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 1c48876d35a751539e186250d29b2df6003e8fec..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) 1997-2003, International Business Machines
+*   Copyright (C) 1997-2016, International Business Machines

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


@@ -22,8 +24,14 @@
 
 #include "unicode/utypes.h"
 
 
 #include "unicode/utypes.h"
 

+/**
+ * \file
+ * \brief C++ API: Break Iterator.
+ */
+

 #if UCONFIG_NO_BREAK_ITERATION
 
 #if UCONFIG_NO_BREAK_ITERATION
 

+#if U_SHOW_CPLUSPLUS_API

 U_NAMESPACE_BEGIN
 
 /*
 U_NAMESPACE_BEGIN
 
 /*


@@ -33,6 +41,7 @@ U_NAMESPACE_BEGIN
 class BreakIterator;
 
 U_NAMESPACE_END
 class BreakIterator;
 
 U_NAMESPACE_END

+#endif // U_SHOW_CPLUSPLUS_API

 
 #else
 
 
 #else
 


@@ -42,32 +51,33 @@ U_NAMESPACE_END
 #include "unicode/locid.h"
 #include "unicode/ubrk.h"
 #include "unicode/strenum.h"
 #include "unicode/locid.h"
 #include "unicode/ubrk.h"
 #include "unicode/strenum.h"

+#include "unicode/utext.h"
+#include "unicode/umisc.h"

 
 

+#if U_SHOW_CPLUSPLUS_API

 U_NAMESPACE_BEGIN
 
 U_NAMESPACE_BEGIN
 

-typedef const void* URegistryKey;
-

 /**
  * The BreakIterator class implements methods for finding the location
  * of boundaries in text. BreakIterator is an abstract base class.
  * Instances of BreakIterator maintain a current position and scan over
  * text returning the index of characters where boundaries occur.
 /**
  * The BreakIterator class implements methods for finding the location
  * of boundaries in text. BreakIterator is an abstract base class.
  * Instances of BreakIterator maintain a current position and scan over
  * text returning the index of characters where boundaries occur.

- * <P>
+ * <p>

  * Line boundary analysis determines where a text string can be broken
  * when line-wrapping. The mechanism correctly handles punctuation and
  * hyphenated words.
  * Line boundary analysis determines where a text string can be broken
  * when line-wrapping. The mechanism correctly handles punctuation and
  * hyphenated words.

- * <P>
+ * <p>

  * Sentence boundary analysis allows selection with correct
  * interpretation of periods within numbers and abbreviations, and
  * trailing punctuation marks such as quotation marks and parentheses.
  * Sentence boundary analysis allows selection with correct
  * interpretation of periods within numbers and abbreviations, and
  * trailing punctuation marks such as quotation marks and parentheses.

- * <P>
+ * <p>

  * Word boundary analysis is used by search and replace functions, as
  * well as within text editing applications that allow the user to
  * select words with a double click. Word selection provides correct
  * interpretation of punctuation marks within and following
  * words. Characters that are not part of a word, such as symbols or
  * punctuation marks, have word-breaks on both sides.
  * Word boundary analysis is used by search and replace functions, as
  * well as within text editing applications that allow the user to
  * select words with a double click. Word selection provides correct
  * interpretation of punctuation marks within and following
  * words. Characters that are not part of a word, such as symbols or
  * punctuation marks, have word-breaks on both sides.

- * <P>
+ * <p>

  * Character boundary analysis allows users to interact with
  * characters as they expect to, for example, when moving the cursor
  * through a text string. Character boundary analysis provides correct
  * Character boundary analysis allows users to interact with
  * characters as they expect to, for example, when moving the cursor
  * through a text string. Character boundary analysis provides correct


@@ -75,126 +85,22 @@ typedef const void* URegistryKey;
  * character is stored.  For example, an accented character might be
  * stored as a base character and a diacritical mark. What users
  * consider to be a character can differ between languages.
  * character is stored.  For example, an accented character might be
  * stored as a base character and a diacritical mark. What users
  * consider to be a character can differ between languages.

