I think my program is correct,but it always gets WA,why?
Here is my program:

program U1048;
const
  digits=100000000;
var
  n1,n2:array[0..200000] of longint;
  i,j,n,a,b:longint;
  s:string;
begin
  readln(n);
  n1[0]:=0;
  for j:=n div 8*8+1 to n do begin
      readln(a,b);
      n1[1]:=n1[1]*10+a;
      n2[1]:=n2[1]*10+b;
  end;
  inc(n1[1],n2[1]);
  for i:=1 to n div 8 do begin
    for j:=1 to 8 do begin
      readln(a,b);
      n1[i+1]:=n1[i+1]*10+a;
      n2[i+1]:=n2[i+1]*10+b;
    end;
    inc(n1[i+1],n2[i+1]);
  end;
  for i:=n div 8+1 downto 1 do
    if n1[i]>=digits then begin
      dec(n1[i],digits);
      inc(n1[i-1]);
    end;
  j:=0;
  while (n1[j]=0)and(j<200000) do inc(j);
  if n1[j]=0 then writeln(0)
  else begin
    write(n1[j]);
    for i:=j+1 to n div 8+1 do begin
      str(n1[i],s);
      while length(s)<8 do s:='0'+s;
      write(s);
    end;
    writeln;
  end;
end.

const    fi     = '1195.inp';

         max    = 3;

         so     = 8;
         line   :array[1..so,1..3,1..2] of byte
                =( ( (1,1), (1,2), (1,3) ),
                   ( (2,1), (2,2), (2,3) ),
                   ( (3,1), (3,2), (3,3) ),
                   ( (1,1), (2,1), (3,1) ),
                   ( (1,2), (2,2), (3,2) ),
                   ( (1,3), (2,3), (3,3) ),
                   ( (1,1), (2,2), (3,3) ),
                   ( (1,3), (2,2), (3,1) ) );

var      a,b        :array[1..3,1..3] of char;
         kq         :byte;


procedure input;
var   f         :text;
      i,j       :byte;
      st        :string;
begin
  {assign(f, fi); reset(f);}
  for i := 1 to 3 do
  begin
    readln({f}, st);
    while st[1] = ' ' do delete(st,1,1);
    while st[ length(st) ] = ' ' do delete(st, length(st), 1);

    for j := 1 to 3 do a[i,j] := st[j];
  end;
  {close(f);}
end;



procedure out;
begin
  case kq of
  0 : writeln('Draw');
  1 : writeln('Crosses win');
  2 : writeln('Ouths win');
  end;
end;



function thang :boolean;
var     i,j    :byte;
        c      :char;
        ok     :boolean;
begin
  thang := true;

  for i := 1 to so do
  begin
   ok := true;

   c  := b[ line[i,1,1], line[i,1,2] ];
   if c = '#' then ok := false;

   for j := 1 to 3 do
     if b[ line[i,j,1], line[i,j,2] ] <> c
     then ok := false;

   if ok then exit;
  end;

  thang := false;
end;




function cross :boolean;
var     i,j             :byte;
begin
  cross := true;

  for i := 1 to 3 do
    for j := 1 to 3 do
     if a[i,j] = '#' then
     begin
       b := a;
       b[i,j] := 'X';
       if thang then
         begin  kq := 1;
                exit;
         end;
     end;

  cross := false;
end;



procedure outh;
var     i,j,u,v         :byte;
        ok              :boolean;
begin
  for i := 1 to 3 do
    for j := 1 to 3 do
     if a[i,j] = '#' then
     begin
       b := a; b[i,j] := 'X';

       ok := false;
       for u := 1 to 3 do
         for v := 1 to 3 do
           if b[u,v] = '#' then
           begin
             b[u,v] := 'O';
             if thang then ok := true;
             b[u,v] := '#';
           end;

       if not ok then exit;
     end;

  kq := 2;
end;



procedure solve;
begin
  kq := 0;
  if cross then exit;
  outh;
end;


begin
  input;
  solve;
  out;
end.

