/src/icu/source/common/uset.cpp

Source (jump to first uncovered line)
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
*
*   Copyright (C) 2002-2011, International Business Machines
*   Corporation and others.  All Rights Reserved.
*
*******************************************************************************
*   file name:  uset.cpp
*   encoding:   UTF-8
*   tab size:   8 (not used)
*   indentation:4
*
*   created on: 2002mar07
*   created by: Markus W. Scherer
*
*   There are functions to efficiently serialize a USet into an array of uint16_t
*   and functions to use such a serialized form efficiently without
*   instantiating a new USet.
*/

#include "unicode/utypes.h"
#include "unicode/uobject.h"
#include "unicode/uset.h"
#include "unicode/uniset.h"
#include "cmemory.h"
#include "unicode/ustring.h"
#include "unicode/parsepos.h"

U_NAMESPACE_USE

U_CAPI USet* U_EXPORT2
uset_openEmpty() {
    return (USet*) new UnicodeSet();
}

U_CAPI USet* U_EXPORT2
uset_open(UChar32 start, UChar32 end) {
    return (USet*) new UnicodeSet(start, end);
}

U_CAPI void U_EXPORT2
uset_close(USet* set) {
    delete (UnicodeSet*) set;
}

U_CAPI USet * U_EXPORT2
uset_clone(const USet *set) {
    return (USet*) (((UnicodeSet*) set)->UnicodeSet::clone());
}

U_CAPI UBool U_EXPORT2
uset_isFrozen(const USet *set) {
    return ((UnicodeSet*) set)->UnicodeSet::isFrozen();
}

U_CAPI void U_EXPORT2
uset_freeze(USet *set) {
    ((UnicodeSet*) set)->UnicodeSet::freeze();
}

U_CAPI USet * U_EXPORT2
uset_cloneAsThawed(const USet *set) {
    return (USet*) (((UnicodeSet*) set)->UnicodeSet::cloneAsThawed());
}

U_CAPI void U_EXPORT2
uset_set(USet* set,
     UChar32 start, UChar32 end) {
    ((UnicodeSet*) set)->UnicodeSet::set(start, end);
}

U_CAPI void U_EXPORT2
uset_addAll(USet* set, const USet *additionalSet) {
    ((UnicodeSet*) set)->UnicodeSet::addAll(*((const UnicodeSet*)additionalSet));
}

U_CAPI void U_EXPORT2
uset_add(USet* set, UChar32 c) {
    ((UnicodeSet*) set)->UnicodeSet::add(c);
}

U_CAPI void U_EXPORT2
uset_addRange(USet* set, UChar32 start, UChar32 end) {
    ((UnicodeSet*) set)->UnicodeSet::add(start, end);    
}

U_CAPI void U_EXPORT2
uset_addString(USet* set, const UChar* str, int32_t strLen) {
    // UnicodeString handles -1 for strLen
    UnicodeString s(strLen<0, str, strLen);
    ((UnicodeSet*) set)->UnicodeSet::add(s);
}

U_CAPI void U_EXPORT2
uset_addAllCodePoints(USet* set, const UChar *str, int32_t strLen) {
    // UnicodeString handles -1 for strLen
    UnicodeString s(str, strLen);
    ((UnicodeSet*) set)->UnicodeSet::addAll(s);
}

U_CAPI void U_EXPORT2
uset_remove(USet* set, UChar32 c) {
    ((UnicodeSet*) set)->UnicodeSet::remove(c);
}

U_CAPI void U_EXPORT2
uset_removeRange(USet* set, UChar32 start, UChar32 end) {
    ((UnicodeSet*) set)->UnicodeSet::remove(start, end);
}

U_CAPI void U_EXPORT2
uset_removeString(USet* set, const UChar* str, int32_t strLen) {
    UnicodeString s(strLen==-1, str, strLen);
    ((UnicodeSet*) set)->UnicodeSet::remove(s);
}

U_CAPI void U_EXPORT2
uset_removeAllCodePoints(USet *set, const UChar *str, int32_t length) {
    UnicodeString s(length==-1, str, length);
    ((UnicodeSet*) set)->UnicodeSet::removeAll(s);
}

U_CAPI void U_EXPORT2
uset_removeAll(USet* set, const USet* remove) {
    ((UnicodeSet*) set)->UnicodeSet::removeAll(*(const UnicodeSet*)remove);
}

U_CAPI void U_EXPORT2
uset_retain(USet* set, UChar32 start, UChar32 end) {
    ((UnicodeSet*) set)->UnicodeSet::retain(start, end);
}

U_CAPI void U_EXPORT2
uset_retainString(USet *set, const UChar *str, int32_t length) {
    UnicodeString s(length==-1, str, length);
    ((UnicodeSet*) set)->UnicodeSet::retain(s);
}

U_CAPI void U_EXPORT2
uset_retainAllCodePoints(USet *set, const UChar *str, int32_t length) {
    UnicodeString s(length==-1, str, length);
    ((UnicodeSet*) set)->UnicodeSet::retainAll(s);
}

U_CAPI void U_EXPORT2
uset_retainAll(USet* set, const USet* retain) {
    ((UnicodeSet*) set)->UnicodeSet::retainAll(*(const UnicodeSet*)retain);
}

U_CAPI void U_EXPORT2
uset_compact(USet* set) {
    ((UnicodeSet*) set)->UnicodeSet::compact();
}

U_CAPI void U_EXPORT2
uset_complement(USet* set) {
    ((UnicodeSet*) set)->UnicodeSet::complement();
}

U_CAPI void U_EXPORT2
uset_complementRange(USet *set, UChar32 start, UChar32 end) {
    ((UnicodeSet*) set)->UnicodeSet::complement(start, end);
}

U_CAPI void U_EXPORT2
uset_complementString(USet *set, const UChar *str, int32_t length) {
    UnicodeString s(length==-1, str, length);
    ((UnicodeSet*) set)->UnicodeSet::complement(s);
}

U_CAPI void U_EXPORT2
uset_complementAllCodePoints(USet *set, const UChar *str, int32_t length) {
    UnicodeString s(length==-1, str, length);
    ((UnicodeSet*) set)->UnicodeSet::complementAll(s);
}

U_CAPI void U_EXPORT2
uset_complementAll(USet* set, const USet* complement) {
    ((UnicodeSet*) set)->UnicodeSet::complementAll(*(const UnicodeSet*)complement);
}

