/* Insert an integer in the intset */
intset *intsetAdd(intset *is, int64_t value, uint8_t *success) {
uint8_t valenc = _intsetValueEncoding(value);
uint32_t pos;
if (success) *success = 1;
/* Upgrade encoding if necessary. If we need to upgrade, we know that
* this value should be either appended (if > 0) or prepended (if < 0),
* because it lies outside the range of existing values. */
if (valenc > is->encoding) {
/* This always succeeds, so we don't need to curry *success. */
return intsetUpgradeAndAdd(is,value);
} else {
/* Abort if the value is already present in the set.
* This call will populate "pos" with the right position to insert
* the value when it cannot be found. */
if (intsetSearch(is,value,&pos)) {
if (success) *success = 0;
return is;
}
is = intsetResize(is,is->length+1);
if (pos < is->length) intsetMoveTail(is,pos,pos+1);
}
_intsetSet(is,pos,value);
is->length++;
return is;
}
/*
* 把 value 从 intset 中移除
*
* 移除成功将 *success 设置为 1 ，失败则设置为 0 。
*
* T = O(n)
*/
intset *intsetRemove(intset *is, int64_t value, int *success) {
uint8_t valenc = _intsetValueEncoding(value);
uint32_t pos;
if (success) *success = 0;
if (valenc <= intrev32ifbe(is->encoding) && // 编码方式匹配
intsetSearch(is,value,&pos)) // 将位置保存到 pos
{
uint32_t len = intrev32ifbe(is->length);
/* We know we can delete */
if (success) *success = 1;
/* Overwrite value with tail and update length */
// 如果 pos 不是 is 的最末尾，那么显式地删除它
// （如果 pos = (len-1) ，那么紧缩空间时值就会自动被『抹除掉』）
if (pos < (len-1)) intsetMoveTail(is,pos+1,pos);
// 紧缩空间，并更新数量计数器
is = intsetResize(is,len-1);
is->length = intrev32ifbe(len-1);
}
return is;
}
/* Determine whether a value belongs to this set
*
* 检查给定值value是否是集合中的元素，
*
* 是返回1，不是返回0。
*
* T = O(logN)
*/
uint8_t intsetFind(intset *is, int64_t value) {
// 计算value的编码
uint8_t valenc = _intsetValueEncoding(value);
// 如果value的编码大于集合的当前编码，那么value一定不存在于集合
// 当value的编码小于等于集合的当前编码时，才再使用intsetSearch进行查找
return valenc <= intrev32ifbe(is->encoding) && intsetSearch(is,value,NULL);
}
/* Delete integer from intset */
intset *intsetRemove(intset *is, int64_t value, int *success) {
uint8_t valenc = _intsetValueEncoding(value);
uint32_t pos;
if (success) *success = 0;
if (valenc <= is->encoding && intsetSearch(is,value,&pos)) {
/* We know we can delete */
if (success) *success = 1;
/* Overwrite value with tail and update length */
if (pos < (is->length-1)) intsetMoveTail(is,pos+1,pos);
is = intsetResize(is,is->length-1);
is->length--;
}
return is;
}
/* Delete integer from intset */
intset *intsetRemove(intset *is, int64_t value, int *success) {
uint8_t valenc = _intsetValueEncoding(value); // value 的编码值
uint32_t pos;
if (success) *success = 0; // 先置 0
if (valenc <= intrev32ifbe(is->encoding) && intsetSearch(is,value,&pos)) {
