1// © 2016 and later: Unicode, Inc. and others.
2// License & terms of use: http://www.unicode.org/copyright.html
3/*
4**********************************************************************
5* Copyright (C) 1998-2016, International Business Machines
6* Corporation and others. All Rights Reserved.
7**********************************************************************
8*
9* File unistr.h
10*
11* Modification History:
12*
13* Date Name Description
14* 09/25/98 stephen Creation.
15* 11/11/98 stephen Changed per 11/9 code review.
16* 04/20/99 stephen Overhauled per 4/16 code review.
17* 11/18/99 aliu Made to inherit from Replaceable. Added method
18* handleReplaceBetween(); other methods unchanged.
19* 06/25/01 grhoten Remove dependency on iostream.
20******************************************************************************
21*/
22
23#ifndef UNISTR_H
24#define UNISTR_H
25
26/**
27 * \file
28 * \brief C++ API: Unicode String
29 */
30
31#include <cstddef>
32#include "unicode/utypes.h"
33#include "unicode/char16ptr.h"
34#include "unicode/rep.h"
35#include "unicode/std_string.h"
36#include "unicode/stringpiece.h"
37#include "unicode/bytestream.h"
38
39struct UConverter; // unicode/ucnv.h
40
41#ifndef USTRING_H
42/**
43 * \ingroup ustring_ustrlen
44 */
45U_STABLE int32_t U_EXPORT2
46u_strlen(const UChar *s);
47#endif
48
49U_NAMESPACE_BEGIN
50
51#if !UCONFIG_NO_BREAK_ITERATION
52class BreakIterator; // unicode/brkiter.h
53#endif
54class Edits;
55
56U_NAMESPACE_END
57
58// Not #ifndef U_HIDE_INTERNAL_API because UnicodeString needs the UStringCaseMapper.
59/**
60 * Internal string case mapping function type.
61 * All error checking must be done.
62 * src and dest must not overlap.
63 * @internal
64 */
65typedef int32_t U_CALLCONV
66UStringCaseMapper(int32_t caseLocale, uint32_t options,
67#if !UCONFIG_NO_BREAK_ITERATION
68 icu::BreakIterator *iter,
69#endif
70 char16_t *dest, int32_t destCapacity,
71 const char16_t *src, int32_t srcLength,
72 icu::Edits *edits,
73 UErrorCode &errorCode);
74
75U_NAMESPACE_BEGIN
76
77class Locale; // unicode/locid.h
78class StringCharacterIterator;
79class UnicodeStringAppendable; // unicode/appendable.h
80
81/* The <iostream> include has been moved to unicode/ustream.h */
82
83/**
84 * Constant to be used in the UnicodeString(char *, int32_t, EInvariant) constructor
85 * which constructs a Unicode string from an invariant-character char * string.
86 * About invariant characters see utypes.h.
87 * This constructor has no runtime dependency on conversion code and is
88 * therefore recommended over ones taking a charset name string
89 * (where the empty string "" indicates invariant-character conversion).
90 *
91 * @stable ICU 3.2
92 */
93#define US_INV icu::UnicodeString::kInvariant
94
95/**
96 * Unicode String literals in C++.
97 *
98 * Note: these macros are not recommended for new code.
99 * Prior to the availability of C++11 and u"unicode string literals",
100 * these macros were provided for portability and efficiency when
101 * initializing UnicodeStrings from literals.
102 *
103 * They work only for strings that contain "invariant characters", i.e.,
104 * only latin letters, digits, and some punctuation.
105 * See utypes.h for details.
106 *
107 * The string parameter must be a C string literal.
108 * The length of the string, not including the terminating
109 * <code>NUL</code>, must be specified as a constant.
110 * @stable ICU 2.0
111 */
112#if !U_CHAR16_IS_TYPEDEF
113# define UNICODE_STRING(cs, _length) icu::UnicodeString(TRUE, u ## cs, _length)
114#else
115# define UNICODE_STRING(cs, _length) icu::UnicodeString(TRUE, (const char16_t*)u ## cs, _length)
116#endif
117
118/**
119 * Unicode String literals in C++.
120 * Dependent on the platform properties, different UnicodeString
121 * constructors should be used to create a UnicodeString object from
122 * a string literal.
123 * The macros are defined for improved performance.
124 * They work only for strings that contain "invariant characters", i.e.,
125 * only latin letters, digits, and some punctuation.
126 * See utypes.h for details.
127 *
128 * The string parameter must be a C string literal.
129 * @stable ICU 2.0
130 */
131#define UNICODE_STRING_SIMPLE(cs) UNICODE_STRING(cs, -1)
132
133/**
134 * \def UNISTR_FROM_CHAR_EXPLICIT
135 * This can be defined to be empty or "explicit".
136 * If explicit, then the UnicodeString(char16_t) and UnicodeString(UChar32)
137 * constructors are marked as explicit, preventing their inadvertent use.
138 * @stable ICU 49
139 */
140#ifndef UNISTR_FROM_CHAR_EXPLICIT
141# if defined(U_COMBINED_IMPLEMENTATION) || defined(U_COMMON_IMPLEMENTATION) || defined(U_I18N_IMPLEMENTATION) || defined(U_IO_IMPLEMENTATION)
142 // Auto-"explicit" in ICU library code.
143# define UNISTR_FROM_CHAR_EXPLICIT explicit
144# else
145 // Empty by default for source code compatibility.
146# define UNISTR_FROM_CHAR_EXPLICIT
147# endif
148#endif
149
150/**
151 * \def UNISTR_FROM_STRING_EXPLICIT
152 * This can be defined to be empty or "explicit".
153 * If explicit, then the UnicodeString(const char *) and UnicodeString(const char16_t *)
154 * constructors are marked as explicit, preventing their inadvertent use.
155 *
156 * In particular, this helps prevent accidentally depending on ICU conversion code
157 * by passing a string literal into an API with a const UnicodeString & parameter.
158 * @stable ICU 49
159 */
160#ifndef UNISTR_FROM_STRING_EXPLICIT
161# if defined(U_COMBINED_IMPLEMENTATION) || defined(U_COMMON_IMPLEMENTATION) || defined(U_I18N_IMPLEMENTATION) || defined(U_IO_IMPLEMENTATION)
162 // Auto-"explicit" in ICU library code.
163# define UNISTR_FROM_STRING_EXPLICIT explicit
164# else
165 // Empty by default for source code compatibility.
166# define UNISTR_FROM_STRING_EXPLICIT
167# endif
168#endif
169
170/**
171 * \def UNISTR_OBJECT_SIZE
172 * Desired sizeof(UnicodeString) in bytes.
173 * It should be a multiple of sizeof(pointer) to avoid unusable space for padding.
174 * The object size may want to be a multiple of 16 bytes,
175 * which is a common granularity for heap allocation.
176 *
177 * Any space inside the object beyond sizeof(vtable pointer) + 2
178 * is available for storing short strings inside the object.
179 * The bigger the object, the longer a string that can be stored inside the object,
180 * without additional heap allocation.
181 *
182 * Depending on a platform's pointer size, pointer alignment requirements,
183 * and struct padding, the compiler will usually round up sizeof(UnicodeString)
184 * to 4 * sizeof(pointer) (or 3 * sizeof(pointer) for P128 data models),
185 * to hold the fields for heap-allocated strings.
186 * Such a minimum size also ensures that the object is easily large enough
187 * to hold at least 2 char16_ts, for one supplementary code point (U16_MAX_LENGTH).
188 *
189 * sizeof(UnicodeString) >= 48 should work for all known platforms.
190 *
191 * For example, on a 64-bit machine where sizeof(vtable pointer) is 8,
192 * sizeof(UnicodeString) = 64 would leave space for
193 * (64 - sizeof(vtable pointer) - 2) / U_SIZEOF_UCHAR = (64 - 8 - 2) / 2 = 27
194 * char16_ts stored inside the object.
195 *
196 * The minimum object size on a 64-bit machine would be
197 * 4 * sizeof(pointer) = 4 * 8 = 32 bytes,
198 * and the internal buffer would hold up to 11 char16_ts in that case.
199 *
200 * @see U16_MAX_LENGTH
201 * @stable ICU 56
202 */
203#ifndef UNISTR_OBJECT_SIZE
204# define UNISTR_OBJECT_SIZE 64
205#endif
206
207/**
208 * UnicodeString is a string class that stores Unicode characters directly and provides
209 * similar functionality as the Java String and StringBuffer/StringBuilder classes.
210 * It is a concrete implementation of the abstract class Replaceable (for transliteration).
211 *
212 * A UnicodeString may also "alias" an external array of characters
213 * (that is, point to it, rather than own the array)
214 * whose lifetime must then at least match the lifetime of the aliasing object.
215 * This aliasing may be preserved when returning a UnicodeString by value,
216 * depending on the compiler and the function implementation,
217 * via Return Value Optimization (RVO) or the move assignment operator.
218 * (However, the copy assignment operator does not preserve aliasing.)
219 * For details see the description of storage models at the end of the class API docs
220 * and in the User Guide chapter linked from there.
221 *
222 * The UnicodeString class is not suitable for subclassing.
223 *
224 * <p>For an overview of Unicode strings in C and C++ see the
225 * <a href="http://userguide.icu-project.org/strings#TOC-Strings-in-C-C-">User Guide Strings chapter</a>.</p>
226 *
227 * <p>In ICU, a Unicode string consists of 16-bit Unicode <em>code units</em>.
228 * A Unicode character may be stored with either one code unit
229 * (the most common case) or with a matched pair of special code units
230 * ("surrogates"). The data type for code units is char16_t.
231 * For single-character handling, a Unicode character code <em>point</em> is a value
232 * in the range 0..0x10ffff. ICU uses the UChar32 type for code points.</p>
233 *
234 * <p>Indexes and offsets into and lengths of strings always count code units, not code points.
235 * This is the same as with multi-byte char* strings in traditional string handling.
236 * Operations on partial strings typically do not test for code point boundaries.
237 * If necessary, the user needs to take care of such boundaries by testing for the code unit
238 * values or by using functions like
239 * UnicodeString::getChar32Start() and UnicodeString::getChar32Limit()
240 * (or, in C, the equivalent macros U16_SET_CP_START() and U16_SET_CP_LIMIT(), see utf.h).</p>
241 *
242 * UnicodeString methods are more lenient with regard to input parameter values
243 * than other ICU APIs. In particular:
244 * - If indexes are out of bounds for a UnicodeString object
245 * (<0 or >length()) then they are "pinned" to the nearest boundary.
246 * - If primitive string pointer values (e.g., const char16_t * or char *)
247 * for input strings are NULL, then those input string parameters are treated
248 * as if they pointed to an empty string.
249 * However, this is <em>not</em> the case for char * parameters for charset names
250 * or other IDs.
251 * - Most UnicodeString methods do not take a UErrorCode parameter because
252 * there are usually very few opportunities for failure other than a shortage
253 * of memory, error codes in low-level C++ string methods would be inconvenient,
254 * and the error code as the last parameter (ICU convention) would prevent
255 * the use of default parameter values.
256 * Instead, such methods set the UnicodeString into a "bogus" state
257 * (see isBogus()) if an error occurs.
258 *
259 * In string comparisons, two UnicodeString objects that are both "bogus"
260 * compare equal (to be transitive and prevent endless loops in sorting),
261 * and a "bogus" string compares less than any non-"bogus" one.
262 *
263 * Const UnicodeString methods are thread-safe. Multiple threads can use
264 * const methods on the same UnicodeString object simultaneously,
265 * but non-const methods must not be called concurrently (in multiple threads)
266 * with any other (const or non-const) methods.
267 *
268 * Similarly, const UnicodeString & parameters are thread-safe.
269 * One object may be passed in as such a parameter concurrently in multiple threads.
270 * This includes the const UnicodeString & parameters for
271 * copy construction, assignment, and cloning.
272 *
273 * <p>UnicodeString uses several storage methods.
274 * String contents can be stored inside the UnicodeString object itself,
275 * in an allocated and shared buffer, or in an outside buffer that is "aliased".
276 * Most of this is done transparently, but careful aliasing in particular provides
277 * significant performance improvements.
278 * Also, the internal buffer is accessible via special functions.
279 * For details see the
280 * <a href="http://userguide.icu-project.org/strings#TOC-Maximizing-Performance-with-the-UnicodeString-Storage-Model">User Guide Strings chapter</a>.</p>
281 *
282 * @see utf.h
283 * @see CharacterIterator
284 * @stable ICU 2.0
285 */
286class U_COMMON_API UnicodeString : public Replaceable
287{
288public:
289
290 /**
291 * Constant to be used in the UnicodeString(char *, int32_t, EInvariant) constructor
292 * which constructs a Unicode string from an invariant-character char * string.
293 * Use the macro US_INV instead of the full qualification for this value.
294 *
295 * @see US_INV
296 * @stable ICU 3.2
297 */
298 enum EInvariant {
299 /**
300 * @see EInvariant
301 * @stable ICU 3.2
302 */
303 kInvariant
304 };
305
306 //========================================
307 // Read-only operations
308 //========================================
309
310 /* Comparison - bitwise only - for international comparison use collation */
311
312 /**
313 * Equality operator. Performs only bitwise comparison.
314 * @param text The UnicodeString to compare to this one.
315 * @return TRUE if <TT>text</TT> contains the same characters as this one,
316 * FALSE otherwise.
317 * @stable ICU 2.0
318 */
319 inline UBool operator== (const UnicodeString& text) const;
320
321 /**
322 * Inequality operator. Performs only bitwise comparison.
323 * @param text The UnicodeString to compare to this one.
324 * @return FALSE if <TT>text</TT> contains the same characters as this one,
325 * TRUE otherwise.
326 * @stable ICU 2.0
327 */
328 inline UBool operator!= (const UnicodeString& text) const;
329
330 /**
331 * Greater than operator. Performs only bitwise comparison.
332 * @param text The UnicodeString to compare to this one.
333 * @return TRUE if the characters in this are bitwise
334 * greater than the characters in <code>text</code>, FALSE otherwise
335 * @stable ICU 2.0
336 */
337 inline UBool operator> (const UnicodeString& text) const;
338
339 /**
340 * Less than operator. Performs only bitwise comparison.
341 * @param text The UnicodeString to compare to this one.
342 * @return TRUE if the characters in this are bitwise
343 * less than the characters in <code>text</code>, FALSE otherwise
344 * @stable ICU 2.0
345 */
346 inline UBool operator< (const UnicodeString& text) const;
347
348 /**
349 * Greater than or equal operator. Performs only bitwise comparison.
350 * @param text The UnicodeString to compare to this one.
351 * @return TRUE if the characters in this are bitwise
352 * greater than or equal to the characters in <code>text</code>, FALSE otherwise
353 * @stable ICU 2.0
354 */
355 inline UBool operator>= (const UnicodeString& text) const;
356
357 /**
358 * Less than or equal operator. Performs only bitwise comparison.
359 * @param text The UnicodeString to compare to this one.
360 * @return TRUE if the characters in this are bitwise
361 * less than or equal to the characters in <code>text</code>, FALSE otherwise
362 * @stable ICU 2.0
363 */
364 inline UBool operator<= (const UnicodeString& text) const;
365
366 /**
367 * Compare the characters bitwise in this UnicodeString to
368 * the characters in <code>text</code>.
369 * @param text The UnicodeString to compare to this one.
370 * @return The result of bitwise character comparison: 0 if this
371 * contains the same characters as <code>text</code>, -1 if the characters in
372 * this are bitwise less than the characters in <code>text</code>, +1 if the
373 * characters in this are bitwise greater than the characters
374 * in <code>text</code>.
375 * @stable ICU 2.0
376 */
377 inline int8_t compare(const UnicodeString& text) const;
378
379 /**
380 * Compare the characters bitwise in the range
381 * [<TT>start</TT>, <TT>start + length</TT>) with the characters
382 * in the <b>entire string</b> <TT>text</TT>.
383 * (The parameters "start" and "length" are not applied to the other text "text".)
384 * @param start the offset at which the compare operation begins
385 * @param length the number of characters of text to compare.
386 * @param text the other text to be compared against this string.
387 * @return The result of bitwise character comparison: 0 if this
388 * contains the same characters as <code>text</code>, -1 if the characters in
389 * this are bitwise less than the characters in <code>text</code>, +1 if the
390 * characters in this are bitwise greater than the characters
391 * in <code>text</code>.
392 * @stable ICU 2.0
393 */
394 inline int8_t compare(int32_t start,
395 int32_t length,
396 const UnicodeString& text) const;
397
398 /**
399 * Compare the characters bitwise in the range
400 * [<TT>start</TT>, <TT>start + length</TT>) with the characters
401 * in <TT>srcText</TT> in the range
402 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
403 * @param start the offset at which the compare operation begins
404 * @param length the number of characters in this to compare.
405 * @param srcText the text to be compared
406 * @param srcStart the offset into <TT>srcText</TT> to start comparison
407 * @param srcLength the number of characters in <TT>src</TT> to compare
408 * @return The result of bitwise character comparison: 0 if this
409 * contains the same characters as <code>srcText</code>, -1 if the characters in
410 * this are bitwise less than the characters in <code>srcText</code>, +1 if the
411 * characters in this are bitwise greater than the characters
412 * in <code>srcText</code>.
413 * @stable ICU 2.0
414 */
415 inline int8_t compare(int32_t start,
416 int32_t length,
417 const UnicodeString& srcText,
418 int32_t srcStart,
419 int32_t srcLength) const;
420
421 /**
422 * Compare the characters bitwise in this UnicodeString with the first
423 * <TT>srcLength</TT> characters in <TT>srcChars</TT>.
424 * @param srcChars The characters to compare to this UnicodeString.
425 * @param srcLength the number of characters in <TT>srcChars</TT> to compare
426 * @return The result of bitwise character comparison: 0 if this
427 * contains the same characters as <code>srcChars</code>, -1 if the characters in
428 * this are bitwise less than the characters in <code>srcChars</code>, +1 if the
429 * characters in this are bitwise greater than the characters
430 * in <code>srcChars</code>.
431 * @stable ICU 2.0
432 */
433 inline int8_t compare(ConstChar16Ptr srcChars,
434 int32_t srcLength) const;
435
436 /**
437 * Compare the characters bitwise in the range
438 * [<TT>start</TT>, <TT>start + length</TT>) with the first
439 * <TT>length</TT> characters in <TT>srcChars</TT>
440 * @param start the offset at which the compare operation begins
441 * @param length the number of characters to compare.
442 * @param srcChars the characters to be compared
443 * @return The result of bitwise character comparison: 0 if this
444 * contains the same characters as <code>srcChars</code>, -1 if the characters in
445 * this are bitwise less than the characters in <code>srcChars</code>, +1 if the
446 * characters in this are bitwise greater than the characters
447 * in <code>srcChars</code>.
448 * @stable ICU 2.0
449 */
450 inline int8_t compare(int32_t start,
451 int32_t length,
452 const char16_t *srcChars) const;
453
454 /**
455 * Compare the characters bitwise in the range
456 * [<TT>start</TT>, <TT>start + length</TT>) with the characters
457 * in <TT>srcChars</TT> in the range
458 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
459 * @param start the offset at which the compare operation begins
460 * @param length the number of characters in this to compare
461 * @param srcChars the characters to be compared
462 * @param srcStart the offset into <TT>srcChars</TT> to start comparison
463 * @param srcLength the number of characters in <TT>srcChars</TT> to compare
464 * @return The result of bitwise character comparison: 0 if this
465 * contains the same characters as <code>srcChars</code>, -1 if the characters in
466 * this are bitwise less than the characters in <code>srcChars</code>, +1 if the
467 * characters in this are bitwise greater than the characters
468 * in <code>srcChars</code>.
469 * @stable ICU 2.0
470 */
471 inline int8_t compare(int32_t start,
472 int32_t length,
473 const char16_t *srcChars,
474 int32_t srcStart,
475 int32_t srcLength) const;
476
477 /**
478 * Compare the characters bitwise in the range
479 * [<TT>start</TT>, <TT>limit</TT>) with the characters
480 * in <TT>srcText</TT> in the range
481 * [<TT>srcStart</TT>, <TT>srcLimit</TT>).
482 * @param start the offset at which the compare operation begins
483 * @param limit the offset immediately following the compare operation
484 * @param srcText the text to be compared
485 * @param srcStart the offset into <TT>srcText</TT> to start comparison
486 * @param srcLimit the offset into <TT>srcText</TT> to limit comparison
487 * @return The result of bitwise character comparison: 0 if this
488 * contains the same characters as <code>srcText</code>, -1 if the characters in
489 * this are bitwise less than the characters in <code>srcText</code>, +1 if the
490 * characters in this are bitwise greater than the characters
491 * in <code>srcText</code>.
492 * @stable ICU 2.0
493 */
494 inline int8_t compareBetween(int32_t start,
495 int32_t limit,
496 const UnicodeString& srcText,
497 int32_t srcStart,
498 int32_t srcLimit) const;
499
500 /**
501 * Compare two Unicode strings in code point order.
502 * The result may be different from the results of compare(), operator<, etc.
503 * if supplementary characters are present:
504 *
505 * In UTF-16, supplementary characters (with code points U+10000 and above) are
506 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
507 * which means that they compare as less than some other BMP characters like U+feff.
508 * This function compares Unicode strings in code point order.
509 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
510 *
511 * @param text Another string to compare this one to.
512 * @return a negative/zero/positive integer corresponding to whether
513 * this string is less than/equal to/greater than the second one
514 * in code point order
515 * @stable ICU 2.0
516 */
517 inline int8_t compareCodePointOrder(const UnicodeString& text) const;
518
519 /**
520 * Compare two Unicode strings in code point order.
521 * The result may be different from the results of compare(), operator<, etc.
522 * if supplementary characters are present:
523 *
524 * In UTF-16, supplementary characters (with code points U+10000 and above) are
525 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
526 * which means that they compare as less than some other BMP characters like U+feff.
527 * This function compares Unicode strings in code point order.
528 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
529 *
530 * @param start The start offset in this string at which the compare operation begins.
531 * @param length The number of code units from this string to compare.
532 * @param srcText Another string to compare this one to.
533 * @return a negative/zero/positive integer corresponding to whether
534 * this string is less than/equal to/greater than the second one
535 * in code point order
536 * @stable ICU 2.0
537 */
538 inline int8_t compareCodePointOrder(int32_t start,
539 int32_t length,
540 const UnicodeString& srcText) const;
541
542 /**
543 * Compare two Unicode strings in code point order.
544 * The result may be different from the results of compare(), operator<, etc.
545 * if supplementary characters are present:
546 *
547 * In UTF-16, supplementary characters (with code points U+10000 and above) are
548 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
549 * which means that they compare as less than some other BMP characters like U+feff.
550 * This function compares Unicode strings in code point order.
551 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
552 *
553 * @param start The start offset in this string at which the compare operation begins.
554 * @param length The number of code units from this string to compare.
555 * @param srcText Another string to compare this one to.
556 * @param srcStart The start offset in that string at which the compare operation begins.
557 * @param srcLength The number of code units from that string to compare.
