pydantic vs. cattrs performance

pydantic_cattrs_timeit.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import urllib.request\n",
    "from dataclasses import dataclass\n",
    "from typing import TypeAlias, TypedDict\n",
    "\n",
    "import orjson\n",
    "import pydantic_core\n",
    "from attrs import define\n",
    "from cattrs import structure\n",
    "from pydantic import BaseModel, TypeAdapter\n",
    "\n",
    "url = \"https://jsonplaceholder.typicode.com/posts\"\n",
    "\n",
    "with urllib.request.urlopen(url) as response:\n",
    "    json_content = response.read()\n",
    "\n",
    "\n",
    "class TypedDictPost(TypedDict):\n",
    "    userId: int\n",
    "    id: int\n",
    "    title: str\n",
    "    body: str\n",
    "\n",
    "\n",
    "TypedDictPosts: TypeAlias = list[TypedDictPost]\n",
    "\n",
    "\n",
    "@dataclass\n",
    "class DataclassPost:\n",
    "    userId: int\n",
    "    id: int\n",
    "    title: str\n",
    "    body: str\n",
    "\n",
    "\n",
    "DataclassPosts: TypeAlias = list[DataclassPost]\n",
    "\n",
    "\n",
    "@define\n",
    "class AttrsPost:\n",
    "    userId: int\n",
    "    id: int\n",
    "    title: str\n",
    "    body: str\n",
    "\n",
    "\n",
    "AttrsPosts: TypeAlias = list[AttrsPost]\n",
    "\n",
    "\n",
    "class PydanticPost(BaseModel):\n",
    "    userId: int\n",
    "    id: int\n",
    "    title: str\n",
    "    body: str\n",
    "\n",
    "\n",
    "PydanticPosts = TypeAdapter(list[PydanticPost])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "113 μs ± 12.9 μs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n",
      "136 μs ± 18.7 μs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n",
      "34.4 μs ± 1.73 μs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n"
     ]
    }
   ],
   "source": [
    "# Parse Any JSON\n",
    "%timeit json.loads(json_content)\n",
    "%timeit pydantic_core.from_json(json_content)\n",
    "%timeit orjson.loads(json_content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "91.3 μs ± 15.8 μs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n",
      "112 μs ± 4.29 μs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n",
      "132 μs ± 23.2 μs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n"
     ]
    }
   ],
   "source": [
    "# Parse JSON with cattrs \n",
    "%timeit structure(orjson.loads(json_content), TypedDictPosts)\n",
    "%timeit structure(orjson.loads(json_content), DataclassPosts)\n",
    "%timeit structure(orjson.loads(json_content), AttrsPosts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "154 μs ± 5.66 μs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n",
      "140 μs ± 20.1 μs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n"
     ]
    }
   ],
   "source": [
    "# Parse JSON with pydantic\n",
    "%timeit PydanticPosts.validate_json(json_content)\n",
    "%timeit PydanticPosts.validate_python(orjson.loads(json_content))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}