Why compilation error? On my Borland Pascal everything is fine, but
Judge system doesn't like my source.

Huge is not accepted in C?
Why I get compilation error?

const
ss:array[1..169] of longint=(
2,3,5,7,11,13,17,19,23,29,31,37,41,43,47,53,59,61,67,71,
73,79,83,89,97,101,103,107,109,113,127,131,137,139,149,151,157,163,167
,173,
179,181,191,193,197,199,211,223,227,229,233,239,241,251,257,263,269,27
1,277,281,
283,293,307,311,313,317,331,337,347,349,353,359,367,373,379,383,389,39
7,401,409,
419,421,431,433,439,443,449,457,461,463,467,479,487,491,499,503,509,52
1,523,541,
547,557,563,569,571,577,587,593,599,601,607,613,617,619,631,641,643,64
7,653,659,
661,673,677,683,691,701,709,719,727,733,739,743,751,757,761,769,773,78
7,797,809,
811,821,823,827,829,839,853,857,859,863,877,881,883,887,907,911,919,92
9,937,941,
947,953,967,971,977,983,991,997,0);
var
  a:array[1..169] of longint;
  i,j,k,x,y:longint;
function judge(m:longint):boolean;
var
  i,j,k,n,q:longint;
begin
  n:=m;
  fillchar(a,sizeof(a),0);
  judge:=false;
  i:=1;
  while (n>1)and(i<169) do begin
    while n mod ss[i]=0 do begin
      inc(a[i]);
      n:=n div ss[i];
    end;
    inc(i);
  end;
  if n>1 then begin ss[169]:=n;a[169]:=1;end;
  k:=1;
  for i:=1 to 169 do if a[i]>0 then begin
    q:=1;
    for j:=1 to a[i] do q:=q*ss[i]+1;
    if m*2<k*q then exit;
    k:=k*q;
  end;
  judge:=true;
end;

begin
  readln(x,y);
  for i:=x to y do
    if judge(i) then begin writeln(i);break;end;
end.

Program n1085;
  Const MaxN = 100;
        MaxM = 100;
        MaxK = 100;
  Var   n,m,k,step : integer;
        i,j,q,w    : integer;
        money      : array [1..MaxK] of integer;
        a          : array [1..MaxM,1..MaxN] of byte;
        b          : array [1..MaxK,1..MaxN] of Record
                                                  sum    : integer;
                                                  step,t : byte;
                                                End;
        lost,num   : longint;

Procedure Init;
  Var stop,l : byte;
      sum    : integer;
  Begin
    Read(n,m);
    FillChar(a,SizeOf(a),0);
    For i:=1 to m Do Begin
      Read(l);
      For j:=1 to l Do Begin
        Read(stop);
        a[i,stop]:=1;
      End;
    End;
    Read(k);
    For i:=1 to k Do
      For j:=1 to n Do Begin
        b[i,j].sum:=0;
        b[i,j].step:=0;
        b[i,j].t:=0;
      End;
    For i:=1 to k Do Begin
      Read(sum,stop,l);
      money[i]:=sum;
      b[i,stop].sum:=sum;
      b[i,stop].step:=1;
      b[i,stop].t:=l;
    End;
  End;

Procedure Solve;
  Var yes : boolean;
  Begin
    step:=1;
    While true Do Begin
      yes:=true;
      For i:=1 to k Do
        For j:=1 to n Do
          If (b[i,j].step=step) and (b[i,j].sum>=4) Then Begin
            yes:=false;
            For q:=1 to m Do
              If a[q,j]=1 Then
                For w:=1 to n Do
                  If (a[q,w]=1) and (b[i,w].step=0) Then Begin
                    b[i,w].step:=step+1;
                    b[i,w].t:=b[i,j].t;
                    If b[i,j].t=0 Then b[i,w].sum:=b[i,j].sum-4
                    Else b[i,w].sum:=b[i,j].sum;
                  End;
          End;
      If yes Then Break;
      step:=step+1;
    End;
  End;

