Condition statement without loops

Question

I'm writing a small function to check if all elements in a list are less than or equal to a limit. This is just for practice, and should be done without using any loops.

def small_enough(a, limit): 
    return all(x <= limit for x in a)
small_enough([66, 101], 200)

Been on this for a while but I can not find any substituting code to replace the for loop.This code works perfectly fine as it is - however, I am trying to get the results without using loops. Trying to write something a bit more 'pythonic'.

Doing anything that examines every element of a list of arbitrary size is *by definition* a loop. — jasonharper, Aug 27 '19 at 14:08
Yes, but not all loops are necessarily multi-line "for x in iterable: ..." loops. What you have is a loop, and it's fine. — kojiro, Aug 27 '19 at 14:08
@pragmaticlearner I'd say "using built-in/library functions with neat comprehensions" — Adam.Er8, Aug 27 '19 at 14:08
In the future, if you have working code and are trying to improve it, consider going to https://codereview.stackexchange.com — kojiro, Aug 27 '19 at 14:09
No matter how you solve it, it involve a loop, either explicit or implicit (if using numpy or something similar). — Vaibhav Vishal, Aug 27 '19 at 14:10
Oh ok! Thanks for the quick responses everyone. @kojiro Will do! — pragmatic learner, Aug 27 '19 at 14:12
You can go with a recursive function, but that's practically a loop too. — Vaibhav Vishal, Aug 27 '19 at 14:12
If you want to really hide the loop, you'd need some higher-order functions like `functools.reduce(operator.and_, map(lambda x: x <= limit, a), True)`. I'm pretty sure most people would consider that *less* Pythonic, though. — chepner, Aug 27 '19 at 14:16
I would consider the `for` keyword in this case to be a generator, rather than a loop, and you solved for the main intent by using `all` rather than an explicit `return False` within a loop and `return True` if you complete the loop — wrschneider, Aug 27 '19 at 14:54

score 4 · Answer 1 · answered Aug 27 '19 at 14:07

4

If using numpy is ok, you can do

import numpy as np

x = np.asarray([66,101])
print(all(x <= 200))

answered Aug 27 '19 at 14:07

user2653663

2,818
1
18
22

score 2 · Accepted Answer · answered Aug 27 '19 at 14:14

2

In general,

f(x) for x in xs == map(f, xs)

so

def small_enough(a, limit): 
    return all(map(lambda x: x <= limit, a))

answered Aug 27 '19 at 14:14

molbdnilo

64,751
3
43
82

Excuse my inexperience,but what does lambda do? – pragmatic learner Aug 27 '19 at 14:16
It's an expression that defines an anonymous function. – chepner Aug 27 '19 at 14:17
@pragmaticlearner Check out [the official tutorial](https://docs.python.org/3/tutorial/controlflow.html#lambda-expressions). – molbdnilo Aug 27 '19 at 14:17

DjaouadNM · Answer 3 · 2019-08-29T11:09:06.040

1

There's always a loop somewhere, even if you can't see it, and here's just one solution without a for loop:

def small_enough(a, limit): 
    return not sum(map(lambda x : x > limit, a))

print(small_enough([66, 101], 200))  #  True
print(small_enough([66, 201], 200))  #  False

edited Aug 29 '19 at 11:09

answered Aug 27 '19 at 14:14

DjaouadNM

22,013
4
33
55

1

Calling dundermethods directly is not a good idea. Here's just one example that doesn't work: `(3).__gt__(2.0)` -> `NotImplemented`. – Aran-Fey Aug 29 '19 at 11:07
@Aran-Fey I just rolled-back my edit, I was using a lambda before, and thanks for your comment, it is really helpful. – DjaouadNM Aug 29 '19 at 14:28

score 1 · Answer 4 · answered Aug 27 '19 at 14:14

1

Of course you can use recursion here:

def small_enough(a, limit):
    if a:
        return a[0] <= limit and small_enough(a[1:], limit)
    else:
        return True

or even

def small_enough(a, limit):
    return not a or (a[0] <= limit and small_enough(a[1:], limit))

But your solution is really much more readable than this. Besides, recursion doesn't actually avoid looping, it just looks different.

answered Aug 27 '19 at 14:14

kojiro

74,557
19
143
201

The recursion is also horrendously inefficient (and, for large enough `a`, will probably produce an exception). – chepner Aug 27 '19 at 14:17
@chepner that said, recursion does force you to give some thought to the construction of the code. In this case, I had to ask myself "does OP really want the answer to be `True` when the input list is empty?" I stuck with what `all` does, but I think it's nevertheless useful to have to think about stuff in more than one way. – kojiro Aug 27 '19 at 14:19
And worth pointing out Python doesn't have tail call optimization, so rewriting loops as tail recursion is an anti-pattern in Python – wrschneider Aug 27 '19 at 14:55

Condition statement without loops

4 Answers4