, int32_t destCapacity,
3410 UConverter *cnv,
3411 UErrorCode &errorCode) const;
3412
3413 /*
3414 * Real constructor for converting from codepage data.
3415 * It assumes that it is called with !fRefCounted.
3416 *
3417 * If <code>codepage==0</code>, then the default converter
3418 * is used for the platform encoding.
3419 * If <code>codepage</code> is an empty string (<code>""</code>),
3420 * then a simple conversion is performed on the codepage-invariant
3421 * subset ("invariant characters") of the platform encoding. See utypes.h.
3422 */
3423 void doCodepageCreate(const char *codepageData,
3424 int32_t dataLength,
3425 const char *codepage);
3426
3427 /*
3428 * Worker function for creating a UnicodeString from
3429 * a codepage string using a UConverter.
3430 */
3431 void
3432 doCodepageCreate(const char *codepageData,
3433 int32_t dataLength,
3434 UConverter *converter,
3435 UErrorCode &status);
3436
3437 #endif
3438
3439 /*
3440 * This function is called when write access to the array
3441 * is necessary.
3442 *
3443 * We need to make a copy of the array if
3444 * the buffer is read-only, or
3445 * the buffer is refCounted (shared), and refCount>1, or
3446 * the buffer is too small.
3447 *
3448 * Return FALSE if memory could not be allocated.
3449 */
3450 UBool cloneArrayIfNeeded(int32_t newCapacity = -1,
3451 int32_t growCapacity = -1,
3452 UBool doCopyArray = TRUE,
3453 int32_t **pBufferToDelete = 0,
3454 UBool forceClone = FALSE);
3455
3456 /**
3457 * Common function for UnicodeString case mappings.
3458 * The stringCaseMapper has the same type UStringCaseMapper
3459 * as in ustr_imp.h for ustrcase_map().
3460 */
3461 UnicodeString &
3462 caseMap(const UCaseMap *csm, UStringCaseMapper *stringCaseMapper);
3463
3464 // ref counting
3465 void addRef(void);
3466 int32_t removeRef(void);
3467 int32_t refCount(void) const;
3468
3469 // constants
3470 enum {
3471 // Set the stack buffer size so that sizeof(UnicodeString) is,
3472 // naturally (without padding), a multiple of sizeof(pointer).
3473 US_STACKBUF_SIZE= sizeof(void *)==4 ? 13 : 15, // Size of stack buffer for short strings
3474 kInvalidUChar=0xffff, // invalid UChar index
3475 kGrowSize=128, // grow size for this buffer
3476 kInvalidHashCode=0, // invalid hash code
3477 kEmptyHashCode=1, // hash code for empty string
3478
3479 // bit flag values for fFlags
3480 kIsBogus=1, // this string is bogus, i.e., not valid or NULL
3481 kUsingStackBuffer=2,// using fUnion.fStackBuffer instead of fUnion.fFields
3482 kRefCounted=4, // there is a refCount field before the characters in fArray
3483 kBufferIsReadonly=8,// do not write to this buffer
3484 kOpenGetBuffer=16, // getBuffer(minCapacity) was called (is "open"),
3485 // and releaseBuffer(newLength) must be called
3486
3487 // combined values for convenience
3488 kShortString=kUsingStackBuffer,
3489 kLongString=kRefCounted,
3490 kReadonlyAlias=kBufferIsReadonly,
3491 kWritableAlias=0
3492 };
3493
3494 friend class StringThreadTest;
3495 friend class UnicodeStringAppendable;
3496
3497 union StackBufferOrFields; // forward declaration necessary before friend declaration
3498 friend union StackBufferOrFields; // make US_STACKBUF_SIZE visible inside fUnion
3499
3500 /*
3501 * The following are all the class fields that are stored
3502 * in each UnicodeString object.
3503 * Note that UnicodeString has virtual functions,
3504 * therefore there is an implicit vtable pointer
3505 * as the first real field.
3506 * The fields should be aligned such that no padding is necessary.
3507 * On 32-bit machines, the size should be 32 bytes,
3508 * on 64-bit machines (8-byte pointers), it should be 40 bytes.
3509 *
3510 * We use a hack to achieve this.
3511 *
3512 * With at least some compilers, each of the following is forced to
3513 * a multiple of sizeof(pointer) [the largest field base unit here is a data pointer],
3514 * rounded up with additional padding if the fields do not already fit that requirement:
3515 * - sizeof(class UnicodeString)
3516 * - offsetof(UnicodeString, fUnion)
3517 * - sizeof(fUnion)
3518 * - sizeof(fFields)
3519 *
3520 * In order to avoid padding, we make sizeof(fStackBuffer)=16 (=8 UChars)
3521 * which is at least as large as sizeof(fFields) on 32-bit and 64-bit machines.
3522 * (Padding at the end of fFields is ok:
3523 * As long as there is no padding after fStackBuffer, it is not wasted space.)
3524 *
3525 * We further assume that the compiler does not reorder the fields,
3526 * so that fRestOfStackBuffer (which holds a few more UChars) immediately follows after fUnion,
3527 * with at most some padding (but no other field) in between.
3528 * (Padding there would be wasted space, but functionally harmless.)
3529 *
3530 * We use a few more sizeof(pointer)'s chunks of space with
3531 * fRestOfStackBuffer, fShortLength and fFlags,
3532 * to get up exactly to the intended sizeof(UnicodeString).
3533 */
3534 // (implicit) *vtable;
3535 union StackBufferOrFields {
3536 // fStackBuffer is used iff (fFlags&kUsingStackBuffer)
3537 // else fFields is used
3538 UChar fStackBuffer[8]; // buffer for short strings, together with fRestOfStackBuffer
3539 struct {
3540 UChar *fArray; // the Unicode data
3541 int32_t fCapacity; // capacity of fArray (in UChars)
3542 int32_t fLength; // number of characters in fArray if >127; else undefined
3543 } fFields;
3544 } fUnion;
3545 UChar fRestOfStackBuffer[US_STACKBUF_SIZE-8];
3546 int8_t fShortLength; // 0..127: length <0: real length is in fUnion.fFields.fLength
3547 uint8_t fFlags; // bit flags: see constants above
3548 };
3549
3550 /**
3551 * Create a new UnicodeString with the concatenation of two others.
