[issue32945] sorted(generator) is slower than sorted(list-comprehension)
Antony Lee
report at bugs.python.org
Sun Feb 25 04:30:03 EST 2018
Antony Lee <anntzer.lee at gmail.com> added the comment:
Feel free to close the issue if that's not the forum for this discussion, but I'm still baffled by what's happening.
In the example that you give, the first case needs to look up the `list` global and that callable (which happens to be the `list` type) needs to figure out what to do with generator passed in. Indeed, we can compare
In [22]: dis.dis(compile("[i for i in [1, 2]]", "<string>", "single"))
1 0 LOAD_CONST 0 (<code object <listcomp> at 0x7f2d3237ec00, file "<string>", line 1>)
2 LOAD_CONST 1 ('<listcomp>')
4 MAKE_FUNCTION 0
6 LOAD_CONST 5 ((1, 2))
8 GET_ITER
10 CALL_FUNCTION 1
12 PRINT_EXPR
14 LOAD_CONST 4 (None)
16 RETURN_VALUE
In [23]: dis.dis(compile("list(i for i in [1, 2])", "<string>", "single"))
1 0 LOAD_NAME 0 (list)
2 LOAD_CONST 0 (<code object <genexpr> at 0x7f2d32392150, file "<string>", line 1>)
4 LOAD_CONST 1 ('<genexpr>')
6 MAKE_FUNCTION 0
8 LOAD_CONST 5 ((1, 2))
10 GET_ITER
12 CALL_FUNCTION 1
14 CALL_FUNCTION 1
16 PRINT_EXPR
18 LOAD_CONST 4 (None)
20 RETURN_VALUE
Note how the latter has an extra function call (to `list`).
In the example I gave, however:
In [24]: dis.dis(compile("sorted([i for i in [1, 2]])", "<string>", "single"))
1 0 LOAD_NAME 0 (sorted)
2 LOAD_CONST 0 (<code object <listcomp> at 0x7f2d3231eb70, file "<string>", line 1>)
4 LOAD_CONST 1 ('<listcomp>')
6 MAKE_FUNCTION 0
8 LOAD_CONST 5 ((1, 2))
10 GET_ITER
12 CALL_FUNCTION 1
14 CALL_FUNCTION 1
16 PRINT_EXPR
18 LOAD_CONST 4 (None)
20 RETURN_VALUE
In [25]: dis.dis(compile("sorted(i for i in [1, 2])", "<string>", "single"))
1 0 LOAD_NAME 0 (sorted)
2 LOAD_CONST 0 (<code object <genexpr> at 0x7f2d32328930, file "<string>", line 1>)
4 LOAD_CONST 1 ('<genexpr>')
6 MAKE_FUNCTION 0
8 LOAD_CONST 5 ((1, 2))
10 GET_ITER
12 CALL_FUNCTION 1
14 CALL_FUNCTION 1
16 PRINT_EXPR
18 LOAD_CONST 4 (None)
20 RETURN_VALUE
so both cases are much more similar -- superficially, at least.
----------
_______________________________________
Python tracker <report at bugs.python.org>
<https://bugs.python.org/issue32945>
_______________________________________
More information about the Python-bugs-list
mailing list