/src/icu/icu4c/source/common/unicode/usetiter.h

Source
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
**********************************************************************
* Copyright (c) 2002-2014, International Business Machines
* Corporation and others.  All Rights Reserved.
**********************************************************************
*/
#ifndef USETITER_H
#define USETITER_H

#include "unicode/utypes.h"

#if U_SHOW_CPLUSPLUS_API

#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 UnicodeSet, if any.
 *
 * This class is not intended for public subclassing.
 *
 * <p>To iterate over code points and strings, use a loop like this:
 * <pre>
 * UnicodeSetIterator it(set);
 * while (it.next()) {
 *     processItem(it.getString());
 * }
 * </pre>
 * <p>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.
 *
 * <p>To iterate over code point ranges, instead of individual code points,
 *    use a loop like this:
 * <pre>
 * UnicodeSetIterator it(set);
 * while (it.nextRange()) {
 *   if (it.isString()) {
 *     processString(it.getString());
 *   } else {
 *     processCodepointRange(it.getCodepoint(), it.getCodepointEnd());
 *   }
 * }
 * </pre>
 *
 * To iterate over only the strings, start with <code>skipToStrings()</code>.
 *
 * @author M. Davis
 * @stable ICU 2.4
 */
class U_COMMON_API UnicodeSetIterator final : public UObject {
    /**
     * Value of <tt>codepoint</tt> if the iterator points to a string.
     * If <tt>codepoint == IS_STRING</tt>, then examine
     * <tt>string</tt> for the current iteration result.
     */
    enum { IS_STRING = -1 };

    /**
     * Current code point, or the special value <tt>IS_STRING</tt>, if
     * the iterator points to a string.
     */
    UChar32 codepoint;

    /**
     * When iterating over ranges using <tt>nextRange()</tt>,
     * <tt>codepointEnd</tt> contains the inclusive end of the
     * iteration range, if <tt>codepoint != IS_STRING</tt>.  If
     * iterating over code points using <tt>next()</tt>, or if
     * <tt>codepoint == IS_STRING</tt>, then the value of
     * <tt>codepointEnd</tt> is undefined.
     */
    UChar32 codepointEnd;

    /**
     * If <tt>codepoint == IS_STRING</tt>, then <tt>string</tt> points
     * to the current string.  If <tt>codepoint != IS_STRING</tt>, the
     * value of <tt>string</tt> is undefined.
     */
    const UnicodeString* string;

 public:

    /**
     * Create an iterator over the given set.  The iterator is valid
     * only so long as <tt>set</tt> is valid.
     * @param set set to iterate over
     * @stable ICU 2.4
     */
    UnicodeSetIterator(const UnicodeSet& set);

    /**
     * Create an iterator over nothing.  <tt>next()</tt> and
     * <tt>nextRange()</tt> return false. This is a convenience
     * constructor allowing the target to be set later.
     * @stable ICU 2.4
     */
    UnicodeSetIterator();

    /**
     * Destructor.
     * @stable ICU 2.4
     */
    virtual ~UnicodeSetIterator();

    /**
     * Returns true if the current element is a string.  If so, the
     * caller can retrieve it with <tt>getString()</tt>.  If this
     * method returns false, the current element is a code point or
     * code point range, depending on whether <tt>next()</tt> or
     * <tt>nextRange()</tt> was called.
     * Elements of types string and codepoint can both be retrieved
     * with the function <tt>getString()</tt>.
     * Elements of type codepoint can also be retrieved with
     * <tt>getCodepoint()</tt>.
     * For ranges, <tt>getCodepoint()</tt> returns the starting codepoint
     * of the range, and <tt>getCodepointEnd()</tt> returns the end
     * of the range.
     * @stable ICU 2.4
     */
    inline UBool isString() const;

    /**
     * Returns the current code point, if <tt>isString()</tt> returned
     * false.  Otherwise returns an undefined result.
     * @stable ICU 2.4
     */
    inline UChar32 getCodepoint() const;

