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
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
|
#!/usr/bin/env python3
# SPDX-License-Identifier: (Apache-2.0 OR MIT)
import dataclasses
import io
import json
import os
from timeit import timeit
from typing import List
import rapidjson
import simplejson
import ujson
from tabulate import tabulate
import orjson
os.sched_setaffinity(os.getpid(), {0, 1})
@dataclasses.dataclass
class Member:
id: int
active: bool
@dataclasses.dataclass
class Object:
id: int
name: str
members: List[Member]
objects_as_dataclass = [
Object(i, str(i) * 3, [Member(j, True) for j in range(0, 10)])
for i in range(100000, 102000)
]
objects_as_dict = [dataclasses.asdict(each) for each in objects_as_dataclass]
output_in_kib = len(orjson.dumps(objects_as_dict)) / 1024
print(f"{output_in_kib:,.0f}KiB output (orjson)")
def default(__obj):
if dataclasses.is_dataclass(__obj):
return dataclasses.asdict(__obj)
headers = ("Library", "dict (ms)", "dataclass (ms)", "vs. orjson")
LIBRARIES = ("orjson", "ujson", "rapidjson", "simplejson", "json")
ITERATIONS = 100
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":
as_dict = timeit(
lambda: json.dumps(objects_as_dict).encode("utf-8"),
number=ITERATIONS,
)
as_dataclass = timeit(
lambda: json.dumps(objects_as_dataclass, default=default).encode("utf-8"),
number=ITERATIONS,
)
elif lib_name == "simplejson":
as_dict = timeit(
lambda: simplejson.dumps(objects_as_dict).encode("utf-8"),
number=ITERATIONS,
)
as_dataclass = timeit(
lambda: simplejson.dumps(objects_as_dataclass, default=default).encode(
"utf-8"
),
number=ITERATIONS,
)
elif lib_name == "ujson":
as_dict = timeit(
lambda: ujson.dumps(objects_as_dict).encode("utf-8"),
number=ITERATIONS,
)
as_dataclass = None
elif lib_name == "rapidjson":
as_dict = timeit(
lambda: rapidjson.dumps(objects_as_dict).encode("utf-8"),
number=ITERATIONS,
)
as_dataclass = timeit(
lambda: rapidjson.dumps(objects_as_dataclass, default=default).encode(
"utf-8"
),
number=ITERATIONS,
)
elif lib_name == "orjson":
as_dict = timeit(lambda: orjson.dumps(objects_as_dict), number=ITERATIONS)
as_dataclass = timeit(
lambda: orjson.dumps(
objects_as_dataclass, None, orjson.OPT_SERIALIZE_DATACLASS
),
number=ITERATIONS,
)
orjson_as_dataclass = per_iter_latency(as_dataclass)
else:
raise NotImplementedError
as_dict = per_iter_latency(as_dict)
as_dataclass = per_iter_latency(as_dataclass)
if lib_name == "orjson":
compared_to_orjson = 1
elif as_dataclass:
compared_to_orjson = int(as_dataclass / orjson_as_dataclass)
else:
compared_to_orjson = None
table.append(
(
lib_name,
f"{as_dict:,.2f}" if as_dict else "",
f"{as_dataclass:,.2f}" if as_dataclass else "",
f"{compared_to_orjson:d}" if compared_to_orjson else "",
)
)
buf = io.StringIO()
buf.write(tabulate(table, headers, tablefmt="github"))
buf.write("\n")
print(buf.getvalue())
|
