Recursion and Python issue

Question

I have been trying to understand recursion. But I don't think I've quite got a hang of it.

Here the outline for my code:

def f():

    used = [anything]
    new = []

    while i < something:
        new.append(i)
        i += 1

    for i in new:
        if i in used:
            f()
        else:
            return new

Now, I don't think I can use this because I'm not iterating and there is no base case. I need to keep running this program till I get a set of values (picked randomly) that are not in used. What would be the best way to achieve that? Create another function?

Any help would be greatly appreciated.

Thanks!

Answer 1

First of all, you need to add parameters, otherwise it's not really recursive. The gist is like this

f(x):
    x-=1
    if x < 5:
        return x
    else:
        f(x)

The point of recursion is to call the function inside itself with a new parameter. x changes value every time so eventually if will drop below 5 and you'll return x (which will be 4). So it would be f(7),subtract 1, f(6), subtract 1, f(5), subtract 1, f(4), return 4.

You also don't define i or something, so you'll have an infinite loop because i will always be less, in fact, I'm surprised the code works, because neither is ever defined.

Answer 2

You should add parameters of the function, and transfer proper parameter to recursive function calls.

For example:

def f(new, used, i):

Answer 3

I think that a regular while loop will solve this problem more sensibly than a recursive call. I don't fully understand what you are trying to achieve with this code, but here it is rewritten with your tropes:

def f(used):

    new = []

    while len(new) == 0 :
        while i < something:
            new.append(i)
            i += 1

        for i in new:
            if i in used:
                new = []
                break

    return new

Answer 4

I think the example you are focusing on is a poor example of recursion. It is almost easier to see an iterative solution to your problem. With a perfect example of recursion, it is hard to see any other solution than a recursive solution.

One of the classic examples of recursion is navigating a tree oriented data structure.

Here is a simple example (hardly tested...):

#!/usr/bin/python

tree = {'text': '1',
        'brnch': [{
                  'text': '1.1',
                  'brnch': [{
                            'text': '1.1.1',
                            'brnch': [{
                                      'text': '1.1.1.1',
                                      'brnch': []}]},
                           {'text': '1.1.2',
                            'brnch': []},
                           {'text': '1.1.3',
                            'brnch': []}]},
                 {'text': '1.2',
                  'brnch': []}]}

def recurse_traverse(tree):
    print ' ' * recurse_traverse.level + tree['text']
    for branch in tree['brnch']:
        recurse_traverse.level += 1
        recurse_traverse(branch)
        recurse_traverse.level -= 1

if __name__ == '__main__':
    import os
    print "testing", os.path.abspath(__file__)   
    recurse_traverse.level = 1
    recurse_traverse(tree)

The fine online book Think Like a Computer Scientist has more examples of recursion .

Answer 5

If I understand correctly, you're asking for a recursive function to generate a list of pseudo-random values that are not included in another list. This recursive function will generate a size number of pseudo-random values that do not exist in used .

from sets import Set
import random

used = Set([1,2,3,4,5])

def f(new, size):
  # Check if we have enough 'new' values that are not in the 'used' set
  if len(new.difference(used)) < size:
    new.add(random.randint(0,100)) # Generate a pseudo-random value and
                                   # add it to the 'new' set
                                   # Values are between 0-100 inclusive
                                   # but you can change that to whatever you like
    new = f(new, size) # Append our results to `new`, this will get the result set
  return new.difference(used) # Return only the different 
                              # values between the two sets

result = f(Set(), 10) # Start with a blank set and a request for a result with 10 values
print(result)