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
|
Performance
===========
Negative values mean that the library failed the test.
The tests are done on my PC. The following libraries are tested:
* `typedload`, the 3 most recent versions. It shines with tagged unions, which is what I mostly use.
* `pydantic2` years of work to rewrite it in Rust, [implemented detection of tagged unions years after I did it](https://github.com/pydantic/pydantic/issues/5163#issuecomment-1619203179), still managing to lose some benchmarks 😅
* `apischema` is slower where there are unions, faster otherwise
Using Python 3.13
-----------------
Using Pypy 7.3.17
-----------------
Run the tests
-------------
Generate the performance chart locally.
```bash
python3 -m venv perfvenv
. perfvenv/bin/activate
pip install apischema pydantic attrs
export PYTHONPATH=$(pwd)
make gnuplot
```
|
