Dijkstra改进算法C语言

/* 9-10(b)-12-13-20.26.c -- 第九章第十题(b) */ #include <stdio.h> #include <stdlib.h> #include "binary_heap.h" int main (void) ; int new_dijkstra (const Adjacenty_List * padj, const Hash_Table * const pht, const int start) ; int main (void) { Adjacenty_List adj ; Hash_Table ht ; int capacity = 5 ; Initialize_H (&ht, capacity * 2) ; Initialize_A (&adj, capacity) ; InitializeALine_A (&adj, &ht, 0, 's', 0, 4, 'a', 2, 'd', 3) ; InitializeALine_A (&adj, &ht, 1, 'a', 0, 2, 'b', 2) ; InitializeALine_A (&adj, &ht, 2, 'b', 0, 2, 'c', 2) ; InitializeALine_A (&adj, &ht, 3, 'c', 0, 2, 'd', 2) ; InitializeALine_A (&adj, &ht, 4, 'd', 0, 2, 'c', 3) ; new_dijkstra (&adj, &ht, 0) ; PrintAdjacenty_List_A (&adj, &ht) ; Release_H (&ht) ; Release_A (&adj) ; return 0 ; } int new_dijkstra (const Adjacenty_List * padj, const Hash_Table * const pht, const int start) { Binary_Heap bh ; Vertex * v ; Adjoin_To_Vertex * w ; int capacity = (*padj) -> capacity, i, v_index, w_index ; int * count ; count = (int *) calloc (sizeof (int), capacity) ; if (NULL == count) return 0 ; Initialize_B (&bh, capacity * 2) ; (*padj) -> list[start].dist = 0 ; for (i = 0; i < capacity; i++) Insert_B (&bh, (*padj) -> list + i) ; while (1) { do { v = DeleteMin_B (&bh) ; if (NEGATIVEINFINITY == v -> dist) break ; } while (TRUE == (*pht) -> lists[v -> hash_value].be_deleted) ; if (NEGATIVEINFINITY == v -> dist) break ; (*pht) -> lists[v -> hash_value].be_deleted = TRUE ; v -> known = TRUE ; w = v -> adjoin_to ; v_index = (*pht) -> lists[v -> hash_value].index_in_adjacenty_list ; while (w) { w_index = (*pht) -> lists[w -> hash_value].index_in_adjacenty_list ; if (FALSE == (*padj) -> list[w_index].known) { if (v -> dist + w -> cvw < (*padj) -> list[w_index].dist) { count[w_index] = count[v_index] + 1 ; (*padj) -> list[w_index].dist = v -> dist + w -> cvw ; Insert_B (&bh, (*padj) -> list + w_index) ; (*padj) -> list[w_index].path = (*pht) -> lists[v -> hash_value].name ; } else if (v -> dist + w -> cvw == (*padj) -> list[w_index].dist) { if (count[v_index] + 1 < count[w_index]) { count[w_index] = count[v_index] + 1 ; (*padj) -> list[w_index].path = (*pht) -> lists[v -> hash_value].name ; } } } w = w -> next ; } } free (count) ; Release_B (&bh) ; return 1 ; } 这个例程,会在多条相同最短路径中选出边数最少的路径.只是改变path域即可在遍历的时候访问到边数少的路径.核心的思想,就是建立一个数组,存放从出发点到当前顶点的最短路径边数.通过每次查看最短路径相同的情况, 即d_v + cvw = d_w.时,如果 Count[v] + 1 < Count[w]时,就执行 Count[w] = Count[v] + 1, 同时 w -> path = c. 就是这样吧,自己想出来的答案跟这个比较贴边.希望,我能记住.或者,消化得好一些.