ICU-66108.tar.gz

[apple/icu.git] / icuSources / i18n / unicode / regex.h

diff --git a/icuSources/i18n/unicode/regex.h b/icuSources/i18n/unicode/regex.h
index d8d7bc0edf794f06c55a5d42c5b48a560b65a9f5..6338eb7c754eff8a097bd864a08088dd28d215b2 100644 (file)
--- a/icuSources/i18n/unicode/regex.h
+++ b/icuSources/i18n/unicode/regex.h
@@ -1,10 +1,12 @@
+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html

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

-*   Copyright (C) 2002-2010, International Business Machines
+*   Copyright (C) 2002-2016, International Business Machines

 *   Corporation and others.  All Rights Reserved.
 **********************************************************************
 *   file name:  regex.h
 *   Corporation and others.  All Rights Reserved.
 **********************************************************************
 *   file name:  regex.h

-*   encoding:   US-ASCII
+*   encoding:   UTF-8

 *   indentation:4
 *
 *   created on: 2002oct22
 *   indentation:4
 *
 *   created on: 2002oct22


@@ -22,28 +24,28 @@
  * \file
  * \brief  C++ API:  Regular Expressions
  *
  * \file
  * \brief  C++ API:  Regular Expressions
  *

- * <h2>Regular Expression API</h2>
- *
- * <p>The ICU API for processing regular expressions consists of two classes,
- *  <code>RegexPattern</code> and <code>RegexMatcher</code>.
- *  <code>RegexPattern</code> objects represent a pre-processed, or compiled
+ * The ICU API for processing regular expressions consists of two classes,
+ *  `RegexPattern` and `RegexMatcher`.
+ *  `RegexPattern` objects represent a pre-processed, or compiled

  *  regular expression.  They are created from a regular expression pattern string,
  *  regular expression.  They are created from a regular expression pattern string,

- *  and can be used to create <code>RegexMatcher</code> objects for the pattern.</p>
+ *  and can be used to create `RegexMatcher` objects for the pattern.

  *
  *

- * <p>Class <code>RegexMatcher</code> bundles together a regular expression
+ * Class `RegexMatcher` bundles together a regular expression

  *  pattern and a target string to which the search pattern will be applied.
  *  pattern and a target string to which the search pattern will be applied.

- *  <code>RegexMatcher</code> includes API for doing plain find or search
+ *  `RegexMatcher` includes API for doing plain find or search

  *  operations, for search and replace operations, and for obtaining detailed
  *  operations, for search and replace operations, and for obtaining detailed

- *  information about bounds of a match. </p>
+ *  information about bounds of a match.

  *
  *

- * <p>Note that by constructing <code>RegexMatcher</code> objects directly from regular
+ * Note that by constructing `RegexMatcher` objects directly from regular

  * expression pattern strings application code can be simplified and the explicit
  * expression pattern strings application code can be simplified and the explicit

- * need for <code>RegexPattern</code> objects can usually be eliminated.
- * </p>
+ * need for `RegexPattern` objects can usually be eliminated.
+ *

  */
 
 #include "unicode/utypes.h"
 
  */
 
 #include "unicode/utypes.h"
 

+#if U_SHOW_CPLUSPLUS_API
+

 #if !UCONFIG_NO_REGULAR_EXPRESSIONS
 
 #include "unicode/uobject.h"
 #if !UCONFIG_NO_REGULAR_EXPRESSIONS
 
 #include "unicode/uobject.h"


@@ -53,57 +55,42 @@
 
 #include "unicode/uregex.h"
 
 
 #include "unicode/uregex.h"
 

-U_NAMESPACE_BEGIN
+// Forward Declarations

 
 

+struct UHashtable;

 
 

-// Forward Declarations...
+U_NAMESPACE_BEGIN

 
 

-class RegexMatcher;
-class RegexPattern;
-class UVector;
-class UVector32;
-class UVector64;
-class UnicodeSet;
-struct REStackFrame;

 struct Regex8BitSet;
 struct Regex8BitSet;

-class  RuleBasedBreakIterator;

 class  RegexCImpl;
 class  RegexCImpl;

-
-
-
-
-/**
- *   RBBIPatternDump   Debug function, displays the compiled form of a pattern.
- *   @internal
- */
-#ifdef REGEX_DEBUG
-U_INTERNAL void U_EXPORT2
-    RegexPatternDump(const RegexPattern *pat);
-#else
-    #undef RegexPatternDump
-    #define RegexPatternDump(pat)
-#endif
-
+class  RegexMatcher;
+class  RegexPattern;
+struct REStackFrame;
+class  RuleBasedBreakIterator;
+class  UnicodeSet;
+class  UVector;
+class  UVector32;
+class  UVector64;

 
 
 /**
 
 
 /**

-  * Class <code>RegexPattern</code> represents a compiled regular expression.  It includes
+  * Class `RegexPattern` represents a compiled regular expression.  It includes

   * factory methods for creating a RegexPattern object from the source (string) form
   * of a regular expression, methods for creating RegexMatchers that allow the pattern
   * to be applied to input text, and a few convenience methods for simple common
   * uses of regular expressions.
   *
   * factory methods for creating a RegexPattern object from the source (string) form
   * of a regular expression, methods for creating RegexMatchers that allow the pattern
   * to be applied to input text, and a few convenience methods for simple common
   * uses of regular expressions.
   *

-  * <p>Class RegexPattern is not intended to be subclassed.</p>
+  * Class RegexPattern is not intended to be subclassed.

   *
   * @stable ICU 2.4
   */
   *
   * @stable ICU 2.4
   */

-class U_I18N_API RegexPattern: public UObject {
+class U_I18N_API RegexPattern U_FINAL : public UObject {

 public:
 
     /**
      * default constructor.  Create a RegexPattern object that refers to no actual
      *   pattern.  Not normally needed; RegexPattern objects are usually
 public:
 
     /**
      * default constructor.  Create a RegexPattern object that refers to no actual
      *   pattern.  Not normally needed; RegexPattern objects are usually

-     *   created using the factory method <code>compile()</code>.
+     *   created using the factory method `compile()`.

      *
      * @stable ICU 2.4
      */
      *
      * @stable ICU 2.4
      */


@@ -126,7 +113,7 @@ public:
 
     /**
      * Comparison operator.  Two RegexPattern objects are considered equal if they
 
     /**
      * Comparison operator.  Two RegexPattern objects are considered equal if they

-     * were constructed from identical source patterns using the same match flag
+     * were constructed from identical source patterns using the same #URegexpFlag

      * settings.
      * @param that a RegexPattern object to compare with "this".
      * @return TRUE if the objects are equivalent.
      * settings.
      * @param that a RegexPattern object to compare with "this".
      * @return TRUE if the objects are equivalent.


@@ -136,13 +123,13 @@ public:
 
     /**
      * Comparison operator.  Two RegexPattern objects are considered equal if they
 
     /**
      * Comparison operator.  Two RegexPattern objects are considered equal if they

-     * were constructed from identical source patterns using the same match flag
+     * were constructed from identical source patterns using the same #URegexpFlag

      * settings.
      * @param that a RegexPattern object to compare with "this".
      * @return TRUE if the objects are different.
      * @stable ICU 2.4
      */
      * settings.
      * @param that a RegexPattern object to compare with "this".
      * @return TRUE if the objects are different.
      * @stable ICU 2.4
      */

-    inline UBool    operator!=(const RegexPattern& that) const {return ! operator ==(that);};
+    inline UBool    operator!=(const RegexPattern& that) const {return ! operator ==(that);}

 
     /**
      * Assignment operator.  After assignment, this RegexPattern will behave identically
 
     /**
      * Assignment operator.  After assignment, this RegexPattern will behave identically


@@ -153,7 +140,7 @@ public:
 
     /**
      * Create an exact copy of this RegexPattern object.  Since RegexPattern is not
 
     /**
      * Create an exact copy of this RegexPattern object.  Since RegexPattern is not

-     * intended to be subclasses, <code>clone()</code> and the copy construction are
+     * intended to be subclassed, <code>clone()</code> and the copy construction are

      * equivalent operations.
      * @return the copy of this RegexPattern
      * @stable ICU 2.4
      * equivalent operations.
      * @return the copy of this RegexPattern
      * @stable ICU 2.4


@@ -166,16 +153,16 @@ public:
     * object.  These compile methods, rather than the constructors, are the usual
     * way that RegexPattern objects are created.
     *
     * object.  These compile methods, rather than the constructors, are the usual
     * way that RegexPattern objects are created.
     *

-    * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+    * Note that RegexPattern objects must not be deleted while RegexMatcher

     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a
     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a

-    * catastrophic error.</p>
+    * catastrophic error.

     *
     *

-    * <p>All pattern match mode flags are set to their default values.</p>
+    * All #URegexpFlag pattern match mode flags are set to their default values.

     *
     *

-    * <p>Note that it is often more convenient to construct a RegexMatcher directly
+    * Note that it is often more convenient to construct a RegexMatcher directly

     *    from a pattern string rather than separately compiling the pattern and
     *    from a pattern string rather than separately compiling the pattern and

-    *    then creating a RegexMatcher object from the pattern.</p>
+    *    then creating a RegexMatcher object from the pattern.

     *
     * @param regex The regular expression to be compiled.
     * @param pe    Receives the position (line and column nubers) of any error
     *
     * @param regex The regular expression to be compiled.
     * @param pe    Receives the position (line and column nubers) of any error


@@ -189,22 +176,21 @@ public:
         UParseError          &pe,
         UErrorCode           &status);
 
         UParseError          &pe,
         UErrorCode           &status);
 

-

    /**
     * Compiles the regular expression in string form into a RegexPattern
     * object.  These compile methods, rather than the constructors, are the usual
     * way that RegexPattern objects are created.
     *
    /**
     * Compiles the regular expression in string form into a RegexPattern
     * object.  These compile methods, rather than the constructors, are the usual
     * way that RegexPattern objects are created.
     *

-    * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+    * Note that RegexPattern objects must not be deleted while RegexMatcher

     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a
     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a

-    * catastrophic error.</p>
+    * catastrophic error.

     *
     *

-    * <p>All pattern match mode flags are set to their default values.</p>
+    * All #URegexpFlag pattern match mode flags are set to their default values.

