ICU-64260.0.1.tar.gz

[apple/icu.git] / icuSources / common / unicode / brkiter.h

diff --git a/icuSources/common/unicode/brkiter.h b/icuSources/common/unicode/brkiter.h
index bdd3cc700f46e62e56f1fa84bfe0a6dcf18406da..42a0fbfa1da3ee002409bee2bfb37d34b2a45896 100644 (file)
--- a/icuSources/common/unicode/brkiter.h
+++ b/icuSources/common/unicode/brkiter.h
@@ -1,6 +1,8 @@
+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html

 /*
 ********************************************************************************
 /*
 ********************************************************************************

-*   Copyright (C) 1997-2010, International Business Machines
+*   Copyright (C) 1997-2016, International Business Machines

 *   Corporation and others.  All Rights Reserved.
 ********************************************************************************
 *
 *   Corporation and others.  All Rights Reserved.
 ********************************************************************************
 *


@@ -29,6 +31,7 @@
 
 #if UCONFIG_NO_BREAK_ITERATION
 
 
 #if UCONFIG_NO_BREAK_ITERATION
 

+#if U_SHOW_CPLUSPLUS_API

 U_NAMESPACE_BEGIN
 
 /*
 U_NAMESPACE_BEGIN
 
 /*


@@ -38,6 +41,7 @@ U_NAMESPACE_BEGIN
 class BreakIterator;
 
 U_NAMESPACE_END
 class BreakIterator;
 
 U_NAMESPACE_END

+#endif // U_SHOW_CPLUSPLUS_API

 
 #else
 
 
 #else
 


@@ -50,6 +54,7 @@ U_NAMESPACE_END
 #include "unicode/utext.h"
 #include "unicode/umisc.h"
 
 #include "unicode/utext.h"
 #include "unicode/umisc.h"
 

+#if U_SHOW_CPLUSPLUS_API

 U_NAMESPACE_BEGIN
 
 /**
 U_NAMESPACE_BEGIN
 
 /**


@@ -168,6 +173,11 @@ public:
     /**
      * Change the text over which this operates. The text boundary is
      * reset to the start.
     /**
      * Change the text over which this operates. The text boundary is
      * reset to the start.

+     *
+     * The BreakIterator will retain a reference to the supplied string.
+     * The caller must not modify or delete the text while the BreakIterator
+     * retains the reference.
+     *

      * @param text The UnicodeString used to change the text.
      * @stable ICU 2.0
      */
      * @param text The UnicodeString used to change the text.
      * @stable ICU 2.0
      */


@@ -182,6 +192,11 @@ public:
      * Utext that was passed as a parameter, but that the underlying text itself
      * must not be altered while being referenced by the break iterator.
      *
      * Utext that was passed as a parameter, but that the underlying text itself
      * must not be altered while being referenced by the break iterator.
      *

+     * All index positions returned by break iterator functions are
+     * native indices from the UText. For example, when breaking UTF-8
+     * encoded text, the break positions returned by next(), previous(), etc.
+     * will be UTF-8 string indices, not UTF-16 positions.
+     *

      * @param text The UText used to change the text.
      * @param status receives any error codes.
      * @stable ICU 3.4
      * @param text The UText used to change the text.
      * @param status receives any error codes.
      * @stable ICU 3.4


@@ -208,19 +223,21 @@ public:
     };
 
     /**
     };
 
     /**

-     * Return the index of the first character in the text being scanned.
+     * Sets the current iteration position to the beginning of the text, position zero.
+     * @return The offset of the beginning of the text, zero.

      * @stable ICU 2.0
      */
     virtual int32_t first(void) = 0;
 
     /**
      * @stable ICU 2.0
      */
     virtual int32_t first(void) = 0;
 
     /**

-     * Return the index immediately BEYOND the last character in the text being scanned.
+     * Set the iterator position to the index immediately BEYOND the last character in the text being scanned.
+     * @return The index immediately BEYOND the last character in the text being scanned.

      * @stable ICU 2.0
      */
     virtual int32_t last(void) = 0;
 
     /**
      * @stable ICU 2.0
      */
     virtual int32_t last(void) = 0;
 
     /**

-     * Return the boundary preceding the current boundary.
+     * Set the iterator position to the boundary preceding the current boundary.

      * @return The character index of the previous text boundary or DONE if all
      * boundaries have been returned.
      * @stable ICU 2.0
      * @return The character index of the previous text boundary or DONE if all
      * boundaries have been returned.
      * @stable ICU 2.0


@@ -228,7 +245,7 @@ public:
     virtual int32_t previous(void) = 0;
 
     /**
     virtual int32_t previous(void) = 0;
 
     /**

-     * Return the boundary following the current boundary.
+     * Advance the iterator to the boundary following the current boundary.