    /**
     * Returns the end of the current code point range, if
     * <tt>isString()</tt> returned false and <tt>nextRange()</tt> was
     * called.  Otherwise returns an undefined result.
     * @stable ICU 2.4
     */
    inline UChar32 getCodepointEnd() const;

    /**
     * Returns the current string, if <tt>isString()</tt> returned
     * 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
     */
    const UnicodeString& getString();

    /**
     * Skips over the remaining code points/ranges, if any.
     * A following call to next() or nextRange() will yield a string, if there is one.
     * No-op if next() would return false, or if it would yield a string anyway.
     *
     * @return *this
     * @stable ICU 70
     * @see UnicodeSet#strings()
     */
    inline UnicodeSetIterator &skipToStrings() {
        // Finish code point/range iteration.
        range = endRange;
        endElement = -1;
        nextElement = 0;
        return *this;
    }

    /**
     * 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.
     *
     * <p>
     * If <tt>isString() == true</tt>, the value is a
     * string, otherwise the value is a
     * single code point.  Elements of either type can be retrieved
     * with the function <tt>getString()</tt>, while elements of
     * consisting of a single code point can be retrieved with
     * <tt>getCodepoint()</tt>
     *
     * <p>The order of iteration is all code points in sorted order,
     * followed by all strings sorted order.    Do not mix
     * calls to <tt>next()</tt> and <tt>nextRange()</tt> without
     * calling <tt>reset()</tt> between them.  The results of doing so
     * are undefined.
     *
     * @return true if there was another element in the set.
     * @stable ICU 2.4
     */
    UBool next();

    /**
     * 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 <tt>isString() == true</tt>, the value is a
     * string and can be accessed with <tt>getString()</tt>.  Otherwise the value is a
     * range of one or more code points from <tt>getCodepoint()</tt> to
     * <tt>getCodepointeEnd()</tt> inclusive.
     *
     * <p>The order of iteration is all code points ranges in sorted
     * order, followed by all strings sorted order.  Ranges are
     * disjoint and non-contiguous.  The value returned from <tt>getString()</tt>
     * is undefined unless <tt>isString() == true</tt>.  Do not mix calls to
     * <tt>next()</tt> and <tt>nextRange()</tt> without calling
     * <tt>reset()</tt> between them.  The results of doing so are
     * undefined.
     *
     * @return true if there was another element in the set.
     * @stable ICU 2.4
     */
    UBool nextRange();

    /**
     * Sets this iterator to visit the elements of the given set and
     * resets it to the start of that set.  The iterator is valid only
     * so long as <tt>set</tt> is valid.
     * @param set the set to iterate over.
     * @stable ICU 2.4
     */
    void reset(const UnicodeSet& set);

    /**
     * Resets this iterator to the start of the set.
     * @stable ICU 2.4
     */
    void reset();

    /**
     * ICU "poor man's RTTI", returns a UClassID for this class.
     *
     * @stable ICU 2.4
     */
    static UClassID U_EXPORT2 getStaticClassID();

    /**
     * ICU "poor man's RTTI", returns a UClassID for the actual class.
     *
     * @stable ICU 2.4
     */
    virtual UClassID getDynamicClassID() const override;

    // ======================= PRIVATES ===========================

private:

    // endElement and nextElements are really UChar32's, but we keep
    // them as signed int32_t's so we can do comparisons with
    // endElement set to -1.  Leave them as int32_t's.
    /** The set
     */
    const UnicodeSet* set;
    /** End range
     */
    int32_t endRange;
    /** Range
     */
    int32_t range;
    /** End element
     */
    int32_t endElement;
    /** Next element
     */
    int32_t nextElement;
    /** Next string
     */
    int32_t nextString;
    /** String count
     */
    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.
     */
    UnicodeString *cpString;