3552 *
3553 * @param s1 The first string to be copied to the new one.
3554 * @param s2 The second string to be copied to the new one, after s1.
3555 * @return UnicodeString(s1).append(s2)
3556 * @stable ICU 2.8
3557 */
3558 U_COMMON_API UnicodeString U_EXPORT2
3559 operator+ (const UnicodeString &s1, const UnicodeString &s2);
3560
3561 //========================================
3562 // Inline members
3563 //========================================
3564
3565 //========================================
3566 // Privates
3567 //========================================
3568
3569 inline void
pinIndex(int32_t & start)3570 UnicodeString::pinIndex(int32_t& start) const
3571 {
3572 // pin index
3573 if(start < 0) {
3574 start = 0;
3575 } else if(start > length()) {
3576 start = length();
3577 }
3578 }
3579
3580 inline void
pinIndices(int32_t & start,int32_t & _length)3581 UnicodeString::pinIndices(int32_t& start,
3582 int32_t& _length) const
3583 {
3584 // pin indices
3585 int32_t len = length();
3586 if(start < 0) {
3587 start = 0;
3588 } else if(start > len) {
3589 start = len;
3590 }
3591 if(_length < 0) {
3592 _length = 0;
3593 } else if(_length > (len - start)) {
3594 _length = (len - start);
3595 }
3596 }
3597
3598 inline UChar*
getArrayStart()3599 UnicodeString::getArrayStart()
3600 { return (fFlags&kUsingStackBuffer) ? fUnion.fStackBuffer : fUnion.fFields.fArray; }
3601
3602 inline const UChar*
getArrayStart()3603 UnicodeString::getArrayStart() const
3604 { return (fFlags&kUsingStackBuffer) ? fUnion.fStackBuffer : fUnion.fFields.fArray; }
3605
3606 //========================================
3607 // Read-only implementation methods
3608 //========================================
3609 inline int32_t
length()3610 UnicodeString::length() const
3611 { return fShortLength>=0 ? fShortLength : fUnion.fFields.fLength; }
3612
3613 inline int32_t
getCapacity()3614 UnicodeString::getCapacity() const
3615 { return (fFlags&kUsingStackBuffer) ? US_STACKBUF_SIZE : fUnion.fFields.fCapacity; }
3616
3617 inline int32_t
hashCode()3618 UnicodeString::hashCode() const
3619 { return doHashCode(); }
3620
3621 inline UBool
isBogus()3622 UnicodeString::isBogus() const
3623 { return (UBool)(fFlags & kIsBogus); }
3624
3625 inline UBool
isWritable()3626 UnicodeString::isWritable() const
3627 { return (UBool)!(fFlags&(kOpenGetBuffer|kIsBogus)); }
3628
3629 inline UBool
isBufferWritable()3630 UnicodeString::isBufferWritable() const
3631 {
3632 return (UBool)(
3633 !(fFlags&(kOpenGetBuffer|kIsBogus|kBufferIsReadonly)) &&
3634 (!(fFlags&kRefCounted) || refCount()==1));
3635 }
3636
3637 inline const UChar *
getBuffer()3638 UnicodeString::getBuffer() const {
3639 if(fFlags&(kIsBogus|kOpenGetBuffer)) {
3640 return 0;
3641 } else if(fFlags&kUsingStackBuffer) {
3642 return fUnion.fStackBuffer;
3643 } else {
3644 return fUnion.fFields.fArray;
3645 }
3646 }
3647
3648 //========================================
3649 // Read-only alias methods
3650 //========================================
3651 inline int8_t
doCompare(int32_t start,int32_t thisLength,const UnicodeString & srcText,int32_t srcStart,int32_t srcLength)3652 UnicodeString::doCompare(int32_t start,
3653 int32_t thisLength,
3654 const UnicodeString& srcText,
3655 int32_t srcStart,
3656 int32_t srcLength) const
3657 {
3658 if(srcText.isBogus()) {
3659 return (int8_t)!isBogus(); // 0 if both are bogus, 1 otherwise
3660 } else {
3661 srcText.pinIndices(srcStart, srcLength);
3662 return doCompare(start, thisLength, srcText.getArrayStart(), srcStart, srcLength);
3663 }
3664 }
3665
3666 inline UBool
3667 UnicodeString::operator== (const UnicodeString& text) const
3668 {
3669 if(isBogus()) {
3670 return text.isBogus();
3671 } else {
3672 int32_t len = length(), textLength = text.length();
3673 return !text.isBogus() && len == textLength && doEquals(text, len);
3674 }
3675 }
3676
3677 inline UBool
3678 UnicodeString::operator!= (const UnicodeString& text) const
3679 { return (! operator==(text)); }
3680
3681 inline UBool
3682 UnicodeString::operator> (const UnicodeString& text) const
3683 { return doCompare(0, length(), text, 0, text.length()) == 1; }
3684
3685 inline UBool
3686 UnicodeString::operator< (const UnicodeString& text) const
3687 { return doCompare(0, length(), text, 0, text.length()) == -1; }
3688
3689 inline UBool
3690 UnicodeString::operator>= (const UnicodeString& text) const
3691 { return doCompare(0, length(), text, 0, text.length()) != -1; }
3692
3693 inline UBool
3694 UnicodeString::operator<= (const UnicodeString& text) const
3695 { return doCompare(0, length(), text, 0, text.length()) != 1; }
3696
3697 inline int8_t
compare(const UnicodeString & text)3698 UnicodeString::compare(const UnicodeString& text) const
3699 { return doCompare(0, length(), text, 0, text.length()); }
3700
3701 inline int8_t
compare(int32_t start,int32_t _length,const UnicodeString & srcText)3702 UnicodeString::compare(int32_t start,
3703 int32_t _length,
3704 const UnicodeString& srcText) const
