chromium/third_party/icu/source/common/uset.cpp

// © 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() { … }

U_CAPI USet* U_EXPORT2
uset_open(UChar32 start, UChar32 end) { … }

U_CAPI void U_EXPORT2
uset_close(USet* set) { … }

U_CAPI USet * U_EXPORT2
uset_clone(const USet *set) { … }

U_CAPI UBool U_EXPORT2
uset_isFrozen(const USet *set) { … }

U_CAPI void U_EXPORT2
uset_freeze(USet *set) { … }

U_CAPI USet * U_EXPORT2
uset_cloneAsThawed(const USet *set) { … }

U_CAPI void U_EXPORT2
uset_set(USet* set,
     UChar32 start, UChar32 end) { … }

U_CAPI void U_EXPORT2
uset_addAll(USet* set, const USet *additionalSet) { … }

U_CAPI void U_EXPORT2
uset_add(USet* set, UChar32 c) { … }

U_CAPI void U_EXPORT2
uset_addRange(USet* set, UChar32 start, UChar32 end) { … }

U_CAPI void U_EXPORT2
uset_addString(USet* set, const char16_t* str, int32_t strLen) { … }

U_CAPI void U_EXPORT2
uset_addAllCodePoints(USet* set, const char16_t *str, int32_t strLen) { … }

U_CAPI void U_EXPORT2
uset_remove(USet* set, UChar32 c) { … }

U_CAPI void U_EXPORT2
uset_removeRange(USet* set, UChar32 start, UChar32 end) { … }

U_CAPI void U_EXPORT2
uset_removeString(USet* set, const char16_t* str, int32_t strLen) { … }

U_CAPI void U_EXPORT2
uset_removeAllCodePoints(USet *set, const char16_t *str, int32_t length) { … }

U_CAPI void U_EXPORT2
uset_removeAll(USet* set, const USet* remove) { … }

U_CAPI void U_EXPORT2
uset_retain(USet* set, UChar32 start, UChar32 end) { … }

U_CAPI void U_EXPORT2
uset_retainString(USet *set, const char16_t *str, int32_t length) { … }

U_CAPI void U_EXPORT2
uset_retainAllCodePoints(USet *set, const char16_t *str, int32_t length) { … }

U_CAPI void U_EXPORT2
uset_retainAll(USet* set, const USet* retain) { … }

U_CAPI void U_EXPORT2
uset_compact(USet* set) { … }

U_CAPI void U_EXPORT2
uset_complement(USet* set) { … }

U_CAPI void U_EXPORT2
uset_complementRange(USet *set, UChar32 start, UChar32 end) { … }

U_CAPI void U_EXPORT2
uset_complementString(USet *set, const char16_t *str, int32_t length) { … }

U_CAPI void U_EXPORT2
uset_complementAllCodePoints(USet *set, const char16_t *str, int32_t length) { … }

U_CAPI void U_EXPORT2
uset_complementAll(USet* set, const USet* complement) { … }

U_CAPI void U_EXPORT2
uset_clear(USet* set) { … }

U_CAPI void U_EXPORT2
uset_removeAllStrings(USet* set) { … }

U_CAPI UBool U_EXPORT2
uset_isEmpty(const USet* set) { … }

U_CAPI UBool U_EXPORT2
uset_hasStrings(const USet* set) { … }

U_CAPI UBool U_EXPORT2
uset_contains(const USet* set, UChar32 c) { … }

U_CAPI UBool U_EXPORT2
uset_containsRange(const USet* set, UChar32 start, UChar32 end) { … }

U_CAPI UBool U_EXPORT2
uset_containsString(const USet* set, const char16_t* str, int32_t strLen) { … }

U_CAPI UBool U_EXPORT2
uset_containsAll(const USet* set1, const USet* set2) { … }

U_CAPI UBool U_EXPORT2
uset_containsAllCodePoints(const USet* set, const char16_t *str, int32_t strLen) { … }

U_CAPI UBool U_EXPORT2
uset_containsNone(const USet* set1, const USet* set2) { … }

U_CAPI UBool U_EXPORT2
uset_containsSome(const USet* set1, const USet* set2) { … }

U_CAPI int32_t U_EXPORT2
uset_span(const USet *set, const char16_t *s, int32_t length, USetSpanCondition spanCondition) { … }

U_CAPI int32_t U_EXPORT2
uset_spanBack(const USet *set, const char16_t *s, int32_t length, USetSpanCondition spanCondition) { … }

U_CAPI int32_t U_EXPORT2
uset_spanUTF8(const USet *set, const char *s, int32_t length, USetSpanCondition spanCondition) { … }

U_CAPI int32_t U_EXPORT2
uset_spanBackUTF8(const USet *set, const char *s, int32_t length, USetSpanCondition spanCondition) { … }

U_CAPI UBool U_EXPORT2
uset_equals(const USet* set1, const USet* set2) { … }

U_CAPI int32_t U_EXPORT2
uset_indexOf(const USet* set, UChar32 c) { … }

U_CAPI UChar32 U_EXPORT2
uset_charAt(const USet* set, int32_t index) { … }

U_CAPI int32_t U_EXPORT2
uset_size(const USet* set) { … }

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 */ { … };
U_NAMESPACE_END

U_CAPI int32_t U_EXPORT2
uset_getRangeCount(const USet *set) { … }

U_CAPI int32_t U_EXPORT2
uset_getItemCount(const USet* uset) { … }

U_CAPI int32_t U_EXPORT2
uset_getItem(const USet* uset, int32_t itemIndex,
             UChar32* start, UChar32* end,
             char16_t* str, int32_t strCapacity,
             UErrorCode* ec) { … }

//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) { … }

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

U_CAPI void U_EXPORT2
uset_setSerializedToOne(USerializedSet* fillSet, UChar32 c) { … }

U_CAPI UBool U_EXPORT2
uset_serializedContains(const USerializedSet* set, UChar32 c) { … }

U_CAPI int32_t U_EXPORT2
uset_getSerializedRangeCount(const USerializedSet* set) { … }

U_CAPI UBool U_EXPORT2
uset_getSerializedRange(const USerializedSet* set, int32_t rangeIndex,
                        UChar32* pStart, UChar32* pEnd) { … }

// 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==nullptr || (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==nullptr) {
//             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);
// }