uint32_t len = intrev32ifbe(is->length);
/* We know we can delete */
if (success) *success = 1; // 找到了 val 的位置
/* Overwrite value with tail and update length */
if (pos < (len-1)) intsetMoveTail(is,pos+1,pos); // pos <- pos+1 覆盖
is = intsetResize(is,len-1); // 重新分配内存( len-1 )
is->length = intrev32ifbe(len-1); // 长度
}
return is;
}
//将值为value的元素从整数集中删除
intset *intsetRemove(intset *is, int64_t value, int *success) {
uint8_t valenc = _intsetValueEncoding(value);
uint32_t pos;
if (success) *success = 0;
if (valenc <= intrev32ifbe(is->encoding) && intsetSearch(is,value,&pos)) {
uint32_t len = intrev32ifbe(is->length);
/* We know we can delete */
if (success) *success = 1;
/* Overwrite value with tail and update length */
//pos为元素所在位置，将从后一位到最后一位得元素都往前移动一位
if (pos < (len-1)) intsetMoveTail(is,pos+1,pos);
//减少内存
is = intsetResize(is,len-1);
is->length = intrev32ifbe(len-1);
}
return is;
}
/* Insert an integer in the intset */
intset *intsetAdd(intset *is, int64_t value, uint8_t *success) {
uint8_t valenc = _intsetValueEncoding(value);
uint32_t pos;
if (success) *success = 1;
/* Upgrade encoding if necessary. If we need to upgrade, we know that
* this value should be either appended (if > 0) or prepended (if < 0),
* because it lies outside the range of existing values. */
// 需要插入整数的所需内存超出了原有集合整数的范围，即内存类型不同，
// 则升级整数类型
if (valenc > intrev32ifbe(is->encoding)) {
/* This always succeeds, so we don't need to curry *success. */
return intsetUpgradeAndAdd(is,value);
// 正常，分配内存，插入
} else {
// intset 内部不允许重复
/* Abort if the value is already present in the set.
* This call will populate "pos" with the right position to insert
* the value when it cannot be found. */
if (intsetSearch(is,value,&pos)) {
if (success) *success = 0;
return is;
}
// realloc
is = intsetResize(is,intrev32ifbe(is->length)+1);
// 迁移内存，腾出空间给新的数据。intsetMoveTail() 完成内存迁移工作
if (pos < intrev32ifbe(is->length)) intsetMoveTail(is,pos,pos+1);
}
// 在腾出的空间中设置新的数据
_intsetSet(is,pos,value);
// 更新 intset size
is->length = intrev32ifbe(intrev32ifbe(is->length)+1);
return is;
}
/*
* 将 value 添加到集合中
*
* 如果元素已经存在， *success 被设置为 0 ，
* 如果元素添加成功， *success 被设置为 1 。
*
* T = O(n)
*/
intset *intsetAdd(intset *is, int64_t value, uint8_t *success) {
uint8_t valenc = _intsetValueEncoding(value);
uint32_t pos;
if (success) *success = 1;
/* Upgrade encoding if necessary. If we need to upgrade, we know that
* this value should be either appended (if > 0) or prepended (if < 0),
* because it lies outside the range of existing values. */
// 如果有需要，进行升级并插入新值
if (valenc > intrev32ifbe(is->encoding)) {
/* This always succeeds, so we don't need to curry *success. */
return intsetUpgradeAndAdd(is,value);
} else {
/* Abort if the value is already present in the set.