558 * @return a negative/zero/positive integer corresponding to whether
559 * this string is less than/equal to/greater than the second one
560 * in code point order
561 * @stable ICU 2.0
562 */
563 inline int8_t compareCodePointOrder(int32_t start,
564 int32_t length,
565 const UnicodeString& srcText,
566 int32_t srcStart,
567 int32_t srcLength) const;
568
569 /**
570 * Compare two Unicode strings in code point order.
571 * The result may be different from the results of compare(), operator<, etc.
572 * if supplementary characters are present:
573 *
574 * In UTF-16, supplementary characters (with code points U+10000 and above) are
575 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
576 * which means that they compare as less than some other BMP characters like U+feff.
577 * This function compares Unicode strings in code point order.
578 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
579 *
580 * @param srcChars A pointer to another string to compare this one to.
581 * @param srcLength The number of code units from that string to compare.
582 * @return a negative/zero/positive integer corresponding to whether
583 * this string is less than/equal to/greater than the second one
584 * in code point order
585 * @stable ICU 2.0
586 */
587 inline int8_t compareCodePointOrder(ConstChar16Ptr srcChars,
588 int32_t srcLength) const;
589
590 /**
591 * Compare two Unicode strings in code point order.
592 * The result may be different from the results of compare(), operator<, etc.
593 * if supplementary characters are present:
594 *
595 * In UTF-16, supplementary characters (with code points U+10000 and above) are
596 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
597 * which means that they compare as less than some other BMP characters like U+feff.
598 * This function compares Unicode strings in code point order.
599 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
600 *
601 * @param start The start offset in this string at which the compare operation begins.
602 * @param length The number of code units from this string to compare.
603 * @param srcChars A pointer to another string to compare this one to.
604 * @return a negative/zero/positive integer corresponding to whether
605 * this string is less than/equal to/greater than the second one
606 * in code point order
607 * @stable ICU 2.0
608 */
609 inline int8_t compareCodePointOrder(int32_t start,
610 int32_t length,
611 const char16_t *srcChars) const;
612
613 /**
614 * Compare two Unicode strings in code point order.
615 * The result may be different from the results of compare(), operator<, etc.
616 * if supplementary characters are present:
617 *
618 * In UTF-16, supplementary characters (with code points U+10000 and above) are
619 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
620 * which means that they compare as less than some other BMP characters like U+feff.
621 * This function compares Unicode strings in code point order.
622 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
623 *
624 * @param start The start offset in this string at which the compare operation begins.
625 * @param length The number of code units from this string to compare.
626 * @param srcChars A pointer to another string to compare this one to.
627 * @param srcStart The start offset in that string at which the compare operation begins.
628 * @param srcLength The number of code units from that string to compare.
629 * @return a negative/zero/positive integer corresponding to whether
630 * this string is less than/equal to/greater than the second one
631 * in code point order
632 * @stable ICU 2.0
633 */
634 inline int8_t compareCodePointOrder(int32_t start,
635 int32_t length,
636 const char16_t *srcChars,
637 int32_t srcStart,
638 int32_t srcLength) const;
639
640 /**
641 * Compare two Unicode strings in code point order.
642 * The result may be different from the results of compare(), operator<, etc.
643 * if supplementary characters are present:
644 *
645 * In UTF-16, supplementary characters (with code points U+10000 and above) are
646 * stored with pairs of surrogate code units. These have values from 0xd800 to 0xdfff,
647 * which means that they compare as less than some other BMP characters like U+feff.
648 * This function compares Unicode strings in code point order.
649 * If either of the UTF-16 strings is malformed (i.e., it contains unpaired surrogates), then the result is not defined.
650 *
651 * @param start The start offset in this string at which the compare operation begins.
652 * @param limit The offset after the last code unit from this string to compare.
653 * @param srcText Another string to compare this one to.
654 * @param srcStart The start offset in that string at which the compare operation begins.
655 * @param srcLimit The offset after the last code unit from that string to compare.
656 * @return a negative/zero/positive integer corresponding to whether
657 * this string is less than/equal to/greater than the second one
658 * in code point order
659 * @stable ICU 2.0
660 */
661 inline int8_t compareCodePointOrderBetween(int32_t start,
662 int32_t limit,
663 const UnicodeString& srcText,
664 int32_t srcStart,
665 int32_t srcLimit) const;
666
667 /**
668 * Compare two strings case-insensitively using full case folding.
669 * This is equivalent to this->foldCase(options).compare(text.foldCase(options)).
670 *
671 * @param text Another string to compare this one to.
672 * @param options A bit set of options:
673 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
674 * Comparison in code unit order with default case folding.
675 *
676 * - U_COMPARE_CODE_POINT_ORDER
677 * Set to choose code point order instead of code unit order
678 * (see u_strCompare for details).
679 *
680 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
681 *
682 * @return A negative, zero, or positive integer indicating the comparison result.
683 * @stable ICU 2.0
684 */
685 inline int8_t caseCompare(const UnicodeString& text, uint32_t options) const;
686
687 /**
688 * Compare two strings case-insensitively using full case folding.
689 * This is equivalent to this->foldCase(options).compare(srcText.foldCase(options)).
690 *
691 * @param start The start offset in this string at which the compare operation begins.
692 * @param length The number of code units from this string to compare.
693 * @param srcText Another string to compare this one to.
694 * @param options A bit set of options:
695 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
696 * Comparison in code unit order with default case folding.
697 *
698 * - U_COMPARE_CODE_POINT_ORDER
699 * Set to choose code point order instead of code unit order
700 * (see u_strCompare for details).
701 *
702 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
703 *
704 * @return A negative, zero, or positive integer indicating the comparison result.
705 * @stable ICU 2.0
706 */
707 inline int8_t caseCompare(int32_t start,
708 int32_t length,
709 const UnicodeString& srcText,
710 uint32_t options) const;
711
712 /**
713 * Compare two strings case-insensitively using full case folding.
714 * This is equivalent to this->foldCase(options).compare(srcText.foldCase(options)).
715 *
716 * @param start The start offset in this string at which the compare operation begins.
717 * @param length The number of code units from this string to compare.
718 * @param srcText Another string to compare this one to.
719 * @param srcStart The start offset in that string at which the compare operation begins.
720 * @param srcLength The number of code units from that string to compare.
721 * @param options A bit set of options:
722 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
723 * Comparison in code unit order with default case folding.
724 *
725 * - U_COMPARE_CODE_POINT_ORDER
726 * Set to choose code point order instead of code unit order
727 * (see u_strCompare for details).
728 *
729 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
730 *
731 * @return A negative, zero, or positive integer indicating the comparison result.
732 * @stable ICU 2.0
733 */
734 inline int8_t caseCompare(int32_t start,
735 int32_t length,
736 const UnicodeString& srcText,
737 int32_t srcStart,
738 int32_t srcLength,
739 uint32_t options) const;
740
741 /**
742 * Compare two strings case-insensitively using full case folding.
743 * This is equivalent to this->foldCase(options).compare(srcChars.foldCase(options)).
744 *
745 * @param srcChars A pointer to another string to compare this one to.
746 * @param srcLength The number of code units from that string to compare.
747 * @param options A bit set of options:
748 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
749 * Comparison in code unit order with default case folding.
750 *
751 * - U_COMPARE_CODE_POINT_ORDER
752 * Set to choose code point order instead of code unit order
753 * (see u_strCompare for details).
754 *
755 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
756 *
757 * @return A negative, zero, or positive integer indicating the comparison result.
758 * @stable ICU 2.0
759 */
760 inline int8_t caseCompare(ConstChar16Ptr srcChars,
761 int32_t srcLength,
762 uint32_t options) const;
763
764 /**
765 * Compare two strings case-insensitively using full case folding.
766 * This is equivalent to this->foldCase(options).compare(srcChars.foldCase(options)).
767 *
768 * @param start The start offset in this string at which the compare operation begins.
769 * @param length The number of code units from this string to compare.
770 * @param srcChars A pointer to another string to compare this one to.
771 * @param options A bit set of options:
772 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
773 * Comparison in code unit order with default case folding.
774 *
775 * - U_COMPARE_CODE_POINT_ORDER
776 * Set to choose code point order instead of code unit order
777 * (see u_strCompare for details).
778 *
779 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
780 *
781 * @return A negative, zero, or positive integer indicating the comparison result.
782 * @stable ICU 2.0
783 */
784 inline int8_t caseCompare(int32_t start,
785 int32_t length,
786 const char16_t *srcChars,
787 uint32_t options) const;
788
789 /**
790 * Compare two strings case-insensitively using full case folding.
791 * This is equivalent to this->foldCase(options).compare(srcChars.foldCase(options)).
792 *
793 * @param start The start offset in this string at which the compare operation begins.
794 * @param length The number of code units from this string to compare.
795 * @param srcChars A pointer to another string to compare this one to.
796 * @param srcStart The start offset in that string at which the compare operation begins.
797 * @param srcLength The number of code units from that string to compare.
798 * @param options A bit set of options:
799 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
800 * Comparison in code unit order with default case folding.
801 *
802 * - U_COMPARE_CODE_POINT_ORDER
803 * Set to choose code point order instead of code unit order
804 * (see u_strCompare for details).
805 *
806 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
807 *
808 * @return A negative, zero, or positive integer indicating the comparison result.
809 * @stable ICU 2.0
810 */
811 inline int8_t caseCompare(int32_t start,
812 int32_t length,
813 const char16_t *srcChars,
814 int32_t srcStart,
815 int32_t srcLength,
816 uint32_t options) const;
817
818 /**
819 * Compare two strings case-insensitively using full case folding.
820 * This is equivalent to this->foldCase(options).compareBetween(text.foldCase(options)).
821 *
822 * @param start The start offset in this string at which the compare operation begins.
823 * @param limit The offset after the last code unit from this string to compare.
824 * @param srcText Another string to compare this one to.
825 * @param srcStart The start offset in that string at which the compare operation begins.
826 * @param srcLimit The offset after the last code unit from that string to compare.
827 * @param options A bit set of options:
828 * - U_FOLD_CASE_DEFAULT or 0 is used for default options:
829 * Comparison in code unit order with default case folding.
830 *
831 * - U_COMPARE_CODE_POINT_ORDER
832 * Set to choose code point order instead of code unit order
833 * (see u_strCompare for details).
834 *
835 * - U_FOLD_CASE_EXCLUDE_SPECIAL_I
836 *
837 * @return A negative, zero, or positive integer indicating the comparison result.
838 * @stable ICU 2.0
839 */
840 inline int8_t caseCompareBetween(int32_t start,
841 int32_t limit,
842 const UnicodeString& srcText,
843 int32_t srcStart,
844 int32_t srcLimit,
845 uint32_t options) const;
846
847 /**
848 * Determine if this starts with the characters in <TT>text</TT>
849 * @param text The text to match.
850 * @return TRUE if this starts with the characters in <TT>text</TT>,
851 * FALSE otherwise
852 * @stable ICU 2.0
853 */
854 inline UBool startsWith(const UnicodeString& text) const;
855
856 /**
857 * Determine if this starts with the characters in <TT>srcText</TT>
858 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
859 * @param srcText The text to match.
860 * @param srcStart the offset into <TT>srcText</TT> to start matching
861 * @param srcLength the number of characters in <TT>srcText</TT> to match
862 * @return TRUE if this starts with the characters in <TT>text</TT>,
863 * FALSE otherwise
864 * @stable ICU 2.0
865 */
866 inline UBool startsWith(const UnicodeString& srcText,
867 int32_t srcStart,
868 int32_t srcLength) const;
869
870 /**
871 * Determine if this starts with the characters in <TT>srcChars</TT>
872 * @param srcChars The characters to match.
873 * @param srcLength the number of characters in <TT>srcChars</TT>
874 * @return TRUE if this starts with the characters in <TT>srcChars</TT>,
875 * FALSE otherwise
876 * @stable ICU 2.0
877 */
878 inline UBool startsWith(ConstChar16Ptr srcChars,
879 int32_t srcLength) const;
880
881 /**
882 * Determine if this ends with the characters in <TT>srcChars</TT>
883 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
884 * @param srcChars The characters to match.
885 * @param srcStart the offset into <TT>srcText</TT> to start matching
886 * @param srcLength the number of characters in <TT>srcChars</TT> to match
887 * @return TRUE if this ends with the characters in <TT>srcChars</TT>, FALSE otherwise
888 * @stable ICU 2.0
889 */
890 inline UBool startsWith(const char16_t *srcChars,
891 int32_t srcStart,
892 int32_t srcLength) const;
893
894 /**
895 * Determine if this ends with the characters in <TT>text</TT>
896 * @param text The text to match.
897 * @return TRUE if this ends with the characters in <TT>text</TT>,
898 * FALSE otherwise
899 * @stable ICU 2.0
900 */
901 inline UBool endsWith(const UnicodeString& text) const;
902
903 /**
904 * Determine if this ends with the characters in <TT>srcText</TT>
905 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
906 * @param srcText The text to match.
907 * @param srcStart the offset into <TT>srcText</TT> to start matching
908 * @param srcLength the number of characters in <TT>srcText</TT> to match
909 * @return TRUE if this ends with the characters in <TT>text</TT>,
910 * FALSE otherwise
911 * @stable ICU 2.0
912 */
913 inline UBool endsWith(const UnicodeString& srcText,
914 int32_t srcStart,
915 int32_t srcLength) const;
916
917 /**
918 * Determine if this ends with the characters in <TT>srcChars</TT>
919 * @param srcChars The characters to match.
920 * @param srcLength the number of characters in <TT>srcChars</TT>
921 * @return TRUE if this ends with the characters in <TT>srcChars</TT>,
922 * FALSE otherwise
923 * @stable ICU 2.0
924 */
925 inline UBool endsWith(ConstChar16Ptr srcChars,
926 int32_t srcLength) const;
927
928 /**
929 * Determine if this ends with the characters in <TT>srcChars</TT>
930 * in the range [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
931 * @param srcChars The characters to match.
932 * @param srcStart the offset into <TT>srcText</TT> to start matching
933 * @param srcLength the number of characters in <TT>srcChars</TT> to match
934 * @return TRUE if this ends with the characters in <TT>srcChars</TT>,
935 * FALSE otherwise
936 * @stable ICU 2.0
937 */
938 inline UBool endsWith(const char16_t *srcChars,
939 int32_t srcStart,
940 int32_t srcLength) const;
941
942
943 /* Searching - bitwise only */
944
945 /**
946 * Locate in this the first occurrence of the characters in <TT>text</TT>,
947 * using bitwise comparison.
948 * @param text The text to search for.
949 * @return The offset into this of the start of <TT>text</TT>,
950 * or -1 if not found.
951 * @stable ICU 2.0
952 */
953 inline int32_t indexOf(const UnicodeString& text) const;
954
955 /**
956 * Locate in this the first occurrence of the characters in <TT>text</TT>
957 * starting at offset <TT>start</TT>, using bitwise comparison.
958 * @param text The text to search for.
959 * @param start The offset at which searching will start.
960 * @return The offset into this of the start of <TT>text</TT>,
961 * or -1 if not found.
962 * @stable ICU 2.0
963 */
964 inline int32_t indexOf(const UnicodeString& text,
965 int32_t start) const;
966
967 /**
968 * Locate in this the first occurrence in the range
969 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
970 * in <TT>text</TT>, using bitwise comparison.
971 * @param text The text to search for.
972 * @param start The offset at which searching will start.
973 * @param length The number of characters to search
974 * @return The offset into this of the start of <TT>text</TT>,
975 * or -1 if not found.
976 * @stable ICU 2.0
977 */
978 inline int32_t indexOf(const UnicodeString& text,
979 int32_t start,
980 int32_t length) const;
981
982 /**
983 * Locate in this the first occurrence in the range
984 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
985 * in <TT>srcText</TT> in the range
986 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>),
987 * using bitwise comparison.
988 * @param srcText The text to search for.
989 * @param srcStart the offset into <TT>srcText</TT> at which
990 * to start matching
991 * @param srcLength the number of characters in <TT>srcText</TT> to match
992 * @param start the offset into this at which to start matching
993 * @param length the number of characters in this to search
994 * @return The offset into this of the start of <TT>text</TT>,
995 * or -1 if not found.
996 * @stable ICU 2.0
997 */
998 inline int32_t indexOf(const UnicodeString& srcText,
999 int32_t srcStart,
1000 int32_t srcLength,
1001 int32_t start,
1002 int32_t length) const;
1003
1004 /**
1005 * Locate in this the first occurrence of the characters in
1006 * <TT>srcChars</TT>
1007 * starting at offset <TT>start</TT>, using bitwise comparison.
1008 * @param srcChars The text to search for.
1009 * @param srcLength the number of characters in <TT>srcChars</TT> to match
1010 * @param start the offset into this at which to start matching
1011 * @return The offset into this of the start of <TT>text</TT>,
1012 * or -1 if not found.
1013 * @stable ICU 2.0
1014 */
1015 inline int32_t indexOf(const char16_t *srcChars,
1016 int32_t srcLength,
1017 int32_t start) const;
1018
1019 /**
1020 * Locate in this the first occurrence in the range
1021 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
1022 * in <TT>srcChars</TT>, using bitwise comparison.
1023 * @param srcChars The text to search for.
1024 * @param srcLength the number of characters in <TT>srcChars</TT>
1025 * @param start The offset at which searching will start.
1026 * @param length The number of characters to search
1027 * @return The offset into this of the start of <TT>srcChars</TT>,
1028 * or -1 if not found.
1029 * @stable ICU 2.0
1030 */
1031 inline int32_t indexOf(ConstChar16Ptr srcChars,
1032 int32_t srcLength,
1033 int32_t start,
1034 int32_t length) const;
1035
1036 /**
1037 * Locate in this the first occurrence in the range
1038 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
1039 * in <TT>srcChars</TT> in the range
1040 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>),
1041 * using bitwise comparison.
1042 * @param srcChars The text to search for.
1043 * @param srcStart the offset into <TT>srcChars</TT> at which
1044 * to start matching
1045 * @param srcLength the number of characters in <TT>srcChars</TT> to match
1046 * @param start the offset into this at which to start matching
1047 * @param length the number of characters in this to search
1048 * @return The offset into this of the start of <TT>text</TT>,
1049 * or -1 if not found.
1050 * @stable ICU 2.0
1051 */
1052 int32_t indexOf(const char16_t *srcChars,
1053 int32_t srcStart,
1054 int32_t srcLength,
1055 int32_t start,
1056 int32_t length) const;
1057
1058 /**
1059 * Locate in this the first occurrence of the BMP code point <code>c</code>,
1060 * using bitwise comparison.
1061 * @param c The code unit to search for.
1062 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1063 * @stable ICU 2.0
1064 */
1065 inline int32_t indexOf(char16_t c) const;
1066
1067 /**
1068 * Locate in this the first occurrence of the code point <TT>c</TT>,
1069 * using bitwise comparison.
1070 *
1071 * @param c The code point to search for.
1072 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1073 * @stable ICU 2.0
1074 */
1075 inline int32_t indexOf(UChar32 c) const;
1076
1077 /**
1078 * Locate in this the first occurrence of the BMP code point <code>c</code>,
1079 * starting at offset <TT>start</TT>, using bitwise comparison.
1080 * @param c The code unit to search for.
1081 * @param start The offset at which searching will start.
1082 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1083 * @stable ICU 2.0
1084 */
1085 inline int32_t indexOf(char16_t c,
1086 int32_t start) const;
1087
1088 /**
1089 * Locate in this the first occurrence of the code point <TT>c</TT>
1090 * starting at offset <TT>start</TT>, using bitwise comparison.
1091 *
1092 * @param c The code point to search for.
1093 * @param start The offset at which searching will start.
1094 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1095 * @stable ICU 2.0
1096 */
1097 inline int32_t indexOf(UChar32 c,
1098 int32_t start) const;
1099
1100 /**
1101 * Locate in this the first occurrence of the BMP code point <code>c</code>
1102 * in the range [<TT>start</TT>, <TT>start + length</TT>),
1103 * using bitwise comparison.
1104 * @param c The code unit to search for.
1105 * @param start the offset into this at which to start matching
1106 * @param length the number of characters in this to search
1107 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1108 * @stable ICU 2.0
1109 */
1110 inline int32_t indexOf(char16_t c,
1111 int32_t start,
1112 int32_t length) const;
1113
1114 /**
1115 * Locate in this the first occurrence of the code point <TT>c</TT>
1116 * in the range [<TT>start</TT>, <TT>start + length</TT>),
1117 * using bitwise comparison.
1118 *
1119 * @param c The code point to search for.
1120 * @param start the offset into this at which to start matching
1121 * @param length the number of characters in this to search
1122 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1123 * @stable ICU 2.0
1124 */
1125 inline int32_t indexOf(UChar32 c,
1126 int32_t start,
1127 int32_t length) const;
1128
1129 /**
1130 * Locate in this the last occurrence of the characters in <TT>text</TT>,
1131 * using bitwise comparison.
1132 * @param text The text to search for.
1133 * @return The offset into this of the start of <TT>text</TT>,
1134 * or -1 if not found.
1135 * @stable ICU 2.0
1136 */
1137 inline int32_t lastIndexOf(const UnicodeString& text) const;
1138
1139 /**
1140 * Locate in this the last occurrence of the characters in <TT>text</TT>
1141 * starting at offset <TT>start</TT>, using bitwise comparison.
1142 * @param text The text to search for.
1143 * @param start The offset at which searching will start.
1144 * @return The offset into this of the start of <TT>text</TT>,
1145 * or -1 if not found.
1146 * @stable ICU 2.0
1147 */
1148 inline int32_t lastIndexOf(const UnicodeString& text,
1149 int32_t start) const;
1150
1151 /**
1152 * Locate in this the last occurrence in the range
1153 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
1154 * in <TT>text</TT>, using bitwise comparison.
1155 * @param text The text to search for.
1156 * @param start The offset at which searching will start.
1157 * @param length The number of characters to search
1158 * @return The offset into this of the start of <TT>text</TT>,
1159 * or -1 if not found.
1160 * @stable ICU 2.0
1161 */
1162 inline int32_t lastIndexOf(const UnicodeString& text,
1163 int32_t start,
1164 int32_t length) const;
1165
1166 /**
1167 * Locate in this the last occurrence in the range
1168 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
1169 * in <TT>srcText</TT> in the range
1170 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>),
1171 * using bitwise comparison.
1172 * @param srcText The text to search for.
1173 * @param srcStart the offset into <TT>srcText</TT> at which
1174 * to start matching
1175 * @param srcLength the number of characters in <TT>srcText</TT> to match
1176 * @param start the offset into this at which to start matching
1177 * @param length the number of characters in this to search
1178 * @return The offset into this of the start of <TT>text</TT>,
1179 * or -1 if not found.
1180 * @stable ICU 2.0
1181 */
1182 inline int32_t lastIndexOf(const UnicodeString& srcText,
1183 int32_t srcStart,
1184 int32_t srcLength,
1185 int32_t start,
1186 int32_t length) const;
1187
1188 /**
1189 * Locate in this the last occurrence of the characters in <TT>srcChars</TT>
1190 * starting at offset <TT>start</TT>, using bitwise comparison.
1191 * @param srcChars The text to search for.
1192 * @param srcLength the number of characters in <TT>srcChars</TT> to match
1193 * @param start the offset into this at which to start matching
1194 * @return The offset into this of the start of <TT>text</TT>,
1195 * or -1 if not found.
