Algorithm template--SAP Network maximum stream 3 (recursive + adjacency matrix)

Implementation features: Same as before

Can see is a lot of beautiful program, code short more than one times, but the speed is and the original adjacency table A level (in the codevs above the grassland drainage that the running time comparison, but obviously the data is very big should be slower than that), the principle is similar, feel DFS inside the back and forth variable is amazing

1 var2 S,t,i,j,k,l,m,n,ans:longint;3A:Array[0.. +,0.. +] ofLongint;4D,DV:Array[0..10000] ofLongint;5 functionmin (x,y:longint): Longint;inline;6          begin7               ifX<y ThenMin:=xElsemin:=y;8          End;9 functionDFS (x,flow:longint): Longint;inline;Ten          varI,j,k,l:longint; One          begin A               ifX=t Thenexit (flow); -dfs:=0; -                fori:=1  toN Do the                   if(a[x,i]>0) and(D[x]= (d[i]+1)) Then -                      begin -K:=dfs (I,min (flow-dfs,a[x,i])); - Dec (a[x,i],k); + Inc (A[I,X],K); - Inc (DFS,K); +                           ifDfs=flow Thenexit; A                      End; at               ifD[s]=n Thenexit; - Dec (dv[d[x]]); -               ifdv[d[x]]=0  Thend[s]:=N; - Inc (D[x]); - Inc (Dv[d[x]); -          End; in begin - readln (n,m,s,t); toFillchar (A,sizeof (a),0); +       fori:=1  toM Do -          begin the readln (j,k,l); * Inc (A[J,K],L); $          End;Panax NotoginsengFillchar (d,sizeof (d),0); -Fillchar (Dv,sizeof (DV),0); thedv[0]:=n;ans:=0; +       whileD[s]<n DoInc (Ans,dfs (s, Maxlongint)); A writeln (ans); the End.

Algorithm template--SAP Network maximum stream 3 (recursive + adjacency matrix)