1 // Copyright (C) 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
4 **********************************************************************
5 * Copyright (c) 2002-2014, International Business Machines
6 * Corporation and others. All Rights Reserved.
7 **********************************************************************
12 #include "unicode/utypes.h"
13 #include "unicode/uobject.h"
14 #include "unicode/unistr.h"
18 * \brief C++ API: UnicodeSetIterator iterates over the contents of a UnicodeSet.
28 * UnicodeSetIterator iterates over the contents of a UnicodeSet. It
29 * iterates over either code points or code point ranges. After all
30 * code points or ranges have been returned, it returns the
31 * multicharacter strings of the UnicodeSet, if any.
33 * This class is not intended to be subclassed. Consider any fields
34 * or methods declared as "protected" to be private. The use of
35 * protected in this class is an artifact of history.
37 * <p>To iterate over code points and strings, use a loop like this:
39 * UnicodeSetIterator it(set);
41 * processItem(it.getString());
44 * <p>Each item in the set is accessed as a string. Set elements
45 * consisting of single code points are returned as strings containing
46 * just the one code point.
48 * <p>To iterate over code point ranges, instead of individual code points,
49 * use a loop like this:
51 * UnicodeSetIterator it(set);
52 * while (it.nextRange()) {
53 * if (it.isString()) {
54 * processString(it.getString());
56 * processCodepointRange(it.getCodepoint(), it.getCodepointEnd());
63 class U_COMMON_API UnicodeSetIterator : public UObject {
68 * Value of <tt>codepoint</tt> if the iterator points to a string.
69 * If <tt>codepoint == IS_STRING</tt>, then examine
70 * <tt>string</tt> for the current iteration result.
73 enum { IS_STRING = -1 };
76 * Current code point, or the special value <tt>IS_STRING</tt>, if
77 * the iterator points to a string.
83 * When iterating over ranges using <tt>nextRange()</tt>,
84 * <tt>codepointEnd</tt> contains the inclusive end of the
85 * iteration range, if <tt>codepoint != IS_STRING</tt>. If
86 * iterating over code points using <tt>next()</tt>, or if
87 * <tt>codepoint == IS_STRING</tt>, then the value of
88 * <tt>codepointEnd</tt> is undefined.
94 * If <tt>codepoint == IS_STRING</tt>, then <tt>string</tt> points
95 * to the current string. If <tt>codepoint != IS_STRING</tt>, the
96 * value of <tt>string</tt> is undefined.
99 const UnicodeString* string;
104 * Create an iterator over the given set. The iterator is valid
105 * only so long as <tt>set</tt> is valid.
106 * @param set set to iterate over
109 UnicodeSetIterator(const UnicodeSet& set);
112 * Create an iterator over nothing. <tt>next()</tt> and
113 * <tt>nextRange()</tt> return false. This is a convenience
114 * constructor allowing the target to be set later.
117 UnicodeSetIterator();
123 virtual ~UnicodeSetIterator();
126 * Returns true if the current element is a string. If so, the
127 * caller can retrieve it with <tt>getString()</tt>. If this
128 * method returns false, the current element is a code point or
129 * code point range, depending on whether <tt>next()</tt> or
130 * <tt>nextRange()</tt> was called.
131 * Elements of types string and codepoint can both be retrieved
132 * with the function <tt>getString()</tt>.
133 * Elements of type codepoint can also be retrieved with
134 * <tt>getCodepoint()</tt>.
135 * For ranges, <tt>getCodepoint()</tt> returns the starting codepoint
136 * of the range, and <tt>getCodepointEnd()</tt> returns the end
140 inline UBool isString() const;
143 * Returns the current code point, if <tt>isString()</tt> returned
144 * false. Otherwise returns an undefined result.
147 inline UChar32 getCodepoint() const;
150 * Returns the end of the current code point range, if
151 * <tt>isString()</tt> returned false and <tt>nextRange()</tt> was
152 * called. Otherwise returns an undefined result.
155 inline UChar32 getCodepointEnd() const;
158 * Returns the current string, if <tt>isString()</tt> returned
159 * true. If the current iteration item is a code point, a UnicodeString
160 * containing that single code point is returned.
162 * Ownership of the returned string remains with the iterator.
163 * The string is guaranteed to remain valid only until the iterator is
164 * advanced to the next item, or until the iterator is deleted.
168 const UnicodeString& getString();
171 * Advances the iteration position to the next element in the set,
172 * which can be either a single code point or a string.
173 * If there are no more elements in the set, return false.
176 * If <tt>isString() == TRUE</tt>, the value is a
177 * string, otherwise the value is a
178 * single code point. Elements of either type can be retrieved
179 * with the function <tt>getString()</tt>, while elements of
180 * consisting of a single code point can be retrieved with
181 * <tt>getCodepoint()</tt>
183 * <p>The order of iteration is all code points in sorted order,
184 * followed by all strings sorted order. Do not mix
185 * calls to <tt>next()</tt> and <tt>nextRange()</tt> without
186 * calling <tt>reset()</tt> between them. The results of doing so
189 * @return true if there was another element in the set.
195 * Returns the next element in the set, either a code point range
196 * or a string. If there are no more elements in the set, return
197 * false. If <tt>isString() == TRUE</tt>, the value is a
198 * string and can be accessed with <tt>getString()</tt>. Otherwise the value is a
199 * range of one or more code points from <tt>getCodepoint()</tt> to
200 * <tt>getCodepointeEnd()</tt> inclusive.
202 * <p>The order of iteration is all code points ranges in sorted
203 * order, followed by all strings sorted order. Ranges are
204 * disjoint and non-contiguous. The value returned from <tt>getString()</tt>
205 * is undefined unless <tt>isString() == TRUE</tt>. Do not mix calls to
206 * <tt>next()</tt> and <tt>nextRange()</tt> without calling
207 * <tt>reset()</tt> between them. The results of doing so are
210 * @return true if there was another element in the set.
216 * Sets this iterator to visit the elements of the given set and
217 * resets it to the start of that set. The iterator is valid only
218 * so long as <tt>set</tt> is valid.
219 * @param set the set to iterate over.
222 void reset(const UnicodeSet& set);
225 * Resets this iterator to the start of the set.
231 * ICU "poor man's RTTI", returns a UClassID for this class.
235 static UClassID U_EXPORT2 getStaticClassID();
238 * ICU "poor man's RTTI", returns a UClassID for the actual class.
242 virtual UClassID getDynamicClassID() const;
244 // ======================= PRIVATES ===========================
248 // endElement and nextElements are really UChar32's, but we keep
249 // them as signed int32_t's so we can do comparisons with
250 // endElement set to -1. Leave them as int32_t's.
254 const UnicodeSet* set;
282 * Points to the string to use when the caller asks for a
283 * string and the current iteration item is a code point, not a string.
286 UnicodeString *cpString;
288 /** Copy constructor. Disallowed.
291 UnicodeSetIterator(const UnicodeSetIterator&); // disallow
293 /** Assignment operator. Disallowed.
296 UnicodeSetIterator& operator=(const UnicodeSetIterator&); // disallow
301 virtual void loadRange(int32_t range);
305 inline UBool UnicodeSetIterator::isString() const {
306 return codepoint == (UChar32)IS_STRING;
309 inline UChar32 UnicodeSetIterator::getCodepoint() const {
313 inline UChar32 UnicodeSetIterator::getCodepointEnd() const {