     *
     *

-    * <p>Note that it is often more convenient to construct a RegexMatcher directly
+    * Note that it is often more convenient to construct a RegexMatcher directly

     *    from a pattern string rather than separately compiling the pattern and
     *    from a pattern string rather than separately compiling the pattern and

-    *    then creating a RegexMatcher object from the pattern.</p>
+    *    then creating a RegexMatcher object from the pattern.

     *
     * @param regex The regular expression to be compiled. Note, the text referred
     *              to by this UText must not be deleted during the lifetime of the
     *
     * @param regex The regular expression to be compiled. Note, the text referred
     *              to by this UText must not be deleted during the lifetime of the


@@ -214,7 +200,7 @@ public:
     * @param status A reference to a UErrorCode to receive any errors.
     * @return      A regexPattern object for the compiled pattern.
     *
     * @param status A reference to a UErrorCode to receive any errors.
     * @return      A regexPattern object for the compiled pattern.
     *

-    * @draft ICU 4.6
+    * @stable ICU 4.6

     */
     static RegexPattern * U_EXPORT2 compile( UText *regex,
         UParseError          &pe,
     */
     static RegexPattern * U_EXPORT2 compile( UText *regex,
         UParseError          &pe,


@@ -222,21 +208,21 @@ public:
 
    /**
     * Compiles the regular expression in string form into a RegexPattern
 
    /**
     * Compiles the regular expression in string form into a RegexPattern

-    * object using the specified match mode flags.  These compile methods,
+    * object using the specified #URegexpFlag match mode flags.  These compile methods,

     * rather than the constructors, are the usual way that RegexPattern objects
     * are created.
     *
     * rather than the constructors, are the usual way that RegexPattern objects
     * are created.
     *

-    * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+    * Note that RegexPattern objects must not be deleted while RegexMatcher

     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a
     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a

-    * catastrophic error.</p>
+    * catastrophic error.

     *
     *

-    * <p>Note that it is often more convenient to construct a RegexMatcher directly
+    * Note that it is often more convenient to construct a RegexMatcher directly

     *    from a pattern string instead of than separately compiling the pattern and
     *    from a pattern string instead of than separately compiling the pattern and

-    *    then creating a RegexMatcher object from the pattern.</p>
+    *    then creating a RegexMatcher object from the pattern.

     *
     * @param regex The regular expression to be compiled.
     *
     * @param regex The regular expression to be compiled.

-    * @param flags The match mode flags to be used.
+    * @param flags The #URegexpFlag match mode flags to be used, e.g. #UREGEX_CASE_INSENSITIVE.

     * @param pe    Receives the position (line and column numbers) of any error
     *              within the regular expression.)
     * @param status   A reference to a UErrorCode to receive any errors.
     * @param pe    Receives the position (line and column numbers) of any error
     *              within the regular expression.)
     * @param status   A reference to a UErrorCode to receive any errors.


@@ -248,57 +234,55 @@ public:
         uint32_t             flags,
         UParseError          &pe,
         UErrorCode           &status);
         uint32_t             flags,
         UParseError          &pe,
         UErrorCode           &status);

-        
-        
+

    /**
     * Compiles the regular expression in string form into a RegexPattern
    /**
     * Compiles the regular expression in string form into a RegexPattern

-    * object using the specified match mode flags.  These compile methods,
+    * object using the specified #URegexpFlag match mode flags.  These compile methods,

     * rather than the constructors, are the usual way that RegexPattern objects
     * are created.
     *
     * rather than the constructors, are the usual way that RegexPattern objects
     * are created.
     *

-    * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+    * Note that RegexPattern objects must not be deleted while RegexMatcher

     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a
     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a

-    * catastrophic error.</p>
+    * catastrophic error.

     *
     *

-    * <p>Note that it is often more convenient to construct a RegexMatcher directly
+    * Note that it is often more convenient to construct a RegexMatcher directly

     *    from a pattern string instead of than separately compiling the pattern and
     *    from a pattern string instead of than separately compiling the pattern and

-    *    then creating a RegexMatcher object from the pattern.</p>
+    *    then creating a RegexMatcher object from the pattern.

     *
     * @param regex The regular expression to be compiled. Note, the text referred
     *              to by this UText must not be deleted during the lifetime of the
     *              RegexPattern object or any RegexMatcher object created from it.
     *
     * @param regex The regular expression to be compiled. Note, the text referred
     *              to by this UText must not be deleted during the lifetime of the
     *              RegexPattern object or any RegexMatcher object created from it.

-    * @param flags The match mode flags to be used.
+    * @param flags The #URegexpFlag match mode flags to be used, e.g. #UREGEX_CASE_INSENSITIVE.

     * @param pe    Receives the position (line and column numbers) of any error
     *              within the regular expression.)
     * @param status   A reference to a UErrorCode to receive any errors.
     * @return      A regexPattern object for the compiled pattern.
     *
     * @param pe    Receives the position (line and column numbers) of any error
     *              within the regular expression.)
     * @param status   A reference to a UErrorCode to receive any errors.
     * @return      A regexPattern object for the compiled pattern.
     *

-    * @draft ICU 4.6
+    * @stable ICU 4.6

     */
     static RegexPattern * U_EXPORT2 compile( UText *regex,
         uint32_t             flags,
         UParseError          &pe,
         UErrorCode           &status);
     */
     static RegexPattern * U_EXPORT2 compile( UText *regex,
         uint32_t             flags,
         UParseError          &pe,
         UErrorCode           &status);

-    

 
    /**
     * Compiles the regular expression in string form into a RegexPattern
 
    /**
     * Compiles the regular expression in string form into a RegexPattern

-    * object using the specified match mode flags.  These compile methods,
+    * object using the specified #URegexpFlag match mode flags.  These compile methods,

     * rather than the constructors, are the usual way that RegexPattern objects
     * are created.
     *
     * rather than the constructors, are the usual way that RegexPattern objects
     * are created.
     *

-    * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+    * Note that RegexPattern objects must not be deleted while RegexMatcher

     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a
     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a

-    * catastrophic error.</p>
+    * catastrophic error.

     *
     *

-    * <p>Note that it is often more convenient to construct a RegexMatcher directly
+    * Note that it is often more convenient to construct a RegexMatcher directly

     *    from a pattern string instead of than separately compiling the pattern and
     *    from a pattern string instead of than separately compiling the pattern and

-    *    then creating a RegexMatcher object from the pattern.</p>
+    *    then creating a RegexMatcher object from the pattern.

     *
     * @param regex The regular expression to be compiled.
     *
     * @param regex The regular expression to be compiled.

-    * @param flags The match mode flags to be used.
+    * @param flags The #URegexpFlag match mode flags to be used, e.g. #UREGEX_CASE_INSENSITIVE.

     * @param status   A reference to a UErrorCode to receive any errors.
     * @return      A regexPattern object for the compiled pattern.
     *
     * @param status   A reference to a UErrorCode to receive any errors.
     * @return      A regexPattern object for the compiled pattern.
     *


@@ -308,39 +292,37 @@ public:
         uint32_t             flags,
         UErrorCode           &status);
 
         uint32_t             flags,
         UErrorCode           &status);
 

-

    /**
     * Compiles the regular expression in string form into a RegexPattern
    /**
     * Compiles the regular expression in string form into a RegexPattern

-    * object using the specified match mode flags.  These compile methods,
+    * object using the specified #URegexpFlag match mode flags.  These compile methods,

     * rather than the constructors, are the usual way that RegexPattern objects
     * are created.
     *
     * rather than the constructors, are the usual way that RegexPattern objects
     * are created.
     *

-    * <p>Note that RegexPattern objects must not be deleted while RegexMatcher
+    * Note that RegexPattern objects must not be deleted while RegexMatcher

     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a
     * objects created from the pattern are active.  RegexMatchers keep a pointer
     * back to their pattern, so premature deletion of the pattern is a

-    * catastrophic error.</p>
+    * catastrophic error.

     *
     *

-    * <p>Note that it is often more convenient to construct a RegexMatcher directly
+    * Note that it is often more convenient to construct a RegexMatcher directly

     *    from a pattern string instead of than separately compiling the pattern and
     *    from a pattern string instead of than separately compiling the pattern and

-    *    then creating a RegexMatcher object from the pattern.</p>
+    *    then creating a RegexMatcher object from the pattern.

     *
     * @param regex The regular expression to be compiled. Note, the text referred
     *              to by this UText must not be deleted during the lifetime of the
     *              RegexPattern object or any RegexMatcher object created from it.
     *
     * @param regex The regular expression to be compiled. Note, the text referred
     *              to by this UText must not be deleted during the lifetime of the
     *              RegexPattern object or any RegexMatcher object created from it.

-    * @param flags The match mode flags to be used.
+    * @param flags The #URegexpFlag match mode flags to be used, e.g. #UREGEX_CASE_INSENSITIVE.

     * @param status   A reference to a UErrorCode to receive any errors.
     * @return      A regexPattern object for the compiled pattern.
     *
     * @param status   A reference to a UErrorCode to receive any errors.
     * @return      A regexPattern object for the compiled pattern.
     *

-    * @draft ICU 4.6
+    * @stable ICU 4.6

     */
     static RegexPattern * U_EXPORT2 compile( UText *regex,
         uint32_t             flags,
         UErrorCode           &status);
     */
     static RegexPattern * U_EXPORT2 compile( UText *regex,
         uint32_t             flags,
         UErrorCode           &status);

-    

 
    /**
 
    /**

-    * Get the match mode flags that were used when compiling this pattern.
-    * @return  the match mode flags
+    * Get the #URegexpFlag match mode flags that were used when compiling this pattern.
+    * @return  the #URegexpFlag match mode flags

     * @stable ICU 2.4
     */
     virtual uint32_t flags() const;
     * @stable ICU 2.4
     */
     virtual uint32_t flags() const;


@@ -350,7 +332,7 @@ public:
     * RegexMatcher can then be used to perform match, find or replace operations
     * on the input.  Note that a RegexPattern object must not be deleted while
     * RegexMatchers created from it still exist and might possibly be used again.
     * RegexMatcher can then be used to perform match, find or replace operations
     * on the input.  Note that a RegexPattern object must not be deleted while
     * RegexMatchers created from it still exist and might possibly be used again.