1196 * @stable ICU 2.0
1197 */
1198 inline int32_t lastIndexOf(const char16_t *srcChars,
1199 int32_t srcLength,
1200 int32_t start) const;
1201
1202 /**
1203 * Locate in this the last occurrence in the range
1204 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
1205 * in <TT>srcChars</TT>, using bitwise comparison.
1206 * @param srcChars The text to search for.
1207 * @param srcLength the number of characters in <TT>srcChars</TT>
1208 * @param start The offset at which searching will start.
1209 * @param length The number of characters to search
1210 * @return The offset into this of the start of <TT>srcChars</TT>,
1211 * or -1 if not found.
1212 * @stable ICU 2.0
1213 */
1214 inline int32_t lastIndexOf(ConstChar16Ptr srcChars,
1215 int32_t srcLength,
1216 int32_t start,
1217 int32_t length) const;
1218
1219 /**
1220 * Locate in this the last occurrence in the range
1221 * [<TT>start</TT>, <TT>start + length</TT>) of the characters
1222 * in <TT>srcChars</TT> in the range
1223 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>),
1224 * using bitwise comparison.
1225 * @param srcChars The text to search for.
1226 * @param srcStart the offset into <TT>srcChars</TT> at which
1227 * to start matching
1228 * @param srcLength the number of characters in <TT>srcChars</TT> to match
1229 * @param start the offset into this at which to start matching
1230 * @param length the number of characters in this to search
1231 * @return The offset into this of the start of <TT>text</TT>,
1232 * or -1 if not found.
1233 * @stable ICU 2.0
1234 */
1235 int32_t lastIndexOf(const char16_t *srcChars,
1236 int32_t srcStart,
1237 int32_t srcLength,
1238 int32_t start,
1239 int32_t length) const;
1240
1241 /**
1242 * Locate in this the last occurrence of the BMP code point <code>c</code>,
1243 * using bitwise comparison.
1244 * @param c The code unit to search for.
1245 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1246 * @stable ICU 2.0
1247 */
1248 inline int32_t lastIndexOf(char16_t c) const;
1249
1250 /**
1251 * Locate in this the last occurrence of the code point <TT>c</TT>,
1252 * using bitwise comparison.
1253 *
1254 * @param c The code point to search for.
1255 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1256 * @stable ICU 2.0
1257 */
1258 inline int32_t lastIndexOf(UChar32 c) const;
1259
1260 /**
1261 * Locate in this the last occurrence of the BMP code point <code>c</code>
1262 * starting at offset <TT>start</TT>, using bitwise comparison.
1263 * @param c The code unit to search for.
1264 * @param start The offset at which searching will start.
1265 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1266 * @stable ICU 2.0
1267 */
1268 inline int32_t lastIndexOf(char16_t c,
1269 int32_t start) const;
1270
1271 /**
1272 * Locate in this the last occurrence of the code point <TT>c</TT>
1273 * starting at offset <TT>start</TT>, using bitwise comparison.
1274 *
1275 * @param c The code point to search for.
1276 * @param start The offset at which searching will start.
1277 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1278 * @stable ICU 2.0
1279 */
1280 inline int32_t lastIndexOf(UChar32 c,
1281 int32_t start) const;
1282
1283 /**
1284 * Locate in this the last occurrence of the BMP code point <code>c</code>
1285 * in the range [<TT>start</TT>, <TT>start + length</TT>),
1286 * using bitwise comparison.
1287 * @param c The code unit to search for.
1288 * @param start the offset into this at which to start matching
1289 * @param length the number of characters in this to search
1290 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1291 * @stable ICU 2.0
1292 */
1293 inline int32_t lastIndexOf(char16_t c,
1294 int32_t start,
1295 int32_t length) const;
1296
1297 /**
1298 * Locate in this the last occurrence of the code point <TT>c</TT>
1299 * in the range [<TT>start</TT>, <TT>start + length</TT>),
1300 * using bitwise comparison.
1301 *
1302 * @param c The code point to search for.
1303 * @param start the offset into this at which to start matching
1304 * @param length the number of characters in this to search
1305 * @return The offset into this of <TT>c</TT>, or -1 if not found.
1306 * @stable ICU 2.0
1307 */
1308 inline int32_t lastIndexOf(UChar32 c,
1309 int32_t start,
1310 int32_t length) const;
1311
1312
1313 /* Character access */
1314
1315 /**
1316 * Return the code unit at offset <tt>offset</tt>.
1317 * If the offset is not valid (0..length()-1) then U+ffff is returned.
1318 * @param offset a valid offset into the text
1319 * @return the code unit at offset <tt>offset</tt>
1320 * or 0xffff if the offset is not valid for this string
1321 * @stable ICU 2.0
1322 */
1323 inline char16_t charAt(int32_t offset) const;
1324
1325 /**
1326 * Return the code unit at offset <tt>offset</tt>.
1327 * If the offset is not valid (0..length()-1) then U+ffff is returned.
1328 * @param offset a valid offset into the text
1329 * @return the code unit at offset <tt>offset</tt>
1330 * @stable ICU 2.0
1331 */
1332 inline char16_t operator[] (int32_t offset) const;
1333
1334 /**
1335 * Return the code point that contains the code unit
1336 * at offset <tt>offset</tt>.
1337 * If the offset is not valid (0..length()-1) then U+ffff is returned.
1338 * @param offset a valid offset into the text
1339 * that indicates the text offset of any of the code units
1340 * that will be assembled into a code point (21-bit value) and returned
1341 * @return the code point of text at <tt>offset</tt>
1342 * or 0xffff if the offset is not valid for this string
1343 * @stable ICU 2.0
1344 */
1345 UChar32 char32At(int32_t offset) const;
1346
1347 /**
1348 * Adjust a random-access offset so that
1349 * it points to the beginning of a Unicode character.
1350 * The offset that is passed in points to
1351 * any code unit of a code point,
1352 * while the returned offset will point to the first code unit
1353 * of the same code point.
1354 * In UTF-16, if the input offset points to a second surrogate
1355 * of a surrogate pair, then the returned offset will point
1356 * to the first surrogate.
1357 * @param offset a valid offset into one code point of the text
1358 * @return offset of the first code unit of the same code point
1359 * @see U16_SET_CP_START
1360 * @stable ICU 2.0
1361 */
1362 int32_t getChar32Start(int32_t offset) const;
1363
1364 /**
1365 * Adjust a random-access offset so that
1366 * it points behind a Unicode character.
1367 * The offset that is passed in points behind
1368 * any code unit of a code point,
1369 * while the returned offset will point behind the last code unit
1370 * of the same code point.
1371 * In UTF-16, if the input offset points behind the first surrogate
1372 * (i.e., to the second surrogate)
1373 * of a surrogate pair, then the returned offset will point
1374 * behind the second surrogate (i.e., to the first surrogate).
1375 * @param offset a valid offset after any code unit of a code point of the text
1376 * @return offset of the first code unit after the same code point
1377 * @see U16_SET_CP_LIMIT
1378 * @stable ICU 2.0
1379 */
1380 int32_t getChar32Limit(int32_t offset) const;
1381
1382 /**
1383 * Move the code unit index along the string by delta code points.
1384 * Interpret the input index as a code unit-based offset into the string,
1385 * move the index forward or backward by delta code points, and
1386 * return the resulting index.
1387 * The input index should point to the first code unit of a code point,
1388 * if there is more than one.
1389 *
1390 * Both input and output indexes are code unit-based as for all
1391 * string indexes/offsets in ICU (and other libraries, like MBCS char*).
1392 * If delta<0 then the index is moved backward (toward the start of the string).
1393 * If delta>0 then the index is moved forward (toward the end of the string).
1394 *
1395 * This behaves like CharacterIterator::move32(delta, kCurrent).
1396 *
1397 * Behavior for out-of-bounds indexes:
1398 * <code>moveIndex32</code> pins the input index to 0..length(), i.e.,
1399 * if the input index<0 then it is pinned to 0;
1400 * if it is index>length() then it is pinned to length().
1401 * Afterwards, the index is moved by <code>delta</code> code points
1402 * forward or backward,
1403 * but no further backward than to 0 and no further forward than to length().
1404 * The resulting index return value will be in between 0 and length(), inclusively.
1405 *
1406 * Examples:
1407 * <pre>
1408 * // s has code points 'a' U+10000 'b' U+10ffff U+2029
1409 * UnicodeString s=UNICODE_STRING("a\\U00010000b\\U0010ffff\\u2029", 31).unescape();
1410 *
1411 * // initial index: position of U+10000
1412 * int32_t index=1;
1413 *
1414 * // the following examples will all result in index==4, position of U+10ffff
1415 *
1416 * // skip 2 code points from some position in the string
1417 * index=s.moveIndex32(index, 2); // skips U+10000 and 'b'
1418 *
1419 * // go to the 3rd code point from the start of s (0-based)
1420 * index=s.moveIndex32(0, 3); // skips 'a', U+10000, and 'b'
1421 *
1422 * // go to the next-to-last code point of s
1423 * index=s.moveIndex32(s.length(), -2); // backward-skips U+2029 and U+10ffff
1424 * </pre>
1425 *
1426 * @param index input code unit index
1427 * @param delta (signed) code point count to move the index forward or backward
1428 * in the string
1429 * @return the resulting code unit index
1430 * @stable ICU 2.0
1431 */
1432 int32_t moveIndex32(int32_t index, int32_t delta) const;
1433
1434 /* Substring extraction */
1435
1436 /**
1437 * Copy the characters in the range
1438 * [<tt>start</tt>, <tt>start + length</tt>) into the array <tt>dst</tt>,
1439 * beginning at <tt>dstStart</tt>.
1440 * If the string aliases to <code>dst</code> itself as an external buffer,
1441 * then extract() will not copy the contents.
1442 *
1443 * @param start offset of first character which will be copied into the array
1444 * @param length the number of characters to extract
1445 * @param dst array in which to copy characters. The length of <tt>dst</tt>
1446 * must be at least (<tt>dstStart + length</tt>).
1447 * @param dstStart the offset in <TT>dst</TT> where the first character
1448 * will be extracted
1449 * @stable ICU 2.0
1450 */
1451 inline void extract(int32_t start,
1452 int32_t length,
1453 Char16Ptr dst,
1454 int32_t dstStart = 0) const;
1455
1456 /**
1457 * Copy the contents of the string into dest.
1458 * This is a convenience function that
1459 * checks if there is enough space in dest,
1460 * extracts the entire string if possible,
1461 * and NUL-terminates dest if possible.
1462 *
1463 * If the string fits into dest but cannot be NUL-terminated
1464 * (length()==destCapacity) then the error code is set to U_STRING_NOT_TERMINATED_WARNING.
1465 * If the string itself does not fit into dest
1466 * (length()>destCapacity) then the error code is set to U_BUFFER_OVERFLOW_ERROR.
1467 *
1468 * If the string aliases to <code>dest</code> itself as an external buffer,
1469 * then extract() will not copy the contents.
1470 *
1471 * @param dest Destination string buffer.
1472 * @param destCapacity Number of char16_ts available at dest.
1473 * @param errorCode ICU error code.
1474 * @return length()
1475 * @stable ICU 2.0
1476 */
1477 int32_t
1478 extract(Char16Ptr dest, int32_t destCapacity,
1479 UErrorCode &errorCode) const;
1480
1481 /**
1482 * Copy the characters in the range
1483 * [<tt>start</tt>, <tt>start + length</tt>) into the UnicodeString
1484 * <tt>target</tt>.
1485 * @param start offset of first character which will be copied
1486 * @param length the number of characters to extract
1487 * @param target UnicodeString into which to copy characters.
1488 * @return A reference to <TT>target</TT>
1489 * @stable ICU 2.0
1490 */
1491 inline void extract(int32_t start,
1492 int32_t length,
1493 UnicodeString& target) const;
1494
1495 /**
1496 * Copy the characters in the range [<tt>start</tt>, <tt>limit</tt>)
1497 * into the array <tt>dst</tt>, beginning at <tt>dstStart</tt>.
1498 * @param start offset of first character which will be copied into the array
1499 * @param limit offset immediately following the last character to be copied
1500 * @param dst array in which to copy characters. The length of <tt>dst</tt>
1501 * must be at least (<tt>dstStart + (limit - start)</tt>).
1502 * @param dstStart the offset in <TT>dst</TT> where the first character
1503 * will be extracted
1504 * @stable ICU 2.0
1505 */
1506 inline void extractBetween(int32_t start,
1507 int32_t limit,
1508 char16_t *dst,
1509 int32_t dstStart = 0) const;
1510
1511 /**
1512 * Copy the characters in the range [<tt>start</tt>, <tt>limit</tt>)
1513 * into the UnicodeString <tt>target</tt>. Replaceable API.
1514 * @param start offset of first character which will be copied
1515 * @param limit offset immediately following the last character to be copied
1516 * @param target UnicodeString into which to copy characters.
1517 * @return A reference to <TT>target</TT>
1518 * @stable ICU 2.0
1519 */
1520 virtual void extractBetween(int32_t start,
1521 int32_t limit,
1522 UnicodeString& target) const;
1523
1524 /**
1525 * Copy the characters in the range
1526 * [<tt>start</TT>, <tt>start + startLength</TT>) into an array of characters.
1527 * All characters must be invariant (see utypes.h).
1528 * Use US_INV as the last, signature-distinguishing parameter.
1529 *
1530 * This function does not write any more than <code>targetCapacity</code>
1531 * characters but returns the length of the entire output string
1532 * so that one can allocate a larger buffer and call the function again
1533 * if necessary.
1534 * The output string is NUL-terminated if possible.
1535 *
1536 * @param start offset of first character which will be copied
1537 * @param startLength the number of characters to extract
1538 * @param target the target buffer for extraction, can be NULL
1539 * if targetLength is 0
1540 * @param targetCapacity the length of the target buffer
1541 * @param inv Signature-distinguishing paramater, use US_INV.
1542 * @return the output string length, not including the terminating NUL
1543 * @stable ICU 3.2
1544 */
1545 int32_t extract(int32_t start,
1546 int32_t startLength,
1547 char *target,
1548 int32_t targetCapacity,
1549 enum EInvariant inv) const;
1550
1551#if U_CHARSET_IS_UTF8 || !UCONFIG_NO_CONVERSION
1552
1553 /**
1554 * Copy the characters in the range
1555 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters
1556 * in the platform's default codepage.
1557 * This function does not write any more than <code>targetLength</code>
1558 * characters but returns the length of the entire output string
1559 * so that one can allocate a larger buffer and call the function again
1560 * if necessary.
1561 * The output string is NUL-terminated if possible.
1562 *
1563 * @param start offset of first character which will be copied
1564 * @param startLength the number of characters to extract
1565 * @param target the target buffer for extraction
1566 * @param targetLength the length of the target buffer
1567 * If <TT>target</TT> is NULL, then the number of bytes required for
1568 * <TT>target</TT> is returned.
1569 * @return the output string length, not including the terminating NUL
1570 * @stable ICU 2.0
1571 */
1572 int32_t extract(int32_t start,
1573 int32_t startLength,
1574 char *target,
1575 uint32_t targetLength) const;
1576
1577#endif
1578
1579#if !UCONFIG_NO_CONVERSION
1580
1581 /**
1582 * Copy the characters in the range
1583 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters
1584 * in a specified codepage.
1585 * The output string is NUL-terminated.
1586 *
1587 * Recommendation: For invariant-character strings use
1588 * extract(int32_t start, int32_t length, char *target, int32_t targetCapacity, enum EInvariant inv) const
1589 * because it avoids object code dependencies of UnicodeString on
1590 * the conversion code.
1591 *
1592 * @param start offset of first character which will be copied
1593 * @param startLength the number of characters to extract
1594 * @param target the target buffer for extraction
1595 * @param codepage the desired codepage for the characters. 0 has
1596 * the special meaning of the default codepage
1597 * If <code>codepage</code> is an empty string (<code>""</code>),
1598 * then a simple conversion is performed on the codepage-invariant
1599 * subset ("invariant characters") of the platform encoding. See utypes.h.
1600 * If <TT>target</TT> is NULL, then the number of bytes required for
1601 * <TT>target</TT> is returned. It is assumed that the target is big enough
1602 * to fit all of the characters.
1603 * @return the output string length, not including the terminating NUL
1604 * @stable ICU 2.0
1605 */
1606 inline int32_t extract(int32_t start,
1607 int32_t startLength,
1608 char *target,
1609 const char *codepage = 0) const;
1610
1611 /**
1612 * Copy the characters in the range
1613 * [<tt>start</TT>, <tt>start + length</TT>) into an array of characters
1614 * in a specified codepage.
1615 * This function does not write any more than <code>targetLength</code>
1616 * characters but returns the length of the entire output string
1617 * so that one can allocate a larger buffer and call the function again
1618 * if necessary.
1619 * The output string is NUL-terminated if possible.
1620 *
1621 * Recommendation: For invariant-character strings use
1622 * extract(int32_t start, int32_t length, char *target, int32_t targetCapacity, enum EInvariant inv) const
1623 * because it avoids object code dependencies of UnicodeString on
1624 * the conversion code.
1625 *
1626 * @param start offset of first character which will be copied
1627 * @param startLength the number of characters to extract
1628 * @param target the target buffer for extraction
1629 * @param targetLength the length of the target buffer
1630 * @param codepage the desired codepage for the characters. 0 has
1631 * the special meaning of the default codepage
1632 * If <code>codepage</code> is an empty string (<code>""</code>),
1633 * then a simple conversion is performed on the codepage-invariant
1634 * subset ("invariant characters") of the platform encoding. See utypes.h.
1635 * If <TT>target</TT> is NULL, then the number of bytes required for
1636 * <TT>target</TT> is returned.
1637 * @return the output string length, not including the terminating NUL
1638 * @stable ICU 2.0
1639 */
1640 int32_t extract(int32_t start,
1641 int32_t startLength,
1642 char *target,
1643 uint32_t targetLength,
1644 const char *codepage) const;
1645
1646 /**
1647 * Convert the UnicodeString into a codepage string using an existing UConverter.
1648 * The output string is NUL-terminated if possible.
1649 *
1650 * This function avoids the overhead of opening and closing a converter if
1651 * multiple strings are extracted.
1652 *
1653 * @param dest destination string buffer, can be NULL if destCapacity==0
1654 * @param destCapacity the number of chars available at dest
1655 * @param cnv the converter object to be used (ucnv_resetFromUnicode() will be called),
1656 * or NULL for the default converter
1657 * @param errorCode normal ICU error code
1658 * @return the length of the output string, not counting the terminating NUL;
1659 * if the length is greater than destCapacity, then the string will not fit
1660 * and a buffer of the indicated length would need to be passed in
1661 * @stable ICU 2.0
1662 */
1663 int32_t extract(char *dest, int32_t destCapacity,
1664 UConverter *cnv,
1665 UErrorCode &errorCode) const;
1666
1667#endif
1668
1669 /**
1670 * Create a temporary substring for the specified range.
1671 * Unlike the substring constructor and setTo() functions,
1672 * the object returned here will be a read-only alias (using getBuffer())
1673 * rather than copying the text.
1674 * As a result, this substring operation is much faster but requires
1675 * that the original string not be modified or deleted during the lifetime
1676 * of the returned substring object.
1677 * @param start offset of the first character visible in the substring
1678 * @param length length of the substring
1679 * @return a read-only alias UnicodeString object for the substring
1680 * @stable ICU 4.4
1681 */
1682 UnicodeString tempSubString(int32_t start=0, int32_t length=INT32_MAX) const;
1683
1684 /**
1685 * Create a temporary substring for the specified range.
1686 * Same as tempSubString(start, length) except that the substring range
1687 * is specified as a (start, limit) pair (with an exclusive limit index)
1688 * rather than a (start, length) pair.
1689 * @param start offset of the first character visible in the substring
1690 * @param limit offset immediately following the last character visible in the substring
1691 * @return a read-only alias UnicodeString object for the substring
1692 * @stable ICU 4.4
1693 */
1694 inline UnicodeString tempSubStringBetween(int32_t start, int32_t limit=INT32_MAX) const;
1695
1696 /**
1697 * Convert the UnicodeString to UTF-8 and write the result
1698 * to a ByteSink. This is called by toUTF8String().
1699 * Unpaired surrogates are replaced with U+FFFD.
1700 * Calls u_strToUTF8WithSub().
1701 *
1702 * @param sink A ByteSink to which the UTF-8 version of the string is written.
1703 * sink.Flush() is called at the end.
1704 * @stable ICU 4.2
1705 * @see toUTF8String
1706 */
1707 void toUTF8(ByteSink &sink) const;
1708
1709 /**
1710 * Convert the UnicodeString to UTF-8 and append the result
1711 * to a standard string.
1712 * Unpaired surrogates are replaced with U+FFFD.
1713 * Calls toUTF8().
1714 *
1715 * @param result A standard string (or a compatible object)
1716 * to which the UTF-8 version of the string is appended.
1717 * @return The string object.
1718 * @stable ICU 4.2
1719 * @see toUTF8
1720 */
1721 template<typename StringClass>
1722 StringClass &toUTF8String(StringClass &result) const {
1723 StringByteSink<StringClass> sbs(&result, length());
1724 toUTF8(sbs);
1725 return result;
1726 }
1727
1728 /**
1729 * Convert the UnicodeString to UTF-32.
1730 * Unpaired surrogates are replaced with U+FFFD.
1731 * Calls u_strToUTF32WithSub().
1732 *
1733 * @param utf32 destination string buffer, can be NULL if capacity==0
1734 * @param capacity the number of UChar32s available at utf32
1735 * @param errorCode Standard ICU error code. Its input value must
1736 * pass the U_SUCCESS() test, or else the function returns
1737 * immediately. Check for U_FAILURE() on output or use with
1738 * function chaining. (See User Guide for details.)
1739 * @return The length of the UTF-32 string.
1740 * @see fromUTF32
1741 * @stable ICU 4.2
1742 */
1743 int32_t toUTF32(UChar32 *utf32, int32_t capacity, UErrorCode &errorCode) const;
1744
1745 /* Length operations */
1746
1747 /**
1748 * Return the length of the UnicodeString object.
1749 * The length is the number of char16_t code units are in the UnicodeString.
1750 * If you want the number of code points, please use countChar32().
1751 * @return the length of the UnicodeString object
1752 * @see countChar32
1753 * @stable ICU 2.0
1754 */
1755 inline int32_t length(void) const;
1756
1757 /**
1758 * Count Unicode code points in the length char16_t code units of the string.
1759 * A code point may occupy either one or two char16_t code units.
1760 * Counting code points involves reading all code units.
1761 *
1762 * This functions is basically the inverse of moveIndex32().
1763 *
1764 * @param start the index of the first code unit to check
1765 * @param length the number of char16_t code units to check
1766 * @return the number of code points in the specified code units
1767 * @see length
1768 * @stable ICU 2.0
1769 */
1770 int32_t
1771 countChar32(int32_t start=0, int32_t length=INT32_MAX) const;
1772
1773 /**
1774 * Check if the length char16_t code units of the string
1775 * contain more Unicode code points than a certain number.
1776 * This is more efficient than counting all code points in this part of the string
1777 * and comparing that number with a threshold.
