NZEC in PRIME1

[k4rt33k]

Hello Everyone,
I am new to python.I am trying to solve PRIME 1 problem using sieve of Eratosthenes.But I am getting NZEC run time error.My submission id is 2478137.I am also submitting the code here

test_cases = int(raw_input(""))
count = 0
while(count < test_cases):
    input = raw_input("")
    input = input.split()
    lowlimit = int(input[0])
    highlimit = int(input[1])
    primelist = range(2,highlimit+1)
    limit = highlimit//2+1
    number = 2
    while(number < limit):
        if(primelist[number-2]):
            primelist[2*(number-1)::number] = [0]*(highlimit//number-1)
        number = number+1
    primelist = filter(None,primelist)
    if(lowlimit!=1):
        primelist = primelist[lowlimit-2:]
    for prime in primelist:
        print prime
    if(count!=test_cases-1):
        print ""
    count = count + 1

What am I currently doing wrong and any pointers in the right direction?

Cheers
Karteek

[numerix]

I'm not sure what causes the error, but you could improve the input and make it somewhat more pythonic:

testcases = int(raw_input())
for t in xrange(testcases):
    low, high = map(int, raw_input())

It is better to use functions, e.g. use a function main() that handles the input and call a function primes(low, high) that
calculates a list of primes and returns that list to main().

Then you should run some tests with function primes(), e.g. low = 9 and high = 13 ...

[k4rt33k]

Hey Numerix,
Thanks for the reply.My guess is that the stack is overflowing.So I am following a double sieve approach to reduce the size of the list that I store.

I'm not sure what causes the error, but you could improve the input and make it somewhat more pythonic:

Thanks for the tip.But same error again.

[numerix]

Hey Numerix,
Thanks for the reply.My guess is that the stack is overflowing.So I am following a double sieve approach to reduce the size of the list that I store.

No, it isn't a stack problem, as there are no stack-operations (e.g. recurrence).

Before posting your code, check it locally with
- boundary cases, e.g. "1 1"
- worst case, i.e. "999900000 1000000000" (time limit! SPOJ machines are slow!)
- some random stuff (e.g. "9 13" for which your code produces the wrong result as I posted before)

If it has passed all these tests, then - maybe - it will get AC.

P.S. Work on it, don't give up. For me, PRIME1 was the first problemset at SPOJ. I spent days on it and until now my solution is the fastest pure python solution (i.e. without psyco).