ICU 73.1 73.1
rbbi.h
Go to the documentation of this file.
1// © 2016 and later: Unicode, Inc. and others.
2// License & terms of use: http://www.unicode.org/copyright.html
3/*
4***************************************************************************
5* Copyright (C) 1999-2016 International Business Machines Corporation *
6* and others. All rights reserved. *
7***************************************************************************
8
9**********************************************************************
10* Date Name Description
11* 10/22/99 alan Creation.
12* 11/11/99 rgillam Complete port from Java.
13**********************************************************************
14*/
15
16#ifndef RBBI_H
17#define RBBI_H
18
19#include "unicode/utypes.h"
20
21#if U_SHOW_CPLUSPLUS_API
22
28#if !UCONFIG_NO_BREAK_ITERATION
29
30#include "unicode/brkiter.h"
31#include "unicode/udata.h"
32#include "unicode/parseerr.h"
33#include "unicode/schriter.h"
34
35struct UCPTrie;
36
37U_NAMESPACE_BEGIN
38
40class LanguageBreakEngine;
41struct RBBIDataHeader;
42class RBBIDataWrapper;
43class UnhandledEngine;
44class UStack;
45
58
59private:
65
66#ifndef U_HIDE_INTERNAL_API
67public:
68#endif /* U_HIDE_INTERNAL_API */
74 RBBIDataWrapper *fData = nullptr;
75
76private:
81 UErrorCode fErrorCode = U_ZERO_ERROR;
82
87 int32_t fPosition = 0;
88
92 int32_t fRuleStatusIndex = 0;
93
97 class BreakCache;
98 BreakCache *fBreakCache = nullptr;
99
104 class DictionaryCache;
105 DictionaryCache *fDictionaryCache = nullptr;
106
114 UStack *fLanguageBreakEngines = nullptr;
115
123 UnhandledEngine *fUnhandledBreakEngine = nullptr;
124
130 uint32_t fDictionaryCharCount = 0;
131
137 CharacterIterator *fCharIter = &fSCharIter;
138
144 UCharCharacterIterator fSCharIter {u"", 0};
145
149 bool fDone = false;
150
154 int32_t *fLookAheadMatches = nullptr;
155
159 UBool fIsPhraseBreaking = false;
160
161 //=======================================================================
162 // constructors
163 //=======================================================================
164
175 RuleBasedBreakIterator(RBBIDataHeader* data, UErrorCode &status);
176
190 RuleBasedBreakIterator(UDataMemory* image, UBool isPhraseBreaking, UErrorCode &status);
191
193 friend class RBBIRuleBuilder;
195 friend class BreakIterator;
196
204
205public:
206
213
221
231 UParseError &parseError,
232 UErrorCode &status);
233
257 RuleBasedBreakIterator(const uint8_t *compiledRules,
258 uint32_t ruleLength,
259 UErrorCode &status);
260
274
280
289
298 virtual bool operator==(const BreakIterator& that) const override;
299
307 inline bool operator!=(const BreakIterator& that) const {
308 return !operator==(that);
309 }
310
321 virtual RuleBasedBreakIterator* clone() const override;
322
328 virtual int32_t hashCode(void) const;
329
335 virtual const UnicodeString& getRules(void) const;
336
337 //=======================================================================
338 // BreakIterator overrides
339 //=======================================================================
340
365 virtual CharacterIterator& getText(void) const override;
366
367
382 virtual UText *getUText(UText *fillIn, UErrorCode &status) const override;
383
391 virtual void adoptText(CharacterIterator* newText) override;
392
404 virtual void setText(const UnicodeString& newText) override;
405
419 virtual void setText(UText *text, UErrorCode &status) override;
420
426 virtual int32_t first(void) override;
427
433 virtual int32_t last(void) override;
434
445 virtual int32_t next(int32_t n) override;
446
452 virtual int32_t next(void) override;
453
459 virtual int32_t previous(void) override;
460
468 virtual int32_t following(int32_t offset) override;
469
477 virtual int32_t preceding(int32_t offset) override;
478
487 virtual UBool isBoundary(int32_t offset) override;
488
497 virtual int32_t current(void) const override;
498
499