-    * <p>
+    *

     * The matcher will retain a reference to the supplied input string, and all regexp
     * pattern matching operations happen directly on this original string.  It is
     * critical that the string not be altered or deleted before use by the regular
     * The matcher will retain a reference to the supplied input string, and all regexp
     * pattern matching operations happen directly on this original string.  It is
     * critical that the string not be altered or deleted before use by the regular


@@ -365,51 +347,20 @@ public:
     virtual RegexMatcher *matcher(const UnicodeString &input,
         UErrorCode          &status) const;
         
     virtual RegexMatcher *matcher(const UnicodeString &input,
         UErrorCode          &status) const;
         

-        
-   /**
-    * Flag to disambiguate RegexPattern::matcher signature
-    * @draft ICU 4.6
-    */
-    enum PatternIsUTextFlag { PATTERN_IS_UTEXT };
-
-   /**
-    * Creates a RegexMatcher that will match the given input against this pattern.  The
-    * RegexMatcher can then be used to perform match, find or replace operations
-    * on the input.  Note that a RegexPattern object must not be deleted while
-    * RegexMatchers created from it still exist and might possibly be used again.
-    * <p>
-    * The matcher will make a shallow clone of the supplied input text, and all regexp
-    * pattern matching operations happen on this clone.  While read-only operations on
-    * the supplied text are permitted, it is critical that the underlying string not be
-    * altered or deleted before use by the regular expression operations is complete.
-    *
-    * @param input    The input text to which the regular expression will be applied.
-    * @param flag     Must be RegexPattern::PATTERN_IS_UTEXT; used to disambiguate
-    *                 method signature.
-    * @param status   A reference to a UErrorCode to receive any errors.
-    * @return         A RegexMatcher object for this pattern and input.
-    *
-    * @draft ICU 4.6
-    */
-    virtual RegexMatcher *matcher(UText *input,
-        PatternIsUTextFlag     flag, 
-        UErrorCode          &status) const;
-

 private:
     /**
 private:
     /**

-     * Cause a compilation error if an application accidently attempts to
-     *   create a matcher with a (UChar *) string as input rather than
+     * Cause a compilation error if an application accidentally attempts to
+     *   create a matcher with a (char16_t *) string as input rather than

      *   a UnicodeString.  Avoids a dangling reference to a temporary string.
      *   a UnicodeString.  Avoids a dangling reference to a temporary string.

-     * <p>
-     * To efficiently work with UChar *strings, wrap the data in a UnicodeString
+     *
+     * To efficiently work with char16_t *strings, wrap the data in a UnicodeString

      * using one of the aliasing constructors, such as
      * using one of the aliasing constructors, such as

-     * <code>UnicodeString(UBool isTerminated, const UChar *text, int32_t textLength);</code>
+     * `UnicodeString(UBool isTerminated, const char16_t *text, int32_t textLength);`

      * or in a UText, using
      * or in a UText, using

-     * <code>utext_openUChars(UText *ut, const UChar *text, int64_t textLength, UErrorCode *status);</code>
+     * `utext_openUChars(UText *ut, const char16_t *text, int64_t textLength, UErrorCode *status);`

      *
      *

-     * @internal

      */
      */

-    RegexMatcher *matcher(const UChar *input,
+    RegexMatcher *matcher(const char16_t *input,

         UErrorCode          &status) const;
 public:
 
         UErrorCode          &status) const;
 public:
 


@@ -430,7 +381,7 @@ public:
 
    /**
     * Test whether a string matches a regular expression.  This convenience function
 
    /**
     * Test whether a string matches a regular expression.  This convenience function

-    * both compiles the reguluar expression and applies it in a single operation.
+    * both compiles the regular expression and applies it in a single operation.

     * Note that if the same pattern needs to be applied repeatedly, this method will be
     * less efficient than creating and reusing a RegexMatcher object.
     *
     * Note that if the same pattern needs to be applied repeatedly, this method will be
     * less efficient than creating and reusing a RegexMatcher object.
     *


@@ -447,10 +398,9 @@ public:
               UParseError     &pe,
               UErrorCode      &status);
 
               UParseError     &pe,
               UErrorCode      &status);
 

-

    /**
     * Test whether a string matches a regular expression.  This convenience function
    /**
     * Test whether a string matches a regular expression.  This convenience function

-    * both compiles the reguluar expression and applies it in a single operation.
+    * both compiles the regular expression and applies it in a single operation.

     * Note that if the same pattern needs to be applied repeatedly, this method will be
     * less efficient than creating and reusing a RegexMatcher object.
     *
     * Note that if the same pattern needs to be applied repeatedly, this method will be
     * less efficient than creating and reusing a RegexMatcher object.
     *


@@ -460,14 +410,13 @@ public:
     * @param status A reference to a UErrorCode to receive any errors.
     * @return True if the regular expression exactly matches the full input string.
     *
     * @param status A reference to a UErrorCode to receive any errors.
     * @return True if the regular expression exactly matches the full input string.
     *

-    * @draft ICU 4.6
+    * @stable ICU 4.6

     */
     static UBool U_EXPORT2 matches(UText *regex,
         UText           *input,
         UParseError     &pe,
         UErrorCode      &status);
 
     */
     static UBool U_EXPORT2 matches(UText *regex,
         UText           *input,
         UParseError     &pe,
         UErrorCode      &status);
 

-

    /**
     * Returns the regular expression from which this pattern was compiled. This method will work
     * even if the pattern was compiled from a UText.
    /**
     * Returns the regular expression from which this pattern was compiled. This method will work
     * even if the pattern was compiled from a UText.


@@ -487,19 +436,67 @@ public:
     * UText, and that UText was modified, the returned UText may no longer reflect the RegexPattern
     * object.
     *
     * UText, and that UText was modified, the returned UText may no longer reflect the RegexPattern
     * object.
     *

-    * @draft ICU 4.6
+    * @stable ICU 4.6

     */
     virtual UText *patternText(UErrorCode      &status) const;
 
 
     /**
     */
     virtual UText *patternText(UErrorCode      &status) const;
 
 
     /**

-     * Split a string into fields.  Somewhat like split() from Perl.
-     * The pattern matches identify delimiters that separate the input
-     *  into fields.  The input data between the matches becomes the
-     *  fields themselves.
-     * <p>
-     *  For the best performance on split() operations,
-     *  <code>RegexMatcher::split</code> is perferable to this function
+     * Get the group number corresponding to a named capture group.
+     * The returned number can be used with any function that access
+     * capture groups by number.
+     *
+     * The function returns an error status if the specified name does not
+     * appear in the pattern.
+     *
+     * @param  groupName   The capture group name.
+     * @param  status      A UErrorCode to receive any errors.
+     *
+     * @stable ICU 55
+     */
+    virtual int32_t groupNumberFromName(const UnicodeString &groupName, UErrorCode &status) const;
+
+
+    /**
+     * Get the group number corresponding to a named capture group.
+     * The returned number can be used with any function that access
+     * capture groups by number.
+     *
+     * The function returns an error status if the specified name does not
+     * appear in the pattern.
+     *
+     * @param  groupName   The capture group name,
+     *                     platform invariant characters only.
+     * @param  nameLength  The length of the name, or -1 if the name is
+     *                     nul-terminated.
+     * @param  status      A UErrorCode to receive any errors.
+     *
+     * @stable ICU 55
+     */
+    virtual int32_t groupNumberFromName(const char *groupName, int32_t nameLength, UErrorCode &status) const;
+
+
+    /**
+     * Split a string into fields.  Somewhat like split() from Perl or Java.
+     * Pattern matches identify delimiters that separate the input
+     * into fields.  The input data between the delimiters becomes the
+     * fields themselves.
+     *
+     * If the delimiter pattern includes capture groups, the captured text will
+     * also appear in the destination array of output strings, interspersed
+     * with the fields.  This is similar to Perl, but differs from Java, 
+     * which ignores the presence of capture groups in the pattern.
+     * 
+     * Trailing empty fields will always be returned, assuming sufficient
+     * destination capacity.  This differs from the default behavior for Java
+     * and Perl where trailing empty fields are not returned.
+     *
+     * The number of strings produced by the split operation is returned.
+     * This count includes the strings from capture groups in the delimiter pattern.
+     * This behavior differs from Java, which ignores capture groups.
+     *
+     * For the best performance on split() operations,
+     * <code>RegexMatcher::split</code> is preferable to this function

      *
      * @param input   The string to be split into fields.  The field delimiters
      *                match the pattern (in the "this" object)
      *
      * @param input   The string to be split into fields.  The field delimiters
      *                match the pattern (in the "this" object)


@@ -524,13 +521,26 @@ public:
 
 
     /**
 
 
     /**

-     * Split a string into fields.  Somewhat like split() from Perl.
-     * The pattern matches identify delimiters that separate the input
-     *  into fields.  The input data between the matches becomes the
-     *  fields themselves.
-     * <p>
+     * Split a string into fields.  Somewhat like %split() from Perl or Java.
+     * Pattern matches identify delimiters that separate the input
+     * into fields.  The input data between the delimiters becomes the
+     * fields themselves.
+     *
+     * If the delimiter pattern includes capture groups, the captured text will
+     * also appear in the destination array of output strings, interspersed
+     * with the fields.  This is similar to Perl, but differs from Java, 
+     * which ignores the presence of capture groups in the pattern.
+     * 
+     * Trailing empty fields will always be returned, assuming sufficient
+     * destination capacity.  This differs from the default behavior for Java
+     * and Perl where trailing empty fields are not returned.
+     *
+     * The number of strings produced by the split operation is returned.
+     * This count includes the strings from capture groups in the delimiter pattern.
+     * This behavior differs from Java, which ignores capture groups.
+     *

      *  For the best performance on split() operations,
      *  For the best performance on split() operations,

-     *  <code>RegexMatcher::split</code> is perferable to this function
+     *  `RegexMatcher::split()` is preferable to this function

      *
      * @param input   The string to be split into fields.  The field delimiters
      *                match the pattern (in the "this" object)
      *
      * @param input   The string to be split into fields.  The field delimiters
      *                match the pattern (in the "this" object)


@@ -544,9 +554,9 @@ public:
      *                of fields, the trailing part of the input string, including any
      *                field delimiters, is placed in the last destination string.
      * @param status  A reference to a UErrorCode to receive any errors.
      *                of fields, the trailing part of the input string, including any
      *                field delimiters, is placed in the last destination string.
      * @param status  A reference to a UErrorCode to receive any errors.