1778 * This function may not need to scan the string at all if the length
1779 * falls within a certain range, and
1780 * never needs to count more than 'number+1' code points.
1781 * Logically equivalent to (countChar32(start, length)>number).
1782 * A Unicode code point may occupy either one or two char16_t code units.
1783 *
1784 * @param start the index of the first code unit to check (0 for the entire string)
1785 * @param length the number of char16_t code units to check
1786 * (use INT32_MAX for the entire string; remember that start/length
1787 * values are pinned)
1788 * @param number The number of code points in the (sub)string is compared against
1789 * the 'number' parameter.
1790 * @return Boolean value for whether the string contains more Unicode code points
1791 * than 'number'. Same as (u_countChar32(s, length)>number).
1792 * @see countChar32
1793 * @see u_strHasMoreChar32Than
1794 * @stable ICU 2.4
1795 */
1796 UBool
1797 hasMoreChar32Than(int32_t start, int32_t length, int32_t number) const;
1798
1799 /**
1800 * Determine if this string is empty.
1801 * @return TRUE if this string contains 0 characters, FALSE otherwise.
1802 * @stable ICU 2.0
1803 */
1804 inline UBool isEmpty(void) const;
1805
1806 /**
1807 * Return the capacity of the internal buffer of the UnicodeString object.
1808 * This is useful together with the getBuffer functions.
1809 * See there for details.
1810 *
1811 * @return the number of char16_ts available in the internal buffer
1812 * @see getBuffer
1813 * @stable ICU 2.0
1814 */
1815 inline int32_t getCapacity(void) const;
1816
1817 /* Other operations */
1818
1819 /**
1820 * Generate a hash code for this object.
1821 * @return The hash code of this UnicodeString.
1822 * @stable ICU 2.0
1823 */
1824 inline int32_t hashCode(void) const;
1825
1826 /**
1827 * Determine if this object contains a valid string.
1828 * A bogus string has no value. It is different from an empty string,
1829 * although in both cases isEmpty() returns TRUE and length() returns 0.
1830 * setToBogus() and isBogus() can be used to indicate that no string value is available.
1831 * For a bogus string, getBuffer() and getTerminatedBuffer() return NULL, and
1832 * length() returns 0.
1833 *
1834 * @return TRUE if the string is bogus/invalid, FALSE otherwise
1835 * @see setToBogus()
1836 * @stable ICU 2.0
1837 */
1838 inline UBool isBogus(void) const;
1839
1840
1841 //========================================
1842 // Write operations
1843 //========================================
1844
1845 /* Assignment operations */
1846
1847 /**
1848 * Assignment operator. Replace the characters in this UnicodeString
1849 * with the characters from <TT>srcText</TT>.
1850 *
1851 * Starting with ICU 2.4, the assignment operator and the copy constructor
1852 * allocate a new buffer and copy the buffer contents even for readonly aliases.
1853 * By contrast, the fastCopyFrom() function implements the old,
1854 * more efficient but less safe behavior
1855 * of making this string also a readonly alias to the same buffer.
1856 *
1857 * If the source object has an "open" buffer from getBuffer(minCapacity),
1858 * then the copy is an empty string.
1859 *
1860 * @param srcText The text containing the characters to replace
1861 * @return a reference to this
1862 * @stable ICU 2.0
1863 * @see fastCopyFrom
1864 */
1865 UnicodeString &operator=(const UnicodeString &srcText);
1866
1867 /**
1868 * Almost the same as the assignment operator.
1869 * Replace the characters in this UnicodeString
1870 * with the characters from <code>srcText</code>.
1871 *
1872 * This function works the same as the assignment operator
1873 * for all strings except for ones that are readonly aliases.
1874 *
1875 * Starting with ICU 2.4, the assignment operator and the copy constructor
1876 * allocate a new buffer and copy the buffer contents even for readonly aliases.
1877 * This function implements the old, more efficient but less safe behavior
1878 * of making this string also a readonly alias to the same buffer.
1879 *
1880 * The fastCopyFrom function must be used only if it is known that the lifetime of
1881 * this UnicodeString does not exceed the lifetime of the aliased buffer
1882 * including its contents, for example for strings from resource bundles
1883 * or aliases to string constants.
1884 *
1885 * If the source object has an "open" buffer from getBuffer(minCapacity),
1886 * then the copy is an empty string.
1887 *
1888 * @param src The text containing the characters to replace.
1889 * @return a reference to this
1890 * @stable ICU 2.4
1891 */
1892 UnicodeString &fastCopyFrom(const UnicodeString &src);
1893
1894#if U_HAVE_RVALUE_REFERENCES
1895 /**
1896 * Move assignment operator, might leave src in bogus state.
1897 * This string will have the same contents and state that the source string had.
1898 * The behavior is undefined if *this and src are the same object.
1899 * @param src source string
1900 * @return *this
1901 * @stable ICU 56
1902 */
1903 UnicodeString &operator=(UnicodeString &&src) U_NOEXCEPT {
1904 return moveFrom(src);
1905 }
1906#endif
1907 // do not use #ifndef U_HIDE_DRAFT_API for moveFrom, needed by non-draft API
1908 /**
1909 * Move assignment, might leave src in bogus state.
1910 * This string will have the same contents and state that the source string had.
1911 * The behavior is undefined if *this and src are the same object.
1912 *
1913 * Can be called explicitly, does not need C++11 support.
1914 * @param src source string
1915 * @return *this
1916 * @draft ICU 56
1917 */
1918 UnicodeString &moveFrom(UnicodeString &src) U_NOEXCEPT;
1919
1920 /**
1921 * Swap strings.
1922 * @param other other string
1923 * @stable ICU 56
1924 */
1925 void swap(UnicodeString &other) U_NOEXCEPT;
1926
1927 /**
1928 * Non-member UnicodeString swap function.
1929 * @param s1 will get s2's contents and state
1930 * @param s2 will get s1's contents and state
1931 * @stable ICU 56
1932 */
1933 friend U_COMMON_API inline void U_EXPORT2
1934 swap(UnicodeString &s1, UnicodeString &s2) U_NOEXCEPT {
1935 s1.swap(s2);
1936 }
1937
1938 /**
1939 * Assignment operator. Replace the characters in this UnicodeString
1940 * with the code unit <TT>ch</TT>.
1941 * @param ch the code unit to replace
1942 * @return a reference to this
1943 * @stable ICU 2.0
1944 */
1945 inline UnicodeString& operator= (char16_t ch);
1946
1947 /**
1948 * Assignment operator. Replace the characters in this UnicodeString
1949 * with the code point <TT>ch</TT>.
1950 * @param ch the code point to replace
1951 * @return a reference to this
1952 * @stable ICU 2.0
1953 */
1954 inline UnicodeString& operator= (UChar32 ch);
1955
1956 /**
1957 * Set the text in the UnicodeString object to the characters
1958 * in <TT>srcText</TT> in the range
1959 * [<TT>srcStart</TT>, <TT>srcText.length()</TT>).
1960 * <TT>srcText</TT> is not modified.
1961 * @param srcText the source for the new characters
1962 * @param srcStart the offset into <TT>srcText</TT> where new characters
1963 * will be obtained
1964 * @return a reference to this
1965 * @stable ICU 2.2
1966 */
1967 inline UnicodeString& setTo(const UnicodeString& srcText,
1968 int32_t srcStart);
1969
1970 /**
1971 * Set the text in the UnicodeString object to the characters
1972 * in <TT>srcText</TT> in the range
1973 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
1974 * <TT>srcText</TT> is not modified.
1975 * @param srcText the source for the new characters
1976 * @param srcStart the offset into <TT>srcText</TT> where new characters
1977 * will be obtained
1978 * @param srcLength the number of characters in <TT>srcText</TT> in the
1979 * replace string.
1980 * @return a reference to this
1981 * @stable ICU 2.0
1982 */
1983 inline UnicodeString& setTo(const UnicodeString& srcText,
1984 int32_t srcStart,
1985 int32_t srcLength);
1986
1987 /**
1988 * Set the text in the UnicodeString object to the characters in
1989 * <TT>srcText</TT>.
1990 * <TT>srcText</TT> is not modified.
1991 * @param srcText the source for the new characters
1992 * @return a reference to this
1993 * @stable ICU 2.0
1994 */
1995 inline UnicodeString& setTo(const UnicodeString& srcText);
1996
1997 /**
1998 * Set the characters in the UnicodeString object to the characters
1999 * in <TT>srcChars</TT>. <TT>srcChars</TT> is not modified.
2000 * @param srcChars the source for the new characters
2001 * @param srcLength the number of Unicode characters in srcChars.
2002 * @return a reference to this
2003 * @stable ICU 2.0
2004 */
2005 inline UnicodeString& setTo(const char16_t *srcChars,
2006 int32_t srcLength);
2007
2008 /**
2009 * Set the characters in the UnicodeString object to the code unit
2010 * <TT>srcChar</TT>.
2011 * @param srcChar the code unit which becomes the UnicodeString's character
2012 * content
2013 * @return a reference to this
2014 * @stable ICU 2.0
2015 */
2016 UnicodeString& setTo(char16_t srcChar);
2017
2018 /**
2019 * Set the characters in the UnicodeString object to the code point
2020 * <TT>srcChar</TT>.
2021 * @param srcChar the code point which becomes the UnicodeString's character
2022 * content
2023 * @return a reference to this
2024 * @stable ICU 2.0
2025 */
2026 UnicodeString& setTo(UChar32 srcChar);
2027
2028 /**
2029 * Aliasing setTo() function, analogous to the readonly-aliasing char16_t* constructor.
2030 * The text will be used for the UnicodeString object, but
2031 * it will not be released when the UnicodeString is destroyed.
2032 * This has copy-on-write semantics:
2033 * When the string is modified, then the buffer is first copied into
2034 * newly allocated memory.
2035 * The aliased buffer is never modified.
2036 *
2037 * In an assignment to another UnicodeString, when using the copy constructor
2038 * or the assignment operator, the text will be copied.
2039 * When using fastCopyFrom(), the text will be aliased again,
2040 * so that both strings then alias the same readonly-text.
2041 *
2042 * @param isTerminated specifies if <code>text</code> is <code>NUL</code>-terminated.
2043 * This must be true if <code>textLength==-1</code>.
2044 * @param text The characters to alias for the UnicodeString.
2045 * @param textLength The number of Unicode characters in <code>text</code> to alias.
2046 * If -1, then this constructor will determine the length
2047 * by calling <code>u_strlen()</code>.
2048 * @return a reference to this
2049 * @stable ICU 2.0
2050 */
2051 UnicodeString &setTo(UBool isTerminated,
2052 ConstChar16Ptr text,
2053 int32_t textLength);
2054
2055 /**
2056 * Aliasing setTo() function, analogous to the writable-aliasing char16_t* constructor.
2057 * The text will be used for the UnicodeString object, but
2058 * it will not be released when the UnicodeString is destroyed.
2059 * This has write-through semantics:
2060 * For as long as the capacity of the buffer is sufficient, write operations
2061 * will directly affect the buffer. When more capacity is necessary, then
2062 * a new buffer will be allocated and the contents copied as with regularly
2063 * constructed strings.
2064 * In an assignment to another UnicodeString, the buffer will be copied.
2065 * The extract(Char16Ptr dst) function detects whether the dst pointer is the same
2066 * as the string buffer itself and will in this case not copy the contents.
2067 *
2068 * @param buffer The characters to alias for the UnicodeString.
2069 * @param buffLength The number of Unicode characters in <code>buffer</code> to alias.
2070 * @param buffCapacity The size of <code>buffer</code> in char16_ts.
2071 * @return a reference to this
2072 * @stable ICU 2.0
2073 */
2074 UnicodeString &setTo(char16_t *buffer,
2075 int32_t buffLength,
2076 int32_t buffCapacity);
2077
2078 /**
2079 * Make this UnicodeString object invalid.
2080 * The string will test TRUE with isBogus().
2081 *
2082 * A bogus string has no value. It is different from an empty string.
2083 * It can be used to indicate that no string value is available.
2084 * getBuffer() and getTerminatedBuffer() return NULL, and
2085 * length() returns 0.
2086 *
2087 * This utility function is used throughout the UnicodeString
2088 * implementation to indicate that a UnicodeString operation failed,
2089 * and may be used in other functions,
2090 * especially but not exclusively when such functions do not
2091 * take a UErrorCode for simplicity.
2092 *
2093 * The following methods, and no others, will clear a string object's bogus flag:
2094 * - remove()
2095 * - remove(0, INT32_MAX)
2096 * - truncate(0)
2097 * - operator=() (assignment operator)
2098 * - setTo(...)
2099 *
2100 * The simplest ways to turn a bogus string into an empty one
2101 * is to use the remove() function.
2102 * Examples for other functions that are equivalent to "set to empty string":
2103 * \code
2104 * if(s.isBogus()) {
2105 * s.remove(); // set to an empty string (remove all), or
2106 * s.remove(0, INT32_MAX); // set to an empty string (remove all), or
2107 * s.truncate(0); // set to an empty string (complete truncation), or
2108 * s=UnicodeString(); // assign an empty string, or
2109 * s.setTo((UChar32)-1); // set to a pseudo code point that is out of range, or
2110 * static const char16_t nul=0;
2111 * s.setTo(&nul, 0); // set to an empty C Unicode string
2112 * }
2113 * \endcode
2114 *
2115 * @see isBogus()
2116 * @stable ICU 2.0
2117 */
2118 void setToBogus();
2119
2120 /**
2121 * Set the character at the specified offset to the specified character.
2122 * @param offset A valid offset into the text of the character to set
2123 * @param ch The new character
2124 * @return A reference to this
2125 * @stable ICU 2.0
2126 */
2127 UnicodeString& setCharAt(int32_t offset,
2128 char16_t ch);
2129
2130
2131 /* Append operations */
2132
2133 /**
2134 * Append operator. Append the code unit <TT>ch</TT> to the UnicodeString
2135 * object.
2136 * @param ch the code unit to be appended
2137 * @return a reference to this
2138 * @stable ICU 2.0
2139 */
2140 inline UnicodeString& operator+= (char16_t ch);
2141
2142 /**
2143 * Append operator. Append the code point <TT>ch</TT> to the UnicodeString
2144 * object.
2145 * @param ch the code point to be appended
2146 * @return a reference to this
2147 * @stable ICU 2.0
2148 */
2149 inline UnicodeString& operator+= (UChar32 ch);
2150
2151 /**
2152 * Append operator. Append the characters in <TT>srcText</TT> to the
2153 * UnicodeString object. <TT>srcText</TT> is not modified.
2154 * @param srcText the source for the new characters
2155 * @return a reference to this
2156 * @stable ICU 2.0
2157 */
2158 inline UnicodeString& operator+= (const UnicodeString& srcText);
2159
2160 /**
2161 * Append the characters
2162 * in <TT>srcText</TT> in the range
2163 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) to the
2164 * UnicodeString object at offset <TT>start</TT>. <TT>srcText</TT>
2165 * is not modified.
2166 * @param srcText the source for the new characters
2167 * @param srcStart the offset into <TT>srcText</TT> where new characters
2168 * will be obtained
2169 * @param srcLength the number of characters in <TT>srcText</TT> in
2170 * the append string
2171 * @return a reference to this
2172 * @stable ICU 2.0
2173 */
2174 inline UnicodeString& append(const UnicodeString& srcText,
2175 int32_t srcStart,
2176 int32_t srcLength);
2177
2178 /**
2179 * Append the characters in <TT>srcText</TT> to the UnicodeString object.
2180 * <TT>srcText</TT> is not modified.
2181 * @param srcText the source for the new characters
2182 * @return a reference to this
2183 * @stable ICU 2.0
2184 */
2185 inline UnicodeString& append(const UnicodeString& srcText);
2186
2187 /**
2188 * Append the characters in <TT>srcChars</TT> in the range
2189 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) to the UnicodeString
2190 * object at offset
2191 * <TT>start</TT>. <TT>srcChars</TT> is not modified.
2192 * @param srcChars the source for the new characters
2193 * @param srcStart the offset into <TT>srcChars</TT> where new characters
2194 * will be obtained
2195 * @param srcLength the number of characters in <TT>srcChars</TT> in
2196 * the append string; can be -1 if <TT>srcChars</TT> is NUL-terminated
2197 * @return a reference to this
2198 * @stable ICU 2.0
2199 */
2200 inline UnicodeString& append(const char16_t *srcChars,
2201 int32_t srcStart,
2202 int32_t srcLength);
2203
2204 /**
2205 * Append the characters in <TT>srcChars</TT> to the UnicodeString object
2206 * at offset <TT>start</TT>. <TT>srcChars</TT> is not modified.
2207 * @param srcChars the source for the new characters
2208 * @param srcLength the number of Unicode characters in <TT>srcChars</TT>;
2209 * can be -1 if <TT>srcChars</TT> is NUL-terminated
2210 * @return a reference to this
2211 * @stable ICU 2.0
2212 */
2213 inline UnicodeString& append(ConstChar16Ptr srcChars,
2214 int32_t srcLength);
2215
2216 /**
2217 * Append the code unit <TT>srcChar</TT> to the UnicodeString object.
2218 * @param srcChar the code unit to append
2219 * @return a reference to this
2220 * @stable ICU 2.0
2221 */
2222 inline UnicodeString& append(char16_t srcChar);
2223
2224 /**
2225 * Append the code point <TT>srcChar</TT> to the UnicodeString object.
2226 * @param srcChar the code point to append
2227 * @return a reference to this
2228 * @stable ICU 2.0
2229 */
2230 UnicodeString& append(UChar32 srcChar);
2231
2232
2233 /* Insert operations */
2234
2235 /**
2236 * Insert the characters in <TT>srcText</TT> in the range
2237 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) into the UnicodeString
2238 * object at offset <TT>start</TT>. <TT>srcText</TT> is not modified.
2239 * @param start the offset where the insertion begins
2240 * @param srcText the source for the new characters
2241 * @param srcStart the offset into <TT>srcText</TT> where new characters
2242 * will be obtained
2243 * @param srcLength the number of characters in <TT>srcText</TT> in
2244 * the insert string
2245 * @return a reference to this
2246 * @stable ICU 2.0
2247 */
2248 inline UnicodeString& insert(int32_t start,
2249 const UnicodeString& srcText,
2250 int32_t srcStart,
2251 int32_t srcLength);
2252
2253 /**
2254 * Insert the characters in <TT>srcText</TT> into the UnicodeString object
2255 * at offset <TT>start</TT>. <TT>srcText</TT> is not modified.
2256 * @param start the offset where the insertion begins
2257 * @param srcText the source for the new characters
2258 * @return a reference to this
2259 * @stable ICU 2.0
2260 */
2261 inline UnicodeString& insert(int32_t start,
2262 const UnicodeString& srcText);
2263
2264 /**
2265 * Insert the characters in <TT>srcChars</TT> in the range
2266 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>) into the UnicodeString
2267 * object at offset <TT>start</TT>. <TT>srcChars</TT> is not modified.
2268 * @param start the offset at which the insertion begins
2269 * @param srcChars the source for the new characters
2270 * @param srcStart the offset into <TT>srcChars</TT> where new characters
2271 * will be obtained
2272 * @param srcLength the number of characters in <TT>srcChars</TT>
2273 * in the insert string
2274 * @return a reference to this
2275 * @stable ICU 2.0
2276 */
2277 inline UnicodeString& insert(int32_t start,
2278 const char16_t *srcChars,
2279 int32_t srcStart,
2280 int32_t srcLength);
2281
2282 /**
2283 * Insert the characters in <TT>srcChars</TT> into the UnicodeString object
2284 * at offset <TT>start</TT>. <TT>srcChars</TT> is not modified.
2285 * @param start the offset where the insertion begins
2286 * @param srcChars the source for the new characters
2287 * @param srcLength the number of Unicode characters in srcChars.
2288 * @return a reference to this
2289 * @stable ICU 2.0
2290 */
2291 inline UnicodeString& insert(int32_t start,
2292 ConstChar16Ptr srcChars,
2293 int32_t srcLength);
2294
2295 /**
2296 * Insert the code unit <TT>srcChar</TT> into the UnicodeString object at
2297 * offset <TT>start</TT>.
2298 * @param start the offset at which the insertion occurs
2299 * @param srcChar the code unit to insert
2300 * @return a reference to this
2301 * @stable ICU 2.0
2302 */
2303 inline UnicodeString& insert(int32_t start,
2304 char16_t srcChar);
2305
2306 /**
2307 * Insert the code point <TT>srcChar</TT> into the UnicodeString object at
2308 * offset <TT>start</TT>.
2309 * @param start the offset at which the insertion occurs
2310 * @param srcChar the code point to insert
2311 * @return a reference to this
2312 * @stable ICU 2.0
2313 */
2314 inline UnicodeString& insert(int32_t start,
2315 UChar32 srcChar);
2316
2317
2318 /* Replace operations */
2319
2320 /**
2321 * Replace the characters in the range
2322 * [<TT>start</TT>, <TT>start + length</TT>) with the characters in
2323 * <TT>srcText</TT> in the range
2324 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>).
2325 * <TT>srcText</TT> is not modified.
2326 * @param start the offset at which the replace operation begins
2327 * @param length the number of characters to replace. The character at
2328 * <TT>start + length</TT> is not modified.
2329 * @param srcText the source for the new characters
2330 * @param srcStart the offset into <TT>srcText</TT> where new characters
2331 * will be obtained
2332 * @param srcLength the number of characters in <TT>srcText</TT> in
2333 * the replace string
2334 * @return a reference to this
2335 * @stable ICU 2.0
2336 */
2337 UnicodeString& replace(int32_t start,
2338 int32_t length,
2339 const UnicodeString& srcText,
2340 int32_t srcStart,
2341 int32_t srcLength);
2342
2343 /**
2344 * Replace the characters in the range
2345 * [<TT>start</TT>, <TT>start + length</TT>)
2346 * with the characters in <TT>srcText</TT>. <TT>srcText</TT> is
2347 * not modified.
2348 * @param start the offset at which the replace operation begins
2349 * @param length the number of characters to replace. The character at
2350 * <TT>start + length</TT> is not modified.
2351 * @param srcText the source for the new characters
2352 * @return a reference to this
2353 * @stable ICU 2.0
2354 */
2355 UnicodeString& replace(int32_t start,
2356 int32_t length,
2357 const UnicodeString& srcText);
2358
2359 /**
2360 * Replace the characters in the range
2361 * [<TT>start</TT>, <TT>start + length</TT>) with the characters in
2362 * <TT>srcChars</TT> in the range
2363 * [<TT>srcStart</TT>, <TT>srcStart + srcLength</TT>). <TT>srcChars</TT>
2364 * is not modified.
2365 * @param start the offset at which the replace operation begins
2366 * @param length the number of characters to replace. The character at
2367 * <TT>start + length</TT> is not modified.
2368 * @param srcChars the source for the new characters
2369 * @param srcStart the offset into <TT>srcChars</TT> where new characters
2370 * will be obtained
2371 * @param srcLength the number of characters in <TT>srcChars</TT>
2372 * in the replace string
2373 * @return a reference to this
2374 * @stable ICU 2.0
2375 */
2376 UnicodeString& replace(int32_t start,
2377 int32_t length,
2378 const char16_t *srcChars,
2379 int32_t srcStart,
2380 int32_t srcLength);
2381
2382 /**
2383 * Replace the characters in the range
2384 * [<TT>start</TT>, <TT>start + length</TT>) with the characters in
2385 * <TT>srcChars</TT>. <TT>srcChars</TT> is not modified.