3705 { return doCompare(start, _length, srcText, 0, srcText.length()); }
3706
3707 inline int8_t
compare(const UChar * srcChars,int32_t srcLength)3708 UnicodeString::compare(const UChar *srcChars,
3709 int32_t srcLength) const
3710 { return doCompare(0, length(), srcChars, 0, srcLength); }
3711
3712 inline int8_t
compare(int32_t start,int32_t _length,const UnicodeString & srcText,int32_t srcStart,int32_t srcLength)3713 UnicodeString::compare(int32_t start,
3714 int32_t _length,
3715 const UnicodeString& srcText,
3716 int32_t srcStart,
3717 int32_t srcLength) const
3718 { return doCompare(start, _length, srcText, srcStart, srcLength); }
3719
3720 inline int8_t
compare(int32_t start,int32_t _length,const UChar * srcChars)3721 UnicodeString::compare(int32_t start,
3722 int32_t _length,
3723 const UChar *srcChars) const
3724 { return doCompare(start, _length, srcChars, 0, _length); }
3725
3726 inline int8_t
compare(int32_t start,int32_t _length,const UChar * srcChars,int32_t srcStart,int32_t srcLength)3727 UnicodeString::compare(int32_t start,
3728 int32_t _length,
3729 const UChar *srcChars,
3730 int32_t srcStart,
3731 int32_t srcLength) const
3732 { return doCompare(start, _length, srcChars, srcStart, srcLength); }
3733
3734 inline int8_t
compareBetween(int32_t start,int32_t limit,const UnicodeString & srcText,int32_t srcStart,int32_t srcLimit)3735 UnicodeString::compareBetween(int32_t start,
3736 int32_t limit,
3737 const UnicodeString& srcText,
3738 int32_t srcStart,
3739 int32_t srcLimit) const
3740 { return doCompare(start, limit - start,
3741 srcText, srcStart, srcLimit - srcStart); }
3742
3743 inline int8_t
doCompareCodePointOrder(int32_t start,int32_t thisLength,const UnicodeString & srcText,int32_t srcStart,int32_t srcLength)3744 UnicodeString::doCompareCodePointOrder(int32_t start,
3745 int32_t thisLength,
3746 const UnicodeString& srcText,
3747 int32_t srcStart,
3748 int32_t srcLength) const
3749 {
3750 if(srcText.isBogus()) {
3751 return (int8_t)!isBogus(); // 0 if both are bogus, 1 otherwise
3752 } else {
3753 srcText.pinIndices(srcStart, srcLength);
3754 return doCompareCodePointOrder(start, thisLength, srcText.getArrayStart(), srcStart, srcLength);
3755 }
3756 }
3757
3758 inline int8_t
compareCodePointOrder(const UnicodeString & text)3759 UnicodeString::compareCodePointOrder(const UnicodeString& text) const
3760 { return doCompareCodePointOrder(0, length(), text, 0, text.length()); }
3761
3762 inline int8_t
compareCodePointOrder(int32_t start,int32_t _length,const UnicodeString & srcText)3763 UnicodeString::compareCodePointOrder(int32_t start,
3764 int32_t _length,
3765 const UnicodeString& srcText) const
3766 { return doCompareCodePointOrder(start, _length, srcText, 0, srcText.length()); }
3767
3768 inline int8_t
compareCodePointOrder(const UChar * srcChars,int32_t srcLength)3769 UnicodeString::compareCodePointOrder(const UChar *srcChars,
3770 int32_t srcLength) const
3771 { return doCompareCodePointOrder(0, length(), srcChars, 0, srcLength); }
3772
3773 inline int8_t
compareCodePointOrder(int32_t start,int32_t _length,const UnicodeString & srcText,int32_t srcStart,int32_t srcLength)3774 UnicodeString::compareCodePointOrder(int32_t start,
3775 int32_t _length,
3776 const UnicodeString& srcText,
3777 int32_t srcStart,
3778 int32_t srcLength) const
3779 { return doCompareCodePointOrder(start, _length, srcText, srcStart, srcLength); }
3780
3781 inline int8_t
compareCodePointOrder(int32_t start,int32_t _length,const UChar * srcChars)3782 UnicodeString::compareCodePointOrder(int32_t start,
3783 int32_t _length,
3784 const UChar *srcChars) const
3785 { return doCompareCodePointOrder(start, _length, srcChars, 0, _length); }
3786
3787 inline int8_t
compareCodePointOrder(int32_t start,int32_t _length,const UChar * srcChars,int32_t srcStart,int32_t srcLength)3788 UnicodeString::compareCodePointOrder(int32_t start,
3789 int32_t _length,
3790 const UChar *srcChars,
3791 int32_t srcStart,
3792 int32_t srcLength) const
3793 { return doCompareCodePointOrder(start, _length, srcChars, srcStart, srcLength); }
3794
3795 inline int8_t
compareCodePointOrderBetween(int32_t start,int32_t limit,const UnicodeString & srcText,int32_t srcStart,int32_t srcLimit)3796 UnicodeString::compareCodePointOrderBetween(int32_t start,
3797 int32_t limit,
3798 const UnicodeString& srcText,
3799 int32_t srcStart,
3800 int32_t srcLimit) const
3801 { return doCompareCodePointOrder(start, limit - start,
3802 srcText, srcStart, srcLimit - srcStart); }
3803
3804 inline int8_t
doCaseCompare(int32_t start,int32_t thisLength,const UnicodeString & srcText,int32_t srcStart,int32_t srcLength,uint32_t options)3805 UnicodeString::doCaseCompare(int32_t start,
3806 int32_t thisLength,
3807 const UnicodeString &srcText,
3808 int32_t srcStart,
3809 int32_t srcLength,
3810 uint32_t options) const
3811 {
3812 if(srcText.isBogus()) {
3813 return (int8_t)!isBogus(); // 0 if both are bogus, 1 otherwise
3814 } else {
3815 srcText.pinIndices(srcStart, srcLength);