-     * @return        The number of fields into which the input string was split.
+     * @return        The number of destination strings used.  

      *
      *

-     * @draft ICU 4.6
+     * @stable ICU 4.6

      */
     virtual int32_t  split(UText *input,
         UText            *dest[],
      */
     virtual int32_t  split(UText *input,
         UText            *dest[],


@@ -602,8 +612,6 @@ private:
     UVector32       *fGroupMap;    // Map from capture group number to position of
                                    //   the group's variables in the matcher stack frame.
 
     UVector32       *fGroupMap;    // Map from capture group number to position of
                                    //   the group's variables in the matcher stack frame.
 

-    int32_t         fMaxCaptureDigits;
-

     UnicodeSet     **fStaticSets;  // Ptr to static (shared) sets for predefined
                                    //   regex character classes, e.g. Word.
 
     UnicodeSet     **fStaticSets;  // Ptr to static (shared) sets for predefined
                                    //   regex character classes, e.g. Word.
 


@@ -618,6 +626,8 @@ private:
     Regex8BitSet   *fInitialChars8;
     UBool           fNeedsAltInput;
 
     Regex8BitSet   *fInitialChars8;
     UBool           fNeedsAltInput;
 

+    UHashtable     *fNamedCaptureMap;  // Map from capture group names to numbers.
+

     friend class RegexCompile;
     friend class RegexMatcher;
     friend class RegexCImpl;
     friend class RegexCompile;
     friend class RegexMatcher;
     friend class RegexCImpl;


@@ -625,19 +635,26 @@ private:
     //
     //  Implementation Methods
     //
     //
     //  Implementation Methods
     //

-    void        init();            // Common initialization, for use by constructors.
-    void        zap();             // Common cleanup
-#ifdef REGEX_DEBUG
+    void        init();                 // Common initialization, for use by constructors.
+    bool        initNamedCaptureMap();  // Lazy init for fNamedCaptureMap.
+    void        zap();                  // Common cleanup
+

     void        dumpOp(int32_t index) const;
     void        dumpOp(int32_t index) const;

-    friend     void U_EXPORT2 RegexPatternDump(const RegexPattern *);
-#endif

 
 

+  public:
+#ifndef U_HIDE_INTERNAL_API
+    /**
+      * Dump a compiled pattern. Internal debug function.
+      * @internal
+      */
+    void        dumpPattern() const;
+#endif  /* U_HIDE_INTERNAL_API */

 };
 
 
 
 /**
 };
 
 
 
 /**

- *  class RegexMatcher bundles together a reular expression pattern and
+ *  class RegexMatcher bundles together a regular expression pattern and

  *  input text to which the expression can be applied.  It includes methods
  *  for testing for matches, and for find and replace operations.
  *
  *  input text to which the expression can be applied.  It includes methods
  *  for testing for matches, and for find and replace operations.
  *


@@ -645,7 +662,7 @@ private:
  *
  * @stable ICU 2.4
  */
  *
  * @stable ICU 2.4
  */

-class U_I18N_API RegexMatcher: public UObject {
+class U_I18N_API RegexMatcher U_FINAL : public UObject {

 public:
 
     /**
 public:
 
     /**


@@ -657,8 +674,7 @@ public:
       * its matcher() method to create the RegexMatcher objects.
       *
       *  @param regexp The Regular Expression to be compiled.
       * its matcher() method to create the RegexMatcher objects.
       *
       *  @param regexp The Regular Expression to be compiled.

-      *  @param flags  Regular expression options, such as case insensitive matching.
-      *                @see UREGEX_CASE_INSENSITIVE
+      *  @param flags  #URegexpFlag options, such as #UREGEX_CASE_INSENSITIVE.

       *  @param status Any errors are reported by setting this UErrorCode variable.
       *  @stable ICU 2.6
       */
       *  @param status Any errors are reported by setting this UErrorCode variable.
       *  @stable ICU 2.6
       */


@@ -673,14 +689,13 @@ public:
       * its matcher() method to create the RegexMatcher objects.
       *
       *  @param regexp The regular expression to be compiled.
       * its matcher() method to create the RegexMatcher objects.
       *
       *  @param regexp The regular expression to be compiled.

-      *  @param flags  Regular expression options, such as case insensitive matching.
-      *                @see UREGEX_CASE_INSENSITIVE
+      *  @param flags  #URegexpFlag options, such as #UREGEX_CASE_INSENSITIVE.

       *  @param status Any errors are reported by setting this UErrorCode variable.
       *
       *  @param status Any errors are reported by setting this UErrorCode variable.
       *

-      *  @draft ICU 4.6
+      *  @stable ICU 4.6

       */
     RegexMatcher(UText *regexp, uint32_t flags, UErrorCode &status);
       */
     RegexMatcher(UText *regexp, uint32_t flags, UErrorCode &status);

-    
+

     /**
       * Construct a RegexMatcher for a regular expression.
       * This is a convenience method that avoids the need to explicitly create
     /**
       * Construct a RegexMatcher for a regular expression.
       * This is a convenience method that avoids the need to explicitly create


@@ -688,7 +703,7 @@ public:
       * created for the same expression, it will be more efficient to
       * separately create and cache a RegexPattern object, and use
       * its matcher() method to create the RegexMatcher objects.
       * created for the same expression, it will be more efficient to
       * separately create and cache a RegexPattern object, and use
       * its matcher() method to create the RegexMatcher objects.

-      * <p>
+      *

       * The matcher will retain a reference to the supplied input string, and all regexp
       * pattern matching operations happen directly on the original string.  It is
       * critical that the string not be altered or deleted before use by the regular
       * The matcher will retain a reference to the supplied input string, and all regexp
       * pattern matching operations happen directly on the original string.  It is
       * critical that the string not be altered or deleted before use by the regular


@@ -697,8 +712,7 @@ public:
       *  @param regexp The Regular Expression to be compiled.
       *  @param input  The string to match.  The matcher retains a reference to the
       *                caller's string; mo copy is made.
       *  @param regexp The Regular Expression to be compiled.
       *  @param input  The string to match.  The matcher retains a reference to the
       *                caller's string; mo copy is made.

-      *  @param flags  Regular expression options, such as case insensitive matching.
-      *                @see UREGEX_CASE_INSENSITIVE
+      *  @param flags  #URegexpFlag options, such as #UREGEX_CASE_INSENSITIVE.

       *  @param status Any errors are reported by setting this UErrorCode variable.
       *  @stable ICU 2.6
       */
       *  @param status Any errors are reported by setting this UErrorCode variable.
       *  @stable ICU 2.6
       */


@@ -712,7 +726,7 @@ public:
       * created for the same expression, it will be more efficient to
       * separately create and cache a RegexPattern object, and use
       * its matcher() method to create the RegexMatcher objects.
       * created for the same expression, it will be more efficient to
       * separately create and cache a RegexPattern object, and use
       * its matcher() method to create the RegexMatcher objects.

-      * <p>
+      *

       * The matcher will make a shallow clone of the supplied input text, and all regexp
       * pattern matching operations happen on this clone.  While read-only operations on
       * the supplied text are permitted, it is critical that the underlying string not be
       * The matcher will make a shallow clone of the supplied input text, and all regexp
       * pattern matching operations happen on this clone.  While read-only operations on
       * the supplied text are permitted, it is critical that the underlying string not be


@@ -720,30 +734,27 @@ public:
       *
       *  @param regexp The Regular Expression to be compiled.
       *  @param input  The string to match.  The matcher retains a shallow clone of the text.
       *
       *  @param regexp The Regular Expression to be compiled.
       *  @param input  The string to match.  The matcher retains a shallow clone of the text.

-      *  @param flags  Regular expression options, such as case insensitive matching.
-      *                @see UREGEX_CASE_INSENSITIVE
+      *  @param flags  #URegexpFlag options, such as #UREGEX_CASE_INSENSITIVE.

       *  @param status Any errors are reported by setting this UErrorCode variable.
       *
       *  @param status Any errors are reported by setting this UErrorCode variable.
       *

-      *  @draft ICU 4.6
+      *  @stable ICU 4.6

       */
     RegexMatcher(UText *regexp, UText *input,
         uint32_t flags, UErrorCode &status);
 
 private:
     /**
       */
     RegexMatcher(UText *regexp, UText *input,
         uint32_t flags, UErrorCode &status);
 
 private:
     /**

-     * Cause a compilation error if an application accidently attempts to
-     *   create a matcher with a (UChar *) string as input rather than
+     * Cause a compilation error if an application accidentally attempts to
+     *   create a matcher with a (char16_t *) string as input rather than

      *   a UnicodeString.    Avoids a dangling reference to a temporary string.
      *   a UnicodeString.    Avoids a dangling reference to a temporary string.

-     * <p>
-     * To efficiently work with UChar *strings, wrap the data in a UnicodeString
+     *
+     * To efficiently work with char16_t *strings, wrap the data in a UnicodeString

      * using one of the aliasing constructors, such as
      * using one of the aliasing constructors, such as

-     * <code>UnicodeString(UBool isTerminated, const UChar *text, int32_t textLength);</code>
+     * `UnicodeString(UBool isTerminated, const char16_t *text, int32_t textLength);`

      * or in a UText, using
      * or in a UText, using

-     * <code>utext_openUChars(UText *ut, const UChar *text, int64_t textLength, UErrorCode *status);</code>
-     *
-     * @internal
+     * `utext_openUChars(UText *ut, const char16_t *text, int64_t textLength, UErrorCode *status);`

      */
      */

-    RegexMatcher(const UnicodeString &regexp, const UChar *input,
+    RegexMatcher(const UnicodeString &regexp, const char16_t *input,

         uint32_t flags, UErrorCode &status);
 public:
 
         uint32_t flags, UErrorCode &status);
 public:
 


@@ -784,8 +795,8 @@ public:
     *   always starts at the beginning of the input region;
     *   unlike that function, it does not require that the entire region be matched.
     *
     *   always starts at the beginning of the input region;
     *   unlike that function, it does not require that the entire region be matched.
     *

-    *   <p>If the match succeeds then more information can be obtained via the <code>start()</code>,
-    *     <code>end()</code>, and <code>group()</code> functions.</p>
+    *   If the match succeeds then more information can be obtained via the start(),
+    *   end(), and group() functions.