      * @return The character index of the next text boundary or DONE if all
      * boundaries have been returned.
      * @stable ICU 2.0
      * @return The character index of the next text boundary or DONE if all
      * boundaries have been returned.
      * @stable ICU 2.0


@@ -236,14 +253,14 @@ public:
     virtual int32_t next(void) = 0;
 
     /**
     virtual int32_t next(void) = 0;
 
     /**

-     * Return character index of the current interator position within the text.
+     * Return character index of the current iterator position within the text.

      * @return The boundary most recently returned.
      * @stable ICU 2.0
      */
     virtual int32_t current(void) const = 0;
 
     /**
      * @return The boundary most recently returned.
      * @stable ICU 2.0
      */
     virtual int32_t current(void) const = 0;
 
     /**

-     * Return the first boundary following the specified offset.
+     * Advance the iterator to the first boundary following the specified offset.

      * The value returned is always greater than the offset or
      * the value BreakIterator.DONE
      * @param offset the offset to begin scanning.
      * The value returned is always greater than the offset or
      * the value BreakIterator.DONE
      * @param offset the offset to begin scanning.


@@ -253,7 +270,7 @@ public:
     virtual int32_t following(int32_t offset) = 0;
 
     /**
     virtual int32_t following(int32_t offset) = 0;
 
     /**

-     * Return the first boundary preceding the specified offset.
+     * Set the iterator position to the first boundary preceding the specified offset.

      * The value returned is always smaller than the offset or
      * the value BreakIterator.DONE
      * @param offset the offset to begin scanning.
      * The value returned is always smaller than the offset or
      * the value BreakIterator.DONE
      * @param offset the offset to begin scanning.


@@ -263,7 +280,7 @@ public:
     virtual int32_t preceding(int32_t offset) = 0;
 
     /**
     virtual int32_t preceding(int32_t offset) = 0;
 
     /**

-     * Return true if the specfied position is a boundary position.
+     * Return true if the specified position is a boundary position.

      * As a side effect, the current position of the iterator is set
      * to the first boundary position at or following the specified offset.
      * @param offset the offset to check.
      * As a side effect, the current position of the iterator is set
      * to the first boundary position at or following the specified offset.
      * @param offset the offset to check.


@@ -273,16 +290,61 @@ public:
     virtual UBool isBoundary(int32_t offset) = 0;
 
     /**
     virtual UBool isBoundary(int32_t offset) = 0;
 
     /**

-     * Return the nth boundary from the current boundary
-     * @param n which boundary to return.  A value of 0
+     * Set the iterator position to the nth boundary from the current boundary
+     * @param n the number of boundaries to move by.  A value of 0

      * does nothing.  Negative values move to previous boundaries
      * and positive values move to later boundaries.
      * does nothing.  Negative values move to previous boundaries
      * and positive values move to later boundaries.

-     * @return The index of the nth boundary from the current position, or
-     * DONE if there are fewer than |n| boundaries in the specfied direction.
+     * @return The new iterator position, or
+     * DONE if there are fewer than |n| boundaries in the specified direction.

      * @stable ICU 2.0
      */
     virtual int32_t next(int32_t n) = 0;
 
      * @stable ICU 2.0
      */
     virtual int32_t next(int32_t n) = 0;
 

+   /**
+     * For RuleBasedBreakIterators, return the status tag from the break rule
+     * that determined the boundary at the current iteration position.
+     * <p>
+     * For break iterator types that do not support a rule status,
+     * a default value of 0 is returned.
+     * <p>
+     * @return the status from the break rule that determined the boundary at
+     *         the current iteration position.
+     * @see RuleBaseBreakIterator::getRuleStatus()
+     * @see UWordBreak
+     * @stable ICU 52
+     */
+    virtual int32_t getRuleStatus() const;
+
+   /**
+    * For RuleBasedBreakIterators, get the status (tag) values from the break rule(s)
+    * that determined the boundary at the current iteration position.
+    * <p>
+    * For break iterator types that do not support rule status,
+    * no values are returned.
+    * <p>
+    * The returned status value(s) are stored into an array provided by the caller.
+    * The values are stored in sorted (ascending) order.
+    * If the capacity of the output array is insufficient to hold the data,
+    *  the output will be truncated to the available length, and a
+    *  U_BUFFER_OVERFLOW_ERROR will be signaled.
+    * <p>
+    * @see RuleBaseBreakIterator::getRuleStatusVec
+    *
+    * @param fillInVec an array to be filled in with the status values.
+    * @param capacity  the length of the supplied vector.  A length of zero causes
+    *                  the function to return the number of status values, in the
+    *                  normal way, without attempting to store any values.
+    * @param status    receives error codes.
+    * @return          The number of rule status values from rules that determined
+    *                  the boundary at the current iteration position.
+    *                  In the event of a U_BUFFER_OVERFLOW_ERROR, the return value
+    *                  is the total number of status values that were available,
+    *                  not the reduced number that were actually returned.
+    * @see getRuleStatus
+    * @stable ICU 52
+    */
+    virtual int32_t getRuleStatusVec(int32_t *fillInVec, int32_t capacity, UErrorCode &status);
+