3816 return doCaseCompare(start, thisLength, srcText.getArrayStart(), srcStart, srcLength, options);
3817 }
3818 }
3819
3820 inline int8_t
caseCompare(const UnicodeString & text,uint32_t options)3821 UnicodeString::caseCompare(const UnicodeString &text, uint32_t options) const {
3822 return doCaseCompare(0, length(), text, 0, text.length(), options);
3823 }
3824
3825 inline int8_t
caseCompare(int32_t start,int32_t _length,const UnicodeString & srcText,uint32_t options)3826 UnicodeString::caseCompare(int32_t start,
3827 int32_t _length,
3828 const UnicodeString &srcText,
3829 uint32_t options) const {
3830 return doCaseCompare(start, _length, srcText, 0, srcText.length(), options);
3831 }
3832
3833 inline int8_t
caseCompare(const UChar * srcChars,int32_t srcLength,uint32_t options)3834 UnicodeString::caseCompare(const UChar *srcChars,
3835 int32_t srcLength,
3836 uint32_t options) const {
3837 return doCaseCompare(0, length(), srcChars, 0, srcLength, options);
3838 }
3839
3840 inline int8_t
caseCompare(int32_t start,int32_t _length,const UnicodeString & srcText,int32_t srcStart,int32_t srcLength,uint32_t options)3841 UnicodeString::caseCompare(int32_t start,
3842 int32_t _length,
3843 const UnicodeString &srcText,
3844 int32_t srcStart,
3845 int32_t srcLength,
3846 uint32_t options) const {
3847 return doCaseCompare(start, _length, srcText, srcStart, srcLength, options);
3848 }
3849
3850 inline int8_t
caseCompare(int32_t start,int32_t _length,const UChar * srcChars,uint32_t options)3851 UnicodeString::caseCompare(int32_t start,
3852 int32_t _length,
3853 const UChar *srcChars,
3854 uint32_t options) const {
3855 return doCaseCompare(start, _length, srcChars, 0, _length, options);
3856 }
3857
3858 inline int8_t
caseCompare(int32_t start,int32_t _length,const UChar * srcChars,int32_t srcStart,int32_t srcLength,uint32_t options)3859 UnicodeString::caseCompare(int32_t start,
3860 int32_t _length,
3861 const UChar *srcChars,
3862 int32_t srcStart,
3863 int32_t srcLength,
3864 uint32_t options) const {
3865 return doCaseCompare(start, _length, srcChars, srcStart, srcLength, options);
3866 }
3867
3868 inline int8_t
caseCompareBetween(int32_t start,int32_t limit,const UnicodeString & srcText,int32_t srcStart,int32_t srcLimit,uint32_t options)3869 UnicodeString::caseCompareBetween(int32_t start,
3870 int32_t limit,
3871 const UnicodeString &srcText,
3872 int32_t srcStart,
3873 int32_t srcLimit,
3874 uint32_t options) const {
3875 return doCaseCompare(start, limit - start, srcText, srcStart, srcLimit - srcStart, options);
3876 }
3877
3878 inline int32_t
indexOf(const UnicodeString & srcText,int32_t srcStart,int32_t srcLength,int32_t start,int32_t _length)3879 UnicodeString::indexOf(const UnicodeString& srcText,
3880 int32_t srcStart,
3881 int32_t srcLength,
3882 int32_t start,
3883 int32_t _length) const
3884 {
3885 if(!srcText.isBogus()) {
3886 srcText.pinIndices(srcStart, srcLength);
3887 if(srcLength > 0) {
3888 return indexOf(srcText.getArrayStart(), srcStart, srcLength, start, _length);
3889 }
3890 }
3891 return -1;
3892 }
3893
3894 inline int32_t
indexOf(const UnicodeString & text)3895 UnicodeString::indexOf(const UnicodeString& text) const
3896 { return indexOf(text, 0, text.length(), 0, length()); }
3897
3898 inline int32_t
indexOf(const UnicodeString & text,int32_t start)3899 UnicodeString::indexOf(const UnicodeString& text,
3900 int32_t start) const {
3901 pinIndex(start);
3902 return indexOf(text, 0, text.length(), start, length() - start);
3903 }
3904
3905 inline int32_t
indexOf(const UnicodeString & text,int32_t start,int32_t _length)3906 UnicodeString::indexOf(const UnicodeString& text,
3907 int32_t start,
3908 int32_t _length) const
3909 { return indexOf(text, 0, text.length(), start, _length); }
3910
3911 inline int32_t
indexOf(const UChar * srcChars,int32_t srcLength,int32_t start)3912 UnicodeString::indexOf(const UChar *srcChars,
3913 int32_t srcLength,
3914 int32_t start) const {
3915 pinIndex(start);
3916 return indexOf(srcChars, 0, srcLength, start, length() - start);
3917 }
3918
3919 inline int32_t
indexOf(const UChar * srcChars,int32_t srcLength,int32_t start,int32_t _length)3920 UnicodeString::indexOf(const UChar *srcChars,
3921 int32_t srcLength,
3922 int32_t start,
3923 int32_t _length) const
3924 { return indexOf(srcChars, 0, srcLength, start, _length); }
3925
3926 inline int32_t
indexOf(UChar c,int32_t start,int32_t _length)3927 UnicodeString::indexOf(UChar c,
3928 int32_t start,
3929 int32_t _length) const
3930 { return doIndexOf(c, start, _length); }
3931
3932 inline int32_t
indexOf(UChar32 c,int32_t start,int32_t _length)3933 UnicodeString::indexOf(UChar32 c,
3934 int32_t start,
3935 int32_t _length) const
3936 { return doIndexOf(c, start, _length); }
3937
3938 inline int32_t
indexOf(UChar c)3939 UnicodeString::indexOf(UChar c) const
3940 { return doIndexOf(c, 0, length()); }
3941
3942 inline int32_t
indexOf(UChar32 c)3943 UnicodeString::indexOf(UChar32 c) const
3944 { return indexOf(c, 0, length()); }
3945
3946 inline int32_t
