Re: Python

Allez, on dégaine un iPython, c'est parti.

In [4]: some_list = 'setanéstebnsetanustebnésatenastubnestpnesanuastnbtsnastbnstanstbnsteanutsenbstanstbnstaà' * 1000
In [7]: %%timeit
 ...: 
 ...: list_a = []
 ...: list_b = []
 ...: 
 ...: for s in some_list:
 ...: if s[0] == 'a':
 ...: list_a.append(s)
 ...: elif s[0] == 'b':
 ...: list_b.append(s)
 ...: 
 ...: 
6.08 ms ± 142 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)
In [8]: %%timeit
 ...: 
 ...: list_a = [s for s in some_list if s.startswith('a')]
 ...: list_b = [s for s in some_list if s.startswith('b')]
 ...: 
 ...: 
11.8 ms ± 159 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)
In [10]: def starts_with_a(s):
 ...: return s[0] == 'a'
 ...: 
 ...: def starts_with_b(s):
 ...: return s[0] == 'b'
 ...: 
In [11]: %%timeit
 ...: 
 ...: list_a = list(filter(starts_with_a, some_list))
 ...: list_b = list(filter(starts_with_b, some_list))
 ...: 
 ...: 
10.4 ms ± 187 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)
In [12]: %%timeit
 ...: generator_a = (s for s in some_list if s.startswith('a'))
 ...: generator_b = (s for s in some_list if s.startswith('b'))
 ...: 
 ...: 
401 ns ± 1.17 ns per loop (mean ± std. dev. of 7 runs, 1000000 loops each)
In [13]: %%timeit
 ...: generator_a = (s for s in some_list if s.startswith('a'))
 ...: generator_b = (s for s in some_list if s.startswith('b'))
 ...: 
 ...: len(list(generator_a))
 ...: len(list(generator_b))
 ...: 
 ...: 
12 ms ± 113 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)
In [14]: len(some_list)
Out[14]: 88000

Sur une telle liste, moyennement grosse, les générateurs ont l'air d'aller légèrement plus vite. C'était un test rapide et sûrement pourrave.