     *
     *    @param   status     A reference to a UErrorCode to receive any errors.
     *    @return  TRUE if there is a match at the start of the input string.
     *
     *    @param   status     A reference to a UErrorCode to receive any errors.
     *    @return  TRUE if there is a match at the start of the input string.


@@ -799,8 +810,8 @@ public:
     *   The match may be of any length, and is not required to extend to the end
     *   of the input string.  Contrast with match().
     *
     *   The match may be of any length, and is not required to extend to the end
     *   of the input string.  Contrast with match().
     *

-    *   <p>If the match succeeds then more information can be obtained via the <code>start()</code>,
-    *     <code>end()</code>, and <code>group()</code> functions.</p>
+    *   If the match succeeds then more information can be obtained via the start(),
+    *   end(), and group() functions.

     *
     *    @param   startIndex The input string (native) index at which to begin matching.
     *    @param   status     A reference to a UErrorCode to receive any errors.
     *
     *    @param   startIndex The input string (native) index at which to begin matching.
     *    @param   status     A reference to a UErrorCode to receive any errors.


@@ -814,17 +825,33 @@ public:
     *  Find the next pattern match in the input string.
     *  The find begins searching the input at the location following the end of
     *  the previous match, or at the start of the string if there is no previous match.
     *  Find the next pattern match in the input string.
     *  The find begins searching the input at the location following the end of
     *  the previous match, or at the start of the string if there is no previous match.

-    *  If a match is found, <code>start(), end()</code> and <code>group()</code>
+    *  If a match is found, `start()`, `end()` and `group()`

     *  will provide more information regarding the match.
     *  will provide more information regarding the match.

-    *  <p>Note that if the input string is changed by the application,
+    *  Note that if the input string is changed by the application,

     *     use find(startPos, status) instead of find(), because the saved starting
     *     use find(startPos, status) instead of find(), because the saved starting

-    *     position may not be valid with the altered input string.</p>
+    *     position may not be valid with the altered input string.

     *  @return  TRUE if a match is found.
     *  @stable ICU 2.4
     */
     virtual UBool find();
 
 
     *  @return  TRUE if a match is found.
     *  @stable ICU 2.4
     */
     virtual UBool find();
 
 

+   /**
+    *  Find the next pattern match in the input string.
+    *  The find begins searching the input at the location following the end of
+    *  the previous match, or at the start of the string if there is no previous match.
+    *  If a match is found, `start()`, `end()` and `group()`
+    *  will provide more information regarding the match.
+    *
+    *  Note that if the input string is changed by the application,
+    *  use find(startPos, status) instead of find(), because the saved starting
+    *  position may not be valid with the altered input string.
+    *  @param   status  A reference to a UErrorCode to receive any errors.
+    *  @return  TRUE if a match is found.
+    * @stable ICU 55
+    */
+    virtual UBool find(UErrorCode &status);
+

    /**
     *   Resets this RegexMatcher and then attempts to find the next substring of the
     *   input string that matches the pattern, starting at the specified index.
    /**
     *   Resets this RegexMatcher and then attempts to find the next substring of the
     *   input string that matches the pattern, starting at the specified index.


@@ -853,6 +880,11 @@ public:
     *    Returns a string containing the text captured by the given group
     *    during the previous match operation.  Group(0) is the entire match.
     *
     *    Returns a string containing the text captured by the given group
     *    during the previous match operation.  Group(0) is the entire match.
     *

+    *    A zero length string is returned both for capture groups that did not
+    *    participate in the match and for actual zero length matches.
+    *    To distinguish between these two cases use the function start(),
+    *    which returns -1 for non-participating groups.
+    *

     *    @param groupNum the capture group number
     *    @param   status     A reference to a UErrorCode to receive any errors.
     *                        Possible errors are  U_REGEX_INVALID_STATE if no match
     *    @param groupNum the capture group number
     *    @param   status     A reference to a UErrorCode to receive any errors.
     *                        Possible errors are  U_REGEX_INVALID_STATE if no match


@@ -863,7 +895,6 @@ public:
     */
     virtual UnicodeString group(int32_t groupNum, UErrorCode &status) const;
 
     */
     virtual UnicodeString group(int32_t groupNum, UErrorCode &status) const;
 

-

    /**
     *   Returns the number of capturing groups in this matcher's pattern.
     *   @return the number of capture groups
    /**
     *   Returns the number of capturing groups in this matcher's pattern.
     *   @return the number of capture groups


@@ -875,8 +906,8 @@ public:
    /**
     *   Returns a shallow clone of the entire live input string with the UText current native index
     *   set to the beginning of the requested group.
    /**
     *   Returns a shallow clone of the entire live input string with the UText current native index
     *   set to the beginning of the requested group.

-    *   Note that copying the entire input string may cause significant performance and memory issues.
-    *   @param   dest        The UText into which the input should be copied, or NULL to create a new UText
+    *
+    *   @param   dest        The UText into which the input should be cloned, or NULL to create a new UText

     *   @param   group_len   A reference to receive the length of the desired capture group
     *   @param   status      A reference to a UErrorCode to receive any errors.
     *                        Possible errors are  U_REGEX_INVALID_STATE if no match
     *   @param   group_len   A reference to receive the length of the desired capture group
     *   @param   status      A reference to a UErrorCode to receive any errors.
     *                        Possible errors are  U_REGEX_INVALID_STATE if no match


@@ -884,32 +915,31 @@ public:
     *                        U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number.
     *   @return dest if non-NULL, a shallow copy of the input text otherwise
     *
     *                        U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number.
     *   @return dest if non-NULL, a shallow copy of the input text otherwise
     *

-    *   @draft ICU 4.6
+    *   @stable ICU 4.6

     */
     virtual UText *group(UText *dest, int64_t &group_len, UErrorCode &status) const; 
 
    /**
     */
     virtual UText *group(UText *dest, int64_t &group_len, UErrorCode &status) const; 
 
    /**

-    * @draft ICU 4.6
-    */
-    virtual UText *group(int32_t groupNum, UText *dest, int64_t &group_len, UErrorCode &status) const;
-
-   /**
-    *   Returns a string containing the text captured by the given group
-    *   during the previous match operation.  Group(0) is the entire match.
+    *   Returns a shallow clone of the entire live input string with the UText current native index
+    *   set to the beginning of the requested group.

     *
     *

-    *   @param   groupNum    the capture group number
-    *   @param   dest        A mutable UText in which the matching text is placed.
-    *                        If NULL, a new UText will be created (which may not be mutable).
+    *   A group length of zero is returned both for capture groups that did not
+    *   participate in the match and for actual zero length matches.
+    *   To distinguish between these two cases use the function start(),
+    *   which returns -1 for non-participating groups.
+    *
+    *   @param   groupNum   The capture group number.
+    *   @param   dest        The UText into which the input should be cloned, or NULL to create a new UText.
+    *   @param   group_len   A reference to receive the length of the desired capture group

     *   @param   status      A reference to a UErrorCode to receive any errors.
     *                        Possible errors are  U_REGEX_INVALID_STATE if no match
     *   @param   status      A reference to a UErrorCode to receive any errors.
     *                        Possible errors are  U_REGEX_INVALID_STATE if no match

-    *                        has been attempted or the last match failed.
-    *   @return  A string containing the matched input text. If a pre-allocated UText
-    *            was provided, it will always be used and returned.
+    *                        has been attempted or the last match failed and
+    *                        U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number.
+    *   @return dest if non-NULL, a shallow copy of the input text otherwise

     *
     *

-    *   @internal ICU 4.4 technology preview
+    *   @stable ICU 4.6

     */
     */

-    virtual UText *group(int32_t groupNum, UText *dest, UErrorCode &status) const;
-
+    virtual UText *group(int32_t groupNum, UText *dest, int64_t &group_len, UErrorCode &status) const;

 
    /**
     *   Returns the index in the input string of the start of the text matched
 
    /**
     *   Returns the index in the input string of the start of the text matched


@@ -921,7 +951,11 @@ public:
     virtual int32_t start(UErrorCode &status) const;
 
    /**
     virtual int32_t start(UErrorCode &status) const;
 
    /**

-    *   @draft ICU 4.6
+    *   Returns the index in the input string of the start of the text matched
+    *   during the previous match operation.
+    *    @param   status      a reference to a UErrorCode to receive any errors.
+    *    @return              The (native) position in the input string of the start of the last match.
+    *   @stable ICU 4.6

     */
     virtual int64_t start64(UErrorCode &status) const;
 
     */
     virtual int64_t start64(UErrorCode &status) const;
 


@@ -942,27 +976,47 @@ public:
     virtual int32_t start(int32_t group, UErrorCode &status) const;
 
    /**
     virtual int32_t start(int32_t group, UErrorCode &status) const;
 
    /**

-    *   @draft ICU 4.6
+    *   Returns the index in the input string of the start of the text matched by the
+    *    specified capture group during the previous match operation.  Return -1 if
+    *    the capture group exists in the pattern, but was not part of the last match.
+    *
+    *    @param  group       the capture group number.
+    *    @param  status      A reference to a UErrorCode to receive any errors.  Possible
+    *                        errors are  U_REGEX_INVALID_STATE if no match has been
+    *                        attempted or the last match failed, and
+    *                        U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number.
+    *    @return the (native) start position of substring matched by the specified group.
+    *    @stable ICU 4.6

     */
     virtual int64_t start64(int32_t group, UErrorCode &status) const;
 
     */
     virtual int64_t start64(int32_t group, UErrorCode &status) const;
 

-

    /**
     *    Returns the index in the input string of the first character following the
     *    text matched during the previous match operation.
    /**
     *    Returns the index in the input string of the first character following the
     *    text matched during the previous match operation.