indexOf(UChar c,int32_t start)3947 UnicodeString::indexOf(UChar c,
3948 int32_t start) const {
3949 pinIndex(start);
3950 return doIndexOf(c, start, length() - start);
3951 }
3952
3953 inline int32_t
indexOf(UChar32 c,int32_t start)3954 UnicodeString::indexOf(UChar32 c,
3955 int32_t start) const {
3956 pinIndex(start);
3957 return indexOf(c, start, length() - start);
3958 }
3959
3960 inline int32_t
lastIndexOf(const UChar * srcChars,int32_t srcLength,int32_t start,int32_t _length)3961 UnicodeString::lastIndexOf(const UChar *srcChars,
3962 int32_t srcLength,
3963 int32_t start,
3964 int32_t _length) const
3965 { return lastIndexOf(srcChars, 0, srcLength, start, _length); }
3966
3967 inline int32_t
lastIndexOf(const UChar * srcChars,int32_t srcLength,int32_t start)3968 UnicodeString::lastIndexOf(const UChar *srcChars,
3969 int32_t srcLength,
3970 int32_t start) const {
3971 pinIndex(start);
3972 return lastIndexOf(srcChars, 0, srcLength, start, length() - start);
3973 }
3974
3975 inline int32_t
lastIndexOf(const UnicodeString & srcText,int32_t srcStart,int32_t srcLength,int32_t start,int32_t _length)3976 UnicodeString::lastIndexOf(const UnicodeString& srcText,
3977 int32_t srcStart,
3978 int32_t srcLength,
3979 int32_t start,
3980 int32_t _length) const
3981 {
3982 if(!srcText.isBogus()) {
3983 srcText.pinIndices(srcStart, srcLength);
3984 if(srcLength > 0) {
3985 return lastIndexOf(srcText.getArrayStart(), srcStart, srcLength, start, _length);
3986 }
3987 }
3988 return -1;
3989 }
3990
3991 inline int32_t
lastIndexOf(const UnicodeString & text,int32_t start,int32_t _length)3992 UnicodeString::lastIndexOf(const UnicodeString& text,
3993 int32_t start,
3994 int32_t _length) const
3995 { return lastIndexOf(text, 0, text.length(), start, _length); }
3996
3997 inline int32_t
lastIndexOf(const UnicodeString & text,int32_t start)3998 UnicodeString::lastIndexOf(const UnicodeString& text,
3999 int32_t start) const {
4000 pinIndex(start);
4001 return lastIndexOf(text, 0, text.length(), start, length() - start);
4002 }
4003
4004 inline int32_t
lastIndexOf(const UnicodeString & text)4005 UnicodeString::lastIndexOf(const UnicodeString& text) const
4006 { return lastIndexOf(text, 0, text.length(), 0, length()); }
4007
4008 inline int32_t
lastIndexOf(UChar c,int32_t start,int32_t _length)4009 UnicodeString::lastIndexOf(UChar c,
4010 int32_t start,
4011 int32_t _length) const
4012 { return doLastIndexOf(c, start, _length); }
4013
4014 inline int32_t
lastIndexOf(UChar32 c,int32_t start,int32_t _length)4015 UnicodeString::lastIndexOf(UChar32 c,
4016 int32_t start,
4017 int32_t _length) const {
4018 return doLastIndexOf(c, start, _length);
4019 }
4020
4021 inline int32_t
lastIndexOf(UChar c)4022 UnicodeString::lastIndexOf(UChar c) const
4023 { return doLastIndexOf(c, 0, length()); }
4024
4025 inline int32_t
lastIndexOf(UChar32 c)4026 UnicodeString::lastIndexOf(UChar32 c) const {
4027 return lastIndexOf(c, 0, length());
4028 }
4029
4030 inline int32_t
lastIndexOf(UChar c,int32_t start)4031 UnicodeString::lastIndexOf(UChar c,
4032 int32_t start) const {
4033 pinIndex(start);
4034 return doLastIndexOf(c, start, length() - start);
4035 }
4036
4037 inline int32_t
lastIndexOf(UChar32 c,int32_t start)4038 UnicodeString::lastIndexOf(UChar32 c,
4039 int32_t start) const {
4040 pinIndex(start);
4041 return lastIndexOf(c, start, length() - start);
4042 }
4043
4044 inline UBool
startsWith(const UnicodeString & text)4045 UnicodeString::startsWith(const UnicodeString& text) const
4046 { return compare(0, text.length(), text, 0, text.length()) == 0; }
4047
4048 inline UBool
startsWith(const UnicodeString & srcText,int32_t srcStart,int32_t srcLength)4049 UnicodeString::startsWith(const UnicodeString& srcText,
4050 int32_t srcStart,
4051 int32_t srcLength) const
4052 { return doCompare(0, srcLength, srcText, srcStart, srcLength) == 0; }
4053
4054 inline UBool
startsWith(const UChar * srcChars,int32_t srcLength)4055 UnicodeString::startsWith(const UChar *srcChars, int32_t srcLength) const {
4056 if(srcLength < 0) {
4057 srcLength = u_strlen(srcChars);
4058 }
4059 return doCompare(0, srcLength, srcChars, 0, srcLength) == 0;
4060 }
4061
4062 inline UBool
startsWith(const UChar * srcChars,int32_t srcStart,int32_t srcLength)4063 UnicodeString::startsWith(const UChar *srcChars, int32_t srcStart, int32_t srcLength) const {
4064 if(srcLength < 0) {
4065 srcLength = u_strlen(srcChars);
4066 }
4067 return doCompare(0, srcLength, srcChars, srcStart, srcLength) == 0;
4068 }
4069
4070 inline UBool
endsWith(const UnicodeString & text)4071 UnicodeString::endsWith(const UnicodeString& text) const
4072 { return doCompare(length() - text.length(), text.length(),
4073 text, 0, text.length()) == 0; }
4074
4075 inline UBool
endsWith(const UnicodeString & srcText,int32_t srcStart,int32_t srcLength)4076 UnicodeString::endsWith(const UnicodeString& srcText,
4077 int32_t srcStart,
4078 int32_t srcLength) const {
4079 srcText.pinIndices(srcStart, srcLength);
4080 return doCompare(length() - srcLength, srcLength,