     /**
      * Create BreakIterator for word-breaks using the given locale.
      * Returns an instance of a BreakIterator implementing word breaks.
     /**
      * Create BreakIterator for word-breaks using the given locale.
      * Returns an instance of a BreakIterator implementing word breaks.


@@ -372,12 +434,13 @@ public:
     static BreakIterator* U_EXPORT2
     createSentenceInstance(const Locale& where, UErrorCode& status);
 
     static BreakIterator* U_EXPORT2
     createSentenceInstance(const Locale& where, UErrorCode& status);
 

+#ifndef U_HIDE_DEPRECATED_API

     /**
      * Create BreakIterator for title-casing breaks using the specified locale
      * Returns an instance of a BreakIterator implementing title breaks.
      * The iterator returned locates title boundaries as described for
      * Unicode 3.2 only. For Unicode 4.0 and above title boundary iteration,
     /**
      * Create BreakIterator for title-casing breaks using the specified locale
      * Returns an instance of a BreakIterator implementing title breaks.
      * The iterator returned locates title boundaries as described for
      * Unicode 3.2 only. For Unicode 4.0 and above title boundary iteration,

-     * please use Word Boundary iterator.{@link #createWordInstance }
+     * please use a word boundary iterator. See {@link #createWordInstance }.

      *
      * @param where the locale.
      * @param status The error code.
      *
      * @param where the locale.
      * @param status The error code.


@@ -392,10 +455,11 @@ public:
      * used; neither the requested locale nor any of its fall back locales
      * could be found.
      * The caller owns the returned object and is responsible for deleting it.
      * used; neither the requested locale nor any of its fall back locales
      * could be found.
      * The caller owns the returned object and is responsible for deleting it.

-     * @stable ICU 2.1
+     * @deprecated ICU 64 Use createWordInstance instead.

      */
     static BreakIterator* U_EXPORT2
     createTitleInstance(const Locale& where, UErrorCode& status);
      */
     static BreakIterator* U_EXPORT2
     createTitleInstance(const Locale& where, UErrorCode& status);

+#endif /* U_HIDE_DEPRECATED_API */

 
     /**
      * Get the set of Locales for which TextBoundaries are installed.
 
     /**
      * Get the set of Locales for which TextBoundaries are installed.


@@ -409,7 +473,7 @@ public:
     static const Locale* U_EXPORT2 getAvailableLocales(int32_t& count);
 
     /**
     static const Locale* U_EXPORT2 getAvailableLocales(int32_t& count);
 
     /**

-     * Get name of the object for the desired Locale, in the desired langauge.
+     * Get name of the object for the desired Locale, in the desired language.

      * @param objectLocale must be from getAvailableLocales.
      * @param displayLocale specifies the desired locale for output.
      * @param name the fill-in parameter of the return value
      * @param objectLocale must be from getAvailableLocales.
      * @param displayLocale specifies the desired locale for output.
      * @param name the fill-in parameter of the return value


@@ -422,7 +486,7 @@ public:
                                          UnicodeString& name);
 
     /**
                                          UnicodeString& name);
 
     /**

-     * Get name of the object for the desired Locale, in the langauge of the
+     * Get name of the object for the desired Locale, in the language of the

      * default locale.
      * @param objectLocale must be from getMatchingLocales
      * @param name the fill-in parameter of the return value
      * default locale.
      * @param objectLocale must be from getMatchingLocales
      * @param name the fill-in parameter of the return value


@@ -433,6 +497,8 @@ public:
                                          UnicodeString& name);
 
     /**
                                          UnicodeString& name);
 
     /**

+     * Deprecated functionality. Use clone() instead.
+     *

      * Thread safe client-buffer-based cloning operation
      *    Do NOT call delete on a safeclone, since 'new' is not used to create it.
      * @param stackBuffer user allocated space for the new clone. If NULL new memory will be allocated.
      * Thread safe client-buffer-based cloning operation
      *    Do NOT call delete on a safeclone, since 'new' is not used to create it.
      * @param stackBuffer user allocated space for the new clone. If NULL new memory will be allocated.