* This call will populate "pos" with the right position to insert
* the value when it cannot be found. */
// 如果值已经存在，那么直接返回
// 如果不存在，那么设置 *pos 设置为新元素添加的位置
if (intsetSearch(is,value,&pos)) {
if (success) *success = 0;
return is;
}
// 扩张 is ，准备添加新元素
is = intsetResize(is,intrev32ifbe(is->length)+1);
// 如果 pos 不是数组中最后一个位置，
// 那么对数组中的原有元素进行移动
if (pos < intrev32ifbe(is->length)) intsetMoveTail(is,pos,pos+1);
}
// 添加新元素
_intsetSet(is,pos,value);
// 更新元素数量
is->length = intrev32ifbe(intrev32ifbe(is->length)+1);
return is;
}
/* Delete integer from intset
*
* 从整数集合中删除值value。
*
* *success的值指示删除是否成功：
* -因值不存在而造成删除失败时该值为0。
* -删除成功时该值为1。
*
* T = O(N)
*/
intset *intsetRemove(intset *is, int64_t value, int *success) {
// 计算value的编码方式
uint8_t valenc = _intsetValueEncoding(value);
uint32_t pos;
// 默认设置标识值为删除失败
if (success) *success = 0;
// 当value的编码大小小于或等于集合的当前编码方式(说明value有可能存在于集合)
// 并且intsetSearch的结果为真，那么执行删除
// T = O(N)
if (valenc <= intrev32ifbe(is->encoding) && intsetSearch(is,value,&pos)) {
// 取出集合当前的元素数量
uint32_t len = intrev32ifbe(is->length);
/* We know we can delete */
// 设置标识值为删除成功
if (success) *success = 1;
/* Overwrite value with tail and update length */
// 如果value不是位于数组的末尾
// 那么需要对原本位于value之后的元素进行移动
//
// 举个例子，如果数组表示如下，而b为删除的目标
// | a | b | c | d |
// 那么intsetMoveTail将b之后的所有数据向前移动一个元素的空间，覆盖b原来的数据
// | a | c | d | d |
// 之后intsetResize缩小内存大小时，数组末尾多出来的一个元素的空间将被移除
// | a | c | d |
if (pos < (len-1)) intsetMoveTail(is,pos+1,pos);
// 缩小数组的大小，移除被删除元素占用的空间
// T = O(N)
is = intsetResize(is,len-1);
// 更新集合的元素数量
is->length = intrev32ifbe(len-1);
}
return is;
}
//将value插入到整数集中
intset *intsetAdd(intset *is, int64_t value, uint8_t *success) {
uint8_t valenc = _intsetValueEncoding(value);
uint32_t pos;
if (success) *success = 1;
/* Upgrade encoding if necessary. If we need to upgrade, we know that
* this value should be either appended (if > 0) or prepended (if < 0),
* because it lies outside the range of existing values. */
if (valenc > intrev32ifbe(is->encoding)) {
//如果表示value所需编码比现整数集的编码要大，升级整数集并将元素插入
/* This always succeeds, so we don't need to curry *success. */
return intsetUpgradeAndAdd(is,value);
} else {
/* Abort if the value is already present in the set.
* This call will populate "pos" with the right position to insert
* the value when it cannot be found. */
//调用intsetSearch查找元素，如果已经存在则返回0
if (intsetSearch(is,value,&pos)) {
if (success) *success = 0;
return is;
}
//扩大内存空间
is = intsetResize(is,intrev32ifbe(is->length)+1);
//将原来从pos到末尾的元素都向后移动一位
if (pos < intrev32ifbe(is->length)) intsetMoveTail(is,pos,pos+1);
}
//将value设到pos位上
_intsetSet(is,pos,value);
//更新元素个数
is->length = intrev32ifbe(intrev32ifbe(is->length)+1);
return is;
}
int main(int argc, char **argv) {
uint8_t success;
int i;
intset *is;
sranddev();
printf("Value encodings: "); {
assert(_intsetValueEncoding(-32768) == INTSET_ENC_INT16);
assert(_intsetValueEncoding(+32767) == INTSET_ENC_INT16);
assert(_intsetValueEncoding(-32769) == INTSET_ENC_INT32);
assert(_intsetValueEncoding(+32768) == INTSET_ENC_INT32);