Procedure SearchAns;
  Var yes : boolean;
      los : longint;
  Begin
    lost:=MaxLongInt;
    num:=0;
    For i:=1 to n Do Begin
      yes:=true;
      los:=0;
      For j:=1 to k Do
        If b[j,i].step=0 Then Begin
          yes:=false;
          Break;
        End
        Else los:=los+money[j]-b[j,i].sum;
      If yes and (los<lost) Then Begin
        lost:=los;
        num:=i;
      End;
    End;
  End;

Procedure WriteIt;
  Begin
    If num<>0 Then Write(num,' ',lost)
    Else Write('0');
  End;

Begin
  Init;
  Solve;
  SearchAns;
  WriteIt;
End.

type
  info=record
     fri:array[1..101]of integer;
     no:integer;
     count:integer;
  end;
label endof, endla;
var
  n,sum,i,j:integer;
  node:array[1..100]of info;
  temp:integer;
begin
assign(input,'input.txt');
assign(output,'output.txt');
  readln(input,n);
  for i:=1 to n do
    begin
      read(input,temp);
      j:=1;
      while temp<>0 do
         begin
            node[i].fri[j]:=temp;
            inc(j);
            read(input,temp);
         end;
      dec(j);
      node[i].count:=j;
      node[i].no:=0;
      if j=0 then goto endof;
      readln;
    end;
  sum:=0;
  node[1].no:=1;
  inc(sum);
  for i:=1 to node[1].count do
     node[node[1].fri[i]].no:=2;
  for i:=2 to n do
    begin
    if node[i].no<>2 then
       begin
         node[i].no:=1;
         inc(sum);
         for j:=1 to node[i].count do
            node[node[i].fri[j]].no:=2;
       end;
    end;
    writeln(output,sum);
    for i:=1 to n do
     begin
      if node[i].no=1 then
        begin
         write(output,i);
         write(output,' ');
        end;
     end;
     goto endla;
endof: write(output,0);
endla:  { TODO -oUser -cConsole Main : Insert code here }
end.

somebody see a mistake?



#define MAXINT 2000000000
#include<stdio.h>
#include<math.h>
#include<stdlib.h>


typedef struct{
    int x,y;
    int ID;
    double ang;
}ponto_t;
ponto_t tb[10000];

int sort( const void *a, const void *b){
    ponto_t* ta = (ponto_t*)a;
    ponto_t* tb = (ponto_t*)b;
    if ((ta->x==0) && (ta->y==0)) return -1;
    if ((tb->x==0) && (tb->y==0)) return 1;
       return (ta->x*tb->y - tb->x*ta->y)>0?1:-1;
}


int main(){
    #ifndef ONLINE_JUDGE
        freopen("1207.in","r",stdin);
        freopen("1207.out","w",stdout);
    #endif
    int n,x=MAXINT,themin=0,k;
    scanf("%d",&n);


    for(k=0;k<n;k++){
        scanf("%d %d",&(tb[k].x),&(tb[k].y));
        tb[k].ID=k+1;
        if (tb[k].x<x){
            x=tb[k].x;
            themin=k;
        }
    }
    for(k=0;k<n;k++){
        tb[k].x-=tb[themin].x;
        tb[k].y-=tb[themin].y;
    }


    qsort(tb,n,sizeof(ponto_t),sort);

    printf("%d %d",tb[0].ID,tb[n/2].ID);

    return 0;
}

type
  info=record
         link:array[1..100]of integer;
         position:integer;
         count:integer;
       end;
label 99;
var
  n,i,j,temp,sum:integer;
  a:array[1..100]of info;
begin
  {assign(input,'input.txt');
  assign(output,'output.txt');
  rewrite(output);
  reset(input);  }
  readln(n);
  for i:=1 to n do
     begin
         read(temp);
         j:=0;
         while temp<>0 do
             begin
                 a[i].link[j]:=temp;
                 inc(j);
                 read(temp);
             end;
         dec(j);

