redis-zset數據結構探索

redis用的人比較多，其中zset你們都熟悉，主要用於排名場景。
zset數據結構，分紅兩部分，一部分是用於排序，一部分用於緩存鍵值。
先看看結構：

typedef struct zset {
    dict *dict;         //緩存
    zskiplist *zsl;     //排序結構
} zset;

 

上面，跳躍表用於排序結構，能夠按照名次，積分查找對應鍵, 時間複雜度: log(n)。
按照名次，積分範圍查找一系列鍵時, 先查詢知足條件的第一個鍵，而後當前鍵查找後續鍵， 時間複雜度: log(n) + o(m), n=總鍵數, m=查詢結果鍵數。

跳躍表結構:

typedef struct zskiplist {
    struct zskiplistNode *header, *tail;        //頭結點:用於順序查詢，經常使用方式; 尾結點:用於倒序簡單查詢。
    unsigned long length;                       //結點數
    int level;                                  //跳躍層級
} zskiplist;

 

結點結構：

typedef struct zskiplistNode {
    robj *obj;                                  //鍵
    double score;                               //積分
    struct zskiplistNode *backward;             //前一個結點, 和level[0]可看做雙鏈表
    struct zskiplistLevel {                     //跳躍層關係, 每層都是單鏈表
        struct zskiplistNode *forward;          //此層下一個結點
        unsigned int span;                      //此層下一個結點和當前結點距離(二者隔了多少結點)
    } level[];                                  //最多32層
} zskiplistNode; 

查詢：
根據名次範圍查詢

void zrangeGenericCommand(client *c, int reverse) {
        ......

        zset *zs = zobj->ptr;           //zset結構變量
        zskiplist *zsl = zs->zsl;       //跳躍表
        zskiplistNode *ln;
        robj *ele;

        /* Check if starting point is trivial, before doing log(N) lookup. */
        if (reverse) {                  //是否倒序查詢
            ln = zsl->tail;             //默認取尾結點
            if (start > 0)
                ln = zslGetElementByRank(zsl,llen-start);       //若是start>0, 則取對應結點
        } else {
            ln = zsl->header->level[0].forward; //默認取第一個結點
            if (start > 0)
                ln = zslGetElementByRank(zsl,start+1);
        }

        while(rangelen--) {             //取rangelen個結點
            serverAssertWithInfo(c,zobj,ln != NULL);
            ele = ln->obj;
            addReplyBulk(c,ele);        //響應鍵名
            if (withscores)
                addReplyDouble(c,ln->score);    //響應鍵值
            ln = reverse ? ln->backward : ln->level[0].forward; //設置下一個結點
        }

        ......
}

/* Finds an element by its rank. The rank argument needs to be 1-based. */
zskiplistNode* zslGetElementByRank(zskiplist *zsl, unsigned long rank) {
    zskiplistNode *x;
    unsigned long traversed = 0;                //當前名次
    int i;

    x = zsl->header;                            //頭結點, 從頭結點的下一個結點遍歷
    for (i = zsl->level-1; i >= 0; i--) {       //從高層到低層鏈表遍歷
        while (x->level[i].forward && (traversed + x->level[i].span) <= rank)   //若是有下一個結點，且下一個結點的名次<=rank
        {
            traversed += x->level[i].span;
            x = x->level[i].forward;
        }
        if (traversed == rank) {                //找到對應名次的結點
            return x;
        }
    }
    return NULL;
}

 zslGetElementByRank()時間複雜度理想值 = log(n)

 
若是有刪除，添加操做，和查詢相似，須要額外維護跳躍表關係。

/* Delete all the elements with score between min and max from the skiplist.
 * Min and max are inclusive, so a score >= min || score <= max is deleted.
 * Note that this function takes the reference to the hash table view of the
 * sorted set, in order to remove the elements from the hash table too. 
 * 根據積分範圍刪除結點
 */
unsigned long zslDeleteRangeByScore(zskiplist *zsl, zrangespec *range, dict *dict) {
    zskiplistNode *update[ZSKIPLIST_MAXLEVEL], *x;      //update維護跳躍表層級關係,用於zslDeleteNode()
    unsigned long removed = 0;
    int i;

    x = zsl->header;
    for (i = zsl->level-1; i >= 0; i--) {
        while (x->level[i].forward && (range->minex ?   //不満足積分條件時，循環
            x->level[i].forward->score <= range->min :
            x->level[i].forward->score < range->min))
                x = x->level[i].forward;
        update[i] = x;                                  //此層最接近於條件的結點
    }

    /* Current node is the last with score < or <= min. */
    x = x->level[0].forward;                            //第一個最可能知足條件的結點

    /* Delete nodes while in range. */
    while (x &&
           (range->maxex ? x->score < range->max : x->score <= range->max))     //知足條件的結點
    {
        zskiplistNode *next = x->level[0].forward;
        zslDeleteNode(zsl,x,update);                    //更新層級關係
        dictDelete(dict,x->obj);                        //刪除緩存
        zslFreeNode(x);
        removed++;
        x = next;
    }
    return removed;
}

/* Internal function used by zslDelete, zslDeleteByScore and zslDeleteByRank */
void zslDeleteNode(zskiplist *zsl, zskiplistNode *x, zskiplistNode **update) {
    int i;
    for (i = 0; i < zsl->level; i++) {                  //更新層級關係
        if (update[i]->level[i].forward == x) {
            update[i]->level[i].span += x->level[i].span - 1;
            update[i]->level[i].forward = x->level[i].forward;
        } else {
            update[i]->level[i].span -= 1;
        }
    }
    if (x->level[0].forward) {                          //維護當前結點的下一個結點
        x->level[0].forward->backward = x->backward;
    } else {
        zsl->tail = x->backward;                        //維護尾結點
    }
    while(zsl->level > 1 && zsl->header->level[zsl->level-1].forward == NULL)
        zsl->level--;                                   //維護層數
    zsl->length--;                                      //維護結點數
}

 

根據鍵名查找積分, 有了zset->dict這個鍵值緩存，只須要時間複雜度0(1)

int zsetScore(robj *zobj, robj *member, double *score) {
    if (!zobj || !member) return C_ERR;

    if (zobj->encoding == OBJ_ENCODING_ZIPLIST) {       //ziplit實現
        if (zzlFind(zobj->ptr, member, score) == NULL) return C_ERR;
    } else if (zobj->encoding == OBJ_ENCODING_SKIPLIST) {
        zset *zs = zobj->ptr;
        dictEntry *de = dictFind(zs->dict, member);     //找到緩存entry
        if (de == NULL) return C_ERR;
        *score = *(double*)dictGetVal(de);              //獲取對應積分  
    } else {
        serverPanic("Unknown sorted set encoding");
    }
    return C_OK;
}

 

若是zset對象用ziplist實現，則查詢和刪除操做時間複雜度 = o(n)

...