2386 * @param start the offset at which the replace operation begins
2387 * @param length number of characters to replace. The character at
2388 * <TT>start + length</TT> is not modified.
2389 * @param srcChars the source for the new characters
2390 * @param srcLength the number of Unicode characters in srcChars
2391 * @return a reference to this
2392 * @stable ICU 2.0
2393 */
2394 inline UnicodeString& replace(int32_t start,
2395 int32_t length,
2396 ConstChar16Ptr srcChars,
2397 int32_t srcLength);
2398
2399 /**
2400 * Replace the characters in the range
2401 * [<TT>start</TT>, <TT>start + length</TT>) with the code unit
2402 * <TT>srcChar</TT>.
2403 * @param start the offset at which the replace operation begins
2404 * @param length the number of characters to replace. The character at
2405 * <TT>start + length</TT> is not modified.
2406 * @param srcChar the new code unit
2407 * @return a reference to this
2408 * @stable ICU 2.0
2409 */
2410 inline UnicodeString& replace(int32_t start,
2411 int32_t length,
2412 char16_t srcChar);
2413
2414 /**
2415 * Replace the characters in the range
2416 * [<TT>start</TT>, <TT>start + length</TT>) with the code point
2417 * <TT>srcChar</TT>.
2418 * @param start the offset at which the replace operation begins
2419 * @param length the number of characters to replace. The character at
2420 * <TT>start + length</TT> is not modified.
2421 * @param srcChar the new code point
2422 * @return a reference to this
2423 * @stable ICU 2.0
2424 */
2425 UnicodeString& replace(int32_t start, int32_t length, UChar32 srcChar);
2426
2427 /**
2428 * Replace the characters in the range [<TT>start</TT>, <TT>limit</TT>)
2429 * with the characters in <TT>srcText</TT>. <TT>srcText</TT> is not modified.
2430 * @param start the offset at which the replace operation begins
2431 * @param limit the offset immediately following the replace range
2432 * @param srcText the source for the new characters
2433 * @return a reference to this
2434 * @stable ICU 2.0
2435 */
2436 inline UnicodeString& replaceBetween(int32_t start,
2437 int32_t limit,
2438 const UnicodeString& srcText);
2439
2440 /**
2441 * Replace the characters in the range [<TT>start</TT>, <TT>limit</TT>)
2442 * with the characters in <TT>srcText</TT> in the range
2443 * [<TT>srcStart</TT>, <TT>srcLimit</TT>). <TT>srcText</TT> is not modified.
2444 * @param start the offset at which the replace operation begins
2445 * @param limit the offset immediately following the replace range
2446 * @param srcText the source for the new characters
2447 * @param srcStart the offset into <TT>srcChars</TT> where new characters
2448 * will be obtained
2449 * @param srcLimit the offset immediately following the range to copy
2450 * in <TT>srcText</TT>
2451 * @return a reference to this
2452 * @stable ICU 2.0
2453 */
2454 inline UnicodeString& replaceBetween(int32_t start,
2455 int32_t limit,
2456 const UnicodeString& srcText,
2457 int32_t srcStart,
2458 int32_t srcLimit);
2459
2460 /**
2461 * Replace a substring of this object with the given text.
2462 * @param start the beginning index, inclusive; <code>0 <= start
2463 * <= limit</code>.
2464 * @param limit the ending index, exclusive; <code>start <= limit
2465 * <= length()</code>.
2466 * @param text the text to replace characters <code>start</code>
2467 * to <code>limit - 1</code>
2468 * @stable ICU 2.0
2469 */
2470 virtual void handleReplaceBetween(int32_t start,
2471 int32_t limit,
2472 const UnicodeString& text);
2473
2474 /**
2475 * Replaceable API
2476 * @return TRUE if it has MetaData
2477 * @stable ICU 2.4
2478 */
2479 virtual UBool hasMetaData() const;
2480
2481 /**
2482 * Copy a substring of this object, retaining attribute (out-of-band)
2483 * information. This method is used to duplicate or reorder substrings.
2484 * The destination index must not overlap the source range.
2485 *
2486 * @param start the beginning index, inclusive; <code>0 <= start <=
2487 * limit</code>.
2488 * @param limit the ending index, exclusive; <code>start <= limit <=
2489 * length()</code>.
2490 * @param dest the destination index. The characters from
2491 * <code>start..limit-1</code> will be copied to <code>dest</code>.
2492 * Implementations of this method may assume that <code>dest <= start ||
2493 * dest >= limit</code>.
2494 * @stable ICU 2.0
2495 */
2496 virtual void copy(int32_t start, int32_t limit, int32_t dest);
2497
2498 /* Search and replace operations */
2499
2500 /**
2501 * Replace all occurrences of characters in oldText with the characters
2502 * in newText
2503 * @param oldText the text containing the search text
2504 * @param newText the text containing the replacement text
2505 * @return a reference to this
2506 * @stable ICU 2.0
2507 */
2508 inline UnicodeString& findAndReplace(const UnicodeString& oldText,
2509 const UnicodeString& newText);
2510
2511 /**
2512 * Replace all occurrences of characters in oldText with characters
2513 * in newText
2514 * in the range [<TT>start</TT>, <TT>start + length</TT>).
2515 * @param start the start of the range in which replace will performed
2516 * @param length the length of the range in which replace will be performed
2517 * @param oldText the text containing the search text
2518 * @param newText the text containing the replacement text
2519 * @return a reference to this
2520 * @stable ICU 2.0
2521 */
2522 inline UnicodeString& findAndReplace(int32_t start,
2523 int32_t length,
2524 const UnicodeString& oldText,
2525 const UnicodeString& newText);
2526
2527 /**
2528 * Replace all occurrences of characters in oldText in the range
2529 * [<TT>oldStart</TT>, <TT>oldStart + oldLength</TT>) with the characters
2530 * in newText in the range
2531 * [<TT>newStart</TT>, <TT>newStart + newLength</TT>)
2532 * in the range [<TT>start</TT>, <TT>start + length</TT>).
2533 * @param start the start of the range in which replace will performed
2534 * @param length the length of the range in which replace will be performed
2535 * @param oldText the text containing the search text
2536 * @param oldStart the start of the search range in <TT>oldText</TT>
2537 * @param oldLength the length of the search range in <TT>oldText</TT>
2538 * @param newText the text containing the replacement text
2539 * @param newStart the start of the replacement range in <TT>newText</TT>
2540 * @param newLength the length of the replacement range in <TT>newText</TT>
2541 * @return a reference to this
2542 * @stable ICU 2.0
2543 */
2544 UnicodeString& findAndReplace(int32_t start,
2545 int32_t length,
2546 const UnicodeString& oldText,
2547 int32_t oldStart,
2548 int32_t oldLength,
2549 const UnicodeString& newText,
2550 int32_t newStart,
2551 int32_t newLength);
2552
2553
2554 /* Remove operations */
2555
2556 /**
2557 * Remove all characters from the UnicodeString object.
2558 * @return a reference to this
2559 * @stable ICU 2.0
2560 */
2561 inline UnicodeString& remove(void);
2562
2563 /**
2564 * Remove the characters in the range
2565 * [<TT>start</TT>, <TT>start + length</TT>) from the UnicodeString object.
2566 * @param start the offset of the first character to remove
2567 * @param length the number of characters to remove
2568 * @return a reference to this
2569 * @stable ICU 2.0
2570 */
2571 inline UnicodeString& remove(int32_t start,
2572 int32_t length = (int32_t)INT32_MAX);
2573
2574 /**
2575 * Remove the characters in the range
2576 * [<TT>start</TT>, <TT>limit</TT>) from the UnicodeString object.
2577 * @param start the offset of the first character to remove
2578 * @param limit the offset immediately following the range to remove
2579 * @return a reference to this
2580 * @stable ICU 2.0
2581 */
2582 inline UnicodeString& removeBetween(int32_t start,
2583 int32_t limit = (int32_t)INT32_MAX);
2584
2585 /**
2586 * Retain only the characters in the range
2587 * [<code>start</code>, <code>limit</code>) from the UnicodeString object.
2588 * Removes characters before <code>start</code> and at and after <code>limit</code>.
2589 * @param start the offset of the first character to retain
2590 * @param limit the offset immediately following the range to retain
2591 * @return a reference to this
2592 * @stable ICU 4.4
2593 */
2594 inline UnicodeString &retainBetween(int32_t start, int32_t limit = INT32_MAX);
2595
2596 /* Length operations */
2597
2598 /**
2599 * Pad the start of this UnicodeString with the character <TT>padChar</TT>.
2600 * If the length of this UnicodeString is less than targetLength,
2601 * length() - targetLength copies of padChar will be added to the
2602 * beginning of this UnicodeString.
2603 * @param targetLength the desired length of the string
2604 * @param padChar the character to use for padding. Defaults to
2605 * space (U+0020)
2606 * @return TRUE if the text was padded, FALSE otherwise.
2607 * @stable ICU 2.0
2608 */
2609 UBool padLeading(int32_t targetLength,
2610 char16_t padChar = 0x0020);
2611
2612 /**
2613 * Pad the end of this UnicodeString with the character <TT>padChar</TT>.
2614 * If the length of this UnicodeString is less than targetLength,
2615 * length() - targetLength copies of padChar will be added to the
2616 * end of this UnicodeString.
2617 * @param targetLength the desired length of the string
2618 * @param padChar the character to use for padding. Defaults to
2619 * space (U+0020)
2620 * @return TRUE if the text was padded, FALSE otherwise.
2621 * @stable ICU 2.0
2622 */
2623 UBool padTrailing(int32_t targetLength,
2624 char16_t padChar = 0x0020);
2625
2626 /**
2627 * Truncate this UnicodeString to the <TT>targetLength</TT>.
2628 * @param targetLength the desired length of this UnicodeString.
2629 * @return TRUE if the text was truncated, FALSE otherwise
2630 * @stable ICU 2.0
2631 */
2632 inline UBool truncate(int32_t targetLength);
2633
2634 /**
2635 * Trims leading and trailing whitespace from this UnicodeString.
2636 * @return a reference to this
2637 * @stable ICU 2.0
2638 */
2639 UnicodeString& trim(void);
2640
2641
2642 /* Miscellaneous operations */
2643
2644 /**
2645 * Reverse this UnicodeString in place.
2646 * @return a reference to this
2647 * @stable ICU 2.0
2648 */
2649 inline UnicodeString& reverse(void);
2650
2651 /**
2652 * Reverse the range [<TT>start</TT>, <TT>start + length</TT>) in
2653 * this UnicodeString.
2654 * @param start the start of the range to reverse
2655 * @param length the number of characters to to reverse
2656 * @return a reference to this
2657 * @stable ICU 2.0
2658 */
2659 inline UnicodeString& reverse(int32_t start,
2660 int32_t length);
2661
2662 /**
2663 * Convert the characters in this to UPPER CASE following the conventions of
2664 * the default locale.
2665 * @return A reference to this.
2666 * @stable ICU 2.0
2667 */
2668 UnicodeString& toUpper(void);
2669
2670 /**
2671 * Convert the characters in this to UPPER CASE following the conventions of
2672 * a specific locale.
2673 * @param locale The locale containing the conventions to use.
2674 * @return A reference to this.
2675 * @stable ICU 2.0
2676 */
2677 UnicodeString& toUpper(const Locale& locale);
2678
2679 /**
2680 * Convert the characters in this to lower case following the conventions of
2681 * the default locale.
2682 * @return A reference to this.
2683 * @stable ICU 2.0
2684 */
2685 UnicodeString& toLower(void);
2686
2687 /**
2688 * Convert the characters in this to lower case following the conventions of
2689 * a specific locale.
2690 * @param locale The locale containing the conventions to use.
2691 * @return A reference to this.
2692 * @stable ICU 2.0
2693 */
2694 UnicodeString& toLower(const Locale& locale);
2695
2696#if !UCONFIG_NO_BREAK_ITERATION
2697
2698 /**
2699 * Titlecase this string, convenience function using the default locale.
2700 *
2701 * Casing is locale-dependent and context-sensitive.
2702 * Titlecasing uses a break iterator to find the first characters of words
2703 * that are to be titlecased. It titlecases those characters and lowercases
2704 * all others.
2705 *
2706 * The titlecase break iterator can be provided to customize for arbitrary
2707 * styles, using rules and dictionaries beyond the standard iterators.
2708 * It may be more efficient to always provide an iterator to avoid
2709 * opening and closing one for each string.
2710 * The standard titlecase iterator for the root locale implements the
2711 * algorithm of Unicode TR 21.
2712 *
2713 * This function uses only the setText(), first() and next() methods of the
2714 * provided break iterator.
2715 *
2716 * @param titleIter A break iterator to find the first characters of words
2717 * that are to be titlecased.
2718 * If none is provided (0), then a standard titlecase
2719 * break iterator is opened.
2720 * Otherwise the provided iterator is set to the string's text.
2721 * @return A reference to this.
2722 * @stable ICU 2.1
2723 */
2724 UnicodeString &toTitle(BreakIterator *titleIter);
2725
2726 /**
2727 * Titlecase this string.
2728 *
2729 * Casing is locale-dependent and context-sensitive.
2730 * Titlecasing uses a break iterator to find the first characters of words
2731 * that are to be titlecased. It titlecases those characters and lowercases
2732 * all others.
2733 *
2734 * The titlecase break iterator can be provided to customize for arbitrary
2735 * styles, using rules and dictionaries beyond the standard iterators.
2736 * It may be more efficient to always provide an iterator to avoid
2737 * opening and closing one for each string.
2738 * The standard titlecase iterator for the root locale implements the
2739 * algorithm of Unicode TR 21.
2740 *
2741 * This function uses only the setText(), first() and next() methods of the
2742 * provided break iterator.
2743 *
2744 * @param titleIter A break iterator to find the first characters of words
2745 * that are to be titlecased.
2746 * If none is provided (0), then a standard titlecase
2747 * break iterator is opened.
2748 * Otherwise the provided iterator is set to the string's text.
2749 * @param locale The locale to consider.
2750 * @return A reference to this.
2751 * @stable ICU 2.1
2752 */
2753 UnicodeString &toTitle(BreakIterator *titleIter, const Locale &locale);
2754
2755 /**
2756 * Titlecase this string, with options.
2757 *
2758 * Casing is locale-dependent and context-sensitive.
2759 * Titlecasing uses a break iterator to find the first characters of words
2760 * that are to be titlecased. It titlecases those characters and lowercases
2761 * all others. (This can be modified with options.)
2762 *
2763 * The titlecase break iterator can be provided to customize for arbitrary
2764 * styles, using rules and dictionaries beyond the standard iterators.
2765 * It may be more efficient to always provide an iterator to avoid
2766 * opening and closing one for each string.
2767 * The standard titlecase iterator for the root locale implements the
2768 * algorithm of Unicode TR 21.
2769 *
2770 * This function uses only the setText(), first() and next() methods of the
2771 * provided break iterator.
2772 *
2773 * @param titleIter A break iterator to find the first characters of words
2774 * that are to be titlecased.
2775 * If none is provided (0), then a standard titlecase
2776 * break iterator is opened.
2777 * Otherwise the provided iterator is set to the string's text.
2778 * @param locale The locale to consider.
2779 * @param options Options bit set, usually 0. See U_TITLECASE_NO_LOWERCASE,
2780 * U_TITLECASE_NO_BREAK_ADJUSTMENT, U_TITLECASE_ADJUST_TO_CASED,
2781 * U_TITLECASE_WHOLE_STRING, U_TITLECASE_SENTENCES.
2782 * @param options Options bit set, see ucasemap_open().
2783 * @return A reference to this.
2784 * @stable ICU 3.8
2785 */
2786 UnicodeString &toTitle(BreakIterator *titleIter, const Locale &locale, uint32_t options);
2787
2788#endif
2789
2790 /**
2791 * Case-folds the characters in this string.
2792 *
2793 * Case-folding is locale-independent and not context-sensitive,
2794 * but there is an option for whether to include or exclude mappings for dotted I
2795 * and dotless i that are marked with 'T' in CaseFolding.txt.
2796 *
2797 * The result may be longer or shorter than the original.
2798 *
2799 * @param options Either U_FOLD_CASE_DEFAULT or U_FOLD_CASE_EXCLUDE_SPECIAL_I
2800 * @return A reference to this.
2801 * @stable ICU 2.0
2802 */
2803 UnicodeString &foldCase(uint32_t options=0 /*U_FOLD_CASE_DEFAULT*/);
2804
2805 //========================================
2806 // Access to the internal buffer
2807 //========================================
2808
2809 /**
2810 * Get a read/write pointer to the internal buffer.
2811 * The buffer is guaranteed to be large enough for at least minCapacity char16_ts,
2812 * writable, and is still owned by the UnicodeString object.
2813 * Calls to getBuffer(minCapacity) must not be nested, and
2814 * must be matched with calls to releaseBuffer(newLength).
2815 * If the string buffer was read-only or shared,
2816 * then it will be reallocated and copied.
2817 *
2818 * An attempted nested call will return 0, and will not further modify the
2819 * state of the UnicodeString object.
2820 * It also returns 0 if the string is bogus.
2821 *
2822 * The actual capacity of the string buffer may be larger than minCapacity.
2823 * getCapacity() returns the actual capacity.
2824 * For many operations, the full capacity should be used to avoid reallocations.
2825 *
2826 * While the buffer is "open" between getBuffer(minCapacity)
2827 * and releaseBuffer(newLength), the following applies:
2828 * - The string length is set to 0.
2829 * - Any read API call on the UnicodeString object will behave like on a 0-length string.
2830 * - Any write API call on the UnicodeString object is disallowed and will have no effect.
2831 * - You can read from and write to the returned buffer.
2832 * - The previous string contents will still be in the buffer;
2833 * if you want to use it, then you need to call length() before getBuffer(minCapacity).
2834 * If the length() was greater than minCapacity, then any contents after minCapacity
2835 * may be lost.
2836 * The buffer contents is not NUL-terminated by getBuffer().
2837 * If length()<getCapacity() then you can terminate it by writing a NUL
2838 * at index length().
2839 * - You must call releaseBuffer(newLength) before and in order to
2840 * return to normal UnicodeString operation.
2841 *
2842 * @param minCapacity the minimum number of char16_ts that are to be available
2843 * in the buffer, starting at the returned pointer;
2844 * default to the current string capacity if minCapacity==-1
2845 * @return a writable pointer to the internal string buffer,
2846 * or nullptr if an error occurs (nested calls, out of memory)
2847 *
2848 * @see releaseBuffer
2849 * @see getTerminatedBuffer()
2850 * @stable ICU 2.0
2851 */
2852 char16_t *getBuffer(int32_t minCapacity);
2853
2854 /**
2855 * Release a read/write buffer on a UnicodeString object with an
2856 * "open" getBuffer(minCapacity).
2857 * This function must be called in a matched pair with getBuffer(minCapacity).
2858 * releaseBuffer(newLength) must be called if and only if a getBuffer(minCapacity) is "open".
2859 *
2860 * It will set the string length to newLength, at most to the current capacity.
2861 * If newLength==-1 then it will set the length according to the
2862 * first NUL in the buffer, or to the capacity if there is no NUL.
2863 *
2864 * After calling releaseBuffer(newLength) the UnicodeString is back to normal operation.
2865 *
2866 * @param newLength the new length of the UnicodeString object;
2867 * defaults to the current capacity if newLength is greater than that;
2868 * if newLength==-1, it defaults to u_strlen(buffer) but not more than
2869 * the current capacity of the string
2870 *
2871 * @see getBuffer(int32_t minCapacity)
2872 * @stable ICU 2.0
2873 */
2874 void releaseBuffer(int32_t newLength=-1);
2875
2876 /**
2877 * Get a read-only pointer to the internal buffer.
2878 * This can be called at any time on a valid UnicodeString.
2879 *
2880 * It returns 0 if the string is bogus, or
2881 * during an "open" getBuffer(minCapacity).
2882 *
2883 * It can be called as many times as desired.
2884 * The pointer that it returns will remain valid until the UnicodeString object is modified,
2885 * at which time the pointer is semantically invalidated and must not be used any more.
2886 *
2887 * The capacity of the buffer can be determined with getCapacity().
2888 * The part after length() may or may not be initialized and valid,
2889 * depending on the history of the UnicodeString object.
2890 *
2891 * The buffer contents is (probably) not NUL-terminated.
2892 * You can check if it is with
2893 * <code>(s.length()<s.getCapacity() && buffer[s.length()]==0)</code>.
2894 * (See getTerminatedBuffer().)
2895 *
2896 * The buffer may reside in read-only memory. Its contents must not
2897 * be modified.
2898 *
2899 * @return a read-only pointer to the internal string buffer,
2900 * or nullptr if the string is empty or bogus
2901 *
2902 * @see getBuffer(int32_t minCapacity)
2903 * @see getTerminatedBuffer()
2904 * @stable ICU 2.0
2905 */
2906 inline const char16_t *getBuffer() const;
2907
2908 /**
2909 * Get a read-only pointer to the internal buffer,
2910 * making sure that it is NUL-terminated.
2911 * This can be called at any time on a valid UnicodeString.
2912 *
2913 * It returns 0 if the string is bogus, or
2914 * during an "open" getBuffer(minCapacity), or if the buffer cannot
2915 * be NUL-terminated (because memory allocation failed).
2916 *
2917 * It can be called as many times as desired.
2918 * The pointer that it returns will remain valid until the UnicodeString object is modified,
2919 * at which time the pointer is semantically invalidated and must not be used any more.
2920 *
2921 * The capacity of the buffer can be determined with getCapacity().
2922 * The part after length()+1 may or may not be initialized and valid,
2923 * depending on the history of the UnicodeString object.
2924 *
2925 * The buffer contents is guaranteed to be NUL-terminated.
2926 * getTerminatedBuffer() may reallocate the buffer if a terminating NUL
2927 * is written.
2928 * For this reason, this function is not const, unlike getBuffer().
2929 * Note that a UnicodeString may also contain NUL characters as part of its contents.
2930 *
2931 * The buffer may reside in read-only memory. Its contents must not
2932 * be modified.
2933 *
2934 * @return a read-only pointer to the internal string buffer,
2935 * or 0 if the string is empty or bogus
2936 *
2937 * @see getBuffer(int32_t minCapacity)
2938 * @see getBuffer()
2939 * @stable ICU 2.2
2940 */
2941 const char16_t *getTerminatedBuffer();
2942
2943 //========================================
2944 // Constructors
2945 //========================================
2946
2947 /** Construct an empty UnicodeString.
2948 * @stable ICU 2.0
2949 */
2950 inline UnicodeString();
2951
2952 /**
2953 * Construct a UnicodeString with capacity to hold <TT>capacity</TT> char16_ts
2954 * @param capacity the number of char16_ts this UnicodeString should hold
2955 * before a resize is necessary; if count is greater than 0 and count
2956 * code points c take up more space than capacity, then capacity is adjusted
2957 * accordingly.