- * <P>
- * This is the interface for all text boundaries.
- * <P>
- * Examples:
- * <P>
- * Helper function to output text
- * <pre>
- * \code
- *    void printTextRange( BreakIterator& iterator, int32_t start, int32_t end )
- *    {
- *        UnicodeString textBuffer, temp;
- *        CharacterIterator *strIter = iterator.createText();
- *        strIter->getText(temp);
- *        cout << " " << start << " " << end << " |"
- *             << temp.extractBetween(start, end, textBuffer)
- *             << "|" << endl;
- *        delete strIter;
- *    }
- * \endcode
- * </pre>
- * Print each element in order:
- * <pre>
- * \code
- *    void printEachForward( BreakIterator& boundary)
- *    {
- *       int32_t start = boundary.first();
- *       for (int32_t end = boundary.next();
- *         end != BreakIterator::DONE;
- *         start = end, end = boundary.next())
- *         {
- *             printTextRange( boundary, start, end );
- *         }
- *    }
- * \code
- * </pre>
- * Print each element in reverse order:
- * <pre>
- * \code
- *    void printEachBackward( BreakIterator& boundary)
- *    {
- *       int32_t end = boundary.last();
- *       for (int32_t start = boundary.previous();
- *         start != BreakIterator::DONE;
- *         end = start, start = boundary.previous())
- *         {
- *             printTextRange( boundary, start, end );
- *         }
- *    }
- * \endcode
- * </pre>
- * Print first element
- * <pre>
- * \code
- *    void printFirst(BreakIterator& boundary)
- *    {
- *        int32_t start = boundary.first();
- *        int32_t end = boundary.next();
- *        printTextRange( boundary, start, end );
- *    }
- * \endcode
- * </pre>
- * Print last element
- * <pre>
- *  \code
- *    void printLast(BreakIterator& boundary)
- *    {
- *        int32_t end = boundary.last();
- *        int32_t start = boundary.previous();
- *        printTextRange( boundary, start, end );
- *    }
- * \endcode
- * </pre>
- * Print the element at a specified position
- * <pre>
- * \code
- *    void printAt(BreakIterator &boundary, int32_t pos )
- *    {
- *        int32_t end = boundary.following(pos);
- *        int32_t start = boundary.previous();
- *        printTextRange( boundary, start, end );
- *    }
- * \endcode
- * </pre>
- * Creating and using text boundaries
- * <pre>
- * \code
- *       void BreakIterator_Example( void )
- *       {
- *           BreakIterator* boundary;
- *           UnicodeString stringToExamine("Aaa bbb ccc. Ddd eee fff.");
- *           cout << "Examining: " << stringToExamine << endl;
+ * <p>
+ * The text boundary positions are found according to the rules
+ * described in Unicode Standard Annex #29, Text Boundaries, and
+ * Unicode Standard Annex #14, Line Breaking Properties.  These
+ * are available at http://www.unicode.org/reports/tr14/ and
+ * http://www.unicode.org/reports/tr29/.
+ * <p>
+ * In addition to the C++ API defined in this header file, a
+ * plain C API with equivalent functionality is defined in the
+ * file ubrk.h
+ * <p>
+ * Code snippets illustrating the use of the Break Iterator APIs
+ * are available in the ICU User Guide,
+ * http://icu-project.org/userguide/boundaryAnalysis.html
+ * and in the sample program icu/source/samples/break/break.cpp

  *
  *

- *           //print each sentence in forward and reverse order
- *           boundary = BreakIterator::createSentenceInstance( Locale::US );
- *           boundary->setText(stringToExamine);
- *           cout << "----- forward: -----------" << endl;
- *           printEachForward(*boundary);
- *           cout << "----- backward: ----------" << endl;
- *           printEachBackward(*boundary);
- *           delete boundary;
- *
- *           //print each word in order
- *           boundary = BreakIterator::createWordInstance();
- *           boundary->setText(stringToExamine);
- *           cout << "----- forward: -----------" << endl;
- *           printEachForward(*boundary);
- *           //print first element
- *           cout << "----- first: -------------" << endl;
- *           printFirst(*boundary);
- *           //print last element
- *           cout << "----- last: --------------" << endl;
- *           printLast(*boundary);
- *           //print word at charpos 10
- *           cout << "----- at pos 10: ---------" << endl;
- *           printAt(*boundary, 10 );
- *
- *           delete boundary;
- *       }
- * \endcode
- * </pre>

  */
 class U_COMMON_API BreakIterator : public UObject {
 public:
  */
 class U_COMMON_API BreakIterator : public UObject {
 public:


@@ -243,49 +149,95 @@ public:
 
     /**
      * Return a CharacterIterator over the text being analyzed.
 
     /**
      * Return a CharacterIterator over the text being analyzed.

-     * Changing the state of the returned iterator can have undefined consequences
-     * on the operation of the break iterator.  If you need to change it, clone it first.