U_CAPI void U_EXPORT2
uset_clear(USet* set) {
    ((UnicodeSet*) set)->UnicodeSet::clear();
}

U_CAPI void U_EXPORT2
uset_removeAllStrings(USet* set) {
    ((UnicodeSet*) set)->UnicodeSet::removeAllStrings();
}

U_CAPI UBool U_EXPORT2
uset_isEmpty(const USet* set) {
    return ((const UnicodeSet*) set)->UnicodeSet::isEmpty();
}

U_CAPI UBool U_EXPORT2
uset_hasStrings(const USet* set) {
    return ((const UnicodeSet*) set)->UnicodeSet::hasStrings();
}

U_CAPI UBool U_EXPORT2
uset_contains(const USet* set, UChar32 c) {
    return ((const UnicodeSet*) set)->UnicodeSet::contains(c);
}

U_CAPI UBool U_EXPORT2
uset_containsRange(const USet* set, UChar32 start, UChar32 end) {
    return ((const UnicodeSet*) set)->UnicodeSet::contains(start, end);
}

U_CAPI UBool U_EXPORT2
uset_containsString(const USet* set, const UChar* str, int32_t strLen) {
    UnicodeString s(strLen==-1, str, strLen);
    return ((const UnicodeSet*) set)->UnicodeSet::contains(s);
}

U_CAPI UBool U_EXPORT2
uset_containsAll(const USet* set1, const USet* set2) {
    return ((const UnicodeSet*) set1)->UnicodeSet::containsAll(* (const UnicodeSet*) set2);
}

U_CAPI UBool U_EXPORT2
uset_containsAllCodePoints(const USet* set, const UChar *str, int32_t strLen) {
    // Create a string alias, since nothing is being added to the set.
    UnicodeString s(strLen==-1, str, strLen);
    return ((const UnicodeSet*) set)->UnicodeSet::containsAll(s);
}

U_CAPI UBool U_EXPORT2
uset_containsNone(const USet* set1, const USet* set2) {
    return ((const UnicodeSet*) set1)->UnicodeSet::containsNone(* (const UnicodeSet*) set2);
}

U_CAPI UBool U_EXPORT2
uset_containsSome(const USet* set1, const USet* set2) {
    return ((const UnicodeSet*) set1)->UnicodeSet::containsSome(* (const UnicodeSet*) set2);
}

U_CAPI int32_t U_EXPORT2
uset_span(const USet *set, const UChar *s, int32_t length, USetSpanCondition spanCondition) {
    return ((UnicodeSet*) set)->UnicodeSet::span(s, length, spanCondition);
}

U_CAPI int32_t U_EXPORT2
uset_spanBack(const USet *set, const UChar *s, int32_t length, USetSpanCondition spanCondition) {
    return ((UnicodeSet*) set)->UnicodeSet::spanBack(s, length, spanCondition);
}

U_CAPI int32_t U_EXPORT2
uset_spanUTF8(const USet *set, const char *s, int32_t length, USetSpanCondition spanCondition) {
    return ((UnicodeSet*) set)->UnicodeSet::spanUTF8(s, length, spanCondition);
}

U_CAPI int32_t U_EXPORT2
uset_spanBackUTF8(const USet *set, const char *s, int32_t length, USetSpanCondition spanCondition) {
    return ((UnicodeSet*) set)->UnicodeSet::spanBackUTF8(s, length, spanCondition);
}

U_CAPI UBool U_EXPORT2
uset_equals(const USet* set1, const USet* set2) {
    return *(const UnicodeSet*)set1 == *(const UnicodeSet*)set2;
}

U_CAPI int32_t U_EXPORT2
uset_indexOf(const USet* set, UChar32 c) {
    return ((UnicodeSet*) set)->UnicodeSet::indexOf(c);
}

U_CAPI UChar32 U_EXPORT2
uset_charAt(const USet* set, int32_t index) {
    return ((UnicodeSet*) set)->UnicodeSet::charAt(index);
}

U_CAPI int32_t U_EXPORT2
uset_size(const USet* set) {
    return ((const UnicodeSet*) set)->UnicodeSet::size();
}

U_NAMESPACE_BEGIN
/**
 * This class only exists to provide access to the UnicodeSet private
 * USet support API.  Declaring a class a friend is more portable than
 * trying to declare extern "C" functions as friends.
 */
class USetAccess /* not : public UObject because all methods are static */ {
public:
    /* Try to have the compiler inline these*/
    inline static int32_t getStringCount(const UnicodeSet& set) {
        return set.stringsSize();
    }
    inline static const UnicodeString* getString(const UnicodeSet& set,
                                                 int32_t i) {
        return set.getString(i);
    }
private:
    /* do not instantiate*/
    USetAccess();
};
U_NAMESPACE_END

U_CAPI int32_t U_EXPORT2
uset_getRangeCount(const USet *set) {
    return ((const UnicodeSet *)set)->UnicodeSet::getRangeCount();
}

U_CAPI int32_t U_EXPORT2
uset_getItemCount(const USet* uset) {
    const UnicodeSet& set = *(const UnicodeSet*)uset;
    return set.getRangeCount() + USetAccess::getStringCount(set);
}

U_CAPI int32_t U_EXPORT2
uset_getItem(const USet* uset, int32_t itemIndex,
             UChar32* start, UChar32* end,
             UChar* str, int32_t strCapacity,
             UErrorCode* ec) {
    if (U_FAILURE(*ec)) return 0;
    const UnicodeSet& set = *(const UnicodeSet*)uset;
    int32_t rangeCount;

    if (itemIndex < 0) {
        *ec = U_ILLEGAL_ARGUMENT_ERROR;
        return -1;
    } else if (itemIndex < (rangeCount = set.getRangeCount())) {
        *start = set.getRangeStart(itemIndex);
        *end = set.getRangeEnd(itemIndex);
        return 0;
    } else {
        itemIndex -= rangeCount;
        if (itemIndex < USetAccess::getStringCount(set)) {
            const UnicodeString* s = USetAccess::getString(set, itemIndex);
            return s->extract(str, strCapacity, *ec);
        } else {
            *ec = U_INDEX_OUTOFBOUNDS_ERROR;
            return -1;
        }
    }
}

//U_CAPI UBool U_EXPORT2
//uset_getRange(const USet* set, int32_t rangeIndex,
//              UChar32* pStart, UChar32* pEnd) {
//    if ((uint32_t) rangeIndex >= (uint32_t) uset_getRangeCount(set)) {
//        return FALSE;
//    }
//    const UnicodeSet* us = (const UnicodeSet*) set;
//    *pStart = us->getRangeStart(rangeIndex);
//    *pEnd = us->getRangeEnd(rangeIndex);
//    return TRUE;
//}