+    *

     *   @param   status      A reference to a UErrorCode to receive any errors.  Possible
     *                        errors are  U_REGEX_INVALID_STATE if no match has been
     *                        attempted or the last match failed.
     *    @return the index of the last character matched, plus one.
     *                        The index value returned is a native index, corresponding to
     *                        code units for the underlying encoding type, for example,
     *   @param   status      A reference to a UErrorCode to receive any errors.  Possible
     *                        errors are  U_REGEX_INVALID_STATE if no match has been
     *                        attempted or the last match failed.
     *    @return the index of the last character matched, plus one.
     *                        The index value returned is a native index, corresponding to
     *                        code units for the underlying encoding type, for example,

-    *                        a byte index for UTF8.
+    *                        a byte index for UTF-8.

     *   @stable ICU 2.4
     */
     virtual int32_t end(UErrorCode &status) const;
 
    /**
     *   @stable ICU 2.4
     */
     virtual int32_t end(UErrorCode &status) const;
 
    /**

-    *   @draft ICU 4.6
+    *    Returns the index in the input string of the first character following the
+    *    text matched during the previous match operation.
+    *
+    *   @param   status      A reference to a UErrorCode to receive any errors.  Possible
+    *                        errors are  U_REGEX_INVALID_STATE if no match has been
+    *                        attempted or the last match failed.
+    *    @return the index of the last character matched, plus one.
+    *                        The index value returned is a native index, corresponding to
+    *                        code units for the underlying encoding type, for example,
+    *                        a byte index for UTF-8.
+    *   @stable ICU 4.6

     */
     virtual int64_t end64(UErrorCode &status) const;
 
     */
     virtual int64_t end64(UErrorCode &status) const;
 


@@ -970,13 +1024,14 @@ public:
    /**
     *    Returns the index in the input string of the character following the
     *    text matched by the specified capture group during the previous match operation.
    /**
     *    Returns the index in the input string of the character following the
     *    text matched by the specified capture group during the previous match operation.

+    *

     *    @param group  the capture group number
     *    @param   status      A reference to a UErrorCode to receive any errors.  Possible
     *                        errors are  U_REGEX_INVALID_STATE if no match has been
     *                        attempted or the last match failed and
     *                        U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number
     *    @return  the index of the first character following the text
     *    @param group  the capture group number
     *    @param   status      A reference to a UErrorCode to receive any errors.  Possible
     *                        errors are  U_REGEX_INVALID_STATE if no match has been
     *                        attempted or the last match failed and
     *                        U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number
     *    @return  the index of the first character following the text

-    *              captured by the specifed group during the previous match operation.
+    *              captured by the specified group during the previous match operation.

     *              Return -1 if the capture group exists in the pattern but was not part of the match.
     *              The index value returned is a native index, corresponding to
     *              code units for the underlying encoding type, for example,
     *              Return -1 if the capture group exists in the pattern but was not part of the match.
     *              The index value returned is a native index, corresponding to
     *              code units for the underlying encoding type, for example,


@@ -986,11 +1041,24 @@ public:
     virtual int32_t end(int32_t group, UErrorCode &status) const;
 
    /**
     virtual int32_t end(int32_t group, UErrorCode &status) const;
 
    /**

-    *   @draft ICU 4.6
+    *    Returns the index in the input string of the character following the
+    *    text matched by the specified capture group during the previous match operation.
+    *
+    *    @param group  the capture group number
+    *    @param   status      A reference to a UErrorCode to receive any errors.  Possible
+    *                        errors are  U_REGEX_INVALID_STATE if no match has been
+    *                        attempted or the last match failed and
+    *                        U_INDEX_OUTOFBOUNDS_ERROR for a bad capture group number
+    *    @return  the index of the first character following the text
+    *              captured by the specified group during the previous match operation.
+    *              Return -1 if the capture group exists in the pattern but was not part of the match.
+    *              The index value returned is a native index, corresponding to
+    *              code units for the underlying encoding type, for example,
+    *              a byte index for UTF8.
+    *   @stable ICU 4.6

     */
     virtual int64_t end64(int32_t group, UErrorCode &status) const;
 
     */
     virtual int64_t end64(int32_t group, UErrorCode &status) const;
 

-

    /**
     *   Resets this matcher.  The effect is to remove any memory of previous matches,
     *       and to cause subsequent find() operations to begin at the beginning of
    /**
     *   Resets this matcher.  The effect is to remove any memory of previous matches,
     *       and to cause subsequent find() operations to begin at the beginning of


@@ -1007,10 +1075,10 @@ public:
     *   The effect is to remove any memory of previous matches,
     *       and to cause subsequent find() operations to begin at
     *       the specified (native) position in the input string.
     *   The effect is to remove any memory of previous matches,
     *       and to cause subsequent find() operations to begin at
     *       the specified (native) position in the input string.

-    * <p>
+    *

     *   The matcher's region is reset to its default, which is the entire
     *   input string.
     *   The matcher's region is reset to its default, which is the entire
     *   input string.

-    * <p>
+    *

     *   An alternative to this function is to set a match region
     *   beginning at the desired index.
     *
     *   An alternative to this function is to set a match region
     *   beginning at the desired index.
     *


@@ -1051,25 +1119,51 @@ public:
     *                until after regexp operations on it are done.
     *   @return this RegexMatcher.
     *
     *                until after regexp operations on it are done.
     *   @return this RegexMatcher.
     *

-    *   @draft ICU 4.6
+    *   @stable ICU 4.6

     */
     virtual RegexMatcher &reset(UText *input);
 
     */
     virtual RegexMatcher &reset(UText *input);
 

+
+  /**
+    *  Set the subject text string upon which the regular expression is looking for matches
+    *  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 regular expression implementation never copies the underlying text
+    *  of a string being matched, 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.
+    *
+    * @stable ICU 4.8 
+    */
+    virtual RegexMatcher &refreshInputText(UText *input, UErrorCode &status);
+

 private:
     /**
 private:
     /**

-     * Cause a compilation error if an application accidently attempts to
-     *   reset a matcher with a (UChar *) string as input rather than
+     * Cause a compilation error if an application accidentally attempts to
+     *   reset a matcher with a (char16_t *) string as input rather than

      *   a UnicodeString.    Avoids a dangling reference to a temporary string.
      *   a UnicodeString.    Avoids a dangling reference to a temporary string.

-     * <p>
-     * To efficiently work with UChar *strings, wrap the data in a UnicodeString
+     *
+     * To efficiently work with char16_t *strings, wrap the data in a UnicodeString

      * using one of the aliasing constructors, such as
      * using one of the aliasing constructors, such as

-     * <code>UnicodeString(UBool isTerminated, const UChar *text, int32_t textLength);</code>
+     * `UnicodeString(UBool isTerminated, const char16_t *text, int32_t textLength);`

      * or in a UText, using
      * or in a UText, using

-     * <code>utext_openUChars(UText *ut, const UChar *text, int64_t textLength, UErrorCode *status);</code>
+     * `utext_openUChars(UText *ut, const char16_t *text, int64_t textLength, UErrorCode *status);`

      *
      *

-     * @internal

      */
      */

-    RegexMatcher &reset(const UChar *input);
+    RegexMatcher &reset(const char16_t *input);

 public:
 
    /**
 public:
 
    /**


@@ -1087,7 +1181,7 @@ public:
     *   a UnicodeString.
     *   @return the input text
     *
     *   a UnicodeString.
     *   @return the input text
     *

-    *   @draft ICU 4.6
+    *   @stable ICU 4.6

     */
     virtual UText *inputText() const;
     
     */
     virtual UText *inputText() const;
     


@@ -1096,9 +1190,10 @@ public:
     *   UText parameter or by returning a shallow clone of the live input. Note that copying
     *   the entire input may cause significant performance and memory issues.
     *   @param dest The UText into which the input should be copied, or NULL to create a new UText
     *   UText parameter or by returning a shallow clone of the live input. Note that copying
     *   the entire input may cause significant performance and memory issues.
     *   @param dest The UText into which the input should be copied, or NULL to create a new UText

+    *   @param status error code

     *   @return dest if non-NULL, a shallow copy of the input text otherwise
     *
     *   @return dest if non-NULL, a shallow copy of the input text otherwise
     *

-    *   @draft ICU 4.6
+    *   @stable ICU 4.6

     */
     virtual UText *getInput(UText *dest, UErrorCode &status) const;
     
     */
     virtual UText *getInput(UText *dest, UErrorCode &status) const;
     


@@ -1126,11 +1221,13 @@ public:
    /** 
      * Identical to region(start, limit, status) but also allows a start position without
      *  resetting the region state.
    /** 
      * Identical to region(start, limit, status) but also allows a start position without
      *  resetting the region state.

+     * @param regionStart The region start
+     * @param regionLimit the limit of the region

      * @param startIndex  The (native) index within the region bounds at which to begin searches.
      * @param status A reference to a UErrorCode to receive any errors.
      *                If startIndex is not within the specified region bounds, 
      *                U_INDEX_OUTOFBOUNDS_ERROR is returned.
      * @param startIndex  The (native) index within the region bounds at which to begin searches.
      * @param status A reference to a UErrorCode to receive any errors.
      *                If startIndex is not within the specified region bounds, 
      *                U_INDEX_OUTOFBOUNDS_ERROR is returned.

-     * @draft ICU 4.6
+     * @stable ICU 4.6

      */
      virtual RegexMatcher &region(int64_t regionStart, int64_t regionLimit, int64_t startIndex, UErrorCode &status);
 
      */
      virtual RegexMatcher &region(int64_t regionStart, int64_t regionLimit, int64_t startIndex, UErrorCode &status);
 


@@ -1145,8 +1242,13 @@ public:
      virtual int32_t regionStart() const;
 
    /**
      virtual int32_t regionStart() const;
 
    /**

-    *   @draft ICU 4.6
-    */
+     * Reports the start index of this matcher's region. The searches this matcher
+     * conducts are limited to finding matches within regionStart (inclusive) and
+     * regionEnd (exclusive).
+     *
+     * @return The starting (native) index of this matcher's region.
+     * @stable ICU 4.6
+     */

      virtual int64_t regionStart64() const;
 
 
      virtual int64_t regionStart64() const;
 
 


@@ -1161,8 +1263,13 @@ public:
       virtual int32_t regionEnd() const;
 
    /**
       virtual int32_t regionEnd() const;
 
    /**

-    *   @draft ICU 4.6
-    */
+     * Reports the end (limit) index (exclusive) of this matcher's region. The searches
+     * this matcher conducts are limited to finding matches within regionStart
+     * (inclusive) and regionEnd (exclusive).
+     *
+     * @return The ending point (native) of this matcher's region.
+     * @stable ICU 4.6
+     */

       virtual int64_t regionEnd64() const;
 
     /**
       virtual int64_t regionEnd64() const;
 
     /**


@@ -1198,7 +1305,7 @@ public:
      
     /**
       * Return true if this matcher is using anchoring bounds.
      