      * @stable ICU 2.0
      */
      * @stable ICU 2.0
      */

-    virtual const CharacterIterator& getText(void) const = 0;
+    virtual CharacterIterator& getText(void) const = 0;
+
+
+    /**
+      *  Get a UText for the text being analyzed.
+      *  The returned UText is a shallow clone of the UText used internally
+      *  by the break iterator implementation.  It can safely be used to
+      *  access the text without impacting any break iterator operations,
+      *  but the underlying text itself must not be altered.
+      *
+      * @param fillIn A UText to be filled in.  If NULL, a new UText will be
+      *           allocated to hold the result.
+      * @param status receives any error codes.
+      * @return   The current UText for this break iterator.  If an input
+      *           UText was provided, it will always be returned.
+      * @stable ICU 3.4
+      */
+     virtual UText *getUText(UText *fillIn, UErrorCode &status) const = 0;

 
     /**
      * 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
      */
     virtual void  setText(const UnicodeString &text) = 0;
 
      * @param text The UnicodeString used to change the text.
      * @stable ICU 2.0
      */
     virtual void  setText(const UnicodeString &text) = 0;
 

+    /**
+     * Reset the break iterator to operate over the text represented by
+     * the UText.  The iterator position is reset to the start.
+     *
+     * This function makes a shallow clone of the supplied UText.  This means
+     * that the caller is free to immediately close or otherwise reuse the
+     * 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
+     */
+    virtual void  setText(UText *text, UErrorCode &status) = 0;
+

     /**
      * 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.

+     * Note that setText(UText *) provides similar functionality to this function,
+     * and is more efficient.

      * @param it The CharacterIterator used to change the text.
      * @stable ICU 2.0
      */
     virtual void  adoptText(CharacterIterator* it) = 0;
 
      * @param it The CharacterIterator used to change the text.
      * @stable ICU 2.0
      */
     virtual void  adoptText(CharacterIterator* it) = 0;
 

-    /**
-     * DONE is returned by previous() and next() after all valid
-     * boundaries have been returned.
-     * @stable ICU 2.0
-     */
-    static const int32_t DONE;
+    enum {
+        /**
+         * DONE is returned by previous() and next() after all valid
+         * boundaries have been returned.
+         * @stable ICU 2.0
+         */
+        DONE = (int32_t)-1
+    };

 
     /**
 
     /**

-     * 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


@@ -293,7 +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


@@ -301,14 +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.


@@ -318,7 +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.


@@ -328,7 +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.


@@ -338,16 +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.


@@ -367,8 +364,8 @@ public:
      * The caller owns the returned object and is responsible for deleting it.
      * @stable ICU 2.0
      */
      * The caller owns the returned object and is responsible for deleting it.
      * @stable ICU 2.0
      */

-    static BreakIterator* createWordInstance(const Locale& where,
-                                                   UErrorCode& status);
+    static BreakIterator* U_EXPORT2
+    createWordInstance(const Locale& where, UErrorCode& status);

 
     /**
      * Create BreakIterator for line-breaks using specified locale.
 
     /**
      * Create BreakIterator for line-breaks using specified locale.


@@ -391,8 +388,8 @@ public:
      * The caller owns the returned object and is responsible for deleting it.
      * @stable ICU 2.0
      */
      * The caller owns the returned object and is responsible for deleting it.
      * @stable ICU 2.0
      */

-    static BreakIterator* createLineInstance(const Locale& where,
-                                                   UErrorCode& status);
+    static BreakIterator* U_EXPORT2
+    createLineInstance(const Locale& where, UErrorCode& status);

 
     /**
      * Create BreakIterator for character-breaks using specified locale
 
     /**
      * Create BreakIterator for character-breaks using specified locale


@@ -413,8 +410,8 @@ public:
      * The caller owns the returned object and is responsible for deleting it.
      * @stable ICU 2.0
      */
      * The caller owns the returned object and is responsible for deleting it.
      * @stable ICU 2.0
      */

-    static BreakIterator* createCharacterInstance(const Locale& where,
-                                                        UErrorCode& status);
+    static BreakIterator* U_EXPORT2
+    createCharacterInstance(const Locale& where, UErrorCode& status);

 
     /**
      * Create BreakIterator for sentence-breaks using specified locale
 
     /**
      * Create BreakIterator for sentence-breaks using specified locale


@@ -434,15 +431,16 @@ public:
      * The caller owns the returned object and is responsible for deleting it.
      * @stable ICU 2.0
      */
      * The caller owns the returned object and is responsible for deleting it.
      * @stable ICU 2.0
      */

-    static BreakIterator* 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.
     /**
      * 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 
+     * The iterator returned locates title boundaries as described for

      * Unicode 3.2 only. For Unicode 4.0 and above title boundary iteration,
      * 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.


