function [mydistance,mypath]=mydijkstra(a,sb,db)%寻找i,j两点最短路径% 输入：a—邻接矩阵，a（i，j)是指i到j之间的距离，可以是有向的% sb—起点的标号, db—终点的标号% 输出：mydistance—最短路的距离, mypath—最短路的路径n=size(a,1); visited(1:n) = 0;distance(1:n) = inf; distance(sb) = 0; %起点到各顶点距离的初始化visited(sb)=1; u=sb;  %u为最新的P标号顶点parent(1:n) = 0; %前驱顶点的初始化for i = 1: n-1    id=find(visited==0); %查找未标号的顶点    for v = id        if  a(u, v) + distance(u) < distance(v)            distance(v) = distance(u) + a(u, v);  %修改标号值            parent(v) = u;        end    end    temp=distance;    temp(visited==1)=inf;  %已标号点的距离换成无穷    [t, u] = min(temp);  %找标号值最小的顶点    visited(u) = 1;       %标记已经标号的顶点endmypath = [];if parent(db) ~= 0   %如果存在路!    t = db; mypath = [db];    while t ~= sb        p = parent(t);        mypath = [p mypath];        t = p;    endendmydistance = distance(db);end