     /**
       * Return true if this matcher is using anchoring bounds.

-      * By default, matchers use anchoring region boounds.
+      * By default, matchers use anchoring region bounds.

       *
       * @return TRUE if this matcher is using anchoring bounds.
       * @stable ICU 4.0
       *
       * @return TRUE if this matcher is using anchoring bounds.
       * @stable ICU 4.0


@@ -1222,9 +1329,9 @@ public:
 
 
     /**
 
 
     /**

-      * Return TRUE if the most recent matching operation touched the
-      *  end of the text being processed.  In this case, additional input text could
-      *  change the results of that match.
+      * Return TRUE if the most recent matching operation attempted to access
+      *  additional input beyond the available input text.
+      *  In this case, additional input text could change the results of the match.

       *
       *  hitEnd() is defined for both successful and unsuccessful matches.
       *  In either case hitEnd() will return TRUE if if the end of the text was
       *
       *  hitEnd() is defined for both successful and unsuccessful matches.
       *  In either case hitEnd() will return TRUE if if the end of the text was


@@ -1292,7 +1399,7 @@ public:
     *    @return              a string containing the results of the find and replace.
     *                          If a pre-allocated UText was provided, it will always be used and returned.
     *
     *    @return              a string containing the results of the find and replace.
     *                          If a pre-allocated UText was provided, it will always be used and returned.
     *

-    *    @draft ICU 4.6
+    *    @stable ICU 4.6

     */
     virtual UText *replaceAll(UText *replacement, UText *dest, UErrorCode &status);
     
     */
     virtual UText *replaceAll(UText *replacement, UText *dest, UErrorCode &status);
     


@@ -1302,15 +1409,15 @@ public:
     * the pattern with the replacement string.   This is a convenience
     * function that provides a complete find-and-replace operation.
     *
     * the pattern with the replacement string.   This is a convenience
     * function that provides a complete find-and-replace operation.
     *

-    * <p>This function first resets this RegexMatcher. It then scans the input string
+    * This function first resets this RegexMatcher. It then scans the input string

     * looking for a match of the pattern. Input that is not part
     * of the match is appended directly to the result string; the match is replaced
     * in the result by the replacement string. The replacement string may contain
     * looking for a match of the pattern. Input that is not part
     * of the match is appended directly to the result string; the match is replaced
     * in the result by the replacement string. The replacement string may contain

-    * references to captured groups.</p>
+    * references to captured groups.

     *
     *

-    * <p>The state of the matcher (the position at which a subsequent find()
+    * The state of the matcher (the position at which a subsequent find()

     *    would begin) after completing a replaceFirst() is not specified.  The
     *    would begin) after completing a replaceFirst() is not specified.  The

-    *    RegexMatcher should be reset before doing additional find() operations.</p>
+    *    RegexMatcher should be reset before doing additional find() operations.

     *
     *    @param   replacement a string containing the replacement text.
     *    @param   status      a reference to a UErrorCode to receive any errors.
     *
     *    @param   replacement a string containing the replacement text.
     *    @param   status      a reference to a UErrorCode to receive any errors.


@@ -1325,15 +1432,15 @@ public:
     * the pattern with the replacement string.   This is a convenience
     * function that provides a complete find-and-replace operation.
     *
     * the pattern with the replacement string.   This is a convenience
     * function that provides a complete find-and-replace operation.
     *

-    * <p>This function first resets this RegexMatcher. It then scans the input string
+    * This function first resets this RegexMatcher. It then scans the input string

     * looking for a match of the pattern. Input that is not part
     * of the match is appended directly to the result string; the match is replaced
     * in the result by the replacement string. The replacement string may contain
     * looking for a match of the pattern. Input that is not part
     * of the match is appended directly to the result string; the match is replaced
     * in the result by the replacement string. The replacement string may contain

-    * references to captured groups.</p>
+    * references to captured groups.

     *
     *

-    * <p>The state of the matcher (the position at which a subsequent find()
+    * The state of the matcher (the position at which a subsequent find()

     *    would begin) after completing a replaceFirst() is not specified.  The
     *    would begin) after completing a replaceFirst() is not specified.  The

-    *    RegexMatcher should be reset before doing additional find() operations.</p>
+    *    RegexMatcher should be reset before doing additional find() operations.

     *
     *    @param   replacement a string containing the replacement text.
     *    @param   dest        a mutable UText in which the results are placed.
     *
     *    @param   replacement a string containing the replacement text.
     *    @param   dest        a mutable UText in which the results are placed.


@@ -1342,7 +1449,7 @@ public:
     *    @return              a string containing the results of the find and replace.
     *                          If a pre-allocated UText was provided, it will always be used and returned.
     *
     *    @return              a string containing the results of the find and replace.
     *                          If a pre-allocated UText was provided, it will always be used and returned.
     *

-    *    @draft ICU 4.6
+    *    @stable ICU 4.6

     */
     virtual UText *replaceFirst(UText *replacement, UText *dest, UErrorCode &status);
     
     */
     virtual UText *replaceFirst(UText *replacement, UText *dest, UErrorCode &status);
     


@@ -1351,13 +1458,13 @@ public:
     *   Implements a replace operation intended to be used as part of an
     *   incremental find-and-replace.
     *
     *   Implements a replace operation intended to be used as part of an
     *   incremental find-and-replace.
     *

-    *   <p>The input string, starting from the end of the previous replacement and ending at
+    *   The input string, starting from the end of the previous replacement and ending at

     *   the start of the current match, is appended to the destination string.  Then the
     *   replacement string is appended to the output string,
     *   the start of the current match, is appended to the destination string.  Then the
     *   replacement string is appended to the output string,

-    *   including handling any substitutions of captured text.</p>
+    *   including handling any substitutions of captured text.

     *
     *

-    *   <p>For simple, prepackaged, non-incremental find-and-replace
-    *   operations, see replaceFirst() or replaceAll().</p>
+    *   For simple, prepackaged, non-incremental find-and-replace
+    *   operations, see replaceFirst() or replaceAll().

     *
     *   @param   dest        A UnicodeString to which the results of the find-and-replace are appended.
     *   @param   replacement A UnicodeString that provides the text to be substituted for
     *
     *   @param   dest        A UnicodeString to which the results of the find-and-replace are appended.
     *   @param   replacement A UnicodeString that provides the text to be substituted for


@@ -1382,13 +1489,13 @@ public:
     *   Implements a replace operation intended to be used as part of an
     *   incremental find-and-replace.
     *
     *   Implements a replace operation intended to be used as part of an
     *   incremental find-and-replace.
     *

-    *   <p>The input string, starting from the end of the previous replacement and ending at
+    *   The input string, starting from the end of the previous replacement and ending at

     *   the start of the current match, is appended to the destination string.  Then the
     *   replacement string is appended to the output string,
     *   the start of the current match, is appended to the destination string.  Then the
     *   replacement string is appended to the output string,

-    *   including handling any substitutions of captured text.</p>
+    *   including handling any substitutions of captured text.

     *
     *

-    *   <p>For simple, prepackaged, non-incremental find-and-replace
-    *   operations, see replaceFirst() or replaceAll().</p>
+    *   For simple, prepackaged, non-incremental find-and-replace
+    *   operations, see replaceFirst() or replaceAll().

     *
     *   @param   dest        A mutable UText to which the results of the find-and-replace are appended.
     *                         Must not be NULL.
     *
     *   @param   dest        A mutable UText to which the results of the find-and-replace are appended.
     *                         Must not be NULL.


@@ -1403,7 +1510,7 @@ public:
     *
     *   @return  this  RegexMatcher
     *
     *
     *   @return  this  RegexMatcher
     *

-    *   @draft ICU 4.6
+    *   @stable ICU 4.6

     */
     virtual RegexMatcher &appendReplacement(UText *dest,
         UText *replacement, UErrorCode &status);
     */
     virtual RegexMatcher &appendReplacement(UText *dest,
         UText *replacement, UErrorCode &status);


@@ -1412,8 +1519,8 @@ public:
    /**
     * As the final step in a find-and-replace operation, append the remainder
     * of the input string, starting at the position following the last appendReplacement(),
    /**
     * As the final step in a find-and-replace operation, append the remainder
     * of the input string, starting at the position following the last appendReplacement(),

-    * to the destination string. <code>appendTail()</code> is intended to be invoked after one
-    * or more invocations of the <code>RegexMatcher::appendReplacement()</code>.
+    * to the destination string. `appendTail()` is intended to be invoked after one
+    * or more invocations of the `RegexMatcher::appendReplacement()`.

     *
     *  @param dest A UnicodeString to which the results of the find-and-replace are appended.
     *  @return  the destination string.
     *
     *  @param dest A UnicodeString to which the results of the find-and-replace are appended.
     *  @return  the destination string.


@@ -1425,20 +1532,21 @@ public:
    /**
     * As the final step in a find-and-replace operation, append the remainder
     * of the input string, starting at the position following the last appendReplacement(),
    /**
     * As the final step in a find-and-replace operation, append the remainder
     * of the input string, starting at the position following the last appendReplacement(),

-    * to the destination string. <code>appendTail()</code> is intended to be invoked after one
-    * or more invocations of the <code>RegexMatcher::appendReplacement()</code>.
+    * to the destination string. `appendTail()` is intended to be invoked after one
+    * or more invocations of the `RegexMatcher::appendReplacement()`.

     *
     *  @param dest A mutable UText to which the results of the find-and-replace are appended.
     *               Must not be NULL.
     *
     *  @param dest A mutable UText to which the results of the find-and-replace are appended.
     *               Must not be NULL.