/*
 * Serialize a USet into 16-bit units.
 * Store BMP code points as themselves with one 16-bit unit each.
 *
 * Important: the code points in the array are in ascending order,
 * therefore all BMP code points precede all supplementary code points.
 *
 * Store each supplementary code point in 2 16-bit units,
 * simply with higher-then-lower 16-bit halves.
 *
 * Precede the entire list with the length.
 * If there are supplementary code points, then set bit 15 in the length
 * and add the bmpLength between it and the array.
 *
 * In other words:
 * - all BMP:            (length=bmpLength) BMP, .., BMP
 * - some supplementary: (length|0x8000) (bmpLength<length) BMP, .., BMP, supp-high, supp-low, ..
 */
U_CAPI int32_t U_EXPORT2
uset_serialize(const USet* set, uint16_t* dest, int32_t destCapacity, UErrorCode* ec) {
    if (ec==NULL || U_FAILURE(*ec)) {
        return 0;
    }

    return ((const UnicodeSet*) set)->UnicodeSet::serialize(dest, destCapacity,* ec);
}

U_CAPI UBool U_EXPORT2
uset_getSerializedSet(USerializedSet* fillSet, const uint16_t* src, int32_t srcLength) {
    int32_t length;

    if(fillSet==NULL) {
        return FALSE;
    }
    if(src==NULL || srcLength<=0) {
        fillSet->length=fillSet->bmpLength=0;
        return FALSE;
    }

    length=*src++;
    if(length&0x8000) {
        /* there are supplementary values */
        length&=0x7fff;
        if(srcLength<(2+length)) {
            fillSet->length=fillSet->bmpLength=0;
            return FALSE;
        }
        fillSet->bmpLength=*src++;
    } else {
        /* only BMP values */
        if(srcLength<(1+length)) {
            fillSet->length=fillSet->bmpLength=0;
            return FALSE;
        }
        fillSet->bmpLength=length;
    }
    fillSet->array=src;
    fillSet->length=length;
    return TRUE;
}

U_CAPI void U_EXPORT2
uset_setSerializedToOne(USerializedSet* fillSet, UChar32 c) {
    if(fillSet==NULL || (uint32_t)c>0x10ffff) {
        return;
    }

    fillSet->array=fillSet->staticArray;
    if(c<0xffff) {
        fillSet->bmpLength=fillSet->length=2;
        fillSet->staticArray[0]=(uint16_t)c;
        fillSet->staticArray[1]=(uint16_t)c+1;
    } else if(c==0xffff) {
        fillSet->bmpLength=1;
        fillSet->length=3;
        fillSet->staticArray[0]=0xffff;
        fillSet->staticArray[1]=1;
        fillSet->staticArray[2]=0;
    } else if(c<0x10ffff) {
        fillSet->bmpLength=0;
        fillSet->length=4;
        fillSet->staticArray[0]=(uint16_t)(c>>16);
        fillSet->staticArray[1]=(uint16_t)c;
        ++c;
        fillSet->staticArray[2]=(uint16_t)(c>>16);
        fillSet->staticArray[3]=(uint16_t)c;
    } else /* c==0x10ffff */ {
        fillSet->bmpLength=0;
        fillSet->length=2;
        fillSet->staticArray[0]=0x10;
        fillSet->staticArray[1]=0xffff;
    }
}

U_CAPI UBool U_EXPORT2
uset_serializedContains(const USerializedSet* set, UChar32 c) {
    const uint16_t* array;

    if(set==NULL || (uint32_t)c>0x10ffff) {
        return FALSE;
    }

    array=set->array;
    if(c<=0xffff) {
        /* find c in the BMP part */
        int32_t lo = 0;
        int32_t hi = set->bmpLength-1;
        if (c < array[0]) {
            hi = 0;
        } else if (c < array[hi]) {
            for(;;) {
                int32_t i = (lo + hi) >> 1;
                if (i == lo) {
                    break;  // Done!
                } else if (c < array[i]) {
                    hi = i;
                } else {
                    lo = i;
                }
            }
        } else {
            hi += 1;
        }
        return (UBool)(hi&1);
    } else {
        /* find c in the supplementary part */
        uint16_t high=(uint16_t)(c>>16), low=(uint16_t)c;
        int32_t base = set->bmpLength;
        int32_t lo = 0;
        int32_t hi = set->length - 2 - base;
        if (high < array[base] || (high==array[base] && low<array[base+1])) {
            hi = 0;
        } else if (high < array[base+hi] || (high==array[base+hi] && low<array[base+hi+1])) {
            for (;;) {
                int32_t i = ((lo + hi) >> 1) & ~1;  // Guarantee even result
                int32_t iabs = i + base;
                if (i == lo) {
                    break;  // Done!
                } else if (high < array[iabs] || (high==array[iabs] && low<array[iabs+1])) {
                    hi = i;
                } else {
                    lo = i;
                }
            }
        } else {
            hi += 2;
        }
        /* count pairs of 16-bit units even per BMP and check if the number of pairs is odd */
        return (UBool)(((hi+(base<<1))&2)!=0);
    }
}

U_CAPI int32_t U_EXPORT2
uset_getSerializedRangeCount(const USerializedSet* set) {
    if(set==NULL) {
        return 0;
    }

    return (set->bmpLength+(set->length-set->bmpLength)/2+1)/2;
}

U_CAPI UBool U_EXPORT2
uset_getSerializedRange(const USerializedSet* set, int32_t rangeIndex,
                        UChar32* pStart, UChar32* pEnd) {
    const uint16_t* array;
    int32_t bmpLength, length;

    if(set==NULL || rangeIndex<0 || pStart==NULL || pEnd==NULL) {
        return FALSE;
    }

    array=set->array;
    length=set->length;
    bmpLength=set->bmpLength;

    rangeIndex*=2; /* address start/limit pairs */
    if(rangeIndex<bmpLength) {
        *pStart=array[rangeIndex++];
        if(rangeIndex<bmpLength) {
            *pEnd=array[rangeIndex]-1;
        } else if(rangeIndex<length) {
            *pEnd=((((int32_t)array[rangeIndex])<<16)|array[rangeIndex+1])-1;
        } else {
            *pEnd=0x10ffff;
        }
        return TRUE;
    } else {
        rangeIndex-=bmpLength;
        rangeIndex*=2; /* address pairs of pairs of units */
        length-=bmpLength;
        if(rangeIndex<length) {
            array+=bmpLength;
            *pStart=(((int32_t)array[rangeIndex])<<16)|array[rangeIndex+1];
            rangeIndex+=2;
            if(rangeIndex<length) {
                *pEnd=((((int32_t)array[rangeIndex])<<16)|array[rangeIndex+1])-1;
            } else {
                *pEnd=0x10ffff;
            }
            return TRUE;
        } else {
            return FALSE;
        }
    }
}

// TODO The old, internal uset.c had an efficient uset_containsOne function.
// Returned the one and only code point, or else -1 or something.
// Consider adding such a function to both C and C++ UnicodeSet/uset.
// See tools/gennorm/store.c for usage, now usetContainsOne there.

