## 10110 - Light, more light

All about problems in Volume 101. If there is a thread about your problem, please use it. If not, create one with its number in the subject.

Moderator: Board moderators

C8H10N4O2
Experienced poster
Posts: 137
Joined: Wed Feb 27, 2002 2:00 am
Location: Pasadena, CA

### 10110 - Light, more light

Works for all test cases. Don't know why it WAs. Some help from them JAVA people would be nice.
[java]import java.io.*;
import java.lang.*;

class Main
{
public static void main(String args[])
{
StringBuffer B;
String S;
int N,Length,i;
char c;
while(true)
{
try
{
B=new StringBuffer();
while((c=(char)System.in.read())!='\n'&&c!=-1)
B.append(c);
}
catch(IOException e)
{
break;
}
S=B.toString();
N = Integer.parseInt(S.substring(0,S.length()-1));
if(N==0||c==-1)
break;
if(Math.pow(Math.floor(Math.sqrt(N)),2)==N)
System.out.println("yes");
else
System.out.println("no");
}
}
};[/java]
wyvmak
Experienced poster
Posts: 110
Joined: Thu Dec 13, 2001 2:00 am
there's a statement in the problem "Which is less then or equals 2^32-1"
i code my solution in C++, but i guess in Java, int would be the same that it cannot hold more than 2^31-1. so maybe try another data type
C8H10N4O2
Experienced poster
Posts: 137
Joined: Wed Feb 27, 2002 2:00 am
Location: Pasadena, CA
Good point. Should have used JAVA long or unsigned int. Thanks.
Stefan Pochmann
A great helper
Posts: 284
Joined: Thu Feb 28, 2002 2:00 am
Location: Germany
Contact:
As far as I remember, Java doesn't have unsigned types. This is also exactly the one single reason one of my friends always mentions to explain why Java is crap
C8H10N4O2
Experienced poster
Posts: 137
Joined: Wed Feb 27, 2002 2:00 am
Location: Pasadena, CA
That program ONLY took me 7 hours in JAVA:O I am starting to see why without the good classes, JAVA is horribly complicated and slow. Although the error checking compiler is quite nice. It is much more strict than C++ and won't let you access out of bounds and such. The other thing I noticed is that the compiler on the 24 hour judge doesn't support all the features of even simple classes; there are wierd little discrepancies. However, I think it would be useful to know from things like TopCoder where you need to analyze other people's code as well as use the nice BigInt. It is silly how some people think StringTokenizer is better than strtok; they haven't mastered strtok. Strings and general input/output is SOOO much easier in C++ (e.g. SSCANF).
ante
New poster
Posts: 8
Joined: Wed Mar 20, 2002 2:00 am

### 10110 More Light's Please

Why I receive WA ??? What's the problem ???

[cpp]
#include<stdio.h>
#include[itex]
void main()
{
int n;
while (1)
{
scanf("%d",&n);
if (n == 0) break;
if (((int)sqrt(n))==sqrt(n))
printf("yes\n");
else
printf("no\n");
}
}
[/cpp]

Thanx in advance
Ivor
Experienced poster
Posts: 150
Joined: Wed Dec 26, 2001 2:00 am
Location: Tallinn, Estonia
Maybe you have a roundng problem?

(int)sqrt could give you a wrong answer. For example 4.999999... would be just 4, not 5. As far as I know at least.

Ivor
ante
New poster
Posts: 8
Joined: Wed Mar 20, 2002 2:00 am

### More Pascal Light also

I tried this in Pascal but it didn't work also, i need to check if N has integer square root .
Please help !

[pascal]
program p10110;
var
n: integer;
begin
while not eof(input) do
begin
readln(n);
if (n = 0) then break;
if frac(sqrt(n))<0.000001 then writeln('yes') else
writeln('no');
end;
end.
[/pascal]

P.S: also tried
[pascal]

if abs(trunc(sqrt(n))-sqrt(n))<0.000001 then writeln('yes') else writeln('no');

if trunc(sqrt(n))*trunc(sqrt(n))=n then writeln('yes') else writeln('no');
[/pascal]
Ivor
Experienced poster
Posts: 150
Joined: Wed Dec 26, 2001 2:00 am
Location: Tallinn, Estonia
Why don't you try generating all squares and then search the array for n.

You'll have 65535 squares at max. So, using binary search you could determine, whteher sqrt(n) is an integer or not. The array won't be that big, but it could be not the fastest way.

Ivor
P.S I haven't solved the problem myself, so I'm just giving some ideas that popped up my mind.
ante
New poster
Posts: 8
Joined: Wed Mar 20, 2002 2:00 am

### UNSIGNED PROBLEM

I didn't read problem carefully enough , 2^32 - 1 is 4,294,967,295 and this must be UNSIGNED int .

Thanx anyway !
Last edited by ante on Wed Apr 24, 2002 11:42 pm, edited 1 time in total.
Ivor
Experienced poster
Posts: 150
Joined: Wed Dec 26, 2001 2:00 am
Location: Tallinn, Estonia
Want to know a secret?

2^31 - 1 = 2147483647.

Ivor
C8H10N4O2
Experienced poster
Posts: 137
Joined: Wed Feb 27, 2002 2:00 am
Location: Pasadena, CA
Isn't ints in pascal 2 bytes? I don't know; I am C++ coder. Have you tried longs? You need unsigned longs in C++ or int64. This problem requires a unsigned 4 byte integer.
xenon
Learning poster
Posts: 100
Joined: Fri May 24, 2002 10:35 am
Location: Scheveningen, Holland

### use type cardinal in pascal

type integer is signed 32 bits.
type cardinal is unsigned 32 bits (could be more, I don't know).
this code works:
[pascal]var
n,r:cardinal;
begin
repeat
readln(n);
if (n=0) then break;
r:=trunc(sqrt(n));
if (r*r=n) then writeln('yes') else writeln('no');
until false;
end.
[/pascal]
so there's no rounding problem
Cubist
New poster
Posts: 17
Joined: Sun May 26, 2002 7:56 am
Location: San Diego, CA

### 10110 - Lights, more lights (slow and fat)

How do you reduce either the running time or the memory usage for
this problem? I was astonished to see my tiny Pascal program using
0.670 seconds and something like 468K! This code doesn't use
anything that other programs I'm submitted, which all ran in 0.000
and 64K.

Any ideas?
--
Chris Long, Mathematics Department, Rutgers University

Mozart needs food. Beethoven now has extra fight power. Bach is about to die.
Betovsky
New poster
Posts: 26
Joined: Wed Jun 05, 2002 7:30 pm
Location: Portugal
read the rulz, there is already a topic about this prob plz post ur questions in there ...

and we dont know nothing about ur code do u want us to guess out ???