+    *  @param status error cod

     *  @return  the destination string.
     *
     *  @return  the destination string.
     *

-    *  @draft ICU 4.6
+    *  @stable ICU 4.6

     */
     virtual UText *appendTail(UText *dest, UErrorCode &status);
 
 
     /**
     */
     virtual UText *appendTail(UText *dest, UErrorCode &status);
 
 
     /**

-     * Split a string into fields.  Somewhat like split() from Perl.
+     * Split a string into fields.  Somewhat like %split() from Perl.

      * The pattern matches identify delimiters that separate the input
      *  into fields.  The input data between the matches becomes the
      *  fields themselves.
      * The pattern matches identify delimiters that separate the input
      *  into fields.  The input data between the matches becomes the
      *  fields themselves.


@@ -1467,7 +1575,7 @@ public:
 
 
     /**
 
 
     /**

-     * Split a string into fields.  Somewhat like split() from Perl.
+     * Split a string into fields.  Somewhat like %split() from Perl.

      * The pattern matches identify delimiters that separate the input
      *  into fields.  The input data between the matches becomes the
      *  fields themselves.
      * The pattern matches identify delimiters that separate the input
      *  into fields.  The input data between the matches becomes the
      *  fields themselves.


@@ -1487,7 +1595,7 @@ public:
      * @param status  A reference to a UErrorCode to receive any errors.
      * @return        The number of fields into which the input string was split.
      *
      * @param status  A reference to a UErrorCode to receive any errors.
      * @return        The number of fields into which the input string was split.
      *

-     * @draft ICU 4.6
+     * @stable ICU 4.6

      */
     virtual int32_t  split(UText *input,
         UText           *dest[],
      */
     virtual int32_t  split(UText *input,
         UText           *dest[],


@@ -1502,14 +1610,14 @@ public:
     *   infinite loop.
     *   When a limit is set a match operation will fail with an error if the
     *   limit is exceeded.
     *   infinite loop.
     *   When a limit is set a match operation will fail with an error if the
     *   limit is exceeded.

-    *   <p>
+    *

     *   The units of the limit are steps of the match engine.
     *   Correspondence with actual processor time will depend on the speed
     *   of the processor and the details of the specific pattern, but will
     *   typically be on the order of milliseconds.
     *   The units of the limit are steps of the match engine.
     *   Correspondence with actual processor time will depend on the speed
     *   of the processor and the details of the specific pattern, but will
     *   typically be on the order of milliseconds.

-    *   <p>
+    *

     *   By default, the matching time is not limited.
     *   By default, the matching time is not limited.

-    *   <p>
+    *

     *
     *   @param   limit       The limit value, or 0 for no limit.
     *   @param   status      A reference to a UErrorCode to receive any errors.
     *
     *   @param   limit       The limit value, or 0 for no limit.
     *   @param   status      A reference to a UErrorCode to receive any errors.


@@ -1526,18 +1634,18 @@ public:
     virtual int32_t getTimeLimit() const;
 
   /**
     virtual int32_t getTimeLimit() const;
 
   /**

-    *  Set the amount of heap storage avaliable for use by the match backtracking stack.
+    *  Set the amount of heap storage available for use by the match backtracking stack.

     *  The matcher is also reset, discarding any results from previous matches.
     *  The matcher is also reset, discarding any results from previous matches.

-    *  <p>
+    *

     *  ICU uses a backtracking regular expression engine, with the backtrack stack
     *  maintained on the heap.  This function sets the limit to the amount of memory
     *  ICU uses a backtracking regular expression engine, with the backtrack stack
     *  maintained on the heap.  This function sets the limit to the amount of memory

-    *  that can be used  for this purpose.  A backtracking stack overflow will
+    *  that can be used for this purpose.  A backtracking stack overflow will

     *  result in an error from the match operation that caused it.
     *  result in an error from the match operation that caused it.

-    *  <p>
+    *

     *  A limit is desirable because a malicious or poorly designed pattern can use
     *  excessive memory, potentially crashing the process.  A limit is enabled
     *  by default.
     *  A limit is desirable because a malicious or poorly designed pattern can use
     *  excessive memory, potentially crashing the process.  A limit is enabled
     *  by default.

-    *  <p>
+    *

     *  @param limit  The maximum size, in bytes, of the matching backtrack stack.
     *                A value of zero means no limit.
     *                The limit must be greater or equal to zero.
     *  @param limit  The maximum size, in bytes, of the matching backtrack stack.
     *                A value of zero means no limit.
     *                The limit must be greater or equal to zero.


@@ -1579,7 +1687,7 @@ public:
   /**
     *  Get the callback function for this URegularExpression.
     *
   /**
     *  Get the callback function for this URegularExpression.
     *

-    *    @param   callback    Out paramater, receives a pointer to the user-supplied 
+    *    @param   callback    Out parameter, receives a pointer to the user-supplied 

     *                         callback function.
     *    @param   context     Out parameter, receives the user context pointer that
     *                         was set when uregex_setMatchCallback() was called.
     *                         callback function.
     *    @param   context     Out parameter, receives the user context pointer that
     *                         was set when uregex_setMatchCallback() was called.


@@ -1602,7 +1710,7 @@ public:
     *                         time the callback function is set will be saved
     *                         and passed to the callback each time that it is called.
     *    @param   status      A reference to a UErrorCode to receive any errors.
     *                         time the callback function is set will be saved
     *                         and passed to the callback each time that it is called.
     *    @param   status      A reference to a UErrorCode to receive any errors.

-    *    @draft ICU 4.6
+    *    @stable ICU 4.6

     */
     virtual void setFindProgressCallback(URegexFindProgressCallback      *callback,
                                               const void                              *context,
     */
     virtual void setFindProgressCallback(URegexFindProgressCallback      *callback,
                                               const void                              *context,


@@ -1612,25 +1720,25 @@ public:
   /**
     *  Get the find progress callback function for this URegularExpression.
     *
   /**
     *  Get the find progress callback function for this URegularExpression.
     *

-    *    @param   callback    Out paramater, receives a pointer to the user-supplied 
+    *    @param   callback    Out parameter, receives a pointer to the user-supplied 

     *                         callback function.
     *    @param   context     Out parameter, receives the user context pointer that
     *                         was set when uregex_setFindProgressCallback() was called.
     *    @param   status      A reference to a UErrorCode to receive any errors.
     *                         callback function.
     *    @param   context     Out parameter, receives the user context pointer that
     *                         was set when uregex_setFindProgressCallback() was called.
     *    @param   status      A reference to a UErrorCode to receive any errors.

-    *    @draft ICU 4.6
+    *    @stable ICU 4.6

     */
     virtual void getFindProgressCallback(URegexFindProgressCallback      *&callback,
                                               const void                      *&context,
                                               UErrorCode                      &status);
 
     */
     virtual void getFindProgressCallback(URegexFindProgressCallback      *&callback,
                                               const void                      *&context,
                                               UErrorCode                      &status);
 

-
+#ifndef U_HIDE_INTERNAL_API

    /**
      *   setTrace   Debug function, enable/disable tracing of the matching engine.
      *              For internal ICU development use only.  DO NO USE!!!!
      *   @internal
      */
     void setTrace(UBool state);
    /**
      *   setTrace   Debug function, enable/disable tracing of the matching engine.
      *              For internal ICU development use only.  DO NO USE!!!!
      *   @internal
      */
     void setTrace(UBool state);

-
+#endif  /* U_HIDE_INTERNAL_API */

 
     /**
     * ICU "poor man's RTTI", returns a UClassID for this class.
 
     /**
     * ICU "poor man's RTTI", returns a UClassID for this class.


@@ -1659,8 +1767,10 @@ private:
     friend class RegexPattern;
     friend class RegexCImpl;
 public:
     friend class RegexPattern;
     friend class RegexCImpl;
 public:

+#ifndef U_HIDE_INTERNAL_API

     /** @internal  */
     void resetPreserveRegion();  // Reset matcher state, but preserve any region.
     /** @internal  */
     void resetPreserveRegion();  // Reset matcher state, but preserve any region.

+#endif  /* U_HIDE_INTERNAL_API */

 private:
 
     //
 private:
 
     //


@@ -1674,11 +1784,13 @@ private:
     REStackFrame        *resetStack();
     inline REStackFrame *StateSave(REStackFrame *fp, int64_t savePatIdx, UErrorCode &status);
     void                 IncrementTime(UErrorCode &status);
     REStackFrame        *resetStack();
     inline REStackFrame *StateSave(REStackFrame *fp, int64_t savePatIdx, UErrorCode &status);
     void                 IncrementTime(UErrorCode &status);

-    UBool                ReportFindProgress(int64_t matchIndex, UErrorCode &status);
+
+    // Call user find callback function, if set. Return TRUE if operation should be interrupted.
+    inline UBool         findProgressInterrupt(int64_t matchIndex, UErrorCode &status);

     
     int64_t              appendGroup(int32_t groupNum, UText *dest, UErrorCode &status) const;
     
     
     int64_t              appendGroup(int32_t groupNum, UText *dest, UErrorCode &status) const;
     

-    UBool                findUsingChunk();
+    UBool                findUsingChunk(UErrorCode &status);

     void                 MatchChunkAt(int32_t startIdx, UBool toEnd, UErrorCode &status);
     UBool                isChunkWordBoundary(int32_t pos);
 
     void                 MatchChunkAt(int32_t startIdx, UBool toEnd, UErrorCode &status);
     UBool                isChunkWordBoundary(int32_t pos);
 


@@ -1763,10 +1875,11 @@ private:
                                            //   reported, or that permanently disables this matcher.
 
     RuleBasedBreakIterator  *fWordBreakItr;
                                            //   reported, or that permanently disables this matcher.
 
     RuleBasedBreakIterator  *fWordBreakItr;

-
-

 };
 
 U_NAMESPACE_END
 #endif  // UCONFIG_NO_REGULAR_EXPRESSIONS
 };
 
 U_NAMESPACE_END
 #endif  // UCONFIG_NO_REGULAR_EXPRESSIONS

+
+#endif /* U_SHOW_CPLUSPLUS_API */
+

 #endif
 #endif