// TODO Investigate incorporating this code into UnicodeSet to improve
// efficiency.
// ---
// #define USET_GROW_DELTA 20
// 
// static int32_t
// findChar(const UChar32* array, int32_t length, UChar32 c) {
//     int32_t i;
// 
//     /* check the last range limit first for more efficient appending */
//     if(length>0) {
//         if(c>=array[length-1]) {
//             return length;
//         }
// 
//         /* do not check the last range limit again in the loop below */
//         --length;
//     }
// 
//     for(i=0; i<length && c>=array[i]; ++i) {}
//     return i;
// }
// 
// static UBool
// addRemove(USet* set, UChar32 c, int32_t doRemove) {
//     int32_t i, length, more;
// 
//     if(set==NULL || (uint32_t)c>0x10ffff) {
//         return FALSE;
//     }
// 
//     length=set->length;
//     i=findChar(set->array, length, c);
//     if((i&1)^doRemove) {
//         /* c is already in the set */
//         return TRUE;
//     }
// 
//     /* how many more array items do we need? */
//     if(i<length && (c+1)==set->array[i]) {
//         /* c is just before the following range, extend that in-place by one */
//         set->array[i]=c;
//         if(i>0) {
//             --i;
//             if(c==set->array[i]) {
//                 /* the previous range collapsed, remove it */
//                 set->length=length-=2;
//                 if(i<length) {
//                     uprv_memmove(set->array+i, set->array+i+2, (length-i)*4);
//                 }
//             }
//         }
//         return TRUE;
//     } else if(i>0 && c==set->array[i-1]) {
//         /* c is just after the previous range, extend that in-place by one */
//         if(++c<=0x10ffff) {
//             set->array[i-1]=c;
//             if(i<length && c==set->array[i]) {
//                 /* the following range collapsed, remove it */
//                 --i;
//                 set->length=length-=2;
//                 if(i<length) {
//                     uprv_memmove(set->array+i, set->array+i+2, (length-i)*4);
//                 }
//             }
//         } else {
//             /* extend the previous range (had limit 0x10ffff) to the end of Unicode */
//             set->length=i-1;
//         }
//         return TRUE;
//     } else if(i==length && c==0x10ffff) {
//         /* insert one range limit c */
//         more=1;
//     } else {
//         /* insert two range limits c, c+1 */
//         more=2;
//     }
// 
//     /* insert <more> range limits */
//     if(length+more>set->capacity) {
//         /* reallocate */
//         int32_t newCapacity=set->capacity+set->capacity/2+USET_GROW_DELTA;
//         UChar32* newArray=(UChar32* )uprv_malloc(newCapacity*4);
//         if(newArray==NULL) {
//             return FALSE;
//         }
//         set->capacity=newCapacity;
//         uprv_memcpy(newArray, set->array, length*4);
// 
//         if(set->array!=set->staticBuffer) {
//             uprv_free(set->array);
//         }
//         set->array=newArray;
//     }
// 
//     if(i<length) {
//         uprv_memmove(set->array+i+more, set->array+i, (length-i)*4);
//     }
//     set->array[i]=c;
//     if(more==2) {
//         set->array[i+1]=c+1;
//     }
//     set->length+=more;
// 
//     return TRUE;
// }
// 
// U_CAPI UBool U_EXPORT2
// uset_add(USet* set, UChar32 c) {
//     return addRemove(set, c, 0);
// }
// 
// U_CAPI void U_EXPORT2
// uset_remove(USet* set, UChar32 c) {
//     addRemove(set, c, 1);
// }