         a[i].count:=j;
         if a[i].count=0 then
            goto 99;
         a[i].position:=0;
         readln;
     end;
   a[1].position:=1;
   sum:=1;
   for i:=1 to a[1].count do
      a[a[1].link[i]].position:=2;
   for i:=2 to n do
      begin
         if a[i].position=0 then
            begin
               a[i].position:=1;
               inc(sum);
               for j:=1 to a[i].count do
                  a[a[i].link[j]].position:=2;
            end;
      end;
   writeln(sum);
   writeln(output,sum);
   for i:=1 to n do
     if a[i].position=1 then
        begin
           write(i);
           write(' ');
        end;
   {close(input);
   close(output);}
99: writeln(0);
   readln;
end.

If I ignore the branches with 0 apples (but why ?!)
then I won't get a BINARY tree !
Please, somebody, explain me.

type
  num=array[0..10000] of longint;
function max(a,b:longint):longint;
begin
  if a>b then max:=a else max:=b;
end;

procedure chen(x:num;y:longint;var ans:num);
var
  i,j,k:longint;
begin
  fillchar(ans,sizeof(ans),0);ans[0]:=1;
  if y=0 then exit;
  for i:=1 to x[0] do begin
    inc(ans[i],x[i]*y);
    if ans[i]>=100 then begin
      ans[i+1]:=ans[i] div 100;
      ans[i]:=ans[i] mod 100;
    end;
  end;
  if ans[i+1]>0 then ans[0]:=i+1 else ans[0]:=i;
end;
procedure plus(x,y:num;var ans:num);
var
  i:longint;
begin
  fillchar(ans,sizeof(ans),0);
  for i:=1 to max(x[0],y[0]) do begin
    inc(ans[i],x[i]+y[i]);
    if ans[i]>=100 then begin
      ans[i+1]:=ans[i] div 100;
      ans[i]:=ans[i] mod 100;
    end;
  end;
  if ans[i+1]>0 then ans[0]:=i+1 else ans[0]:=i;
end;
function bigger(x,y:num):longint;
var
  i:longint;
begin
  bigger:=0;
  if x[0]>y[0] then begin bigger:=1;exit;end else if x[0]<y[0] then
begin bigger:=-1;exit;end;
  for i:=x[0] downto 1 do
    if x[i]>y[i] then begin bigger:=1;exit;end else if x[i]<y[i] then
begin bigger:=-1;exit;end;
end;
procedure minus(x,y:num;var ans:num);
var
  i,j,k:longint;
begin
  fillchar(ans,sizeof(ans),0);
  for i:=1 to x[0] do begin
    inc(ans[i],x[i]-y[i]);
    if ans[i]<0 then begin inc(ans[i],100);dec(ans[i+1]);end;
  end;
  ans[0]:=i;
  while (ans[0]>1)and(ans[ans[0]]=0) do dec(ans[0]);
end;
procedure work(x:num);
var
  i,j,k,y:longint;
  a,b,c,d:num;
begin
  fillchar(a,sizeof(a),0);a[0]:=1;b:=a;c:=a;
  a[1]:=2*trunc(sqrt(x[x[0]]));
  c[2]:=x[x[0]]-sqr(a[1] div 2);c[1]:=x[x[0]-1];
  write(a[1] div 2);
  if c[2]=0 then c[0]:=1 else c[0]:=2;
  for i:=x[0]-1 downto 1 do begin
    chen(a,10,a);
    for y:=9 downto 0 do begin
      fillchar(b,sizeof(b),0);b[0]:=1;
      b[1]:=y;
      plus(a,b,b);
      chen(b,y,b);
      if bigger(c,b)>=0 then break;
    end;
    write(y);
    minus(c,b,c);
    fillchar(d,sizeof(d),0);d[0]:=1;d[1]:=y*2;
    plus(a,d,a);
    for j:=c[0] downto 1 do c[j+1]:=c[j];inc(c[0]);
    c[1]:=x[i-1];
    while (c[0]>1)and(c[c[0]]=0) do dec(c[0]);
  end;
  writeln;
end;
var
  a:num;
  i,j,k:longint;
  z:char;
  l:array[0..5000] of longint;
begin
  k:=0;
  while not(eoln) do begin
    read(z);
    inc(k);
    l[k]:=ord(z)-ord('0');
  end;
  a[0]:=0;
  for i:=1 to (k+1) div 2 do a[i]:=l[k-i*2+2]+10*l[k-i*2+1];
  a[0]:=(k+1) div 2;
  chen(a,2,a);
  work(a);
end.

