Kruskal算法（并查集+Qsort优化）  
 program busy_city;{Kruskal} 
const maxn=300; maxm=maxn*maxn; 
type 
        edge=record 
                v1,v2,cost:integer; 
        end; 
        arr=array[1..maxn]of integer; 
        edges=array[1..maxm]of edge; 
var 
        p:edges; 
        n,m,ans:integer; 

function find(father:arr; v:integer):integer;{并查集，寻找根结点的函数} 
var t:integer; 
begin 
        t:=v; 
        while(father[t]>=0) do t:=father[t]; 
        find:=t; 
end; 

procedure init;{以边目录方式输入} 
var i:integer; 
begin 
        readln(n,m); 
        for i:=1 to m do with p[i] do readln(v1,v2,cost); 
end; 

procedure qsort(l,r:integer);{预处理：快速排序} 
var 
        i,j,mid:integer; 
        t:edge; 
begin 
        i:=l; 
        j:=r; 
        mid:=p[(l+r)shr 1].cost; 
        repeat 
                while(p[i].cost<mid)do inc(i); 
                while(p[j].cost>mid)do dec(j); 
                if i<=j then begin 
                        t:=p[i]; 
                        p[i]:=p[j]; 
                        p[j]:=t; 
                        inc(i); 
                        dec(j); 
                end; 
        until i>j; 
        if l<j then qsort(l,j); 
        if i<r then qsort(i,r); 
end; 


procedure kruskal;{Kruskal算法} 
var 
        father:arr; 
        i,j,k,vf1,vf2:integer; 
begin 
        qsort(1,m); 
        for i:=1 to n do father[i]:=-1; 
        i:=1; 
        j:=1; 
        while((i<=maxm) and (j<=n-1)) do begin 
                vf1:=find(father,p[i].v1); 
                vf2:=find(father,p[i].v2); 
                if vf1<>vf2 then begin 
                        father[vf2]:=vf1; 
                        inc(j); 
                        writeln(p[i].v1,'---',p[i].v2);{输出MST中的边} 
                        ans:=ans+p[i].cost; 
                end; 
                inc(i); 
        end; 
end; 

begin 
        init; 
        kruskal; 
        writeln('MST=',ans);{输出MST的最小代价} 
        readln 
end.