Coverage Report

Created: 2025-01-28 06:38

Line	Count	Source (jump to first uncovered line)
1		// © 2016 and later: Unicode, Inc. and others.
2		// License & terms of use: http://www.unicode.org/copyright.html
3		/*
4		*******************************************************************************
5		*
6		* Copyright (C) 2002-2011, International Business Machines
7		* Corporation and others. All Rights Reserved.
8		*
9		*******************************************************************************
10		* file name: uset.cpp
11		* encoding: UTF-8
12		* tab size: 8 (not used)
13		* indentation:4
14		*
15		* created on: 2002mar07
16		* created by: Markus W. Scherer
17		*
18		* There are functions to efficiently serialize a USet into an array of uint16_t
19		* and functions to use such a serialized form efficiently without
20		* instantiating a new USet.
21		*/
22
23		#include "unicode/utypes.h"
24		#include "unicode/uobject.h"
25		#include "unicode/uset.h"
26		#include "unicode/uniset.h"
27		#include "cmemory.h"
28		#include "unicode/ustring.h"
29		#include "unicode/parsepos.h"
30
31		U_NAMESPACE_USE
32
33		U_CAPI USet* U_EXPORT2
34	0	uset_openEmpty() {
35	0	return (USet*) new UnicodeSet();
36	0	}
37
38		U_CAPI USet* U_EXPORT2
39	0	uset_open(UChar32 start, UChar32 end) {
40	0	return (USet*) new UnicodeSet(start, end);
41	0	}
42
43		U_CAPI void U_EXPORT2
44	0	uset_close(USet* set) {
45	0	delete (UnicodeSet*) set;
46	0	}
47
48		U_CAPI USet * U_EXPORT2
49	0	uset_clone(const USet *set) {
50	0	return (USet) (((UnicodeSet) set)->UnicodeSet::clone());
51	0	}
52
53		U_CAPI UBool U_EXPORT2
54	0	uset_isFrozen(const USet *set) {
55	0	return ((UnicodeSet*) set)->UnicodeSet::isFrozen();
56	0	}
57
58		U_CAPI void U_EXPORT2
59	0	uset_freeze(USet *set) {
60	0	((UnicodeSet*) set)->UnicodeSet::freeze();
61	0	}
62
63		U_CAPI USet * U_EXPORT2
64	0	uset_cloneAsThawed(const USet *set) {
65	0	return (USet) (((UnicodeSet) set)->UnicodeSet::cloneAsThawed());
66	0	}
67
68		U_CAPI void U_EXPORT2
69		uset_set(USet* set,
70	0	UChar32 start, UChar32 end) {
71	0	((UnicodeSet*) set)->UnicodeSet::set(start, end);
72	0	}
73
74		U_CAPI void U_EXPORT2
75	0	uset_addAll(USet* set, const USet *additionalSet) {
76	0	((UnicodeSet) set)->UnicodeSet::addAll(((const UnicodeSet*)additionalSet));
77	0	}
78
79		U_CAPI void U_EXPORT2
80	0	uset_add(USet* set, UChar32 c) {
81	0	((UnicodeSet*) set)->UnicodeSet::add(c);
82	0	}
83
84		U_CAPI void U_EXPORT2
85	0	uset_addRange(USet* set, UChar32 start, UChar32 end) {
86	0	((UnicodeSet*) set)->UnicodeSet::add(start, end);
87	0	}
88
89		U_CAPI void U_EXPORT2
90	0	uset_addString(USet* set, const UChar* str, int32_t strLen) {
91		// UnicodeString handles -1 for strLen
92	0	UnicodeString s(strLen<0, str, strLen);
93	0	((UnicodeSet*) set)->UnicodeSet::add(s);
94	0	}
95
96		U_CAPI void U_EXPORT2
97	0	uset_addAllCodePoints(USet* set, const UChar *str, int32_t strLen) {
98		// UnicodeString handles -1 for strLen
99	0	UnicodeString s(str, strLen);
100	0	((UnicodeSet*) set)->UnicodeSet::addAll(s);
101	0	}
102
103		U_CAPI void U_EXPORT2
104	0	uset_remove(USet* set, UChar32 c) {
105	0	((UnicodeSet*) set)->UnicodeSet::remove(c);
106	0	}
107
108		U_CAPI void U_EXPORT2
109	0	uset_removeRange(USet* set, UChar32 start, UChar32 end) {
110	0	((UnicodeSet*) set)->UnicodeSet::remove(start, end);
111	0	}
112
113		U_CAPI void U_EXPORT2
114	0	uset_removeString(USet* set, const UChar* str, int32_t strLen) {
115	0	UnicodeString s(strLen==-1, str, strLen);
116	0	((UnicodeSet*) set)->UnicodeSet::remove(s);
117	0	}
118
119		U_CAPI void U_EXPORT2
120	0	uset_removeAllCodePoints(USet set, const UChar str, int32_t length) {
121	0	UnicodeString s(length==-1, str, length);
122	0	((UnicodeSet*) set)->UnicodeSet::removeAll(s);
123	0	}
124
125		U_CAPI void U_EXPORT2
126	0	uset_removeAll(USet* set, const USet* remove) {
127	0	((UnicodeSet) set)->UnicodeSet::removeAll((const UnicodeSet*)remove);
128	0	}
129
130		U_CAPI void U_EXPORT2
131	0	uset_retain(USet* set, UChar32 start, UChar32 end) {
132	0	((UnicodeSet*) set)->UnicodeSet::retain(start, end);
133	0	}
134
135		U_CAPI void U_EXPORT2
136	0	uset_retainString(USet set, const UChar str, int32_t length) {
137	0	UnicodeString s(length==-1, str, length);
138	0	((UnicodeSet*) set)->UnicodeSet::retain(s);
139	0	}
140
141		U_CAPI void U_EXPORT2
142	0	uset_retainAllCodePoints(USet set, const UChar str, int32_t length) {
143	0	UnicodeString s(length==-1, str, length);
144	0	((UnicodeSet*) set)->UnicodeSet::retainAll(s);
145	0	}
146
147		U_CAPI void U_EXPORT2
148	0	uset_retainAll(USet* set, const USet* retain) {
149	0	((UnicodeSet) set)->UnicodeSet::retainAll((const UnicodeSet*)retain);
150	0	}
151
152		U_CAPI void U_EXPORT2
153	0	uset_compact(USet* set) {
154	0	((UnicodeSet*) set)->UnicodeSet::compact();
155	0	}
156
157		U_CAPI void U_EXPORT2
158	0	uset_complement(USet* set) {
159	0	((UnicodeSet*) set)->UnicodeSet::complement();
160	0	}
161
162		U_CAPI void U_EXPORT2
163	0	uset_complementRange(USet *set, UChar32 start, UChar32 end) {
164	0	((UnicodeSet*) set)->UnicodeSet::complement(start, end);
165	0	}
166
167		U_CAPI void U_EXPORT2
168	0	uset_complementString(USet set, const UChar str, int32_t length) {
169	0	UnicodeString s(length==-1, str, length);
170	0	((UnicodeSet*) set)->UnicodeSet::complement(s);
171	0	}
