Question

I’m having trouble debugging this code excerpt. Never mind the fact that it doesn’t actually return what the author wants, I already explained that, my question is different.

def factors(n):
    result = []
    for x in xrange(2,n):
        print "\t%i,foo" % x
        if n % x == 0:
            isPrime = True
            print "\t\t%i,bar" % x
            for factor in result:
                print "\t\t%i %% %i = %i" % (x,factor,x % factor)
                if x % factor == 0:
                        isPrime = False
                        print "\t\t\t%i,foobar" % x
                        subFactors = factors(x)
                        result.extend(subFactors)
            if isPrime:
                result.append(x)
print ""
return result

def main():
    factor = dict()
    for i in xrange(1,100):
        factor[i] = factors(i)
        factor[i].insert(0,1)
        factor[i].append(i)
        print "%i: %s" % (i,factor[i])

if __name__ == "__main__":
     main()

That code is infinite looping! Specifically, repeatedly outputs the following:

2,foo
    2,bar
3,foo

    4 % 2 = 0
        4,foobar

Note that on the second iteration it doesn’t print the “bar” print statement, and the value of x changes from the “bar” print statement to the “mod” print statement.

I’m unable to explain this behavior to the author. Can one of you kind people?

Answer 1

The issue is that you’re extending the result list while you’re iterating over its contents. Here are the two critical lines, without the other distracting bits around them:

for factor in result:
    result.extend(subFactors)

This has undefined behavior, according to the python spec. I’m pretty sure it is the cause of the infinite loop.