Time complexity of string slice

Question

What\'s the time complexity of slicing a Python string? Given that Python strings are immutable, I can imagine slicing them being either O(1) or O(n)

Answer 1

It all depends on how big your slices are. I threw together the following two benchmarks. The first slices the entire string and the second only a little bit. Curve fitting with this tool gives

# s[1:-1]
y = 0.09 x^2 + 10.66 x - 3.25

# s[1:1000]
y = -0.15 x + 17.13706461

The first looks quite linear for slices of strings up to 4MB. I guess this really measures the time taken to construct a second string. The second is pretty constant, although it's so fast it's probably not that stable.

import time

def go(n):
    start = time.time()
    s = "abcd" * n
    for j in xrange(50000):

        #benchmark one
        a = s[1:-1]

        #benchmark two
        a = s[1:1000]

    end = time.time()
    return (end - start) * 1000

for n in range(1000, 100000, 5000):
    print n/1000.0, go(n)