172
173		U_CAPI void U_EXPORT2
174	0	uset_complementAllCodePoints(USet set, const UChar str, int32_t length) {
175	0	UnicodeString s(length==-1, str, length);
176	0	((UnicodeSet*) set)->UnicodeSet::complementAll(s);
177	0	}
178
179		U_CAPI void U_EXPORT2
180	0	uset_complementAll(USet* set, const USet* complement) {
181	0	((UnicodeSet) set)->UnicodeSet::complementAll((const UnicodeSet*)complement);
182	0	}
183
184		U_CAPI void U_EXPORT2
185	0	uset_clear(USet* set) {
186	0	((UnicodeSet*) set)->UnicodeSet::clear();
187	0	}
188
189		U_CAPI void U_EXPORT2
190	0	uset_removeAllStrings(USet* set) {
191	0	((UnicodeSet*) set)->UnicodeSet::removeAllStrings();
192	0	}
193
194		U_CAPI UBool U_EXPORT2
195	0	uset_isEmpty(const USet* set) {
196	0	return ((const UnicodeSet*) set)->UnicodeSet::isEmpty();
197	0	}
198
199		U_CAPI UBool U_EXPORT2
200	0	uset_hasStrings(const USet* set) {
201	0	return ((const UnicodeSet*) set)->UnicodeSet::hasStrings();
202	0	}
203
204		U_CAPI UBool U_EXPORT2
205	0	uset_contains(const USet* set, UChar32 c) {
206	0	return ((const UnicodeSet*) set)->UnicodeSet::contains(c);
207	0	}
208
209		U_CAPI UBool U_EXPORT2
210	0	uset_containsRange(const USet* set, UChar32 start, UChar32 end) {
211	0	return ((const UnicodeSet*) set)->UnicodeSet::contains(start, end);
212	0	}
213
214		U_CAPI UBool U_EXPORT2
215	0	uset_containsString(const USet* set, const UChar* str, int32_t strLen) {
216	0	UnicodeString s(strLen==-1, str, strLen);
217	0	return ((const UnicodeSet*) set)->UnicodeSet::contains(s);
218	0	}
219
220		U_CAPI UBool U_EXPORT2
221	0	uset_containsAll(const USet* set1, const USet* set2) {
222	0	return ((const UnicodeSet) set1)->UnicodeSet::containsAll( (const UnicodeSet*) set2);
223	0	}
224
225		U_CAPI UBool U_EXPORT2
226	0	uset_containsAllCodePoints(const USet* set, const UChar *str, int32_t strLen) {
227		// Create a string alias, since nothing is being added to the set.
228	0	UnicodeString s(strLen==-1, str, strLen);
229	0	return ((const UnicodeSet*) set)->UnicodeSet::containsAll(s);
230	0	}
231
232		U_CAPI UBool U_EXPORT2
233	0	uset_containsNone(const USet* set1, const USet* set2) {
234	0	return ((const UnicodeSet) set1)->UnicodeSet::containsNone( (const UnicodeSet*) set2);
235	0	}
236
237		U_CAPI UBool U_EXPORT2
238	0	uset_containsSome(const USet* set1, const USet* set2) {
239	0	return ((const UnicodeSet) set1)->UnicodeSet::containsSome( (const UnicodeSet*) set2);
240	0	}
241
242		U_CAPI int32_t U_EXPORT2
243	0	uset_span(const USet set, const UChar s, int32_t length, USetSpanCondition spanCondition) {
244	0	return ((UnicodeSet*) set)->UnicodeSet::span(s, length, spanCondition);
245	0	}
246
247		U_CAPI int32_t U_EXPORT2
248	0	uset_spanBack(const USet set, const UChar s, int32_t length, USetSpanCondition spanCondition) {
249	0	return ((UnicodeSet*) set)->UnicodeSet::spanBack(s, length, spanCondition);
250	0	}
251
252		U_CAPI int32_t U_EXPORT2
253	0	uset_spanUTF8(const USet set, const char s, int32_t length, USetSpanCondition spanCondition) {
254	0	return ((UnicodeSet*) set)->UnicodeSet::spanUTF8(s, length, spanCondition);
255	0	}
256
257		U_CAPI int32_t U_EXPORT2
258	0	uset_spanBackUTF8(const USet set, const char s, int32_t length, USetSpanCondition spanCondition) {
259	0	return ((UnicodeSet*) set)->UnicodeSet::spanBackUTF8(s, length, spanCondition);
260	0	}
261
262		U_CAPI UBool U_EXPORT2
263	0	uset_equals(const USet* set1, const USet* set2) {
264	0	return (const UnicodeSet)set1 == (const UnicodeSet)set2;
265	0	}
266
267		U_CAPI int32_t U_EXPORT2
268	0	uset_indexOf(const USet* set, UChar32 c) {
269	0	return ((UnicodeSet*) set)->UnicodeSet::indexOf(c);
270	0	}
271
272		U_CAPI UChar32 U_EXPORT2
273	0	uset_charAt(const USet* set, int32_t index) {
274	0	return ((UnicodeSet*) set)->UnicodeSet::charAt(index);
275	0	}
276
277		U_CAPI int32_t U_EXPORT2
278	0	uset_size(const USet* set) {
279	0	return ((const UnicodeSet*) set)->UnicodeSet::size();
280	0	}
281
282		U_NAMESPACE_BEGIN
283		/**
284		* This class only exists to provide access to the UnicodeSet private
285		* USet support API. Declaring a class a friend is more portable than
286		* trying to declare extern "C" functions as friends.
287		*/
288		class USetAccess /* not : public UObject because all methods are static */ {
289		public:
290		/* Try to have the compiler inline these*/
291	0	inline static int32_t getStringCount(const UnicodeSet& set) {
292	0	return set.stringsSize();
293	0	}
294		inline static const UnicodeString* getString(const UnicodeSet& set,
295	0	int32_t i) {
296	0	return set.getString(i);
297	0	}
298		private:
299		/* do not instantiate*/
300		USetAccess();
301		};
302		U_NAMESPACE_END
303
304		U_CAPI int32_t U_EXPORT2
305	0	uset_getRangeCount(const USet *set) {
306	0	return ((const UnicodeSet *)set)->UnicodeSet::getRangeCount();
307	0	}
308
309		U_CAPI int32_t U_EXPORT2
310	0	uset_getItemCount(const USet* uset) {
311	0	const UnicodeSet& set = (const UnicodeSet)uset;
312	0	return set.getRangeCount() + USetAccess::getStringCount(set);
313	0	}
314
315		U_CAPI int32_t U_EXPORT2
316		uset_getItem(const USet* uset, int32_t itemIndex,
317		UChar32* start, UChar32* end,
318		UChar* str, int32_t strCapacity,
319	0	UErrorCode* ec) {
320	0	if (U_FAILURE(*ec)) return 0;
321	0	const UnicodeSet& set = (const UnicodeSet)uset;
322	0	int32_t rangeCount;