531 virtual int32_t getRuleStatus() const override;
532
556 virtual int32_t getRuleStatusVec(int32_t *fillInVec, int32_t capacity, UErrorCode &status) override;
557
569 virtual UClassID getDynamicClassID(void) const override;
570
582 static UClassID U_EXPORT2 getStaticClassID(void);
583
584#ifndef U_FORCE_HIDE_DEPRECATED_API
611 virtual RuleBasedBreakIterator *createBufferClone(void *stackBuffer,
612 int32_t &BufferSize,
613 UErrorCode &status) override;
614#endif // U_FORCE_HIDE_DEPRECATED_API
615
633 virtual const uint8_t *getBinaryRules(uint32_t &length);
634
660 virtual RuleBasedBreakIterator &refreshInputText(UText *input, UErrorCode &status) override;
661
662
663private:
664 //=======================================================================
665 // implementation
666 //=======================================================================
676 int32_t handleSafePrevious(int32_t fromPosition);
677
690 int32_t handleNext();
691
692 /*
693 * Templatized version of handleNext() and handleSafePrevious().
694 *
695 * There will be exactly four instantiations, two each for 8 and 16 bit tables,
696 * two each for 8 and 16 bit trie.
697 * Having separate instantiations for the table types keeps conditional tests of
698 * the table type out of the inner loops, at the expense of replicated code.
699 *
700 * The template parameter for the Trie access function is a value, not a type.
701 * Doing it this way, the compiler will inline the Trie function in the
702 * expanded functions. (Both the 8 and 16 bit access functions have the same type
703 * signature)
704 */
705
706 typedef uint16_t (*PTrieFunc)(const UCPTrie *, UChar32);
707
708 template<typename RowType, PTrieFunc trieFunc>
709 int32_t handleSafePrevious(int32_t fromPosition);
710
711 template<typename RowType, PTrieFunc trieFunc>
712 int32_t handleNext();
713
714
721 const LanguageBreakEngine *getLanguageBreakEngine(UChar32 c);
722
723 public:
724#ifndef U_HIDE_INTERNAL_API
729 void dumpCache();
730
736#endif /* U_HIDE_INTERNAL_API */
737};
738
739U_NAMESPACE_END
740
741#endif /* #if !UCONFIG_NO_BREAK_ITERATION */
742
743#endif /* U_SHOW_CPLUSPLUS_API */
744
745#endif
C++ API: Break Iterator.
The BreakIterator class implements methods for finding the location of boundaries in text.
Definition: brkiter.h:106
virtual bool operator==(const BreakIterator &) const =0
Return true if another object is semantically equal to this one.
Abstract class that defines an API for iteration on text objects.
Definition: chariter.h:361
A subclass of BreakIterator whose behavior is specified using a list of rules.
Definition: rbbi.h:57
virtual int32_t current(void) const override
Returns the current iteration position.
virtual int32_t following(int32_t offset) override
Sets the iterator to refer to the first boundary position following the specified position.
virtual bool operator==(const BreakIterator &that) const override
Equality operator.
virtual RuleBasedBreakIterator * clone() const override
Returns a newly-constructed RuleBasedBreakIterator with the same behavior, and iterating over the sam...
virtual RuleBasedBreakIterator * createBufferClone(void *stackBuffer, int32_t &BufferSize, UErrorCode &status) override
Deprecated functionality.
void dumpTables()
Debugging function only.
virtual UClassID getDynamicClassID(void) const override
Returns a unique class ID POLYMORPHICALLY.
virtual int32_t last(void) override
Sets the current iteration position to the end of the text.
virtual int32_t getRuleStatus() const override
Return the status tag from the break rule that determined the boundary at the current iteration posit...
virtual UText * getUText(UText *fillIn, UErrorCode &status) const override
Get a UText for the text being analyzed.
virtual UBool isBoundary(int32_t offset) override
Returns true if the specified position is a boundary position.
bool operator!=(const BreakIterator &that) const
Not-equal operator.