2958 * @param c is used to initially fill the string
2959 * @param count specifies how many code points c are to be written in the
2960 * string
2961 * @stable ICU 2.0
2962 */
2963 UnicodeString(int32_t capacity, UChar32 c, int32_t count);
2964
2965 /**
2966 * Single char16_t (code unit) constructor.
2967 *
2968 * It is recommended to mark this constructor "explicit" by
2969 * <code>-DUNISTR_FROM_CHAR_EXPLICIT=explicit</code>
2970 * on the compiler command line or similar.
2971 * @param ch the character to place in the UnicodeString
2972 * @stable ICU 2.0
2973 */
2974 UNISTR_FROM_CHAR_EXPLICIT UnicodeString(char16_t ch);
2975
2976 /**
2977 * Single UChar32 (code point) constructor.
2978 *
2979 * It is recommended to mark this constructor "explicit" by
2980 * <code>-DUNISTR_FROM_CHAR_EXPLICIT=explicit</code>
2981 * on the compiler command line or similar.
2982 * @param ch the character to place in the UnicodeString
2983 * @stable ICU 2.0
2984 */
2985 UNISTR_FROM_CHAR_EXPLICIT UnicodeString(UChar32 ch);
2986
2987 /**
2988 * char16_t* constructor.
2989 *
2990 * It is recommended to mark this constructor "explicit" by
2991 * <code>-DUNISTR_FROM_STRING_EXPLICIT=explicit</code>
2992 * on the compiler command line or similar.
2993 * @param text The characters to place in the UnicodeString. <TT>text</TT>
2994 * must be NULL (U+0000) terminated.
2995 * @stable ICU 2.0
2996 */
2997 UNISTR_FROM_STRING_EXPLICIT UnicodeString(const char16_t *text);
2998
2999 /*
3000 * Do not use #ifndef U_HIDE_DRAFT_API for the following constructor,
3001 * it should always be available regardless of U_HIDE_DRAFT_API status
3002 */
3003#if !U_CHAR16_IS_TYPEDEF
3004 /**
3005 * uint16_t * constructor.
3006 * Delegates to UnicodeString(const char16_t *).
3007 *
3008 * It is recommended to mark this constructor "explicit" by
3009 * <code>-DUNISTR_FROM_STRING_EXPLICIT=explicit</code>
3010 * on the compiler command line or similar.
3011 * @param text NUL-terminated UTF-16 string
3012 * @draft ICU 59
3013 */
3014 UNISTR_FROM_STRING_EXPLICIT UnicodeString(const uint16_t *text) :
3015 UnicodeString(ConstChar16Ptr(text)) {}
3016#endif
3017
3018 /*
3019 * Do not use #ifndef U_HIDE_DRAFT_API for the following constructor,
3020 * it should always be available regardless of U_HIDE_DRAFT_API status
3021 */
3022#if U_SIZEOF_WCHAR_T==2 || defined(U_IN_DOXYGEN)
3023 /**
3024 * wchar_t * constructor.
3025 * (Only defined if U_SIZEOF_WCHAR_T==2.)
3026 * Delegates to UnicodeString(const char16_t *).
3027 *
3028 * It is recommended to mark this constructor "explicit" by
3029 * <code>-DUNISTR_FROM_STRING_EXPLICIT=explicit</code>
3030 * on the compiler command line or similar.
3031 * @param text NUL-terminated UTF-16 string
3032 * @draft ICU 59
3033 */
3034 UNISTR_FROM_STRING_EXPLICIT UnicodeString(const wchar_t *text) :
3035 UnicodeString(ConstChar16Ptr(text)) {}
3036#endif
3037
3038 /*
3039 * Do not use #ifndef U_HIDE_DRAFT_API for the following constructor,
3040 * it should always be available regardless of U_HIDE_DRAFT_API status
3041 */
3042 /**
3043 * nullptr_t constructor.
3044 * Effectively the same as the default constructor, makes an empty string object.
3045 *
3046 * It is recommended to mark this constructor "explicit" by
3047 * <code>-DUNISTR_FROM_STRING_EXPLICIT=explicit</code>
3048 * on the compiler command line or similar.
3049 * @param text nullptr
3050 * @draft ICU 59
3051 */
3052 UNISTR_FROM_STRING_EXPLICIT inline UnicodeString(const std::nullptr_t text);
3053
3054 /**
3055 * char16_t* constructor.
3056 * @param text The characters to place in the UnicodeString.
3057 * @param textLength The number of Unicode characters in <TT>text</TT>
3058 * to copy.
3059 * @stable ICU 2.0
3060 */
3061 UnicodeString(const char16_t *text,
3062 int32_t textLength);
3063
3064 /*
3065 * Do not use #ifndef U_HIDE_DRAFT_API for the following constructor,
3066 * it should always be available regardless of U_HIDE_DRAFT_API status
3067 */
3068#if !U_CHAR16_IS_TYPEDEF
3069 /**
3070 * uint16_t * constructor.
3071 * Delegates to UnicodeString(const char16_t *, int32_t).
3072 * @param text UTF-16 string
3073 * @param length string length
3074 * @draft ICU 59
3075 */
3076 UnicodeString(const uint16_t *text, int32_t length) :
3077 UnicodeString(ConstChar16Ptr(text), length) {}
3078#endif
3079
3080 /*
3081 * Do not use #ifndef U_HIDE_DRAFT_API for the following constructor,
3082 * it should always be available regardless of U_HIDE_DRAFT_API status
3083 */
3084#if U_SIZEOF_WCHAR_T==2 || defined(U_IN_DOXYGEN)
3085 /**
3086 * wchar_t * constructor.
3087 * (Only defined if U_SIZEOF_WCHAR_T==2.)
3088 * Delegates to UnicodeString(const char16_t *, int32_t).
3089 * @param text NUL-terminated UTF-16 string
3090 * @param length string length
3091 * @draft ICU 59
3092 */
3093 UnicodeString(const wchar_t *text, int32_t length) :
3094 UnicodeString(ConstChar16Ptr(text), length) {}
3095#endif
3096
3097 /*
3098 * Do not use #ifndef U_HIDE_DRAFT_API for the following constructor,
3099 * it should always be available regardless of U_HIDE_DRAFT_API status
3100 */
3101 /**
3102 * nullptr_t constructor.
3103 * Effectively the same as the default constructor, makes an empty string object.
3104 * @param text nullptr
3105 * @param length ignored
3106 * @draft ICU 59
3107 */
3108 inline UnicodeString(const std::nullptr_t text, int32_t length);
3109
3110 /**
3111 * Readonly-aliasing char16_t* constructor.
3112 * The text will be used for the UnicodeString object, but
3113 * it will not be released when the UnicodeString is destroyed.
3114 * This has copy-on-write semantics:
3115 * When the string is modified, then the buffer is first copied into
3116 * newly allocated memory.
3117 * The aliased buffer is never modified.
3118 *
3119 * In an assignment to another UnicodeString, when using the copy constructor
3120 * or the assignment operator, the text will be copied.
3121 * When using fastCopyFrom(), the text will be aliased again,
3122 * so that both strings then alias the same readonly-text.
3123 *
3124 * @param isTerminated specifies if <code>text</code> is <code>NUL</code>-terminated.
3125 * This must be true if <code>textLength==-1</code>.
3126 * @param text The characters to alias for the UnicodeString.
3127 * @param textLength The number of Unicode characters in <code>text</code> to alias.
3128 * If -1, then this constructor will determine the length
3129 * by calling <code>u_strlen()</code>.
3130 * @stable ICU 2.0
3131 */
3132 UnicodeString(UBool isTerminated,
3133 ConstChar16Ptr text,
3134 int32_t textLength);
3135
3136 /**
3137 * Writable-aliasing char16_t* constructor.
3138 * The text will be used for the UnicodeString object, but
3139 * it will not be released when the UnicodeString is destroyed.
3140 * This has write-through semantics:
3141 * For as long as the capacity of the buffer is sufficient, write operations
3142 * will directly affect the buffer. When more capacity is necessary, then
3143 * a new buffer will be allocated and the contents copied as with regularly
3144 * constructed strings.
3145 * In an assignment to another UnicodeString, the buffer will be copied.
3146 * The extract(Char16Ptr dst) function detects whether the dst pointer is the same
3147 * as the string buffer itself and will in this case not copy the contents.
3148 *
3149 * @param buffer The characters to alias for the UnicodeString.
3150 * @param buffLength The number of Unicode characters in <code>buffer</code> to alias.
3151 * @param buffCapacity The size of <code>buffer</code> in char16_ts.
3152 * @stable ICU 2.0
3153 */
3154 UnicodeString(char16_t *buffer, int32_t buffLength, int32_t buffCapacity);
3155
3156 /*
3157 * Do not use #ifndef U_HIDE_DRAFT_API for the following constructor,
3158 * it should always be available regardless of U_HIDE_DRAFT_API status
3159 */
3160#if !U_CHAR16_IS_TYPEDEF
3161 /**
3162 * Writable-aliasing uint16_t * constructor.
3163 * Delegates to UnicodeString(const char16_t *, int32_t, int32_t).
3164 * @param buffer writable buffer of/for UTF-16 text
3165 * @param buffLength length of the current buffer contents
3166 * @param buffCapacity buffer capacity
3167 * @draft ICU 59
3168 */
3169 UnicodeString(uint16_t *buffer, int32_t buffLength, int32_t buffCapacity) :
3170 UnicodeString(Char16Ptr(buffer), buffLength, buffCapacity) {}
3171#endif
3172
3173 /*
3174 * Do not use #ifndef U_HIDE_DRAFT_API for the following constructor,
3175 * it should always be available regardless of U_HIDE_DRAFT_API status
3176 */
3177#if U_SIZEOF_WCHAR_T==2 || defined(U_IN_DOXYGEN)
3178 /**
3179 * Writable-aliasing wchar_t * constructor.
3180 * (Only defined if U_SIZEOF_WCHAR_T==2.)
3181 * Delegates to UnicodeString(const char16_t *, int32_t, int32_t).
3182 * @param buffer writable buffer of/for UTF-16 text
3183 * @param buffLength length of the current buffer contents
3184 * @param buffCapacity buffer capacity
3185 * @draft ICU 59
3186 */
3187 UnicodeString(wchar_t *buffer, int32_t buffLength, int32_t buffCapacity) :
3188 UnicodeString(Char16Ptr(buffer), buffLength, buffCapacity) {}
3189#endif
3190
3191 /*
3192 * Do not use #ifndef U_HIDE_DRAFT_API for the following constructor,
3193 * it should always be available regardless of U_HIDE_DRAFT_API status
3194 */
3195 /**
3196 * Writable-aliasing nullptr_t constructor.
3197 * Effectively the same as the default constructor, makes an empty string object.
3198 * @param buffer nullptr
3199 * @param buffLength ignored
3200 * @param buffCapacity ignored
3201 * @draft ICU 59
3202 */
3203 inline UnicodeString(std::nullptr_t buffer, int32_t buffLength, int32_t buffCapacity);
3204
3205#if U_CHARSET_IS_UTF8 || !UCONFIG_NO_CONVERSION
3206
3207 /**
3208 * char* constructor.
3209 * Uses the default converter (and thus depends on the ICU conversion code)
3210 * unless U_CHARSET_IS_UTF8 is set to 1.
3211 *
3212 * For ASCII (really "invariant character") strings it is more efficient to use
3213 * the constructor that takes a US_INV (for its enum EInvariant).
3214 * For ASCII (invariant-character) string literals, see UNICODE_STRING and
3215 * UNICODE_STRING_SIMPLE.
3216 *
3217 * It is recommended to mark this constructor "explicit" by
3218 * <code>-DUNISTR_FROM_STRING_EXPLICIT=explicit</code>
3219 * on the compiler command line or similar.
3220 * @param codepageData an array of bytes, null-terminated,
3221 * in the platform's default codepage.
3222 * @stable ICU 2.0
3223 * @see UNICODE_STRING
3224 * @see UNICODE_STRING_SIMPLE
3225 */
3226 UNISTR_FROM_STRING_EXPLICIT UnicodeString(const char *codepageData);
3227
3228 /**
3229 * char* constructor.
3230 * Uses the default converter (and thus depends on the ICU conversion code)
3231 * unless U_CHARSET_IS_UTF8 is set to 1.
3232 * @param codepageData an array of bytes in the platform's default codepage.
3233 * @param dataLength The number of bytes in <TT>codepageData</TT>.
3234 * @stable ICU 2.0
3235 */
3236 UnicodeString(const char *codepageData, int32_t dataLength);
3237
3238#endif
3239
3240#if !UCONFIG_NO_CONVERSION
3241
3242 /**
3243 * char* constructor.
3244 * @param codepageData an array of bytes, null-terminated
3245 * @param codepage the encoding of <TT>codepageData</TT>. The special
3246 * value 0 for <TT>codepage</TT> indicates that the text is in the
3247 * platform's default codepage.
3248 *
3249 * If <code>codepage</code> is an empty string (<code>""</code>),
3250 * then a simple conversion is performed on the codepage-invariant
3251 * subset ("invariant characters") of the platform encoding. See utypes.h.
3252 * Recommendation: For invariant-character strings use the constructor
3253 * UnicodeString(const char *src, int32_t length, enum EInvariant inv)
3254 * because it avoids object code dependencies of UnicodeString on
3255 * the conversion code.
3256 *
3257 * @stable ICU 2.0
3258 */
3259 UnicodeString(const char *codepageData, const char *codepage);
3260
3261 /**
3262 * char* constructor.
3263 * @param codepageData an array of bytes.
3264 * @param dataLength The number of bytes in <TT>codepageData</TT>.
3265 * @param codepage the encoding of <TT>codepageData</TT>. The special
3266 * value 0 for <TT>codepage</TT> indicates that the text is in the
3267 * platform's default codepage.
3268 * If <code>codepage</code> is an empty string (<code>""</code>),
3269 * then a simple conversion is performed on the codepage-invariant
3270 * subset ("invariant characters") of the platform encoding. See utypes.h.
3271 * Recommendation: For invariant-character strings use the constructor
3272 * UnicodeString(const char *src, int32_t length, enum EInvariant inv)
3273 * because it avoids object code dependencies of UnicodeString on
3274 * the conversion code.
3275 *
3276 * @stable ICU 2.0
3277 */
3278 UnicodeString(const char *codepageData, int32_t dataLength, const char *codepage);
3279
3280 /**
3281 * char * / UConverter constructor.
3282 * This constructor uses an existing UConverter object to
3283 * convert the codepage string to Unicode and construct a UnicodeString
3284 * from that.
3285 *
3286 * The converter is reset at first.
3287 * If the error code indicates a failure before this constructor is called,
3288 * or if an error occurs during conversion or construction,
3289 * then the string will be bogus.
3290 *
3291 * This function avoids the overhead of opening and closing a converter if
3292 * multiple strings are constructed.
3293 *
3294 * @param src input codepage string
3295 * @param srcLength length of the input string, can be -1 for NUL-terminated strings
3296 * @param cnv converter object (ucnv_resetToUnicode() will be called),
3297 * can be NULL for the default converter
3298 * @param errorCode normal ICU error code
3299 * @stable ICU 2.0
3300 */
3301 UnicodeString(
3302 const char *src, int32_t srcLength,
3303 UConverter *cnv,
3304 UErrorCode &errorCode);
3305
3306#endif
3307
3308 /**
3309 * Constructs a Unicode string from an invariant-character char * string.
3310 * About invariant characters see utypes.h.
3311 * This constructor has no runtime dependency on conversion code and is
3312 * therefore recommended over ones taking a charset name string
3313 * (where the empty string "" indicates invariant-character conversion).
3314 *
3315 * Use the macro US_INV as the third, signature-distinguishing parameter.
3316 *
3317 * For example:
3318 * \code
3319 * void fn(const char *s) {
3320 * UnicodeString ustr(s, -1, US_INV);
3321 * // use ustr ...
3322 * }
3323 * \endcode
3324 *
3325 * @param src String using only invariant characters.
3326 * @param length Length of src, or -1 if NUL-terminated.
3327 * @param inv Signature-distinguishing paramater, use US_INV.
3328 *
3329 * @see US_INV
3330 * @stable ICU 3.2
3331 */
3332 UnicodeString(const char *src, int32_t length, enum EInvariant inv);
3333
3334
3335 /**
3336 * Copy constructor.
3337 *
3338 * Starting with ICU 2.4, the assignment operator and the copy constructor
3339 * allocate a new buffer and copy the buffer contents even for readonly aliases.
3340 * By contrast, the fastCopyFrom() function implements the old,
3341 * more efficient but less safe behavior
3342 * of making this string also a readonly alias to the same buffer.
3343 *
3344 * If the source object has an "open" buffer from getBuffer(minCapacity),
3345 * then the copy is an empty string.
3346 *
3347 * @param that The UnicodeString object to copy.
3348 * @stable ICU 2.0
3349 * @see fastCopyFrom
3350 */
3351 UnicodeString(const UnicodeString& that);
3352
3353#if U_HAVE_RVALUE_REFERENCES
3354 /**
3355 * Move constructor, might leave src in bogus state.
3356 * This string will have the same contents and state that the source string had.
3357 * @param src source string
3358 * @stable ICU 56
3359 */
3360 UnicodeString(UnicodeString &&src) U_NOEXCEPT;
3361#endif
3362
3363 /**
3364 * 'Substring' constructor from tail of source string.
3365 * @param src The UnicodeString object to copy.
3366 * @param srcStart The offset into <tt>src</tt> at which to start copying.
3367 * @stable ICU 2.2
3368 */
3369 UnicodeString(const UnicodeString& src, int32_t srcStart);
3370
3371 /**
3372 * 'Substring' constructor from subrange of source string.
3373 * @param src The UnicodeString object to copy.
3374 * @param srcStart The offset into <tt>src</tt> at which to start copying.
3375 * @param srcLength The number of characters from <tt>src</tt> to copy.
3376 * @stable ICU 2.2
3377 */
3378 UnicodeString(const UnicodeString& src, int32_t srcStart, int32_t srcLength);
3379
3380 /**
3381 * Clone this object, an instance of a subclass of Replaceable.
3382 * Clones can be used concurrently in multiple threads.
3383 * If a subclass does not implement clone(), or if an error occurs,
3384 * then NULL is returned.
3385 * The clone functions in all subclasses return a pointer to a Replaceable
3386 * because some compilers do not support covariant (same-as-this)
3387 * return types; cast to the appropriate subclass if necessary.
3388 * The caller must delete the clone.
3389 *
3390 * @return a clone of this object
3391 *
3392 * @see Replaceable::clone
3393 * @see getDynamicClassID
3394 * @stable ICU 2.6
3395 */
3396 virtual Replaceable *clone() const;
3397
3398 /** Destructor.
3399 * @stable ICU 2.0
3400 */
3401 virtual ~UnicodeString();
3402
3403 /**
3404 * Create a UnicodeString from a UTF-8 string.
3405 * Illegal input is replaced with U+FFFD. Otherwise, errors result in a bogus string.
3406 * Calls u_strFromUTF8WithSub().
3407 *
3408 * @param utf8 UTF-8 input string.
3409 * Note that a StringPiece can be implicitly constructed
3410 * from a std::string or a NUL-terminated const char * string.
3411 * @return A UnicodeString with equivalent UTF-16 contents.
3412 * @see toUTF8
3413 * @see toUTF8String
3414 * @stable ICU 4.2
3415 */
3416 static UnicodeString fromUTF8(StringPiece utf8);
3417
3418 /**
3419 * Create a UnicodeString from a UTF-32 string.
3420 * Illegal input is replaced with U+FFFD. Otherwise, errors result in a bogus string.
3421 * Calls u_strFromUTF32WithSub().
3422 *
3423 * @param utf32 UTF-32 input string. Must not be NULL.
3424 * @param length Length of the input string, or -1 if NUL-terminated.
3425 * @return A UnicodeString with equivalent UTF-16 contents.
3426 * @see toUTF32
3427 * @stable ICU 4.2
3428 */
3429 static UnicodeString fromUTF32(const UChar32 *utf32, int32_t length);
3430
3431 /* Miscellaneous operations */
3432
3433 /**
3434 * Unescape a string of characters and return a string containing
3435 * the result. The following escape sequences are recognized:
3436 *
3437 * \\uhhhh 4 hex digits; h in [0-9A-Fa-f]
3438 * \\Uhhhhhhhh 8 hex digits
3439 * \\xhh 1-2 hex digits
3440 * \\ooo 1-3 octal digits; o in [0-7]
3441 * \\cX control-X; X is masked with 0x1F
3442 *
3443 * as well as the standard ANSI C escapes:
3444 *
3445 * \\a => U+0007, \\b => U+0008, \\t => U+0009, \\n => U+000A,
3446 * \\v => U+000B, \\f => U+000C, \\r => U+000D, \\e => U+001B,
3447 * \\&quot; => U+0022, \\' => U+0027, \\? => U+003F, \\\\ => U+005C
3448 *
3449 * Anything else following a backslash is generically escaped. For
3450 * example, "[a\\-z]" returns "[a-z]".
3451 *
3452 * If an escape sequence is ill-formed, this method returns an empty
3453 * string. An example of an ill-formed sequence is "\\u" followed by
3454 * fewer than 4 hex digits.
3455 *
3456 * This function is similar to u_unescape() but not identical to it.
3457 * The latter takes a source char*, so it does escape recognition
3458 * and also invariant conversion.
3459 *
3460 * @return a string with backslash escapes interpreted, or an
3461 * empty string on error.
3462 * @see UnicodeString#unescapeAt()
3463 * @see u_unescape()
3464 * @see u_unescapeAt()
3465 * @stable ICU 2.0
3466 */
3467 UnicodeString unescape() const;
3468
3469 /**
3470 * Unescape a single escape sequence and return the represented
3471 * character. See unescape() for a listing of the recognized escape
3472 * sequences. The character at offset-1 is assumed (without
3473 * checking) to be a backslash. If the escape sequence is
3474 * ill-formed, or the offset is out of range, U_SENTINEL=-1 is
3475 * returned.
3476 *
3477 * @param offset an input output parameter. On input, it is the
3478 * offset into this string where the escape sequence is located,
3479 * after the initial backslash. On output, it is advanced after the
3480 * last character parsed. On error, it is not advanced at all.
3481 * @return the character represented by the escape sequence at
3482 * offset, or U_SENTINEL=-1 on error.
3483 * @see UnicodeString#unescape()
3484 * @see u_unescape()
3485 * @see u_unescapeAt()
3486 * @stable ICU 2.0
3487 */
3488 UChar32 unescapeAt(int32_t &offset) const;
3489
3490 /**
3491 * ICU "poor man's RTTI", returns a UClassID for this class.
3492 *
3493 * @stable ICU 2.2
3494 */
3495 static UClassID U_EXPORT2 getStaticClassID();
3496
3497 /**
3498 * ICU "poor man's RTTI", returns a UClassID for the actual class.
3499 *
3500 * @stable ICU 2.2
3501 */
3502 virtual UClassID getDynamicClassID() const;
3503
3504 //========================================
3505 // Implementation methods
3506 //========================================
3507
3508protected:
3509 /**
3510 * Implement Replaceable::getLength() (see jitterbug 1027).
