X-Git-Url: https://git.saurik.com/apple/icu.git/blobdiff_plain/374ca955a76ecab1204ca8bfa63ff9238d998416..46f4442e9a5a4f3b98b7c1083586332f6a8a99a4:/icuSources/common/unicode/usetiter.h diff --git a/icuSources/common/unicode/usetiter.h b/icuSources/common/unicode/usetiter.h index 9584b748..defa75cd 100644 --- a/icuSources/common/unicode/usetiter.h +++ b/icuSources/common/unicode/usetiter.h @@ -1,6 +1,6 @@ /* ********************************************************************** -* Copyright (c) 2002-2004, International Business Machines +* Copyright (c) 2002-2006, International Business Machines * Corporation and others. All Rights Reserved. ********************************************************************** */ @@ -11,30 +11,40 @@ #include "unicode/uobject.h" #include "unicode/unistr.h" +/** + * \file + * \brief C++ API: UnicodeSetIterator iterates over the contents of a UnicodeSet. + */ + U_NAMESPACE_BEGIN class UnicodeSet; class UnicodeString; /** + * * UnicodeSetIterator iterates over the contents of a UnicodeSet. It * iterates over either code points or code point ranges. After all * code points or ranges have been returned, it returns the * multicharacter strings of the UnicodSet, if any. * - *

To iterate over code points, use a loop like this: + * This class is not intended to be subclassed. Consider any fields + * or methods declared as "protected" to be private. The use of + * protected in this class is an artifact of history. + * + *

To iterate over code points and strings, use a loop like this: *

  * UnicodeSetIterator it(set);
  * while (set.next()) {
- *   if (set.isString()) {
- *     processString(set.getString());
- *   } else {
- *     processCodepoint(set.getCodepoint());
- *   }
+ *     processItem(set.getString());
  * }
  * 
+ *

Each item in the set is accessed as a string. Set elements + * consisting of single code points are returned as strings containing + * just the one code point. * - *

To iterate over code point ranges, use a loop like this: + *

To iterate over code point ranges, instead of individual code points, + * use a loop like this: *

  * UnicodeSetIterator it(set);
  * while (it.nextRange()) {
@@ -115,9 +125,14 @@ class U_COMMON_API UnicodeSetIterator : public UObject {
      * caller can retrieve it with getString().  If this
      * method returns false, the current element is a code point or
      * code point range, depending on whether next() or
-     * nextRange() was called, and the caller can retrieve it
-     * with getCodepoint() and, for a range,
-     * getCodepointEnd().
+     * nextRange() was called.
+     * Elements of types string and codepoint can both be retrieved
+     * with the function getString().
+     * Elements of type codepoint can also be retrieved with
+     * getCodepoint().
+     * For ranges, getCodepoint() returns the starting codepoint
+     * of the range, and getCodepointEnd() returns the end
+     * of the range.
      * @stable ICU 2.4
      */
     inline UBool isString() const;
@@ -139,28 +154,37 @@ class U_COMMON_API UnicodeSetIterator : public UObject {
 
     /**
      * Returns the current string, if isString() returned
-     * true.  Otherwise returns an undefined result.
+     * true.  If the current iteration item is a code point, a UnicodeString
+     * containing that single code point is returned.
+     *
+     * Ownership of the returned string remains with the iterator.
+     * The string is guaranteed to remain valid only until the iterator is
+     *   advanced to the next item, or until the iterator is deleted.
+     * 
      * @stable ICU 2.4
      */
-    inline const UnicodeString& getString() const;
+    const UnicodeString& getString();
 
     /**
-     * Returns the next element in the set, either a single code point
-     * or a string.  If there are no more elements in the set, return
-     * false.  If codepoint == IS_STRING, the value is a
-     * string in the string field.  Otherwise the value is a
-     * single code point in the codepoint field.
+     * Advances the iteration position to the next element in the set, 
+     * which can be either a single code point or a string.  
+     * If there are no more elements in the set, return false.
+     *
+     * 

+ * If isString() == TRUE, the value is a + * string, otherwise the value is a + * single code point. Elements of either type can be retrieved + * with the function getString(), while elements of + * consisting of a single code point can be retrieved with + * getCodepoint() * *

The order of iteration is all code points in sorted order, - * followed by all strings sorted order. codepointEnd is - * undefined after calling this method. string is - * undefined unless codepoint == IS_STRING. Do not mix + * followed by all strings sorted order. Do not mix * calls to next() and nextRange() without * calling reset() between them. The results of doing so * are undefined. * - * @return true if there was another element in the set and this - * object contains the element. + * @return true if there was another element in the set. * @stable ICU 2.4 */ UBool next(); @@ -168,21 +192,20 @@ class U_COMMON_API UnicodeSetIterator : public UObject { /** * Returns the next element in the set, either a code point range * or a string. If there are no more elements in the set, return - * false. If codepoint == IS_STRING, the value is a - * string in the string field. Otherwise the value is a - * range of one or more code points from codepoint to - * codepointeEnd inclusive. + * false. If isString() == TRUE, the value is a + * string and can be accessed with getString(). Otherwise the value is a + * range of one or more code points from getCodepoint() to + * getCodepointeEnd() inclusive. * *

The order of iteration is all code points ranges in sorted * order, followed by all strings sorted order. Ranges are - * disjoint and non-contiguous. string is undefined - * unless codepoint == IS_STRING. Do not mix calls to + * disjoint and non-contiguous. The value returned from getString() + * is undefined unless isString() == TRUE. Do not mix calls to * next() and nextRange() without calling * reset() between them. The results of doing so are * undefined. * - * @return true if there was another element in the set and this - * object contains the element. + * @return true if there was another element in the set. * @stable ICU 2.4 */ UBool nextRange(); @@ -253,6 +276,13 @@ class U_COMMON_API UnicodeSetIterator : public UObject { */ int32_t stringCount; + /** + * Points to the string to use when the caller asks for a + * string and the current iteration item is a code point, not a string. + * @internal + */ + UnicodeString *cpString; + /** Copy constructor. Disallowed. * @stable ICU 2.4 */ @@ -282,9 +312,6 @@ inline UChar32 UnicodeSetIterator::getCodepointEnd() const { return codepointEnd; } -inline const UnicodeString& UnicodeSetIterator::getString() const { - return *string; -} U_NAMESPACE_END