4081 srcText, srcStart, srcLength) == 0;
4082 }
4083
4084 inline UBool
endsWith(const UChar * srcChars,int32_t srcLength)4085 UnicodeString::endsWith(const UChar *srcChars,
4086 int32_t srcLength) const {
4087 if(srcLength < 0) {
4088 srcLength = u_strlen(srcChars);
4089 }
4090 return doCompare(length() - srcLength, srcLength,
4091 srcChars, 0, srcLength) == 0;
4092 }
4093
4094 inline UBool
endsWith(const UChar * srcChars,int32_t srcStart,int32_t srcLength)4095 UnicodeString::endsWith(const UChar *srcChars,
4096 int32_t srcStart,
4097 int32_t srcLength) const {
4098 if(srcLength < 0) {
4099 srcLength = u_strlen(srcChars + srcStart);
4100 }
4101 return doCompare(length() - srcLength, srcLength,
4102 srcChars, srcStart, srcLength) == 0;
4103 }
4104
4105 //========================================
4106 // replace
4107 //========================================
4108 inline UnicodeString&
replace(int32_t start,int32_t _length,const UnicodeString & srcText)4109 UnicodeString::replace(int32_t start,
4110 int32_t _length,
4111 const UnicodeString& srcText)
4112 { return doReplace(start, _length, srcText, 0, srcText.length()); }
4113
4114 inline UnicodeString&
replace(int32_t start,int32_t _length,const UnicodeString & srcText,int32_t srcStart,int32_t srcLength)4115 UnicodeString::replace(int32_t start,
4116 int32_t _length,
4117 const UnicodeString& srcText,
4118 int32_t srcStart,
4119 int32_t srcLength)
4120 { return doReplace(start, _length, srcText, srcStart, srcLength); }
4121
4122 inline UnicodeString&
replace(int32_t start,int32_t _length,const UChar * srcChars,int32_t srcLength)4123 UnicodeString::replace(int32_t start,
4124 int32_t _length,
4125 const UChar *srcChars,
4126 int32_t srcLength)
4127 { return doReplace(start, _length, srcChars, 0, srcLength); }
4128
4129 inline UnicodeString&
replace(int32_t start,int32_t _length,const UChar * srcChars,int32_t srcStart,int32_t srcLength)4130 UnicodeString::replace(int32_t start,
4131 int32_t _length,
4132 const UChar *srcChars,
4133 int32_t srcStart,
4134 int32_t srcLength)
4135 { return doReplace(start, _length, srcChars, srcStart, srcLength); }
4136
4137 inline UnicodeString&
replace(int32_t start,int32_t _length,UChar srcChar)4138 UnicodeString::replace(int32_t start,
4139 int32_t _length,
4140 UChar srcChar)
4141 { return doReplace(start, _length, &srcChar, 0, 1); }
4142
4143 inline UnicodeString&
replaceBetween(int32_t start,int32_t limit,const UnicodeString & srcText)4144 UnicodeString::replaceBetween(int32_t start,
4145 int32_t limit,
4146 const UnicodeString& srcText)
4147 { return doReplace(start, limit - start, srcText, 0, srcText.length()); }
4148
4149 inline UnicodeString&
replaceBetween(int32_t start,int32_t limit,const UnicodeString & srcText,int32_t srcStart,int32_t srcLimit)4150 UnicodeString::replaceBetween(int32_t start,
4151 int32_t limit,
4152 const UnicodeString& srcText,
4153 int32_t srcStart,
4154 int32_t srcLimit)
4155 { return doReplace(start, limit - start, srcText, srcStart, srcLimit - srcStart); }
4156
4157 inline UnicodeString&
findAndReplace(const UnicodeString & oldText,const UnicodeString & newText)4158 UnicodeString::findAndReplace(const UnicodeString& oldText,
4159 const UnicodeString& newText)
4160 { return findAndReplace(0, length(), oldText, 0, oldText.length(),
4161 newText, 0, newText.length()); }
4162
4163 inline UnicodeString&
findAndReplace(int32_t start,int32_t _length,const UnicodeString & oldText,const UnicodeString & newText)4164 UnicodeString::findAndReplace(int32_t start,
4165 int32_t _length,
4166 const UnicodeString& oldText,
4167 const UnicodeString& newText)
4168 { return findAndReplace(start, _length, oldText, 0, oldText.length(),
4169 newText, 0, newText.length()); }
4170
4171 // ============================
4172 // extract
4173 // ============================
4174 inline void
doExtract(int32_t start,int32_t _length,UnicodeString & target)4175 UnicodeString::doExtract(int32_t start,
4176 int32_t _length,
4177 UnicodeString& target) const
4178 { target.replace(0, target.length(), *this, start, _length); }
4179
4180 inline void
extract(int32_t start,int32_t _length,UChar * target,int32_t targetStart)4181 UnicodeString::extract(int32_t start,
4182 int32_t _length,
4183 UChar *target,
4184 int32_t targetStart) const
4185 { doExtract(start, _length, target, targetStart); }
4186
4187 inline void
extract(int32_t start,int32_t _length,UnicodeString & target)4188 UnicodeString::extract(int32_t start,
4189 int32_t _length,
4190 UnicodeString& target) const
4191 { doExtract(start, _length, target); }
4192
4193 #if !UCONFIG_NO_CONVERSION
4194
4195 inline int32_t
extract(int32_t start,int32_t _length,char * dst,const char * codepage)4196 UnicodeString::extract(int32_t start,
4197 int32_t _length,
4198 char *dst,
4199 const char *codepage) const
4200
4201 {
4202 // This dstSize value will be checked explicitly
4203 return extract(start, _length, dst, dst!=0 ? 0xffffffff : 0, codepage);
4204 }
4205
4206 #endif
4207