3511 * @stable ICU 2.4
3512 */
3513 virtual int32_t getLength() const;
3514
3515 /**
3516 * The change in Replaceable to use virtual getCharAt() allows
3517 * UnicodeString::charAt() to be inline again (see jitterbug 709).
3518 * @stable ICU 2.4
3519 */
3520 virtual char16_t getCharAt(int32_t offset) const;
3521
3522 /**
3523 * The change in Replaceable to use virtual getChar32At() allows
3524 * UnicodeString::char32At() to be inline again (see jitterbug 709).
3525 * @stable ICU 2.4
3526 */
3527 virtual UChar32 getChar32At(int32_t offset) const;
3528
3529private:
3530 // For char* constructors. Could be made public.
3531 UnicodeString &setToUTF8(StringPiece utf8);
3532 // For extract(char*).
3533 // We could make a toUTF8(target, capacity, errorCode) public but not
3534 // this version: New API will be cleaner if we make callers create substrings
3535 // rather than having start+length on every method,
3536 // and it should take a UErrorCode&.
3537 int32_t
3538 toUTF8(int32_t start, int32_t len,
3539 char *target, int32_t capacity) const;
3540
3541 /**
3542 * Internal string contents comparison, called by operator==.
3543 * Requires: this & text not bogus and have same lengths.
3544 */
3545 UBool doEquals(const UnicodeString &text, int32_t len) const;
3546
3547 inline int8_t
3548 doCompare(int32_t start,
3549 int32_t length,
3550 const UnicodeString& srcText,
3551 int32_t srcStart,
3552 int32_t srcLength) const;
3553
3554 int8_t doCompare(int32_t start,
3555 int32_t length,
3556 const char16_t *srcChars,
3557 int32_t srcStart,
3558 int32_t srcLength) const;
3559
3560 inline int8_t
3561 doCompareCodePointOrder(int32_t start,
3562 int32_t length,
3563 const UnicodeString& srcText,
3564 int32_t srcStart,
3565 int32_t srcLength) const;
3566
3567 int8_t doCompareCodePointOrder(int32_t start,
3568 int32_t length,
3569 const char16_t *srcChars,
3570 int32_t srcStart,
3571 int32_t srcLength) const;
3572
3573 inline int8_t
3574 doCaseCompare(int32_t start,
3575 int32_t length,
3576 const UnicodeString &srcText,
3577 int32_t srcStart,
3578 int32_t srcLength,
3579 uint32_t options) const;
3580
3581 int8_t
3582 doCaseCompare(int32_t start,
3583 int32_t length,
3584 const char16_t *srcChars,
3585 int32_t srcStart,
3586 int32_t srcLength,
3587 uint32_t options) const;
3588
3589 int32_t doIndexOf(char16_t c,
3590 int32_t start,
3591 int32_t length) const;
3592
3593 int32_t doIndexOf(UChar32 c,
3594 int32_t start,
3595 int32_t length) const;
3596
3597 int32_t doLastIndexOf(char16_t c,
3598 int32_t start,
3599 int32_t length) const;
3600
3601 int32_t doLastIndexOf(UChar32 c,
3602 int32_t start,
3603 int32_t length) const;
3604
3605 void doExtract(int32_t start,
3606 int32_t length,
3607 char16_t *dst,
3608 int32_t dstStart) const;
3609
3610 inline void doExtract(int32_t start,
3611 int32_t length,
3612 UnicodeString& target) const;
3613
3614 inline char16_t doCharAt(int32_t offset) const;
3615
3616 UnicodeString& doReplace(int32_t start,
3617 int32_t length,
3618 const UnicodeString& srcText,
3619 int32_t srcStart,
3620 int32_t srcLength);
3621
3622 UnicodeString& doReplace(int32_t start,
3623 int32_t length,
3624 const char16_t *srcChars,
3625 int32_t srcStart,
3626 int32_t srcLength);
3627
3628 UnicodeString& doAppend(const UnicodeString& src, int32_t srcStart, int32_t srcLength);
3629 UnicodeString& doAppend(const char16_t *srcChars, int32_t srcStart, int32_t srcLength);
3630
3631 UnicodeString& doReverse(int32_t start,
3632 int32_t length);
3633
3634 // calculate hash code
3635 int32_t doHashCode(void) const;
3636
3637 // get pointer to start of array
3638 // these do not check for kOpenGetBuffer, unlike the public getBuffer() function
3639 inline char16_t* getArrayStart(void);
3640 inline const char16_t* getArrayStart(void) const;
3641
3642 inline UBool hasShortLength() const;
3643 inline int32_t getShortLength() const;
3644
3645 // A UnicodeString object (not necessarily its current buffer)
3646 // is writable unless it isBogus() or it has an "open" getBuffer(minCapacity).
3647 inline UBool isWritable() const;
3648
3649 // Is the current buffer writable?
3650 inline UBool isBufferWritable() const;
3651
3652 // None of the following does releaseArray().
3653 inline void setZeroLength();
3654 inline void setShortLength(int32_t len);
3655 inline void setLength(int32_t len);
3656 inline void setToEmpty();
3657 inline void setArray(char16_t *array, int32_t len, int32_t capacity); // sets length but not flags
3658
3659 // allocate the array; result may be the stack buffer
3660 // sets refCount to 1 if appropriate
3661 // sets fArray, fCapacity, and flags
3662 // sets length to 0
3663 // returns boolean for success or failure
3664 UBool allocate(int32_t capacity);
3665
3666 // release the array if owned
3667 void releaseArray(void);
3668
3669 // turn a bogus string into an empty one
3670 void unBogus();
3671
3672 // implements assigment operator, copy constructor, and fastCopyFrom()
3673 UnicodeString &copyFrom(const UnicodeString &src, UBool fastCopy=FALSE);
3674
3675 // Copies just the fields without memory management.
3676 void copyFieldsFrom(UnicodeString &src, UBool setSrcToBogus) U_NOEXCEPT;
3677
3678 // Pin start and limit to acceptable values.
3679 inline void pinIndex(int32_t& start) const;
3680 inline void pinIndices(int32_t& start,
3681 int32_t& length) const;
3682
3683#if !UCONFIG_NO_CONVERSION
3684
3685 /* Internal extract() using UConverter. */
3686 int32_t doExtract(int32_t start, int32_t length,
3687 char *dest, int32_t destCapacity,
3688 UConverter *cnv,
3689 UErrorCode &errorCode) const;
3690
3691 /*
3692 * Real constructor for converting from codepage data.
3693 * It assumes that it is called with !fRefCounted.
3694 *
3695 * If <code>codepage==0</code>, then the default converter
3696 * is used for the platform encoding.
3697 * If <code>codepage</code> is an empty string (<code>""</code>),
3698 * then a simple conversion is performed on the codepage-invariant
3699 * subset ("invariant characters") of the platform encoding. See utypes.h.
3700 */
3701 void doCodepageCreate(const char *codepageData,
3702 int32_t dataLength,
3703 const char *codepage);
3704
3705 /*
3706 * Worker function for creating a UnicodeString from
3707 * a codepage string using a UConverter.
3708 */
3709 void
3710 doCodepageCreate(const char *codepageData,
3711 int32_t dataLength,
3712 UConverter *converter,
3713 UErrorCode &status);
3714
3715#endif
3716
3717 /*
3718 * This function is called when write access to the array
3719 * is necessary.
3720 *
3721 * We need to make a copy of the array if
3722 * the buffer is read-only, or
3723 * the buffer is refCounted (shared), and refCount>1, or
3724 * the buffer is too small.
3725 *
3726 * Return FALSE if memory could not be allocated.
3727 */
3728 UBool cloneArrayIfNeeded(int32_t newCapacity = -1,
3729 int32_t growCapacity = -1,
3730 UBool doCopyArray = TRUE,
3731 int32_t **pBufferToDelete = 0,
3732 UBool forceClone = FALSE);
3733
3734 /**
3735 * Common function for UnicodeString case mappings.
3736 * The stringCaseMapper has the same type UStringCaseMapper
3737 * as in ustr_imp.h for ustrcase_map().
3738 */
3739 UnicodeString &
3740 caseMap(int32_t caseLocale, uint32_t options,
3741#if !UCONFIG_NO_BREAK_ITERATION
3742 BreakIterator *iter,
3743#endif
3744 UStringCaseMapper *stringCaseMapper);
3745
3746 // ref counting
3747 void addRef(void);
3748 int32_t removeRef(void);
3749 int32_t refCount(void) const;
3750
3751 // constants
3752 enum {
3753 /**
3754 * Size of stack buffer for short strings.
3755 * Must be at least U16_MAX_LENGTH for the single-code point constructor to work.
3756 * @see UNISTR_OBJECT_SIZE
3757 */
3758 US_STACKBUF_SIZE=(int32_t)(UNISTR_OBJECT_SIZE-sizeof(void *)-2)/U_SIZEOF_UCHAR,
3759 kInvalidUChar=0xffff, // U+FFFF returned by charAt(invalid index)
3760 kInvalidHashCode=0, // invalid hash code
3761 kEmptyHashCode=1, // hash code for empty string
3762
3763 // bit flag values for fLengthAndFlags
3764 kIsBogus=1, // this string is bogus, i.e., not valid or NULL
3765 kUsingStackBuffer=2,// using fUnion.fStackFields instead of fUnion.fFields
3766 kRefCounted=4, // there is a refCount field before the characters in fArray
3767 kBufferIsReadonly=8,// do not write to this buffer
3768 kOpenGetBuffer=16, // getBuffer(minCapacity) was called (is "open"),
3769 // and releaseBuffer(newLength) must be called
3770 kAllStorageFlags=0x1f,
3771
3772 kLengthShift=5, // remaining 11 bits for non-negative short length, or negative if long
3773 kLength1=1<<kLengthShift,
3774 kMaxShortLength=0x3ff, // max non-negative short length (leaves top bit 0)
3775 kLengthIsLarge=0xffe0, // short length < 0, real length is in fUnion.fFields.fLength
3776
3777 // combined values for convenience
3778 kShortString=kUsingStackBuffer,
3779 kLongString=kRefCounted,
3780 kReadonlyAlias=kBufferIsReadonly,
3781 kWritableAlias=0
3782 };
3783
3784 friend class UnicodeStringAppendable;
3785
3786 union StackBufferOrFields; // forward declaration necessary before friend declaration
3787 friend union StackBufferOrFields; // make US_STACKBUF_SIZE visible inside fUnion
3788
3789 /*
3790 * The following are all the class fields that are stored
3791 * in each UnicodeString object.
3792 * Note that UnicodeString has virtual functions,
3793 * therefore there is an implicit vtable pointer
3794 * as the first real field.
3795 * The fields should be aligned such that no padding is necessary.
3796 * On 32-bit machines, the size should be 32 bytes,
3797 * on 64-bit machines (8-byte pointers), it should be 40 bytes.
3798 *
3799 * We use a hack to achieve this.
3800 *
3801 * With at least some compilers, each of the following is forced to
3802 * a multiple of sizeof(pointer) [the largest field base unit here is a data pointer],
3803 * rounded up with additional padding if the fields do not already fit that requirement:
3804 * - sizeof(class UnicodeString)
3805 * - offsetof(UnicodeString, fUnion)
3806 * - sizeof(fUnion)
3807 * - sizeof(fStackFields)
3808 *
3809 * We optimize for the longest possible internal buffer for short strings.
3810 * fUnion.fStackFields begins with 2 bytes for storage flags
3811 * and the length of relatively short strings,
3812 * followed by the buffer for short string contents.
3813 * There is no padding inside fStackFields.
3814 *
3815 * Heap-allocated and aliased strings use fUnion.fFields.
3816 * Both fStackFields and fFields must begin with the same fields for flags and short length,
3817 * that is, those must have the same memory offsets inside the object,
3818 * because the flags must be inspected in order to decide which half of fUnion is being used.
3819 * We assume that the compiler does not reorder the fields.
3820 *
3821 * (Padding at the end of fFields is ok:
3822 * As long as it is no larger than fStackFields, it is not wasted space.)
3823 *
3824 * For some of the history of the UnicodeString class fields layout, see
3825 * - ICU ticket #11551 "longer UnicodeString contents in stack buffer"
3826 * - ICU ticket #11336 "UnicodeString: recombine stack buffer arrays"
3827 * - ICU ticket #8322 "why is sizeof(UnicodeString)==48?"
3828 */
3829 // (implicit) *vtable;
3830 union StackBufferOrFields {
3831 // fStackFields is used iff (fLengthAndFlags&kUsingStackBuffer) else fFields is used.
3832 // Each struct of the union must begin with fLengthAndFlags.
3833 struct {
3834 int16_t fLengthAndFlags; // bit fields: see constants above
3835 char16_t fBuffer[US_STACKBUF_SIZE]; // buffer for short strings
3836 } fStackFields;
3837 struct {
3838 int16_t fLengthAndFlags; // bit fields: see constants above
3839 int32_t fLength; // number of characters in fArray if >127; else undefined
3840 int32_t fCapacity; // capacity of fArray (in char16_ts)
3841 // array pointer last to minimize padding for machines with P128 data model
3842 // or pointer sizes that are not a power of 2
3843 char16_t *fArray; // the Unicode data
3844 } fFields;
3845 } fUnion;
3846};
3847
3848/**
3849 * Create a new UnicodeString with the concatenation of two others.
3850 *
3851 * @param s1 The first string to be copied to the new one.
3852 * @param s2 The second string to be copied to the new one, after s1.
3853 * @return UnicodeString(s1).append(s2)
3854 * @stable ICU 2.8
3855 */
3856U_COMMON_API UnicodeString U_EXPORT2
3857operator+ (const UnicodeString &s1, const UnicodeString &s2);
3858
3859//========================================
3860// Inline members
3861//========================================
3862
3863//========================================
3864// Privates
3865//========================================
3866
3867inline void
3868UnicodeString::pinIndex(int32_t& start) const
3869{
3870 // pin index
3871 if(start < 0) {
3872 start = 0;
3873 } else if(start > length()) {
3874 start = length();
3875 }
3876}
3877
3878inline void
3879UnicodeString::pinIndices(int32_t& start,
3880 int32_t& _length) const
3881{
3882 // pin indices
3883 int32_t len = length();
3884 if(start < 0) {
3885 start = 0;
3886 } else if(start > len) {
3887 start = len;
3888 }
3889 if(_length < 0) {
3890 _length = 0;
3891 } else if(_length > (len - start)) {
3892 _length = (len - start);
3893 }
3894}
3895
3896inline char16_t*
3897UnicodeString::getArrayStart() {
3898 return (fUnion.fFields.fLengthAndFlags&kUsingStackBuffer) ?
3899 fUnion.fStackFields.fBuffer : fUnion.fFields.fArray;
3900}
3901
3902inline const char16_t*
3903UnicodeString::getArrayStart() const {
3904 return (fUnion.fFields.fLengthAndFlags&kUsingStackBuffer) ?
3905 fUnion.fStackFields.fBuffer : fUnion.fFields.fArray;
3906}
3907
3908//========================================
3909// Default constructor
3910//========================================
3911
3912inline
3913UnicodeString::UnicodeString() {
3914 fUnion.fStackFields.fLengthAndFlags=kShortString;
3915}
3916
3917inline UnicodeString::UnicodeString(const std::nullptr_t /*text*/) {
3918 fUnion.fStackFields.fLengthAndFlags=kShortString;
3919}
3920
3921inline UnicodeString::UnicodeString(const std::nullptr_t /*text*/, int32_t /*length*/) {
3922 fUnion.fStackFields.fLengthAndFlags=kShortString;
3923}
3924
3925inline UnicodeString::UnicodeString(std::nullptr_t /*buffer*/, int32_t /*buffLength*/, int32_t /*buffCapacity*/) {
3926 fUnion.fStackFields.fLengthAndFlags=kShortString;
3927}
3928
3929//========================================
3930// Read-only implementation methods
3931//========================================
3932inline UBool
3933UnicodeString::hasShortLength() const {
3934 return fUnion.fFields.fLengthAndFlags>=0;
3935}
3936
3937inline int32_t
3938UnicodeString::getShortLength() const {
3939 // fLengthAndFlags must be non-negative -> short length >= 0
3940 // and arithmetic or logical shift does not matter.
3941 return fUnion.fFields.fLengthAndFlags>>kLengthShift;
3942}
3943
3944inline int32_t
3945UnicodeString::length() const {
3946 return hasShortLength() ? getShortLength() : fUnion.fFields.fLength;
3947}
3948
3949inline int32_t
3950UnicodeString::getCapacity() const {
3951 return (fUnion.fFields.fLengthAndFlags&kUsingStackBuffer) ?