323
324	0	if (itemIndex < 0) {
325	0	*ec = U_ILLEGAL_ARGUMENT_ERROR;
326	0	return -1;
327	0	} else if (itemIndex < (rangeCount = set.getRangeCount())) {
328	0	*start = set.getRangeStart(itemIndex);
329	0	*end = set.getRangeEnd(itemIndex);
330	0	return 0;
331	0	} else {
332	0	itemIndex -= rangeCount;
333	0	if (itemIndex < USetAccess::getStringCount(set)) {
334	0	const UnicodeString* s = USetAccess::getString(set, itemIndex);
335	0	return s->extract(str, strCapacity, *ec);
336	0	} else {
337	0	*ec = U_INDEX_OUTOFBOUNDS_ERROR;
338	0	return -1;
339	0	}
340	0	}
341	0	}
342
343		//U_CAPI UBool U_EXPORT2
344		//uset_getRange(const USet* set, int32_t rangeIndex,
345		// UChar32* pStart, UChar32* pEnd) {
346		// if ((uint32_t) rangeIndex >= (uint32_t) uset_getRangeCount(set)) {
347		// return FALSE;
348		// }
349		// const UnicodeSet* us = (const UnicodeSet*) set;
350		// *pStart = us->getRangeStart(rangeIndex);
351		// *pEnd = us->getRangeEnd(rangeIndex);
352		// return TRUE;
353		//}
354
355		/*
356		* Serialize a USet into 16-bit units.
357		* Store BMP code points as themselves with one 16-bit unit each.
358		*
359		* Important: the code points in the array are in ascending order,
360		* therefore all BMP code points precede all supplementary code points.
361		*
362		* Store each supplementary code point in 2 16-bit units,
363		* simply with higher-then-lower 16-bit halves.
364		*
365		* Precede the entire list with the length.
366		* If there are supplementary code points, then set bit 15 in the length
367		* and add the bmpLength between it and the array.
368		*
369		* In other words:
370		* - all BMP: (length=bmpLength) BMP, .., BMP
371		* - some supplementary: (length\|0x8000) (bmpLength<length) BMP, .., BMP, supp-high, supp-low, ..
372		*/
373		U_CAPI int32_t U_EXPORT2
374	0	uset_serialize(const USet* set, uint16_t* dest, int32_t destCapacity, UErrorCode* ec) {
375	0	if (ec==NULL \|\| U_FAILURE(*ec)) {
376	0	return 0;
377	0	}
378
379	0	return ((const UnicodeSet) set)->UnicodeSet::serialize(dest, destCapacity, ec);
380	0	}
381
382		U_CAPI UBool U_EXPORT2
383	0	uset_getSerializedSet(USerializedSet* fillSet, const uint16_t* src, int32_t srcLength) {
384	0	int32_t length;
385
386	0	if(fillSet==NULL) {
387	0	return FALSE;
388	0	}
389	0	if(src==NULL \|\| srcLength<=0) {
390	0	fillSet->length=fillSet->bmpLength=0;
391	0	return FALSE;
392	0	}
393
394	0	length=*src++;
395	0	if(length&0x8000) {
396		/* there are supplementary values */
397	0	length&=0x7fff;
398	0	if(srcLength<(2+length)) {
399	0	fillSet->length=fillSet->bmpLength=0;
400	0	return FALSE;
401	0	}
402	0	fillSet->bmpLength=*src++;
403	0	} else {
404		/* only BMP values */
405	0	if(srcLength<(1+length)) {
406	0	fillSet->length=fillSet->bmpLength=0;
407	0	return FALSE;
408	0	}
409	0	fillSet->bmpLength=length;
410	0	}
411	0	fillSet->array=src;
412	0	fillSet->length=length;
413	0	return TRUE;
414	0	}
415
416		U_CAPI void U_EXPORT2
417	0	uset_setSerializedToOne(USerializedSet* fillSet, UChar32 c) {
418	0	if(fillSet==NULL \|\| (uint32_t)c>0x10ffff) {
419	0	return;
420	0	}
421
422	0	fillSet->array=fillSet->staticArray;
423	0	if(c<0xffff) {
424	0	fillSet->bmpLength=fillSet->length=2;
425	0	fillSet->staticArray[0]=(uint16_t)c;
426	0	fillSet->staticArray[1]=(uint16_t)c+1;
427	0	} else if(c==0xffff) {
428	0	fillSet->bmpLength=1;
429	0	fillSet->length=3;
430	0	fillSet->staticArray[0]=0xffff;
431	0	fillSet->staticArray[1]=1;
432	0	fillSet->staticArray[2]=0;
433	0	} else if(c<0x10ffff) {
434	0	fillSet->bmpLength=0;
435	0	fillSet->length=4;
436	0	fillSet->staticArray[0]=(uint16_t)(c>>16);
437	0	fillSet->staticArray[1]=(uint16_t)c;
438	0	++c;
439	0	fillSet->staticArray[2]=(uint16_t)(c>>16);
440	0	fillSet->staticArray[3]=(uint16_t)c;
441	0	} else /* c==0x10ffff */ {
442	0	fillSet->bmpLength=0;
443	0	fillSet->length=2;
444	0	fillSet->staticArray[0]=0x10;
445	0	fillSet->staticArray[1]=0xffff;
446	0	}
447	0	}
448
449		U_CAPI UBool U_EXPORT2
450	0	uset_serializedContains(const USerializedSet* set, UChar32 c) {
451	0	const uint16_t* array;
452
453	0	if(set==NULL \|\| (uint32_t)c>0x10ffff) {
454	0	return FALSE;
455	0	}
456
457	0	array=set->array;
458	0	if(c<=0xffff) {
459		/* find c in the BMP part */
460	0	int32_t lo = 0;
461	0	int32_t hi = set->bmpLength-1;
462	0	if (c < array[0]) {
463	0	hi = 0;
464	0	} else if (c < array[hi]) {
465	0	for(;;) {
466	0	int32_t i = (lo + hi) >> 1;
467	0	if (i == lo) {
468	0	break; // Done!
469	0	} else if (c < array[i]) {
470	0	hi = i;
471	0	} else {
472	0	lo = i;
473	0	}
474	0	}
475	0	} else {
476	0	hi += 1;
477	0	}
478	0	return (UBool)(hi&1);
479	0	} else {
480		/* find c in the supplementary part */
481	0	uint16_t high=(uint16_t)(c>>16), low=(uint16_t)c;
482	0	int32_t base = set->bmpLength;
483	0	int32_t lo = 0;
484	0	int32_t hi = set->length - 2 - base;
485	0	if (high < array[base] \|\| (high==array[base] && low<array[base+1])) {
486	0	hi = 0;
487	0	} else if (high < array[base+hi] \|\| (high==array[base+hi] && low<array[base+hi+1])) {
488	0	for (;;) {
489	0	int32_t i = ((lo + hi) >> 1) & ~1; // Guarantee even result
490	0	int32_t iabs = i + base;
491	0	if (i == lo) {
492	0	break; // Done!
493	0	} else if (high < array[iabs] \|\| (high==array[iabs] && low<array[iabs+1])) {
494	0	hi = i;
495	0	} else {
496	0	lo = i;
497	0	}
498	0	}
499	0	} else {