I am very sorry that you can proceduce 0 Horn and 0 Hoof!

var
  z,y,x,a,b,tt,i,j,k,p,q,r:longint;
  ss:array[1..10000] of longint;
  s:string;
begin
  tt:=1;ss[1]:=2;
  for i:=3 to trunc(sqrt(1000000)) do begin
    for j:=1 to tt do if (i mod ss[j]=0)or(ss[j]>trunc(sqrt(i))) then
break;
    if i mod ss[j]<>0 then begin
      inc(tt);
      ss[tt]:=i;
    end;
  end;
  readln(a,b);
  for y:=a to b do begin
    q:=1;
    x:=y;
    r:=1;
    k:=1;
    while (x<>1 )and(k<>tt) do begin
      for j:=k to tt do
        if x mod ss[j]=0 then break;
      k:=j;
      if x mod ss[j]=0 then begin
        z:=1;
        p:=1;
        while x mod ss[j]=0 do begin
          z:=z*ss[j];
          p:=z+p;
          q:=q*ss[j];
          x:=x div ss[j];
        end;
        r:=r*p;
        if (r>=2*q) then break;
      end;
    end;
    if x<>1 then begin r:=r*(x+1);q:=y;end;
    if r<2*q then begin

      writeln(y);
      halt;
    end;
  end;
end.

I just have passed this problem, but I find some strange thing.
I use pascal language, and I used 'Ch=EOF' to judge the end of file
in this problem. And I always got WA. After reading some other codes,
I changed this line to 'Ch=#26', and got AC. I don't know why.

Maybe the input is wrong but i don't know how to do this
 help
thanks !

#include<iostream.h>
#include<math.h>
#include<stdio.h>
const double pi=3.1415926;
const double R=3437.5;

double x_ship,y_ship,z_ship;
double x_ice,y_ice,z_ice;
double dist(double x1,double y1,double z1,double x2,double y2,double
z2)
{
 double ans;
 ans=sqrt((x1-x2)*(x1-x2)+(y1-y2)*(y1-y2)+(z1-z2)*(z1-z2));
 return ans;
}

void main()
{
 double X1,X2,X3,Y1,Y2,Y3;
 double A1,A2,A3,B1,B2,B3;
 double ans,sita,d,r1,r2;
 int location_ship1,location_ship2,location_ice1,location_ice2;
 char ch[200];
 char c;

 //north=1;south=-1;east=1;west=-1;

 //input
 gets(ch);
 gets(ch);
 gets(ch);
 cin>>X1>>c>>X2>>c>>X3>>c>>ch;
 if(ch[0]=='N') location_ship1=1; else location_ship1=-1;
 cin>>ch>>Y1>>c>>Y2>>c>>Y3>>c>>ch;
 if(ch[0]=='E') location_ship2=1; else location_ship2=-1;
 gets(ch);
 cin>>A1>>c>>A2>>c>>A3>>c>>ch;
 if(ch[0]=='N') location_ice1=1; else location_ice1=-1;
 cin>>ch>>B1>>c>>B2>>c>>B3>>c>>ch;
 if(ch[0]=='E') location_ice2=1; else location_ice2=-1;
 gets(ch);


 z_ship=R*sin((X1+(X2+X3/60)/60)*pi/180);
 z_ship*=location_ship1;
 r1=R*cos((X1+(X2+X3/60)/60)*pi/180);
 z_ice=R*sin((A1+(A2+A3/60)/60)*pi/180);
 z_ice*=location_ice1;
 r2=R*cos((A1+(A2+A3/60)/60)*pi/180);