4208 inline void
extractBetween(int32_t start,int32_t limit,UChar * dst,int32_t dstStart)4209 UnicodeString::extractBetween(int32_t start,
4210 int32_t limit,
4211 UChar *dst,
4212 int32_t dstStart) const {
4213 pinIndex(start);
4214 pinIndex(limit);
4215 doExtract(start, limit - start, dst, dstStart);
4216 }
4217
4218 inline UnicodeString
tempSubStringBetween(int32_t start,int32_t limit)4219 UnicodeString::tempSubStringBetween(int32_t start, int32_t limit) const {
4220 return tempSubString(start, limit - start);
4221 }
4222
4223 inline UChar
doCharAt(int32_t offset)4224 UnicodeString::doCharAt(int32_t offset) const
4225 {
4226 if((uint32_t)offset < (uint32_t)length()) {
4227 return getArrayStart()[offset];
4228 } else {
4229 return kInvalidUChar;
4230 }
4231 }
4232
4233 inline UChar
charAt(int32_t offset)4234 UnicodeString::charAt(int32_t offset) const
4235 { return doCharAt(offset); }
4236
4237 inline UChar
4238 UnicodeString::operator[] (int32_t offset) const
4239 { return doCharAt(offset); }
4240
4241 inline UBool
isEmpty()4242 UnicodeString::isEmpty() const {
4243 return fShortLength == 0;
4244 }
4245
4246 //========================================
4247 // Write implementation methods
4248 //========================================
4249 inline void
setLength(int32_t len)4250 UnicodeString::setLength(int32_t len) {
4251 if(len <= 127) {
4252 fShortLength = (int8_t)len;
4253 } else {
4254 fShortLength = (int8_t)-1;
4255 fUnion.fFields.fLength = len;
4256 }
4257 }
4258
4259 inline void
setToEmpty()4260 UnicodeString::setToEmpty() {
4261 fShortLength = 0;
4262 fFlags = kShortString;
4263 }
4264
4265 inline void
setArray(UChar * array,int32_t len,int32_t capacity)4266 UnicodeString::setArray(UChar *array, int32_t len, int32_t capacity) {
4267 setLength(len);
4268 fUnion.fFields.fArray = array;
4269 fUnion.fFields.fCapacity = capacity;
4270 }
4271
4272 inline const UChar *
getTerminatedBuffer()4273 UnicodeString::getTerminatedBuffer() {
4274 if(!isWritable()) {
4275 return 0;
4276 } else {
4277 UChar *array = getArrayStart();
4278 int32_t len = length();
4279 if(len < getCapacity() && ((fFlags&kRefCounted) == 0 || refCount() == 1)) {
4280 /*
4281 * kRefCounted: Do not write the NUL if the buffer is shared.
4282 * That is mostly safe, except when the length of one copy was modified
4283 * without copy-on-write, e.g., via truncate(newLength) or remove(void).
4284 * Then the NUL would be written into the middle of another copy's string.
4285 */
4286 if(!(fFlags&kBufferIsReadonly)) {
4287 /*
4288 * We must not write to a readonly buffer, but it is known to be
4289 * NUL-terminated if len<capacity.
4290 * A shared, allocated buffer (refCount()>1) must not have its contents
4291 * modified, but the NUL at [len] is beyond the string contents,
4292 * and multiple string objects and threads writing the same NUL into the
4293 * same location is harmless.
4294 * In all other cases, the buffer is fully writable and it is anyway safe
4295 * to write the NUL.
4296 *
4297 * Note: An earlier version of this code tested whether there is a NUL
4298 * at [len] already, but, while safe, it generated lots of warnings from
4299 * tools like valgrind and Purify.
4300 */
4301 array[len] = 0;
4302 }
4303 return array;
4304 } else if(cloneArrayIfNeeded(len+1)) {
4305 array = getArrayStart();
4306 array[len] = 0;
4307 return array;
4308 } else {
4309 return 0;
4310 }
4311 }
4312 }
4313
4314 inline UnicodeString&
4315 UnicodeString::operator= (UChar ch)
4316 { return doReplace(0, length(), &ch, 0, 1); }
4317
4318 inline UnicodeString&
4319 UnicodeString::operator= (UChar32 ch)
4320 { return replace(0, length(), ch); }
4321
4322 inline UnicodeString&
setTo(const UnicodeString & srcText,int32_t srcStart,int32_t srcLength)4323 UnicodeString::setTo(const UnicodeString& srcText,
4324 int32_t srcStart,
4325 int32_t srcLength)
4326 {
4327 unBogus();
4328 return doReplace(0, length(), srcText, srcStart, srcLength);
4329 }
4330
4331 inline UnicodeString&
setTo(const UnicodeString & srcText,int32_t srcStart)4332 UnicodeString::setTo(const UnicodeString& srcText,
4333 int32_t srcStart)
4334 {
4335 unBogus();
4336 srcText.pinIndex(srcStart);
4337 return doReplace(0, length(), srcText, srcStart, srcText.length() - srcStart);
4338 }
4339
4340 inline UnicodeString&
setTo(const UnicodeString & srcText)4341 UnicodeString::setTo(const UnicodeString& srcText)
4342 {
4343 return copyFrom(srcText);
4344 }
4345
4346 inline UnicodeString&
setTo(const UChar * srcChars,int32_t srcLength)4347 UnicodeString::setTo(const UChar *srcChars,
4348 int32_t srcLength)
4349 {
4350 unBogus();
4351 return doReplace(0, length(), srcChars, 0, srcLength);
4352 }
4353
4354 inline UnicodeString&
setTo(UChar srcChar)4355 UnicodeString::setTo(UChar srcChar)
4356 {
4357 unBogus();
4358 return doReplace(0, length(), &srcChar, 0, 1);
4359 }
4360
4361 inline UnicodeString&
setTo(UChar32 srcChar)4362 UnicodeString::setTo(UChar32 srcChar)
4363 {
4364 unBogus();
4365 return replace(0, length(), srcChar);
