1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
|
#!/usr/bin/env python3
# SPDX-License-Identifier: (Apache-2.0 OR MIT)
import io
import json
import lzma
import os
from pathlib import Path
from timeit import timeit
import rapidjson
import simplejson
import ujson
from tabulate import tabulate
import orjson
os.sched_setaffinity(os.getpid(), {0, 1})
dirname = os.path.join(os.path.dirname(__file__), "..", "data")
def read_fixture_obj(filename):
path = Path(dirname, filename)
if path.suffix == ".xz":
contents = lzma.decompress(path.read_bytes())
else:
contents = path.read_bytes()
return orjson.loads(contents)
data = read_fixture_obj("twitter.json.xz")
headers = ("Library", "unsorted (ms)", "sorted (ms)", "vs. orjson")
LIBRARIES = ("orjson", "ujson", "rapidjson", "simplejson", "json")
ITERATIONS = 500
def per_iter_latency(val):
if val is None:
return None
return (val * 1000) / ITERATIONS
table = []
for lib_name in LIBRARIES:
if lib_name == "json":
time_unsorted = timeit(
lambda: json.dumps(data).encode("utf-8"),
number=ITERATIONS,
)
time_sorted = timeit(
lambda: json.dumps(data, sort_keys=True).encode("utf-8"),
number=ITERATIONS,
)
elif lib_name == "simplejson":
time_unsorted = timeit(
lambda: simplejson.dumps(data).encode("utf-8"),
number=ITERATIONS,
)
time_sorted = timeit(
lambda: simplejson.dumps(data, sort_keys=True).encode("utf-8"),
number=ITERATIONS,
)
elif lib_name == "ujson":
time_unsorted = timeit(
lambda: ujson.dumps(data).encode("utf-8"),
number=ITERATIONS,
)
time_sorted = timeit(
lambda: ujson.dumps(data, sort_keys=True).encode("utf-8"),
number=ITERATIONS,
)
elif lib_name == "rapidjson":
time_unsorted = timeit(
lambda: rapidjson.dumps(data).encode("utf-8"),
number=ITERATIONS,
)
time_sorted = timeit(
lambda: rapidjson.dumps(data, sort_keys=True).encode("utf-8"),
number=ITERATIONS,
)
elif lib_name == "orjson":
time_unsorted = timeit(lambda: orjson.dumps(data), number=ITERATIONS)
time_sorted = timeit(
lambda: orjson.dumps(data, None, orjson.OPT_SORT_KEYS),
number=ITERATIONS,
)
orjson_time_sorted = per_iter_latency(time_sorted)
else:
raise NotImplementedError
time_unsorted = per_iter_latency(time_unsorted)
time_sorted = per_iter_latency(time_sorted)
if lib_name == "orjson":
compared_to_orjson = 1
elif time_unsorted:
compared_to_orjson = time_sorted / orjson_time_sorted
else:
compared_to_orjson = None
table.append(
(
lib_name,
f"{time_unsorted:,.2f}" if time_unsorted else "",
f"{time_sorted:,.2f}" if time_sorted else "",
f"{compared_to_orjson:,.1f}" if compared_to_orjson else "",
)
)
buf = io.StringIO()
buf.write(tabulate(table, headers, tablefmt="github"))
buf.write("\n")
print(buf.getvalue())
|
