OJ Board

Try an input with a large string of the same character with a different character near the end.
e.g.
xxxxxx......xxxy
xxxxxx........xyxx

IMHO the idea behind the solution (and not the implementation) of this problem is somewhat simpler that the one Hamedv referred to.

Of course, what I find find hard (or easy) can quite naturally seem easy (or hard) to someone else .

Thanks Hamedv. That is a much harder problem. The present problem asks that you only consider adding characters at the end of the string, which simplifies matters a lot.

@baodog:
Your code prints an extra unprintable character at the end of each line on my machine. But it looks otherwise ok.

@hamedv:
I'm curious about the O(n^2) DP you mentioned. The simple brute force way of solving this problem is also O(n^2). Could you please post an outline of your idea?

Function calls are certainly not eating up your time. Instead of using two scanners, why don't you use nextInt() on the first scanner? Maybe declaring the Scanner as

Scanner getInput=new Scanner(new BufferedInputStream(System.in));


might also help.

Java is slower than C++, but this does not ...

Suggestions:
Use Scanner for input
Use longs (the problem statement talks of 32bit ints. Java ints are 32 bits but they are signed so they are effectively 31 bits for positive numbers)
Get rid of this line
if(num1>10000||num2>10000||num1<0||num2<0)break;
Input is guaranteed to be valid, so no need ...

You don't need the recommended readLn method. Java uspport at this judge has been upgraded and it now fully supports jdk1.6. So, you can simply use Scanner and/or BufferedReader. Make sure that the class with the main method is named Main.

The file name does not have to be Main.java. The driver class in the file should be named Main. The file name can be anything you want provided that the class Main is not declared public.

I tried this problem with a line sweep algo, but I keep getting WA. The same approach passed in 10245. Can anyone suggest some tricky test cases?

Thanks in advance,
MAK

Try the easier problems first (e.g. ad-hoc, simple geometry/math). Whenever you run into any trouble, consult this forum as well as any other forums/groups/mailing lists. Just looking up the previous posts on the subject should generally be sufficient to point you in the right direction. Whenever ...

it has been a week here in acm.uva.es, i havent solved a single one

That's not really as bad as you think. I got some 20+ compile errors and a few WAs before I got my first accepted solution (I used to code in Java).

MAK wrote:
You migth also have (to borrow) two bottles.


I actually wrote


You migth also have two bottles.

(sorry for the typo :) )

What I meant was that at the end of the simulation you can have either one or two empty bottles left, but never three. So you can either borrow one or two ...

I think I solved it with long long (C++) instead of double. But double works too I guess.

I just saw something very interesting at the Bug Fixes forum. Someone (probably one of the system admins) says that

1. Java uses 190+ MB of RAM just for a single System.out.println

2. Java is 'unsecure'. Java classes may illegally access system data or files.

I'm quite sure that the second ...

hm... as far as I am concerned the only allowed languages should be those allowed in ACM World Finals + Pascal...

Regards


That's what I used to think before I learnt C#, Python and LISP. I don't think there is any reason to keep such a limited set of allowed languages. Probably more people use ...