    /** Copy constructor. Disallowed.
     */
    UnicodeSetIterator(const UnicodeSetIterator&) = delete;

    /** Assignment operator. Disallowed.
     */
    UnicodeSetIterator& operator=(const UnicodeSetIterator&) = delete;

    /** Load range
     */
    void loadRange(int32_t range);
};

inline UBool UnicodeSetIterator::isString() const {
    return codepoint < 0;
}

inline UChar32 UnicodeSetIterator::getCodepoint() const {
    return codepoint;
}

inline UChar32 UnicodeSetIterator::getCodepointEnd() const {
    return codepointEnd;
}


U_NAMESPACE_END

#endif /* U_SHOW_CPLUSPLUS_API */

#endif

Coverage Report

Created: 2025-10-24 06:54

Line	Count	Source
1		// © 2016 and later: Unicode, Inc. and others.
2		// License & terms of use: http://www.unicode.org/copyright.html
3		/*
4		**********************************************************************
5		* Copyright (c) 2002-2014, International Business Machines
6		* Corporation and others. All Rights Reserved.
7		**********************************************************************
8		*/
9		#ifndef USETITER_H
10		#define USETITER_H
11
12		#include "unicode/utypes.h"
13
14		#if U_SHOW_CPLUSPLUS_API
15
16		#include "unicode/uobject.h"
17		#include "unicode/unistr.h"
18
19		/**
20		* \file
21		* \brief C++ API: UnicodeSetIterator iterates over the contents of a UnicodeSet.
22		*/
23
24		U_NAMESPACE_BEGIN
25
26		class UnicodeSet;
27		class UnicodeString;
28
29		/**
30		*
31		* UnicodeSetIterator iterates over the contents of a UnicodeSet. It
32		* iterates over either code points or code point ranges. After all
33		* code points or ranges have been returned, it returns the
34		* multicharacter strings of the UnicodeSet, if any.
35		*
36		* This class is not intended for public subclassing.
37		*
38		* <p>To iterate over code points and strings, use a loop like this:
39		* <pre>
40		* UnicodeSetIterator it(set);
41		* while (it.next()) {
42		* processItem(it.getString());
43		* }
44		* </pre>
45		* <p>Each item in the set is accessed as a string. Set elements
46		* consisting of single code points are returned as strings containing
47		* just the one code point.
48		*
49		* <p>To iterate over code point ranges, instead of individual code points,
50		* use a loop like this:
51		* <pre>
52		* UnicodeSetIterator it(set);
53		* while (it.nextRange()) {
54		* if (it.isString()) {
55		* processString(it.getString());
56		* } else {
57		* processCodepointRange(it.getCodepoint(), it.getCodepointEnd());
58		* }
59		* }
60		* </pre>
61		*
62		* To iterate over only the strings, start with <code>skipToStrings()</code>.
63		*
64		* @author M. Davis
65		* @stable ICU 2.4
66		*/
67		class U_COMMON_API UnicodeSetIterator final : public UObject {
68		/**
69		* Value of <tt>codepoint</tt> if the iterator points to a string.
70		* If <tt>codepoint == IS_STRING</tt>, then examine
71		* <tt>string</tt> for the current iteration result.
72		*/
73		enum { IS_STRING = -1 };
74
75		/**
76		* Current code point, or the special value <tt>IS_STRING</tt>, if
77		* the iterator points to a string.
78		*/
79		UChar32 codepoint;
80
81		/**
82		* When iterating over ranges using <tt>nextRange()</tt>,
83		* <tt>codepointEnd</tt> contains the inclusive end of the
84		* iteration range, if <tt>codepoint != IS_STRING</tt>. If
85		* iterating over code points using <tt>next()</tt>, or if
86		* <tt>codepoint == IS_STRING</tt>, then the value of
87		* <tt>codepointEnd</tt> is undefined.
88		*/
89		UChar32 codepointEnd;
90
91		/**
92		* If <tt>codepoint == IS_STRING</tt>, then <tt>string</tt> points
93		* to the current string. If <tt>codepoint != IS_STRING</tt>, the