Definition: rbbi.h:307
void dumpCache()
Debugging function only.
virtual RuleBasedBreakIterator & refreshInputText(UText *input, UErrorCode &status) override
Set the subject text string upon which the break iterator is operating without changing any other asp...
RuleBasedBreakIterator()
Default constructor.
virtual int32_t getRuleStatusVec(int32_t *fillInVec, int32_t capacity, UErrorCode &status) override
Get the status (tag) values from the break rule(s) that determined the boundary at the current iterat...
virtual int32_t first(void) override
Sets the current iteration position to the beginning of the text, position zero.
RuleBasedBreakIterator & operator=(const RuleBasedBreakIterator &that)
Assignment operator.
static UClassID getStaticClassID(void)
Returns the class ID for this class.
virtual int32_t preceding(int32_t offset) override
Sets the iterator to refer to the last boundary position before the specified position.
RuleBasedBreakIterator(const RuleBasedBreakIterator &that)
Copy constructor.
virtual CharacterIterator & getText(void) const override
virtual int32_t previous(void) override
Moves the iterator backwards, to the last boundary preceding this one.
virtual void adoptText(CharacterIterator *newText) override
Set the iterator to analyze a new piece of text.
RuleBasedBreakIterator(UDataMemory *image, UErrorCode &status)
This constructor uses the udata interface to create a BreakIterator whose internal tables live in a m...
virtual const UnicodeString & getRules(void) const
Returns the description used to create this iterator.
virtual int32_t next(int32_t n) override
Advances the iterator either forward or backward the specified number of steps.
virtual void setText(UText *text, UErrorCode &status) override
Reset the break iterator to operate over the text represented by the UText.
virtual ~RuleBasedBreakIterator()
Destructor.
RuleBasedBreakIterator(const UnicodeString &rules, UParseError &parseError, UErrorCode &status)
Construct a RuleBasedBreakIterator from a set of rules supplied as a string.
virtual const uint8_t * getBinaryRules(uint32_t &length)
Return the binary form of compiled break rules, which can then be used to create a new break iterator...
virtual int32_t hashCode(void) const
Compute a hash code for this BreakIterator.
virtual int32_t next(void) override
Advances the iterator to the next boundary position.
RuleBasedBreakIterator(const uint8_t *compiledRules, uint32_t ruleLength, UErrorCode &status)
Construct a RuleBasedBreakIterator from a set of precompiled binary rules.
virtual void setText(const UnicodeString &newText) override
Set the iterator to analyze a new piece of text.
A concrete subclass of CharacterIterator that iterates over the characters (code units or code points...
Definition: uchriter.h:38
UnicodeString is a string class that stores Unicode characters directly and provides similar function...
Definition: unistr.h:296
C API: Parse Error Information.
C++ API: String Character Iterator.
Immutable Unicode code point trie structure.
Definition: ucptrie.h:59
A UParseError struct is used to returned detailed information about parsing errors.
Definition: parseerr.h:58
UText struct.
Definition: utext.h:1328
C API: Data loading interface.
struct UDataMemory UDataMemory
Forward declaration of the data memory type.
Definition: udata.h:161
int32_t UChar32
Define UChar32 as a type for single Unicode code points.
Definition: umachine.h:435
int8_t UBool
The ICU boolean type, a signed-byte integer.
Definition: umachine.h:247
void * UClassID
UClassID is used to identify classes without using the compiler's RTTI.
Definition: uobject.h:96
#define UTEXT_INITIALIZER
initializer to be used with local (stack) instances of a UText struct.
Definition: utext.h:1558
Basic definitions for ICU, for both C and C++ APIs.
UErrorCode
Standard ICU4C error code type, a substitute for exceptions.
Definition: utypes.h:415
@ U_ZERO_ERROR
No error, no warning.
Definition: utypes.h:449
#define U_COMMON_API
Set to export library symbols from inside the common library, and to import them from outside.
Definition: utypes.h:300