@@ -447,26 +513,33 @@ public:
      *  necessary.
      * @return pointer to the new clone
      *
      *  necessary.
      * @return pointer to the new clone
      *

-     * @stable ICU 2.0
+     * @deprecated ICU 52. Use clone() instead.

      */
     virtual BreakIterator *  createBufferClone(void *stackBuffer,
                                                int32_t &BufferSize,
                                                UErrorCode &status) = 0;
 
      */
     virtual BreakIterator *  createBufferClone(void *stackBuffer,
                                                int32_t &BufferSize,
                                                UErrorCode &status) = 0;
 

+#ifndef U_HIDE_DEPRECATED_API
+

     /**
      *   Determine whether the BreakIterator was created in user memory by
      *   createBufferClone(), and thus should not be deleted.  Such objects
      *   must be closed by an explicit call to the destructor (not delete).
     /**
      *   Determine whether the BreakIterator was created in user memory by
      *   createBufferClone(), and thus should not be deleted.  Such objects
      *   must be closed by an explicit call to the destructor (not delete).

-     *  @stable ICU 2.0
+     * @deprecated ICU 52. Always delete the BreakIterator.

      */
     inline UBool isBufferClone(void);
 
      */
     inline UBool isBufferClone(void);
 

+#endif /* U_HIDE_DEPRECATED_API */
+

 #if !UCONFIG_NO_SERVICE
     /**
      * Register a new break iterator of the indicated kind, to use in the given locale.
      * The break iterator will be adopted.  Clones of the iterator will be returned
      * if a request for a break iterator of the given kind matches or falls back to
      * this locale.
 #if !UCONFIG_NO_SERVICE
     /**
      * Register a new break iterator of the indicated kind, to use in the given locale.
      * The break iterator will be adopted.  Clones of the iterator will be returned
      * if a request for a break iterator of the given kind matches or falls back to
      * this locale.

+     * Because ICU may choose to cache BreakIterators internally, this must
+     * be called at application startup, prior to any calls to
+     * BreakIterator::createXXXInstance to avoid undefined behavior.

      * @param toAdopt the BreakIterator instance to be adopted
      * @param locale the Locale for which this instance is to be registered
      * @param kind the type of iterator for which this instance is to be registered
      * @param toAdopt the BreakIterator instance to be adopted
      * @param locale the Locale for which this instance is to be registered
      * @param kind the type of iterator for which this instance is to be registered


@@ -483,6 +556,9 @@ public:
      * Unregister a previously-registered BreakIterator using the key returned from the
      * register call.  Key becomes invalid after a successful call and should not be used again.
      * The BreakIterator corresponding to the key will be deleted.
      * Unregister a previously-registered BreakIterator using the key returned from the
      * register call.  Key becomes invalid after a successful call and should not be used again.
      * The BreakIterator corresponding to the key will be deleted.

+     * Because ICU may choose to cache BreakIterators internally, this should
+     * be called during application shutdown, after all calls to
+     * BreakIterator::createXXXInstance to avoid undefined behavior.

      * @param key the registry key returned by a previous call to registerInstance
      * @param status the in/out status code, no special meanings are assigned
      * @return TRUE if the iterator for the key was successfully unregistered
      * @param key the registry key returned by a previous call to registerInstance
      * @param status the in/out status code, no special meanings are assigned
      * @return TRUE if the iterator for the key was successfully unregistered


@@ -506,6 +582,7 @@ public:
      */
     Locale getLocale(ULocDataLocaleType type, UErrorCode& status) const;
 
      */
     Locale getLocale(ULocDataLocaleType type, UErrorCode& status) const;
 

+#ifndef U_HIDE_INTERNAL_API

     /** Get the locale for this break iterator object. You can choose between valid and actual locale.
      *  @param type type of the locale we're looking for (valid or actual)
      *  @param status error code for the operation
     /** Get the locale for this break iterator object. You can choose between valid and actual locale.
      *  @param type type of the locale we're looking for (valid or actual)
      *  @param status error code for the operation


@@ -513,9 +590,46 @@ public:
      *  @internal
      */
     const char *getLocaleID(ULocDataLocaleType type, UErrorCode& status) const;
      *  @internal
      */
     const char *getLocaleID(ULocDataLocaleType type, UErrorCode& status) const;

