Code that randomly capitalizes each letter of a string (Code cleaning help)
Using choice
instead, and calling lower
and upper
only once.
from random import choice
def rand_upper(string):
return ''.join(map(choice, zip(string.lower(), string.upper())))
Even better, as Peter commented:
def rand_upper(string):
return ''.join(map(choice, zip(string, string.swapcase())))
Another, based on Olvin_Roght's:
def rand_upper(string):
return ''.join([c if getrandbits(1) else c.swapcase() for c in string])
Two more, mixing our solutions for the fastest so far:
def rand_upper(string):
return ''.join([c if getrandbits(1) else d
for c, d in zip(string, string.swapcase())])
def rand_upper(string):
return ''.join([z[getrandbits(1)] for z in zip(string, string.swapcase())])
Benchmark using string = rand_upper('a' * 1000)
:
739 μs 797 μs 725 μs original
764 μs 787 μs 693 μs original_2
713 μs 691 μs 680 μs Samwise
699 μs 657 μs 682 μs theCoder
477 μs 486 μs 490 μs superb_rain
520 μs 476 μs 489 μs Peter_Wood
135 μs 131 μs 141 μs based_on_Olvin_Roght
120 μs 113 μs 121 μs superb_Peter_Olvin
125 μs 117 μs 118 μs superb_Peter_Olvin_2
(Not including Olvin's original because it's the only one with quadratic instead of linear time, so a comparison with a single size would be misleading.)
Code:
from timeit import repeat
from random import randrange, choice, getrandbits
def original(string):
import random
strList = [l for l in string.lower()]
newList = []
for i in strList:
j = random.randrange(2)
if j == 1:
letter = i.upper()
newList.append(letter)
else:
newList.append(i)
return "".join(newList)
def original_2(string):
strList = [l for l in string.lower()]
newList = []
for i in strList:
j = randrange(2)
if j == 1:
letter = i.upper()
newList.append(letter)
else:
newList.append(i)
return "".join(newList)
def Samwise(string: str) -> str:
return "".join(
c.upper() if randrange(2) else c.lower()
for c in string
)
def theCoder(string):
return ''.join(choice((str.upper, str.lower))(c) for c in string)
def superb_rain(string):
return ''.join(map(choice, zip(string.lower(), string.upper())))
def Peter_Wood(string):
return ''.join(map(choice, zip(string, string.swapcase())))
def based_on_Olvin_Roght(string):
return ''.join([c if getrandbits(1) else c.swapcase() for c in string])
def superb_Peter_Olvin(string):
return ''.join([c if getrandbits(1) else d for c, d in zip(string, string.swapcase())])
def superb_Peter_Olvin_2(string):
return ''.join([z[getrandbits(1)] for z in zip(string, string.swapcase())])
funcs = original, original_2, Samwise, theCoder, superb_rain, Peter_Wood, based_on_Olvin_Roght, superb_Peter_Olvin, superb_Peter_Olvin_2
string = original('a' * 1000)
number = 1000
tss = [[] for _ in funcs]
for _ in range(4):
for func, ts in zip(funcs, tss):
t = min(repeat(lambda: func(string), number=number)) / number
ts.append(t)
print(*('%d μs ' % (1e6 * t) for t in ts[1:]), func.__name__)
print()
Mastering generator expressions is a great way to make code like this shorter:
from random import randrange
def rand_upper(string: str) -> str:
return "".join(
c.upper() if randrange(2) else c.lower()
for c in string
)
>>> rand_upper("Sphinx of black quartz, witness my vow!")
'sPhiNx of BlacK qUARTz, wiTnEsS mY VOw!'
The general trick is that any time you're building a list by append
ing one element at a time, there's probably a way that you could do it more simply as a list comprehension, by writing an expression that generates each element of the list.
If you're not actually returning the list, and are instead passing it to a function that accepts any iterable (e.g. str.join
), you can leave out the list part (the []
) and just pass the generator expression directly to that function.