assert(_intsetValueEncoding(-2147483648) == INTSET_ENC_INT32);
assert(_intsetValueEncoding(+2147483647) == INTSET_ENC_INT32);
assert(_intsetValueEncoding(-2147483649) == INTSET_ENC_INT64);
assert(_intsetValueEncoding(+2147483648) == INTSET_ENC_INT64);
assert(_intsetValueEncoding(-9223372036854775808ull) == INTSET_ENC_INT64);
assert(_intsetValueEncoding(+9223372036854775807ull) == INTSET_ENC_INT64);
ok();
}
printf("Basic adding: "); {
is = intsetNew();
is = intsetAdd(is,5,&success); assert(success);
is = intsetAdd(is,6,&success); assert(success);
is = intsetAdd(is,4,&success); assert(success);
is = intsetAdd(is,4,&success); assert(!success);
ok();
}
printf("Large number of random adds: "); {
int inserts = 0;
is = intsetNew();
for (i = 0; i < 1024; i++) {
is = intsetAdd(is,rand()%0x800,&success);
if (success) inserts++;
}
assert(is->length == inserts);
checkConsistency(is);
ok();
}
printf("Upgrade from int16 to int32: "); {
is = intsetNew();
is = intsetAdd(is,32,NULL);
assert(is->encoding == INTSET_ENC_INT16);
is = intsetAdd(is,65535,NULL);
assert(is->encoding == INTSET_ENC_INT32);
assert(intsetFind(is,32));
assert(intsetFind(is,65535));
checkConsistency(is);
is = intsetNew();
is = intsetAdd(is,32,NULL);
assert(is->encoding == INTSET_ENC_INT16);
is = intsetAdd(is,-65535,NULL);
assert(is->encoding == INTSET_ENC_INT32);
assert(intsetFind(is,32));
assert(intsetFind(is,-65535));
checkConsistency(is);
ok();
}
printf("Upgrade from int16 to int64: "); {
is = intsetNew();
is = intsetAdd(is,32,NULL);
assert(is->encoding == INTSET_ENC_INT16);
is = intsetAdd(is,4294967295,NULL);
assert(is->encoding == INTSET_ENC_INT64);
assert(intsetFind(is,32));
assert(intsetFind(is,4294967295));
checkConsistency(is);
is = intsetNew();
is = intsetAdd(is,32,NULL);
assert(is->encoding == INTSET_ENC_INT16);
is = intsetAdd(is,-4294967295,NULL);
assert(is->encoding == INTSET_ENC_INT64);
assert(intsetFind(is,32));
assert(intsetFind(is,-4294967295));
checkConsistency(is);
ok();
}
printf("Upgrade from int32 to int64: "); {
is = intsetNew();
is = intsetAdd(is,65535,NULL);
assert(is->encoding == INTSET_ENC_INT32);
is = intsetAdd(is,4294967295,NULL);
assert(is->encoding == INTSET_ENC_INT64);
assert(intsetFind(is,65535));
assert(intsetFind(is,4294967295));
checkConsistency(is);
is = intsetNew();
is = intsetAdd(is,65535,NULL);
assert(is->encoding == INTSET_ENC_INT32);
is = intsetAdd(is,-4294967295,NULL);
assert(is->encoding == INTSET_ENC_INT64);
assert(intsetFind(is,65535));
assert(intsetFind(is,-4294967295));
checkConsistency(is);
ok();
}
printf("Stress lookups: "); {
long num = 100000, size = 10000;
int i, bits = 20;
long long start;
is = createSet(bits,size);
checkConsistency(is);
start = usec();
for (i = 0; i < num; i++) intsetSearch(is,rand() % ((1<<bits)-1),NULL);
printf("%ld lookups, %ld element set, %lldusec\n",num,size,usec()-start);
}
printf("Stress add+delete: "); {
int i, v1, v2;
is = intsetNew();
for (i = 0; i < 0xffff; i++) {
v1 = rand() % 0xfff;
is = intsetAdd(is,v1,NULL);
