21 #if U_SHOW_CPLUSPLUS_API 28 #if !UCONFIG_NO_BREAK_ITERATION 40 class LanguageBreakEngine;
41 struct RBBIDataHeader;
42 class RBBIDataWrapper;
43 class UnhandledEngine;
66 #ifndef U_HIDE_INTERNAL_API 74 RBBIDataWrapper *fData =
nullptr;
87 int32_t fPosition = 0;
92 int32_t fRuleStatusIndex = 0;
98 BreakCache *fBreakCache =
nullptr;
104 class DictionaryCache;
105 DictionaryCache *fDictionaryCache =
nullptr;
114 UStack *fLanguageBreakEngines =
nullptr;
123 UnhandledEngine *fUnhandledBreakEngine =
nullptr;
130 uint32_t fDictionaryCharCount = 0;
154 int32_t *fLookAheadMatches =
nullptr;
159 UBool fIsPhraseBreaking =
false;
193 friend class RBBIRuleBuilder;
328 virtual int32_t hashCode(
void)
const;
426 virtual int32_t
first(
void)
override;
433 virtual int32_t
last(
void)
override;
445 virtual int32_t
next(int32_t n)
override;
452 virtual int32_t
next(
void)
override;
459 virtual int32_t
previous(
void)
override;
468 virtual int32_t
following(int32_t offset)
override;
477 virtual int32_t
preceding(int32_t offset)
override;
497 virtual int32_t
current(
void)
const override;
582 static UClassID U_EXPORT2 getStaticClassID(
void);
584 #ifndef U_FORCE_HIDE_DEPRECATED_API 614 #endif // U_FORCE_HIDE_DEPRECATED_API 633 virtual const uint8_t *getBinaryRules(uint32_t &length);
676 int32_t handleSafePrevious(int32_t fromPosition);
690 int32_t handleNext();
708 template<
typename RowType, PTrieFunc trieFunc>
709 int32_t handleSafePrevious(int32_t fromPosition);
711 template<
typename RowType, PTrieFunc trieFunc>
712 int32_t handleNext();
721 const LanguageBreakEngine *getLanguageBreakEngine(
UChar32 c);
724 #ifndef U_HIDE_INTERNAL_API
BreakIterator & operator=(const BreakIterator &other)
virtual int32_t next(void)=0
Advance the iterator to the boundary following the current boundary.
bool operator!=(const BreakIterator &that) const
Not-equal operator.
virtual UBool isBoundary(int32_t offset)=0
Return true if the specified position is a boundary position.
virtual void adoptText(CharacterIterator *it)=0
Change the text over which this operates.
virtual bool operator==(const BreakIterator &) const =0
Return true if another object is semantically equal to this one.
Immutable Unicode code point trie structure.
void * UClassID
UClassID is used to identify classes without using the compiler's RTTI.
virtual CharacterIterator & getText(void) const =0
Return a CharacterIterator over the text being analyzed.
virtual UText * getUText(UText *fillIn, UErrorCode &status) const =0
Get a UText for the text being analyzed.
virtual int32_t first(void)=0
Sets the current iteration position to the beginning of the text, position zero.
virtual int32_t following(int32_t offset)=0
Advance the iterator to the first boundary following the specified offset.
Abstract class that defines an API for iteration on text objects.
C++ API: String Character Iterator.
A concrete subclass of CharacterIterator that iterates over the characters (code units or code points...
The BreakIterator class implements methods for finding the location of boundaries in text...
virtual int32_t last(void)=0
Set the iterator position to the index immediately BEYOND the last character in the text being scanne...
virtual int32_t current(void) const =0
Return character index of the current iterator position within the text.
#define UTEXT_INITIALIZER
initializer to be used with local (stack) instances of a UText struct.
int32_t UChar32
Define UChar32 as a type for single Unicode code points.
virtual int32_t getRuleStatusVec(int32_t *fillInVec, int32_t capacity, UErrorCode &status)
For RuleBasedBreakIterators, get the status (tag) values from the break rule(s) that determined the b...
C API: Data loading interface.
struct UDataMemory UDataMemory
Forward declaration of the data memory type.
virtual int32_t previous(void)=0
Set the iterator position to the boundary preceding the current boundary.
virtual UClassID getDynamicClassID(void) const override=0
Return a polymorphic class ID for this object.
virtual void setText(const UnicodeString &text)=0
Change the text over which this operates.
virtual BreakIterator & refreshInputText(UText *input, UErrorCode &status)=0
Set the subject text string upon which the break iterator is operating without changing any other asp...
C API: Parse Error Information.
UErrorCode
Standard ICU4C error code type, a substitute for exceptions.
virtual BreakIterator * createBufferClone(void *stackBuffer, int32_t &BufferSize, UErrorCode &status)=0
Deprecated functionality.
virtual int32_t getRuleStatus() const
For RuleBasedBreakIterators, return the status tag from the break rule that determined the boundary a...
A subclass of BreakIterator whose behavior is specified using a list of rules.
virtual int32_t preceding(int32_t offset)=0
Set the iterator position to the first boundary preceding the specified offset.
A UParseError struct is used to returned detailed information about parsing errors.
Basic definitions for ICU, for both C and C++ APIs.
#define U_COMMON_API
Set to export library symbols from inside the common library, and to import them from outside...
UnicodeString is a string class that stores Unicode characters directly and provides similar function...
virtual BreakIterator * clone() const =0
Return a polymorphic copy of this object.
int8_t UBool
The ICU boolean type, a signed-byte integer.