The 2nd test on the site does not pass, everything works for me, what should I do?

try this test:
input:
1
2
0 0

output:
NO

because for this sentences in the problem:
"exactly one child has confessed the he (or she) had broken the cup".

I dont notice to this and get WA#2.
sorry for my poor english.

GOOD LUCK!

Edited by author 15.11.2010 23:48

Here are jokes with statuses, such as we go in order along the vertices and if we get to a vertex that we haven't been to, then we start going in a cycle from this vertex, setting the status of this cycle, if we get to the same status, then this is a cycle, if we get to a vertex with a lower status, then everything is fine and we stop the cycle, and if the vertex has no status, then we just opened a new vertex and it turns out that the algorithm is linear

I think this is very easy for 319 points of hardness, you only need to find cycle in directed graph

Why did you got wa on test 3 ? I'm having the same problem... :(

Was wrong. Sorry

Edited by author 26.01.2017 16:13

var g:array[1..25000,1..25000] of integer; v,i,j,n,f,un,uk,t,k,s,x:longint;
Q:array[1..100000] of integer;
p:array[1..25000] of integer;
begin
assign(input,'input.txt');
assign(output,'output.txt');
reset(input);rewrite(output);
readln(t);
for i:=1 to t do
begin
  readln(n);k:=0;
  fillchar(g,sizeof(g),0);
  for j:=1 to n do begin read(s);
                         if s=0 then begin f:=j; k:=k+1;g[j,j]:=0;end
                                else begin  g[j,s]:=1;g[s,j]:=1;end;
                         if k>1 then break;
                   end;
  readln;
  if k<>1 then begin Writeln('NO');end else
  begin
    for j:=1 to n do p[j]:=-1;
    fillchar(Q,sizeof(Q),0);
    un:=1;
    uk:=2;
    Q[1]:=f;
    p[f]:=0;
    x:=0;
    while un<>uk do
    begin
      v:=Q[un]; un:=un+1;
      for j:=1 to n do
      if (g[v,j]<>0) and (p[j]=-1) then begin
                                        Q[uk]:=j;
                                        uk:=uk+1;
                                        p[j]:=p[v]+1;
                                        end;

    end;
    for j:=1 to n do
      if p[j]=-1 then begin x:=1;break;end;
      if x=0 then writeln('YES') else writeln('NO');
  end;
end;
end.

#include <iostream>
using namespace std;
bool Chekcker(int A[],int k)
{
    int i,j,v;
    bool circle=true;
    for(j=1;j<=k;j++)
        for(v=1;v<=k;v++)
        {
            if(A[v]==0 && A[j]==v )
                A[j]=0;
        }

    for(j=1;j<=k;j++)
        for(v=1;v<=k;v++)
        {
            if(A[j]==0 && A[v]==j )
                A[v]=0;
        }
    for(v=1;v<=k;v++)
        if(A[v]==0)
        { circle=true;}
        else{circle=false;break;}

    return circle;
}

int main() {
    int n;
    cin>>n;
    int i,j,k = 0,v;
    int temK;
    bool circle;
    int conf;
    int A[30000];
    for(i=1;i<=n;i++)
    {
        cin>>k;
        j=1;
        v=1;
        temK=k;
        circle=true;
        conf=0;

        for(j=1;j<=k;j++)
        {
            cin>>A[j];

        }

        for(j=1;j<=k;j++)
        {
            if(A[j]==0)
            conf++;
        }



        circle=Chekcker(A,k);

        if(conf==0 || conf>1)
        {circle=false;}

        if(circle==true)
            cout<<"YES"<<endl;
        else
            cout<<"NO"<<endl;

        for(v=1;v<=k;v++)
            A[v]=0;

        }



       return 0;
}


this code get correct on every test in the forum, however fails test #3. Whats Wrong????

I've sent the same solution using c++ and ruby - and ruby's version works 8 times slower than c++ solution and hardly passes TL (however all solutions were accepted, working time 0,125 vs 0,984 sec). I hope it might be caused by the old interpreter's version - can we use any newer version?

This is the fragment of my program:
scanf("%d", &n);
for (int i = 0; i < n; i++)
{
    int p;
    scanf("%d", &p);
    p--;
    if (p >= n)
        throw;

If I remove throw, I'll have WA 3.

So children can say absolutely incorrect information?

16
25000
0 3 4 5 ... 1
25000
0 3 4 5 ... 1
....

....
25000
0 3 4 5 ... 1

Edited by author 21.11.2013 02:11

1
2
0 2

Is answer of second child to himself means "circle" ?

Test:

1
1
1

And what is the answer if yes?

Hope this test can help you:

2
3
2 1 0
1
0
-
NO
YES

Good luck!

var t,n,i,j:longint;
    d:array[1..25000] of integer;
    ok:boolean;
begin
 read(t);
 for i:=1 to t do
 begin
  read(n);ok:=false;
  for j:=1 to n do
  begin
   read(d[j]);
   if d[j]=0 then ok:=true;
  end;
  if not ok then writeln('NO')
   else begin
         for j:=1 to n do
          if d[j]<>0 then
           if d[d[j]]=j then begin ok:=false;writeln('NO');break;end;
         if ok then writeln('YES');
        end;
 end;
end.

Can anybody help me with this problem, I nave WA3.

Give me please some tests, I have WA3.

So... Good problem!
But I can't understand, who is this mother-heroine, which have 25000 children! =)
By the way, they are must eat anything. Can you imagine house, where they are must living? I can't... ;)
Interesting, author thought about this facts? =)