3952 US_STACKBUF_SIZE : fUnion.fFields.fCapacity;
3953}
3954
3955inline int32_t
3956UnicodeString::hashCode() const
3957{ return doHashCode(); }
3958
3959inline UBool
3960UnicodeString::isBogus() const
3961{ return (UBool)(fUnion.fFields.fLengthAndFlags & kIsBogus); }
3962
3963inline UBool
3964UnicodeString::isWritable() const
3965{ return (UBool)!(fUnion.fFields.fLengthAndFlags&(kOpenGetBuffer|kIsBogus)); }
3966
3967inline UBool
3968UnicodeString::isBufferWritable() const
3969{
3970 return (UBool)(
3971 !(fUnion.fFields.fLengthAndFlags&(kOpenGetBuffer|kIsBogus|kBufferIsReadonly)) &&
3972 (!(fUnion.fFields.fLengthAndFlags&kRefCounted) || refCount()==1));
3973}
3974
3975inline const char16_t *
3976UnicodeString::getBuffer() const {
3977 if(fUnion.fFields.fLengthAndFlags&(kIsBogus|kOpenGetBuffer)) {
3978 return nullptr;
3979 } else if(fUnion.fFields.fLengthAndFlags&kUsingStackBuffer) {
3980 return fUnion.fStackFields.fBuffer;
3981 } else {
3982 return fUnion.fFields.fArray;
3983 }
3984}
3985
3986//========================================
3987// Read-only alias methods
3988//========================================
3989inline int8_t
3990UnicodeString::doCompare(int32_t start,
3991 int32_t thisLength,
3992 const UnicodeString& srcText,
3993 int32_t srcStart,
3994 int32_t srcLength) const
3995{
3996 if(srcText.isBogus()) {
3997 return (int8_t)!isBogus(); // 0 if both are bogus, 1 otherwise
3998 } else {
3999 srcText.pinIndices(srcStart, srcLength);
4000 return doCompare(start, thisLength, srcText.getArrayStart(), srcStart, srcLength);
4001 }
4002}
4003
4004inline UBool
4005UnicodeString::operator== (const UnicodeString& text) const
4006{
4007 if(isBogus()) {
4008 return text.isBogus();
4009 } else {
4010 int32_t len = length(), textLength = text.length();
4011 return !text.isBogus() && len == textLength && doEquals(text, len);
4012 }
4013}
4014
4015inline UBool
4016UnicodeString::operator!= (const UnicodeString& text) const
4017{ return (! operator==(text)); }
4018
4019inline UBool
4020UnicodeString::operator> (const UnicodeString& text) const
4021{ return doCompare(0, length(), text, 0, text.length()) == 1; }
4022
4023inline UBool
4024UnicodeString::operator< (const UnicodeString& text) const
4025{ return doCompare(0, length(), text, 0, text.length()) == -1; }
4026
4027inline UBool
4028UnicodeString::operator>= (const UnicodeString& text) const
4029{ return doCompare(0, length(), text, 0, text.length()) != -1; }
4030
4031inline UBool
4032UnicodeString::operator<= (const UnicodeString& text) const
4033{ return doCompare(0, length(), text, 0, text.length()) != 1; }
4034
4035inline int8_t
4036UnicodeString::compare(const UnicodeString& text) const
4037{ return doCompare(0, length(), text, 0, text.length()); }
4038
4039inline int8_t
4040UnicodeString::compare(int32_t start,
4041 int32_t _length,
4042 const UnicodeString& srcText) const
4043{ return doCompare(start, _length, srcText, 0, srcText.length()); }
4044
4045inline int8_t
4046UnicodeString::compare(ConstChar16Ptr srcChars,
4047 int32_t srcLength) const
4048{ return doCompare(0, length(), srcChars, 0, srcLength); }
4049
4050inline int8_t
4051UnicodeString::compare(int32_t start,
4052 int32_t _length,
4053 const UnicodeString& srcText,
4054 int32_t srcStart,
4055 int32_t srcLength) const
4056{ return doCompare(start, _length, srcText, srcStart, srcLength); }
4057
4058inline int8_t
4059UnicodeString::compare(int32_t start,
4060 int32_t _length,
4061 const char16_t *srcChars) const
4062{ return doCompare(start, _length, srcChars, 0, _length); }
4063
4064inline int8_t
4065UnicodeString::compare(int32_t start,
4066 int32_t _length,
4067 const char16_t *srcChars,
4068 int32_t srcStart,
4069 int32_t srcLength) const
4070{ return doCompare(start, _length, srcChars, srcStart, srcLength); }
4071
4072inline int8_t
4073UnicodeString::compareBetween(int32_t start,
4074 int32_t limit,
4075 const UnicodeString& srcText,
4076 int32_t srcStart,
4077 int32_t srcLimit) const
4078{ return doCompare(start, limit - start,
4079 srcText, srcStart, srcLimit - srcStart); }
4080
4081inline int8_t
4082UnicodeString::doCompareCodePointOrder(int32_t start,
4083 int32_t thisLength,
4084 const UnicodeString& srcText,
4085 int32_t srcStart,
4086 int32_t srcLength) const
4087{
4088 if(srcText.isBogus()) {
4089 return (int8_t)!isBogus(); // 0 if both are bogus, 1 otherwise
4090 } else {
4091 srcText.pinIndices(srcStart, srcLength);
4092 return doCompareCodePointOrder(start, thisLength, srcText.getArrayStart(), srcStart, srcLength);
4093 }
4094}
4095
4096inline int8_t
4097UnicodeString::compareCodePointOrder(const UnicodeString& text) const
4098{ return doCompareCodePointOrder(0, length(), text, 0, text.length()); }
4099
4100inline int8_t
4101UnicodeString::compareCodePointOrder(int32_t start,
4102 int32_t _length,
4103 const UnicodeString& srcText) const
4104{ return doCompareCodePointOrder(start, _length, srcText, 0, srcText.length()); }
4105
4106inline int8_t
4107UnicodeString::compareCodePointOrder(ConstChar16Ptr srcChars,
4108 int32_t srcLength) const
4109{ return doCompareCodePointOrder(0, length(), srcChars, 0, srcLength); }
4110
4111inline int8_t
4112UnicodeString::compareCodePointOrder(int32_t start,
4113 int32_t _length,
4114 const UnicodeString& srcText,
4115 int32_t srcStart,
4116 int32_t srcLength) const
4117{ return doCompareCodePointOrder(start, _length, srcText, srcStart, srcLength); }
4118
4119inline int8_t
4120UnicodeString::compareCodePointOrder(int32_t start,
4121 int32_t _length,
4122 const char16_t *srcChars) const
4123{ return doCompareCodePointOrder(start, _length, srcChars, 0, _length); }
4124
4125inline int8_t
4126UnicodeString::compareCodePointOrder(int32_t start,
4127 int32_t _length,
4128 const char16_t *srcChars,
4129 int32_t srcStart,
4130 int32_t srcLength) const
4131{ return doCompareCodePointOrder(start, _length, srcChars, srcStart, srcLength); }
4132
4133inline int8_t
4134UnicodeString::compareCodePointOrderBetween(int32_t start,
4135 int32_t limit,
4136 const UnicodeString& srcText,
4137 int32_t srcStart,
4138 int32_t srcLimit) const
4139{ return doCompareCodePointOrder(start, limit - start,
4140 srcText, srcStart, srcLimit - srcStart); }
4141
4142inline int8_t
4143UnicodeString::doCaseCompare(int32_t start,
4144 int32_t thisLength,
4145 const UnicodeString &srcText,
4146 int32_t srcStart,
4147 int32_t srcLength,
4148 uint32_t options) const
4149{
4150 if(srcText.isBogus()) {
4151 return (int8_t)!isBogus(); // 0 if both are bogus, 1 otherwise
4152 } else {
4153 srcText.pinIndices(srcStart, srcLength);
4154 return doCaseCompare(start, thisLength, srcText.getArrayStart(), srcStart, srcLength, options);
4155 }
4156}
4157
4158inline int8_t
4159UnicodeString::caseCompare(const UnicodeString &text, uint32_t options) const {
4160 return doCaseCompare(0, length(), text, 0, text.length(), options);
4161}
4162
4163inline int8_t
4164UnicodeString::caseCompare(int32_t start,
4165 int32_t _length,
4166 const UnicodeString &srcText,
4167 uint32_t options) const {
4168 return doCaseCompare(start, _length, srcText, 0, srcText.length(), options);
4169}
4170
4171inline int8_t
4172UnicodeString::caseCompare(ConstChar16Ptr srcChars,
4173 int32_t srcLength,
4174 uint32_t options) const {
4175 return doCaseCompare(0, length(), srcChars, 0, srcLength, options);
4176}
4177
4178inline int8_t
4179UnicodeString::caseCompare(int32_t start,
4180 int32_t _length,
4181 const UnicodeString &srcText,
4182 int32_t srcStart,
4183 int32_t srcLength,
4184 uint32_t options) const {
4185 return doCaseCompare(start, _length, srcText, srcStart, srcLength, options);
4186}
4187
4188inline int8_t
4189UnicodeString::caseCompare(int32_t start,
4190 int32_t _length,
4191 const char16_t *srcChars,
4192 uint32_t options) const {
4193 return doCaseCompare(start, _length, srcChars, 0, _length, options);
4194}
4195
4196inline int8_t
4197UnicodeString::caseCompare(int32_t start,
4198 int32_t _length,
4199 const char16_t *srcChars,
4200 int32_t srcStart,
4201 int32_t srcLength,
4202 uint32_t options) const {
4203 return doCaseCompare(start, _length, srcChars, srcStart, srcLength, options);
4204}
4205
4206inline int8_t
4207UnicodeString::caseCompareBetween(int32_t start,
4208 int32_t limit,
4209 const UnicodeString &srcText,
4210 int32_t srcStart,
4211 int32_t srcLimit,
4212 uint32_t options) const {
4213 return doCaseCompare(start, limit - start, srcText, srcStart, srcLimit - srcStart, options);
4214}
4215
4216inline int32_t
4217UnicodeString::indexOf(const UnicodeString& srcText,
4218 int32_t srcStart,
4219 int32_t srcLength,
4220 int32_t start,
4221 int32_t _length) const
4222{
4223 if(!srcText.isBogus()) {
4224 srcText.pinIndices(srcStart, srcLength);
4225 if(srcLength > 0) {
4226 return indexOf(srcText.getArrayStart(), srcStart, srcLength, start, _length);
4227 }
4228 }
4229 return -1;
4230}
4231
4232inline int32_t
4233UnicodeString::indexOf(const UnicodeString& text) const
4234{ return indexOf(text, 0, text.length(), 0, length()); }
4235
4236inline int32_t
4237UnicodeString::indexOf(const UnicodeString& text,
4238 int32_t start) const {
4239 pinIndex(start);
4240 return indexOf(text, 0, text.length(), start, length() - start);
4241}
4242
4243inline int32_t
4244UnicodeString::indexOf(const UnicodeString& text,
4245 int32_t start,
4246 int32_t _length) const
4247{ return indexOf(text, 0, text.length(), start, _length); }
4248
4249inline int32_t
4250UnicodeString::indexOf(const char16_t *srcChars,
4251 int32_t srcLength,
4252 int32_t start) const {
4253 pinIndex(start);
4254 return indexOf(srcChars, 0, srcLength, start, length() - start);
4255}
4256
4257inline int32_t
4258UnicodeString::indexOf(ConstChar16Ptr srcChars,
4259 int32_t srcLength,
4260 int32_t start,
4261 int32_t _length) const
4262{ return indexOf(srcChars, 0, srcLength, start, _length); }
4263
4264inline int32_t
4265UnicodeString::indexOf(char16_t c,
4266 int32_t start,
4267 int32_t _length) const
4268{ return doIndexOf(c, start, _length); }
4269
4270inline int32_t
4271UnicodeString::indexOf(UChar32 c,
4272 int32_t start,
4273 int32_t _length) const
4274{ return doIndexOf(c, start, _length); }
4275
4276inline int32_t
4277UnicodeString::indexOf(char16_t c) const
4278{ return doIndexOf(c, 0, length()); }
4279
4280inline int32_t
4281UnicodeString::indexOf(UChar32 c) const
4282{ return indexOf(c, 0, length()); }
4283
4284inline int32_t
4285UnicodeString::indexOf(char16_t c,
4286 int32_t start) const {
4287 pinIndex(start);
4288 return doIndexOf(c, start, length() - start);
4289}
4290
4291inline int32_t
4292UnicodeString::indexOf(UChar32 c,
4293 int32_t start) const {
4294 pinIndex(start);
4295 return indexOf(c, start, length() - start);
4296}
4297
4298inline int32_t
4299UnicodeString::lastIndexOf(ConstChar16Ptr srcChars,
4300 int32_t srcLength,
4301 int32_t start,
4302 int32_t _length) const
4303{ return lastIndexOf(srcChars, 0, srcLength, start, _length); }
4304
4305inline int32_t
4306UnicodeString::lastIndexOf(const char16_t *srcChars,
4307 int32_t srcLength,
4308 int32_t start) const {
4309 pinIndex(start);
4310 return lastIndexOf(srcChars, 0, srcLength, start, length() - start);
4311}
4312
4313inline int32_t
4314UnicodeString::lastIndexOf(const UnicodeString& srcText,
4315 int32_t srcStart,
4316 int32_t srcLength,
4317 int32_t start,
4318 int32_t _length) const
4319{
4320 if(!srcText.isBogus()) {
4321 srcText.pinIndices(srcStart, srcLength);
4322 if(srcLength > 0) {
4323 return lastIndexOf(srcText.getArrayStart(), srcStart, srcLength, start, _length);
4324 }
4325 }
4326 return -1;
4327}
4328
4329inline int32_t
4330UnicodeString::lastIndexOf(const UnicodeString& text,
4331 int32_t start,
4332 int32_t _length) const
4333{ return lastIndexOf(text, 0, text.length(), start, _length); }
4334
4335inline int32_t
4336UnicodeString::lastIndexOf(const UnicodeString& text,
4337 int32_t start) const {
4338 pinIndex(start);
4339 return lastIndexOf(text, 0, text.length(), start, length() - start);
4340}
4341
4342inline int32_t
4343UnicodeString::lastIndexOf(const UnicodeString& text) const
4344{ return lastIndexOf(text, 0, text.length(), 0, length()); }
4345
4346inline int32_t
4347UnicodeString::lastIndexOf(char16_t c,
4348 int32_t start,
4349 int32_t _length) const
4350{ return doLastIndexOf(c, start, _length); }
4351
4352inline int32_t
4353UnicodeString::lastIndexOf(UChar32 c,
4354 int32_t start,
4355 int32_t _length) const {
4356 return doLastIndexOf(c, start, _length);
4357}
4358
4359inline int32_t
4360UnicodeString::lastIndexOf(char16_t c) const
4361{ return doLastIndexOf(c, 0, length()); }
4362
4363inline int32_t
4364UnicodeString::lastIndexOf(UChar32 c) const {
4365 return lastIndexOf(c, 0, length());
4366}
4367
4368inline int32_t
4369UnicodeString::lastIndexOf(char16_t c,
4370 int32_t start) const {
4371 pinIndex(start);
4372 return doLastIndexOf(c, start, length() - start);
4373}
4374
4375inline int32_t
4376UnicodeString::lastIndexOf(UChar32 c,
4377 int32_t start) const {
4378 pinIndex(start);
4379 return lastIndexOf(c, start, length() - start);
4380}
4381
4382inline UBool
4383UnicodeString::startsWith(const UnicodeString& text) const
4384{ return compare(0, text.length(), text, 0, text.length()) == 0; }
4385
4386inline UBool
4387UnicodeString::startsWith(const UnicodeString& srcText,
4388 int32_t srcStart,
4389 int32_t srcLength) const
4390{ return doCompare(0, srcLength, srcText, srcStart, srcLength) == 0; }
4391
4392inline UBool
4393UnicodeString::startsWith(ConstChar16Ptr srcChars, int32_t srcLength) const {
4394 if(srcLength < 0) {
4395 srcLength = u_strlen(toUCharPtr(srcChars));
4396 }
4397 return doCompare(0, srcLength, srcChars, 0, srcLength) == 0;
4398}
4399
4400inline UBool
4401UnicodeString::startsWith(const char16_t *srcChars, int32_t srcStart, int32_t srcLength) const {
4402 if(srcLength < 0) {
4403 srcLength = u_strlen(toUCharPtr(srcChars));
4404 }
4405 return doCompare(0, srcLength, srcChars, srcStart, srcLength) == 0;
4406}
4407
4408inline UBool
4409UnicodeString::endsWith(const UnicodeString& text) const
4410{ return doCompare(length() - text.length(), text.length(),
4411 text, 0, text.length()) == 0; }
4412
4413inline UBool
4414UnicodeString::endsWith(const UnicodeString& srcText,
4415 int32_t srcStart,
4416 int32_t srcLength) const {
4417 srcText.pinIndices(srcStart, srcLength);
4418 return doCompare(length() - srcLength, srcLength,
4419 srcText, srcStart, srcLength) == 0;
4420}
4421
4422inline UBool
4423UnicodeString::endsWith(ConstChar16Ptr srcChars,
4424 int32_t srcLength) const {
4425 if(srcLength < 0) {
4426 srcLength = u_strlen(toUCharPtr(srcChars));
4427 }
4428 return doCompare(length() - srcLength, srcLength,
4429 srcChars, 0, srcLength) == 0;
4430}
4431
4432inline UBool
4433UnicodeString::endsWith(const char16_t *srcChars,
4434 int32_t srcStart,
4435 int32_t srcLength) const {
4436 if(srcLength < 0) {
4437 srcLength = u_strlen(toUCharPtr(srcChars + srcStart));
4438 }
4439 return doCompare(length() - srcLength, srcLength,
4440 srcChars, srcStart, srcLength) == 0;
4441}
4442
4443//========================================
4444// replace
4445//========================================
4446inline UnicodeString&
4447UnicodeString::replace(int32_t start,
4448 int32_t _length,
4449 const UnicodeString& srcText)
4450{ return doReplace(start, _length, srcText, 0, srcText.length()); }
4451
4452inline UnicodeString&
4453UnicodeString::replace(int32_t start,
4454 int32_t _length,
4455 const UnicodeString& srcText,
4456 int32_t srcStart,
4457 int32_t srcLength)
4458{ return doReplace(start, _length, srcText, srcStart, srcLength); }
4459
4460inline UnicodeString&
4461UnicodeString::replace(int32_t start,
4462 int32_t _length,
4463 ConstChar16Ptr srcChars,
4464 int32_t srcLength)
4465{ return doReplace(start, _length, srcChars, 0, srcLength); }
4466
4467inline UnicodeString&
4468UnicodeString::replace(int32_t start,
4469 int32_t _length,
4470 const char16_t *srcChars,
4471 int32_t srcStart,
4472 int32_t srcLength)
4473{ return doReplace(start, _length, srcChars, srcStart, srcLength); }
4474
4475inline UnicodeString&
4476UnicodeString::replace(int32_t start,
4477 int32_t _length,
4478 char16_t srcChar)
4479{ return doReplace(start, _length, &srcChar, 0, 1); }
4480
4481inline UnicodeString&
4482UnicodeString::replaceBetween(int32_t start,
4483 int32_t limit,
4484 const UnicodeString& srcText)
4485{ return doReplace(start, limit - start, srcText, 0, srcText.length()); }
4486
4487inline UnicodeString&
4488UnicodeString::replaceBetween(int32_t start,
4489 int32_t limit,
4490 const UnicodeString& srcText,
4491 int32_t srcStart,
4492 int32_t srcLimit)
4493{ return doReplace(start, limit - start, srcText, srcStart, srcLimit - srcStart); }
4494
4495inline UnicodeString&
4496UnicodeString::findAndReplace(const UnicodeString& oldText,
4497 const UnicodeString& newText)
4498{ return findAndReplace(0, length(), oldText, 0, oldText.length(),
4499 newText, 0, newText.length()); }
4500
4501inline UnicodeString&
4502UnicodeString::findAndReplace(int32_t start,
4503 int32_t _length,
4504 const UnicodeString& oldText,
4505 const UnicodeString& newText)
4506{ return findAndReplace(start, _length, oldText, 0, oldText.length(),
4507 newText, 0, newText.length()); }
4508
4509// ============================
4510// extract
4511// ============================
4512inline void
4513UnicodeString::doExtract(int32_t start,
4514 int32_t _length,
4515 UnicodeString& target) const
4516{ target.replace(0, target.length(), *this, start, _length); }
4517
4518inline void
4519UnicodeString::extract(int32_t start,
4520 int32_t _length,
4521 Char16Ptr target,
4522 int32_t targetStart) const
4523{ doExtract(start, _length, target, targetStart); }
4524
4525inline void
4526UnicodeString::extract(int32_t start,
4527 int32_t _length,
4528 UnicodeString& target) const
4529{ doExtract(start, _length, target); }
4530
4531#if !UCONFIG_NO_CONVERSION
4532
4533inline int32_t
4534UnicodeString::extract(int32_t start,
4535 int32_t _length,
4536 char *dst,
4537 const char *codepage) const
4538
4539{
4540 // This dstSize value will be checked explicitly
4541 return extract(start, _length, dst, dst!=0 ? 0xffffffff : 0, codepage);
4542}
4543
4544#endif
4545
4546inline void
4547UnicodeString::extractBetween(int32_t start,
4548 int32_t limit,
4549 char16_t *dst,
4550 int32_t dstStart) const {
4551 pinIndex(start);
4552 pinIndex(limit);
4553 doExtract(start, limit - start, dst, dstStart);
4554}
4555
4556inline UnicodeString
4557UnicodeString::tempSubStringBetween(int32_t start, int32_t limit) const {
4558 return tempSubString(start, limit - start);
4559}
4560
4561inline char16_t
4562UnicodeString::doCharAt(int32_t offset) const
4563{
4564 if((uint32_t)offset < (uint32_t)length()) {
4565 return getArrayStart()[offset];
4566 } else {
4567 return kInvalidUChar;
4568 }
4569}
4570
4571inline char16_t
4572UnicodeString::charAt(int32_t offset) const
4573{ return doCharAt(offset); }
4574
4575inline char16_t
4576UnicodeString::operator[] (int32_t offset) const
4577{ return doCharAt(offset); }
4578
4579inline UBool
4580UnicodeString::isEmpty() const {
4581 // Arithmetic or logical right shift does not matter: only testing for 0.
4582 return (fUnion.fFields.fLengthAndFlags>>kLengthShift) == 0;
4583}
4584
4585//========================================
4586// Write implementation methods
4587//========================================
4588inline void
4589UnicodeString::setZeroLength() {
4590 fUnion.fFields.fLengthAndFlags &= kAllStorageFlags;
4591}
4592
4593inline void
4594UnicodeString::setShortLength(int32_t len) {
4595 // requires 0 <= len <= kMaxShortLength
4596 fUnion.fFields.fLengthAndFlags =
4597 (int16_t)((fUnion.fFields.fLengthAndFlags & kAllStorageFlags) | (len << kLengthShift));
4598}
4599
4600inline void
4601UnicodeString::setLength(int32_t len) {
4602 if(len <= kMaxShortLength) {
4603 setShortLength(len);
4604 } else {
4605 fUnion.fFields.fLengthAndFlags |= kLengthIsLarge;
4606 fUnion.fFields.fLength = len;
4607 }
4608}
4609
4610inline void
4611UnicodeString::setToEmpty() {
4612 fUnion.fFields.fLengthAndFlags = kShortString;
4613}
4614
4615inline void
4616UnicodeString::setArray(char16_t *array, int32_t len, int32_t capacity) {
4617 setLength(len);
4618 fUnion.fFields.fArray = array;
4619 fUnion.fFields.fCapacity = capacity;
4620}
4621
4622inline UnicodeString&
4623UnicodeString::operator= (char16_t ch)
4624{ return doReplace(0, length(), &ch, 0, 1); }
4625
4626inline UnicodeString&
4627UnicodeString::operator= (UChar32 ch)
4628{ return replace(0, length(), ch); }
4629
4630inline UnicodeString&
4631UnicodeString::setTo(const UnicodeString& srcText,
4632 int32_t srcStart,
4633 int32_t srcLength)
4634{
4635 unBogus();
4636 return doReplace(0, length(), srcText, srcStart, srcLength);
4637}
4638
4639inline UnicodeString&
4640UnicodeString::setTo(const UnicodeString& srcText,
4641 int32_t srcStart)
4642{
4643 unBogus();
4644 srcText.pinIndex(srcStart);
4645 return doReplace(0, length(), srcText, srcStart, srcText.length() - srcStart);
4646}
4647
4648inline UnicodeString&
4649UnicodeString::setTo(const UnicodeString& srcText)
4650{
4651 return copyFrom(srcText);
4652}
4653
4654inline UnicodeString&
4655UnicodeString::setTo(const char16_t *srcChars,
4656 int32_t srcLength)
4657{
4658 unBogus();
4659 return doReplace(0, length(), srcChars, 0, srcLength);
4660}
4661
4662inline UnicodeString&
4663UnicodeString::setTo(char16_t srcChar)
4664{
4665 unBogus();
4666 return doReplace(0, length(), &srcChar, 0, 1);
4667}
4668
4669inline UnicodeString&
4670UnicodeString::setTo(UChar32 srcChar)
4671{
4672 unBogus();
4673 return replace(0, length(), srcChar);
4674}
4675
4676inline UnicodeString&
4677UnicodeString::append(const UnicodeString& srcText,
4678 int32_t srcStart,
4679 int32_t srcLength)
4680{ return doAppend(srcText, srcStart, srcLength); }
4681
4682inline UnicodeString&
4683UnicodeString::append(const UnicodeString& srcText)
4684{ return doAppend(srcText, 0, srcText.length()); }
4685
4686inline UnicodeString&
4687UnicodeString::append(const char16_t *srcChars,
4688 int32_t srcStart,
4689 int32_t srcLength)
4690{ return doAppend(srcChars, srcStart, srcLength); }
4691
4692inline UnicodeString&
4693UnicodeString::append(ConstChar16Ptr srcChars,
4694 int32_t srcLength)
4695{ return doAppend(srcChars, 0, srcLength); }
4696
4697inline UnicodeString&
4698UnicodeString::append(char16_t srcChar)
4699{ return doAppend(&srcChar, 0, 1); }
4700
4701inline UnicodeString&
4702UnicodeString::operator+= (char16_t ch)
4703{ return doAppend(&ch, 0, 1); }
4704
4705inline UnicodeString&
4706UnicodeString::operator+= (UChar32 ch) {
4707 return append(ch);
4708}
4709
4710inline UnicodeString&
4711UnicodeString::operator+= (const UnicodeString& srcText)
4712{ return doAppend(srcText, 0, srcText.length()); }
4713
4714inline UnicodeString&
4715UnicodeString::insert(int32_t start,
4716 const UnicodeString& srcText,
4717 int32_t srcStart,
4718 int32_t srcLength)
4719{ return doReplace(start, 0, srcText, srcStart, srcLength); }
4720
4721inline UnicodeString&
4722UnicodeString::insert(int32_t start,
4723 const UnicodeString& srcText)
4724{ return doReplace(start, 0, srcText, 0, srcText.length()); }
4725
4726inline UnicodeString&
4727UnicodeString::insert(int32_t start,
4728 const char16_t *srcChars,
4729 int32_t srcStart,
4730 int32_t srcLength)
4731{ return doReplace(start, 0, srcChars, srcStart, srcLength); }
4732
4733inline UnicodeString&
4734UnicodeString::insert(int32_t start,
4735 ConstChar16Ptr srcChars,
4736 int32_t srcLength)
4737{ return doReplace(start, 0, srcChars, 0, srcLength); }
4738
4739inline UnicodeString&
4740UnicodeString::insert(int32_t start,
4741 char16_t srcChar)
4742{ return doReplace(start, 0, &srcChar, 0, 1); }
4743
4744inline UnicodeString&
4745UnicodeString::insert(int32_t start,
4746 UChar32 srcChar)
4747{ return replace(start, 0, srcChar); }
4748
4749
4750inline UnicodeString&
4751UnicodeString::remove()
4752{
4753 // remove() of a bogus string makes the string empty and non-bogus
4754 if(isBogus()) {
4755 setToEmpty();
4756 } else {
4757 setZeroLength();
4758 }
4759 return *this;
4760}
4761
4762inline UnicodeString&
4763UnicodeString::remove(int32_t start,
4764 int32_t _length)
4765{
4766 if(start <= 0 && _length == INT32_MAX) {
4767 // remove(guaranteed everything) of a bogus string makes the string empty and non-bogus
4768 return remove();
4769 }
4770 return doReplace(start, _length, NULL, 0, 0);
4771}
4772
4773inline UnicodeString&
4774UnicodeString::removeBetween(int32_t start,
4775 int32_t limit)
4776{ return doReplace(start, limit - start, NULL, 0, 0); }
4777
4778inline UnicodeString &
4779UnicodeString::retainBetween(int32_t start, int32_t limit) {
4780 truncate(limit);
4781 return doReplace(0, start, NULL, 0, 0);
4782}
4783
4784inline UBool
4785UnicodeString::truncate(int32_t targetLength)
4786{
4787 if(isBogus() && targetLength == 0) {
4788 // truncate(0) of a bogus string makes the string empty and non-bogus
4789 unBogus();
4790 return FALSE;
4791 } else if((uint32_t)targetLength < (uint32_t)length()) {
4792 setLength(targetLength);
4793 return TRUE;
4794 } else {
4795 return FALSE;
4796 }
4797}
4798
4799inline UnicodeString&
4800UnicodeString::reverse()
4801{ return doReverse(0, length()); }
4802
4803inline UnicodeString&
4804UnicodeString::reverse(int32_t start,
4805 int32_t _length)
4806{ return doReverse(start, _length); }
4807
4808U_NAMESPACE_END
4809
4810#endif
4811