4366 }
4367
4368 inline UnicodeString&
append(const UnicodeString & srcText,int32_t srcStart,int32_t srcLength)4369 UnicodeString::append(const UnicodeString& srcText,
4370 int32_t srcStart,
4371 int32_t srcLength)
4372 { return doReplace(length(), 0, srcText, srcStart, srcLength); }
4373
4374 inline UnicodeString&
append(const UnicodeString & srcText)4375 UnicodeString::append(const UnicodeString& srcText)
4376 { return doReplace(length(), 0, srcText, 0, srcText.length()); }
4377
4378 inline UnicodeString&
append(const UChar * srcChars,int32_t srcStart,int32_t srcLength)4379 UnicodeString::append(const UChar *srcChars,
4380 int32_t srcStart,
4381 int32_t srcLength)
4382 { return doReplace(length(), 0, srcChars, srcStart, srcLength); }
4383
4384 inline UnicodeString&
append(const UChar * srcChars,int32_t srcLength)4385 UnicodeString::append(const UChar *srcChars,
4386 int32_t srcLength)
4387 { return doReplace(length(), 0, srcChars, 0, srcLength); }
4388
4389 inline UnicodeString&
append(UChar srcChar)4390 UnicodeString::append(UChar srcChar)
4391 { return doReplace(length(), 0, &srcChar, 0, 1); }
4392
4393 inline UnicodeString&
4394 UnicodeString::operator+= (UChar ch)
4395 { return doReplace(length(), 0, &ch, 0, 1); }
4396
4397 inline UnicodeString&
4398 UnicodeString::operator+= (UChar32 ch) {
4399 return append(ch);
4400 }
4401
4402 inline UnicodeString&
4403 UnicodeString::operator+= (const UnicodeString& srcText)
4404 { return doReplace(length(), 0, srcText, 0, srcText.length()); }
4405
4406 inline UnicodeString&
insert(int32_t start,const UnicodeString & srcText,int32_t srcStart,int32_t srcLength)4407 UnicodeString::insert(int32_t start,
4408 const UnicodeString& srcText,
4409 int32_t srcStart,
4410 int32_t srcLength)
4411 { return doReplace(start, 0, srcText, srcStart, srcLength); }
4412
4413 inline UnicodeString&
insert(int32_t start,const UnicodeString & srcText)4414 UnicodeString::insert(int32_t start,
4415 const UnicodeString& srcText)
4416 { return doReplace(start, 0, srcText, 0, srcText.length()); }
4417
4418 inline UnicodeString&
insert(int32_t start,const UChar * srcChars,int32_t srcStart,int32_t srcLength)4419 UnicodeString::insert(int32_t start,
4420 const UChar *srcChars,
4421 int32_t srcStart,
4422 int32_t srcLength)
4423 { return doReplace(start, 0, srcChars, srcStart, srcLength); }
4424
4425 inline UnicodeString&
insert(int32_t start,const UChar * srcChars,int32_t srcLength)4426 UnicodeString::insert(int32_t start,
4427 const UChar *srcChars,
4428 int32_t srcLength)
4429 { return doReplace(start, 0, srcChars, 0, srcLength); }
4430
4431 inline UnicodeString&
insert(int32_t start,UChar srcChar)4432 UnicodeString::insert(int32_t start,
4433 UChar srcChar)
4434 { return doReplace(start, 0, &srcChar, 0, 1); }
4435
4436 inline UnicodeString&
insert(int32_t start,UChar32 srcChar)4437 UnicodeString::insert(int32_t start,
4438 UChar32 srcChar)
4439 { return replace(start, 0, srcChar); }
4440
4441
4442 inline UnicodeString&
remove()4443 UnicodeString::remove()
4444 {
4445 // remove() of a bogus string makes the string empty and non-bogus
4446 // we also un-alias a read-only alias to deal with NUL-termination
4447 // issues with getTerminatedBuffer()
4448 if(fFlags & (kIsBogus|kBufferIsReadonly)) {
4449 setToEmpty();
4450 } else {
4451 fShortLength = 0;
4452 }
4453 return *this;
4454 }
4455
4456 inline UnicodeString&
remove(int32_t start,int32_t _length)4457 UnicodeString::remove(int32_t start,
4458 int32_t _length)
4459 {
4460 if(start <= 0 && _length == INT32_MAX) {
4461 // remove(guaranteed everything) of a bogus string makes the string empty and non-bogus
4462 return remove();
4463 }
4464 return doReplace(start, _length, NULL, 0, 0);
4465 }
4466
4467 inline UnicodeString&
removeBetween(int32_t start,int32_t limit)4468 UnicodeString::removeBetween(int32_t start,
4469 int32_t limit)
4470 { return doReplace(start, limit - start, NULL, 0, 0); }
4471
4472 inline UnicodeString &
retainBetween(int32_t start,int32_t limit)4473 UnicodeString::retainBetween(int32_t start, int32_t limit) {
4474 truncate(limit);
4475 return doReplace(0, start, NULL, 0, 0);
4476 }
4477
4478 inline UBool
truncate(int32_t targetLength)4479 UnicodeString::truncate(int32_t targetLength)
4480 {
4481 if(isBogus() && targetLength == 0) {
4482 // truncate(0) of a bogus string makes the string empty and non-bogus
4483 unBogus();
4484 return FALSE;
4485 } else if((uint32_t)targetLength < (uint32_t)length()) {
4486 setLength(targetLength);
4487 if(fFlags&kBufferIsReadonly) {
4488 fUnion.fFields.fCapacity = targetLength; // not NUL-terminated any more
4489 }
4490 return TRUE;
4491 } else {
4492 return FALSE;
4493 }
4494 }
4495
4496 inline UnicodeString&
reverse()4497 UnicodeString::reverse()
4498 { return doReverse(0, length()); }
4499
4500 inline UnicodeString&
reverse(int32_t start,int32_t _length)4501 UnicodeString::reverse(int32_t start,
4502 int32_t _length)
4503 { return doReverse(start, _length); }
4504
4505 U_NAMESPACE_END
4506
4507 #endif
4508