+#endif  /* U_HIDE_INTERNAL_API */
+
+    /**
+     *  Set the subject text string upon which the break iterator is operating
+     *  without changing any other aspect of the matching state.
+     *  The new and previous text strings must have the same content.
+     *
+     *  This function is intended for use in environments where ICU is operating on
+     *  strings that may move around in memory.  It provides a mechanism for notifying
+     *  ICU that the string has been relocated, and providing a new UText to access the
+     *  string in its new position.
+     *
+     *  Note that the break iterator implementation never copies the underlying text
+     *  of a string being processed, but always operates directly on the original text
+     *  provided by the user. Refreshing simply drops the references to the old text
+     *  and replaces them with references to the new.
+     *
+     *  Caution:  this function is normally used only by very specialized,
+     *  system-level code.  One example use case is with garbage collection that moves
+     *  the text in memory.
+     *
+     * @param input      The new (moved) text string.
+     * @param status     Receives errors detected by this function.
+     * @return           *this
+     *
+     * @stable ICU 49
+     */
+    virtual BreakIterator &refreshInputText(UText *input, UErrorCode &status) = 0;
+
+#ifndef U_HIDE_INTERNAL_API
+    /**
+     * Set the ULineWordOptions for this break iterator.
+     * @param lineWordOpts The ULineWordOptions to set.
+     * @internal Apple only
+     */
+    void setLineWordOpts(ULineWordOptions lineWordOpts);
+#endif  /* U_HIDE_INTERNAL_API */

 
  private:
 
  private:

-    static BreakIterator* buildInstance(const Locale& loc, const char *type, int32_t kind, UErrorCode& status);
+    static BreakIterator* buildInstance(const Locale& loc, const char *type, UErrorCode& status);

     static BreakIterator* createInstance(const Locale& loc, int32_t kind, UErrorCode& status);
     static BreakIterator* makeInstance(const Locale& loc, int32_t kind, UErrorCode& status);
 
     static BreakIterator* createInstance(const Locale& loc, int32_t kind, UErrorCode& status);
     static BreakIterator* makeInstance(const Locale& loc, int32_t kind, UErrorCode& status);
 


@@ -523,35 +637,45 @@ public:
     friend class ICUBreakIteratorService;
 
 protected:
     friend class ICUBreakIteratorService;
 
 protected:

+    // Do not enclose protected default/copy constructors with #ifndef U_HIDE_INTERNAL_API
+    // or else the compiler will create a public ones.

     /** @internal */
     BreakIterator();
     /** @internal */
     /** @internal */
     BreakIterator();
     /** @internal */

-    UBool fBufferClone;
+    BreakIterator (const BreakIterator &other);
+#ifndef U_HIDE_INTERNAL_API

     /** @internal */
     /** @internal */

-    BreakIterator (const BreakIterator &other) : UObject(other), fBufferClone(FALSE) {}
+    BreakIterator (const Locale& valid, const Locale &actual);
+    /** @internal. Assignment Operator, used by RuleBasedBreakIterator. */
+    BreakIterator &operator = (const BreakIterator &other);
+#endif  /* U_HIDE_INTERNAL_API */
+    ULineWordOptions fLineWordOpts;

 
 private:
 
 
 private:
 

-    /** @internal */
+    /** @internal (private) */

     char actualLocale[ULOC_FULLNAME_CAPACITY];
     char validLocale[ULOC_FULLNAME_CAPACITY];
     char actualLocale[ULOC_FULLNAME_CAPACITY];
     char validLocale[ULOC_FULLNAME_CAPACITY];

-
-    /**
-     * The assignment operator has no real implementation.
-     * It's provided to make the compiler happy. Do not call.
-     */
-    BreakIterator& operator=(const BreakIterator&);

 };
 
 };
 

+inline void BreakIterator::setLineWordOpts(ULineWordOptions lineWordOpts)
+{
+    fLineWordOpts = lineWordOpts;
+}
+
+#ifndef U_HIDE_DEPRECATED_API
+

 inline UBool BreakIterator::isBufferClone()
 {
 inline UBool BreakIterator::isBufferClone()
 {

-    return fBufferClone;
+    return FALSE;

 }
 
 }
 

+#endif /* U_HIDE_DEPRECATED_API */
+

 U_NAMESPACE_END
 U_NAMESPACE_END

+#endif // U_SHOW_CPLUSPLUS_API

 
 #endif /* #if !UCONFIG_NO_BREAK_ITERATION */
 
 
 #endif /* #if !UCONFIG_NO_BREAK_ITERATION */
 

-#endif // _BRKITER
+#endif // BRKITER_H

 //eof
 //eof

-