94		* value of <tt>string</tt> is undefined.
95		*/
96		const UnicodeString* string;
97
98		public:
99
100		/**
101		* Create an iterator over the given set. The iterator is valid
102		* only so long as <tt>set</tt> is valid.
103		* @param set set to iterate over
104		* @stable ICU 2.4
105		*/
106		UnicodeSetIterator(const UnicodeSet& set);
107
108		/**
109		* Create an iterator over nothing. <tt>next()</tt> and
110		* <tt>nextRange()</tt> return false. This is a convenience
111		* constructor allowing the target to be set later.
112		* @stable ICU 2.4
113		*/
114		UnicodeSetIterator();
115
116		/**
117		* Destructor.
118		* @stable ICU 2.4
119		*/
120		virtual ~UnicodeSetIterator();
121
122		/**
123		* Returns true if the current element is a string. If so, the
124		* caller can retrieve it with <tt>getString()</tt>. If this
125		* method returns false, the current element is a code point or
126		* code point range, depending on whether <tt>next()</tt> or
127		* <tt>nextRange()</tt> was called.
128		* Elements of types string and codepoint can both be retrieved
129		* with the function <tt>getString()</tt>.
130		* Elements of type codepoint can also be retrieved with
131		* <tt>getCodepoint()</tt>.
132		* For ranges, <tt>getCodepoint()</tt> returns the starting codepoint
133		* of the range, and <tt>getCodepointEnd()</tt> returns the end
134		* of the range.
135		* @stable ICU 2.4
136		*/
137		inline UBool isString() const;
138
139		/**
140		* Returns the current code point, if <tt>isString()</tt> returned
141		* false. Otherwise returns an undefined result.
142		* @stable ICU 2.4
143		*/
144		inline UChar32 getCodepoint() const;
145
146		/**
147		* Returns the end of the current code point range, if
148		* <tt>isString()</tt> returned false and <tt>nextRange()</tt> was
149		* called. Otherwise returns an undefined result.
150		* @stable ICU 2.4
151		*/
152		inline UChar32 getCodepointEnd() const;
153
154		/**
155		* Returns the current string, if <tt>isString()</tt> returned
156		* true. If the current iteration item is a code point, a UnicodeString
157		* containing that single code point is returned.
158		*
159		* Ownership of the returned string remains with the iterator.
160		* The string is guaranteed to remain valid only until the iterator is
161		* advanced to the next item, or until the iterator is deleted.
162		*
163		* @stable ICU 2.4
164		*/
165		const UnicodeString& getString();
166
167		/**
168		* Skips over the remaining code points/ranges, if any.
169		* A following call to next() or nextRange() will yield a string, if there is one.
170		* No-op if next() would return false, or if it would yield a string anyway.
171		*
172		* @return *this
173		* @stable ICU 70
174		* @see UnicodeSet#strings()
175		*/
176	0	inline UnicodeSetIterator &skipToStrings() {
177	0	// Finish code point/range iteration.
178	0	range = endRange;
179	0	endElement = -1;
180	0	nextElement = 0;
181	0	return *this;
182	0	}
183
184		/**
185		* Advances the iteration position to the next element in the set,
186		* which can be either a single code point or a string.
187		* If there are no more elements in the set, return false.
188		*
189		* <p>
190		* If <tt>isString() == true</tt>, the value is a
191		* string, otherwise the value is a
192		* single code point. Elements of either type can be retrieved
193		* with the function <tt>getString()</tt>, while elements of
194		* consisting of a single code point can be retrieved with
195		* <tt>getCodepoint()</tt>
196		*
197		* <p>The order of iteration is all code points in sorted order,