 y_ship=r1*cos((Y1+(Y2+Y3/60)/60)*pi/180);
 y_ice=r2*cos((B1+(B2+B3/60)/60)*pi/180);

 x_ship=r1*sin((Y1+(Y2+Y3/60)/60)*pi/180);
 x_ship*=location_ship2;
 x_ice=r2*sin((B1+(B2+B3/60)/60)*pi/180);
 x_ice*=location_ice2;

 d=dist(x_ship,y_ship,z_ship,x_ice,y_ice,z_ice);
 sita=acos((R*R*2-d*d)/(2*R*R));
 ans=R*sita;
 printf("The distance to the iceberg: %.2lf miles. \n",ans);
 if(ans<100) printf("DANGER!\n");
}

Hi,
i've got a little bit diffrent alg. than 'normal' one. array[-
32000..32000], where i mark begining and ends of interval. than just
searching all table and write the output. i can't find any test,
where my prog. doesn't work, but i still have wrong answer. here's my
code:


type twartosc=record
       val:shortint;
       p1,p2,k1,k2:boolean;
     end;

var tab:array[-32000..32000] of twartosc;
    n,p,i,k,w,j:integer;
    o1,o2:boolean; {is F1 and F2 definned with given argument?)


begin
  read(n);
if n=0 then begin write(0); exit; end;
  for i:=1 to n do
    begin
      read(p);
      read(k);
      read(w);
      for j:=p to k do tab[j].val:=w;
      tab[p].p1:=true;
      tab[k].k1:=true; {reading F1 and mark it in array}
    end;
  readln;
  read(n);
if n<>0 then
  for i:=1 to n do
    begin
      read(p);
      read(k);
      read(w);
      tab[p].p2:=true;
      tab[k].k2:=true; {the same with F2 (it's value isn't nedded)}
    end;

  j:=0; {let j tell us about number of intervals of output function}
  for i:=-32000 to 32000 do
    begin
      if (tab[i].k1) then o1:=false; {close F1}
      if (tab[i].p2) then o2:=true;  {open F2}
      if tab[i].k2 then
        begin
          if o1 then inc(j); {if F1 is opened and we're closing F2,
new interval is begging}
          o2:=false;
        end;
      if tab[i].p1 then
        begin
          if not o2 then inc(j); {similar to prev}
          o1:=true;
        end;
    end;

  write(j,' '); {write number of intervals, and now we're seeking for
them}
  for i:=-32000 to 32000 do
    begin
      if (tab[i].k1) then
        begin
          o1:=false;
          if not o2 then write(i,' ',tab[i].val,' ');
        end;
      if (tab[i].p2) then
        begin
          o2:=true;
          if (o1) and (not tab[i].k2) then write(i,' ',tab
[i].val,' ');
        end;
      if tab[i].k2 then
        begin
          if (o1) and (not tab[i].p2) then write(i,' ');
          o2:=false;
        end;
      if tab[i].p1 then
        begin
          if not o2 then write(i,' ');;
          o1:=true;
        end;
    end;
end.

It's very strange that my program has got AC (It has worked for ~3
sec).
I've been founding answer with binary search ! :))
If you have really fast solution, please, send it on _my_inbox@mail.ru

Help!!! This code get Compile Error. Tell me why, pls!

var n,k,s,t:longint;
begin
   read(n,k);
   s:=1; t:=0;
   for t:=1 to 31 do
   begin
      s:=s*2;
      if s>n then break;
      if s>k then
      begin
         t:=t+(n-s+k-1) div k;
         break;
      end;
   end;
   writeln(t);
end.

Can anyone send me such algorythm (or post url for related info) for
multiplying big nums? _my_inbox@mail.ru
I heard about n*log(n), but never about n^1.5...
I've solved 1108 with standard n*m complexity without any optimizing
with time ~ 2.5 sec, but i want 0.05 sec without any
precomputations :) .  Also, I guess where is some law in digits order
of the numbers, wich can help to compute each number for O(n).