assert(intsetFind(is,v1));
v2 = rand() % 0xfff;
is = intsetRemove(is,v2,NULL);
assert(!intsetFind(is,v2));
}
checkConsistency(is);
ok();
}
}
/* Determine whether a value belongs to this set */
uint8_t intsetFind(intset *is, int64_t value) {
uint8_t valenc = _intsetValueEncoding(value);
return valenc <= is->encoding && intsetSearch(is,value,NULL);
}
/*
* 查看 value 是否存在于 is
*
* T = O(lg N)
*/
uint8_t intsetFind(intset *is, int64_t value) {
uint8_t valenc = _intsetValueEncoding(value);
return valenc <= intrev32ifbe(is->encoding) && // 编码方式匹配
intsetSearch(is,value,NULL); // 查找 value
}
/* Insert an integer in the intset
*
* 尝试将元素value添加到整数集合中。
*
* *success的值指示添加是否成功：
* -如果添加成功，那么将*success的值设为1。
* -因为元素已存在而造成添加失败时，将*success的值设为0。
*
* T = O(N)
*/
intset *intsetAdd(intset *is, int64_t value, uint8_t *success) {
// 计算编码value所需的长度
uint8_t valenc = _intsetValueEncoding(value);
uint32_t pos;
// 默认设置插入为成功
if (success) *success = 1;
/* Upgrade encoding if necessary. If we need to upgrade, we know that
* this value should be either appended (if > 0) or prepended (if < 0),
* because it lies outside the range of existing values. */
// 如果value的编码比整数集合现在的编码要大
// 那么表示value必然可以添加到整数集合中
// 并且整数集合需要对自身进行升级，才能满足value所需的编码
if (valenc > intrev32ifbe(is->encoding)) {
/* This always succeeds, so we don't need to curry *success. */
// T = O(N)
return intsetUpgradeAndAdd(is,value);
} else {
// 运行到这里，表示整数集合现有的编码方式适用于value
/* Abort if the value is already present in the set.
* This call will populate "pos" with the right position to insert
* the value when it cannot be found. */
// 在整数集合中查找value，看他是否存在：
// -如果存在，那么将*success设置为0，并返回未经改动的整数集合
// -如果不存在，那么可以插入value的位置将被保存到pos指针中
// 等待后续程序使用
if (intsetSearch(is,value,&pos)) {
if (success) *success = 0;
return is;
}
// 运行到这里，表示value不存在于集合中
// 程序需要将value添加到整数集合中
// 为value在集合中分配空间
is = intsetResize(is,intrev32ifbe(is->length)+1);
// 如果新元素不是被添加到底层数组的末端
// 那么需要对现有元素的数据进行移动，空出pos上的位置，用于设置新值
// 举个例子
// 如果数组为：
// | x | y | z | ? |
// |<----->|
// 而新元素n的pos为1，那么数组将移动y和z两个元素
// | x | y | y | z |
// |<----->|
// 这样就可以将新元素设置到pos上了：
// | x | n | y | z |
// T = O(N)
if (pos < intrev32ifbe(is->length)) intsetMoveTail(is,pos,pos+1);
}
// 将新值设置到底层数组的指定位置中
_intsetSet(is,pos,value);
// 增一集合元素数量的计数器
is->length = intrev32ifbe(intrev32ifbe(is->length)+1);
// 返回添加新元素后的整数集合
return is;
/* p.s.上面的代码可以重构成以下更简单的形式：
if (valenc > intrev32(is->encoding)) {
return intsetUpgradeAndAdd(is, value);
}
if (intsetSearch(is, value, &pos)) {
if (success) *success = 0;
return is;
} else {
is = intsetResize(is, intrev32ifbe(is->length) + 1);
if (pos < intrev32ifbe(is->length)) intsetMoveTail(is, pos, pos + 1);
_intsetSet(is, pos, value);
is->length = intrev32ifbe(intrev32ifbe(is->length) + 1);
return is;
}
*/
}
/* Determine whether a value belongs to this set */
uint8_t intsetFind(intset *is, int64_t value) {
uint8_t valenc = _intsetValueEncoding(value);
//valenc小于等于当前编码表示元素可以被该前编码表示，那么再调用intsetSearch寻找元素
return valenc <= intrev32ifbe(is->encoding) && intsetSearch(is,value,NULL);
}