500	0	hi += 2;
501	0	}
502		/* count pairs of 16-bit units even per BMP and check if the number of pairs is odd */
503	0	return (UBool)(((hi+(base<<1))&2)!=0);
504	0	}
505	0	}
506
507		U_CAPI int32_t U_EXPORT2
508	0	uset_getSerializedRangeCount(const USerializedSet* set) {
509	0	if(set==NULL) {
510	0	return 0;
511	0	}
512
513	0	return (set->bmpLength+(set->length-set->bmpLength)/2+1)/2;
514	0	}
515
516		U_CAPI UBool U_EXPORT2
517		uset_getSerializedRange(const USerializedSet* set, int32_t rangeIndex,
518	0	UChar32* pStart, UChar32* pEnd) {
519	0	const uint16_t* array;
520	0	int32_t bmpLength, length;
521
522	0	if(set==NULL \|\| rangeIndex<0 \|\| pStart==NULL \|\| pEnd==NULL) {
523	0	return FALSE;
524	0	}
525
526	0	array=set->array;
527	0	length=set->length;
528	0	bmpLength=set->bmpLength;
529
530	0	rangeIndex=2; / address start/limit pairs */
531	0	if(rangeIndex<bmpLength) {
532	0	*pStart=array[rangeIndex++];
533	0	if(rangeIndex<bmpLength) {
534	0	*pEnd=array[rangeIndex]-1;
535	0	} else if(rangeIndex<length) {
536	0	*pEnd=((((int32_t)array[rangeIndex])<<16)\|array[rangeIndex+1])-1;
537	0	} else {
538	0	*pEnd=0x10ffff;
539	0	}
540	0	return TRUE;
541	0	} else {
542	0	rangeIndex-=bmpLength;
543	0	rangeIndex=2; / address pairs of pairs of units */
544	0	length-=bmpLength;
545	0	if(rangeIndex<length) {
546	0	array+=bmpLength;
547	0	*pStart=(((int32_t)array[rangeIndex])<<16)\|array[rangeIndex+1];
548	0	rangeIndex+=2;
549	0	if(rangeIndex<length) {
550	0	*pEnd=((((int32_t)array[rangeIndex])<<16)\|array[rangeIndex+1])-1;
551	0	} else {
552	0	*pEnd=0x10ffff;
553	0	}
554	0	return TRUE;
555	0	} else {
556	0	return FALSE;
557	0	}
558	0	}
559	0	}
560
561		// TODO The old, internal uset.c had an efficient uset_containsOne function.
562		// Returned the one and only code point, or else -1 or something.
563		// Consider adding such a function to both C and C++ UnicodeSet/uset.
564		// See tools/gennorm/store.c for usage, now usetContainsOne there.
565
566		// TODO Investigate incorporating this code into UnicodeSet to improve
567		// efficiency.
568		// ---
569		// #define USET_GROW_DELTA 20
570		//
571		// static int32_t
572		// findChar(const UChar32* array, int32_t length, UChar32 c) {
573		// int32_t i;
574		//
575		// /* check the last range limit first for more efficient appending */
576		// if(length>0) {
577		// if(c>=array[length-1]) {
578		// return length;
579		// }
580		//
581		// /* do not check the last range limit again in the loop below */
582		// --length;
583		// }
584		//
585		// for(i=0; i<length && c>=array[i]; ++i) {}
586		// return i;
587		// }
588		//
589		// static UBool
590		// addRemove(USet* set, UChar32 c, int32_t doRemove) {
591		// int32_t i, length, more;
592		//
593		// if(set==NULL \|\| (uint32_t)c>0x10ffff) {
594		// return FALSE;
595		// }
596		//
597		// length=set->length;
598		// i=findChar(set->array, length, c);
599		// if((i&1)^doRemove) {
600		// /* c is already in the set */
601		// return TRUE;
602		// }
603		//
604		// /* how many more array items do we need? */
605		// if(i<length && (c+1)==set->array[i]) {
606		// /* c is just before the following range, extend that in-place by one */
607		// set->array[i]=c;
608		// if(i>0) {
609		// --i;
610		// if(c==set->array[i]) {
611		// /* the previous range collapsed, remove it */
612		// set->length=length-=2;
613		// if(i<length) {
614		// uprv_memmove(set->array+i, set->array+i+2, (length-i)*4);
615		// }
616		// }
617		// }
618		// return TRUE;
619		// } else if(i>0 && c==set->array[i-1]) {
620		// /* c is just after the previous range, extend that in-place by one */
621		// if(++c<=0x10ffff) {
622		// set->array[i-1]=c;
623		// if(i<length && c==set->array[i]) {
624		// /* the following range collapsed, remove it */
625		// --i;
626		// set->length=length-=2;
627		// if(i<length) {
628		// uprv_memmove(set->array+i, set->array+i+2, (length-i)*4);
629		// }
630		// }
631		// } else {
632		// /* extend the previous range (had limit 0x10ffff) to the end of Unicode */
633		// set->length=i-1;
634		// }
635		// return TRUE;
636		// } else if(i==length && c==0x10ffff) {
637		// /* insert one range limit c */
638		// more=1;
639		// } else {
640		// /* insert two range limits c, c+1 */
641		// more=2;
642		// }
643		//
644		// /* insert <more> range limits */
645		// if(length+more>set->capacity) {
646		// /* reallocate */
647		// int32_t newCapacity=set->capacity+set->capacity/2+USET_GROW_DELTA;
648		// UChar32* newArray=(UChar32* )uprv_malloc(newCapacity*4);
649		// if(newArray==NULL) {
650		// return FALSE;
651		// }
652		// set->capacity=newCapacity;
653		// uprv_memcpy(newArray, set->array, length*4);
654		//
655		// if(set->array!=set->staticBuffer) {
656		// uprv_free(set->array);
657		// }
658		// set->array=newArray;
659		// }
660		//
661		// if(i<length) {
662		// uprv_memmove(set->array+i+more, set->array+i, (length-i)*4);
663		// }
664		// set->array[i]=c;
665		// if(more==2) {
666		// set->array[i+1]=c+1;
667		// }
668		// set->length+=more;
669		//
670		// return TRUE;
671		// }
672		//
673		// U_CAPI UBool U_EXPORT2
674		// uset_add(USet* set, UChar32 c) {
675		// return addRemove(set, c, 0);
676		// }
677		//
678		// U_CAPI void U_EXPORT2
679		// uset_remove(USet* set, UChar32 c) {
680		// addRemove(set, c, 1);
681		// }