198		* followed by all strings sorted order. Do not mix
199		* calls to <tt>next()</tt> and <tt>nextRange()</tt> without
200		* calling <tt>reset()</tt> between them. The results of doing so
201		* are undefined.
202		*
203		* @return true if there was another element in the set.
204		* @stable ICU 2.4
205		*/
206		UBool next();
207
208		/**
209		* Returns the next element in the set, either a code point range
210		* or a string. If there are no more elements in the set, return
211		* false. If <tt>isString() == true</tt>, the value is a
212		* string and can be accessed with <tt>getString()</tt>. Otherwise the value is a
213		* range of one or more code points from <tt>getCodepoint()</tt> to
214		* <tt>getCodepointeEnd()</tt> inclusive.
215		*
216		* <p>The order of iteration is all code points ranges in sorted
217		* order, followed by all strings sorted order. Ranges are
218		* disjoint and non-contiguous. The value returned from <tt>getString()</tt>
219		* is undefined unless <tt>isString() == true</tt>. Do not mix calls to
220		* <tt>next()</tt> and <tt>nextRange()</tt> without calling
221		* <tt>reset()</tt> between them. The results of doing so are
222		* undefined.
223		*
224		* @return true if there was another element in the set.
225		* @stable ICU 2.4
226		*/
227		UBool nextRange();
228
229		/**
230		* Sets this iterator to visit the elements of the given set and
231		* resets it to the start of that set. The iterator is valid only
232		* so long as <tt>set</tt> is valid.
233		* @param set the set to iterate over.
234		* @stable ICU 2.4
235		*/
236		void reset(const UnicodeSet& set);
237
238		/**
239		* Resets this iterator to the start of the set.
240		* @stable ICU 2.4
241		*/
242		void reset();
243
244		/**
245		* ICU "poor man's RTTI", returns a UClassID for this class.
246		*
247		* @stable ICU 2.4
248		*/
249		static UClassID U_EXPORT2 getStaticClassID();
250
251		/**
252		* ICU "poor man's RTTI", returns a UClassID for the actual class.
253		*
254		* @stable ICU 2.4
255		*/
256		virtual UClassID getDynamicClassID() const override;
257
258		// ======================= PRIVATES ===========================
259
260		private:
261
262		// endElement and nextElements are really UChar32's, but we keep
263		// them as signed int32_t's so we can do comparisons with
264		// endElement set to -1. Leave them as int32_t's.
265		/** The set
266		*/
267		const UnicodeSet* set;
268		/** End range
269		*/
270		int32_t endRange;
271		/** Range
272		*/
273		int32_t range;
274		/** End element
275		*/
276		int32_t endElement;
277		/** Next element
278		*/
279		int32_t nextElement;
280		/** Next string
281		*/
282		int32_t nextString;
283		/** String count
284		*/
285		int32_t stringCount;
286
287		/**
288		* Points to the string to use when the caller asks for a
289		* string and the current iteration item is a code point, not a string.
290		*/
291		UnicodeString *cpString;
292
293		/** Copy constructor. Disallowed.
294		*/
295		UnicodeSetIterator(const UnicodeSetIterator&) = delete;
296
297		/** Assignment operator. Disallowed.
298		*/
299		UnicodeSetIterator& operator=(const UnicodeSetIterator&) = delete;
300
301		/** Load range
302		*/
303		void loadRange(int32_t range);
304		};
305
306	2.56M	inline UBool UnicodeSetIterator::isString() const {
307	2.56M	return codepoint < 0;
308	2.56M	}
309
310	45.5M	inline UChar32 UnicodeSetIterator::getCodepoint() const {
311	45.5M	return codepoint;
312	45.5M	}
313
314	0	inline UChar32 UnicodeSetIterator::getCodepointEnd() const {
315	0	return codepointEnd;
316	0	}
317
318
319		U_NAMESPACE_END
320
321		#endif /* U_SHOW_CPLUSPLUS_API */
322
323		#endif