@@ -457,23 +455,25 @@ 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* 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.
      * <p><b>Note:</b> this will not return locales added through the register
 
     /**
      * Get the set of Locales for which TextBoundaries are installed.
      * <p><b>Note:</b> this will not return locales added through the register

-     * call.</p>
+     * call. To see the registered locales too, use the getAvailableLocales
+     * function that returns a StringEnumeration object </p>

      * @param count the output parameter of number of elements in the locale list
      * @return available locales
      * @stable ICU 2.0
      */
      * @param count the output parameter of number of elements in the locale list
      * @return available locales
      * @stable ICU 2.0
      */

-    static const Locale* 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


@@ -481,22 +481,24 @@ public:
      * @return user-displayable name
      * @stable ICU 2.0
      */
      * @return user-displayable name
      * @stable ICU 2.0
      */

-    static UnicodeString& getDisplayName(const Locale& objectLocale,
+    static UnicodeString& U_EXPORT2 getDisplayName(const Locale& objectLocale,

                                          const Locale& displayLocale,
                                          UnicodeString& name);
 
     /**
                                          const Locale& displayLocale,
                                          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
      * @return user-displayable name
      * @stable ICU 2.0
      */
      * default locale.
      * @param objectLocale must be from getMatchingLocales
      * @param name the fill-in parameter of the return value
      * @return user-displayable name
      * @stable ICU 2.0
      */

-    static UnicodeString& getDisplayName(const Locale& objectLocale,
+    static UnicodeString& U_EXPORT2 getDisplayName(const Locale& objectLocale,

                                          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.


@@ -511,90 +513,169 @@ 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.
     /**
      * Register a new break iterator of the indicated kind, to use in the given locale.

-     * The break iterator will be adoped.  Clones of the iterator will be returned
+     * 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 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 status the in/out status code, no special meanings are assigned
      * @return a registry key that can be used to unregister this instance
      * @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 status the in/out status code, no special meanings are assigned
      * @return a registry key that can be used to unregister this instance

-     * @draft ICU 2.4
+     * @stable ICU 2.4

      */
      */

-    static URegistryKey registerInstance(BreakIterator* toAdopt, const Locale& locale, UBreakIteratorType kind, UErrorCode& status);
+    static URegistryKey U_EXPORT2 registerInstance(BreakIterator* toAdopt,
+                                        const Locale& locale,
+                                        UBreakIteratorType kind,
+                                        UErrorCode& status);

 
     /**
      * 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

-     * @draft ICU 2.4
+     * @stable ICU 2.4

      */
      */

-    static UBool unregister(URegistryKey key, UErrorCode& status);
+    static UBool U_EXPORT2 unregister(URegistryKey key, UErrorCode& status);

 
     /**
 
     /**

-     * Return a StringEnumeration over the locales available at the time of the call, 
+     * Return a StringEnumeration over the locales available at the time of the call,

      * including registered locales.
      * @return a StringEnumeration over the locales available at the time of the call
      * including registered locales.
      * @return a StringEnumeration over the locales available at the time of the call

-     * @draft ICU 2.4
+     * @stable ICU 2.4

      */
      */

-    static StringEnumeration* getAvailableLocales(void);
+    static StringEnumeration* U_EXPORT2 getAvailableLocales(void);
+#endif

 
 

- private:
-    static BreakIterator* makeCharacterInstance(const Locale& loc, UErrorCode& status);
-    static BreakIterator* makeWordInstance(const Locale& loc, UErrorCode& status);
-    static BreakIterator* makeLineInstance(const Locale& loc, UErrorCode& status);
-    static BreakIterator* makeSentenceInstance(const Locale& loc, UErrorCode& status);
-    static BreakIterator* makeTitleInstance(const Locale& loc, UErrorCode& status);
+    /**
+     * Returns the locale for this break iterator. Two flavors are available: valid and
+     * actual locale.
+     * @stable ICU 2.8
+     */
+    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
+     *  @return the locale
+     *  @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;

 
 

-    static BreakIterator* createInstance(const Locale& loc, UBreakIteratorType kind, UErrorCode& status);
+#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:
+    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);
 
     friend class ICUBreakIteratorFactory;
     friend class ICUBreakIteratorService;
 
 protected:
     static BreakIterator* makeInstance(const Locale& loc, int32_t kind, UErrorCode& status);
 
     friend class ICUBreakIteratorFactory;
     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:

-    /**
-     * The assignment operator has no real implementation.
-     * It's provided to make the compiler happy. Do not call.
-     */
-    BreakIterator& operator=(const BreakIterator&) { return *this; }
+
+    /** @internal (private) */
+    char actualLocale[ULOC_FULLNAME_CAPACITY];
+    char validLocale[ULOC_FULLNAME_CAPACITY];

 };
 
 };
 

+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

-