fastval 0.1.2

No project description provided

FastVal

Ultra-fast Cython-based data validation library for Python, optimized for FastAPI. A faster alternative to Pydantic.

Features

• Fast validation using Cython
• Nested model support
• Optional fields with defaults
• JSON serialization with orjson
• Drop-in compatible with FastAPI
• Pydantic-style annotations

Installation

Clone the repo and build:

git clone https://github.com/jncremona2/fastval.git
cd fastval
pip install -e .

Usage

Define models with annotations, just like Pydantic:

from fastval import BaseModel
from typing import List

class User(BaseModel):
    name: str
    age: int = 25
    email: str

class Post(BaseModel):
    title: str
    content: str
    author: User
    tags: List[str] = []

# Validate data
user = User({'name': 'John', 'email': '[e-mail]'})
print(user.dict())  # {'name': 'John', 'age': 25, 'email': '[e-mail]'}
print(user.json())  # JSON string

For FastAPI:

from fastapi import FastAPI

app = FastAPI()

@app.post("/users")
async def create_user(payload: dict):
    user = User(payload)
    return user.dict()

Benchmarks

FastVal outperforms Pydantic in validation and JSON serialization:

• Validation (10,000 users): FastVal 2.65x faster (0.0042s vs 0.0112s)
• JSON Serialization: FastVal 3.07x faster (0.0022s vs 0.0066s)
• Nested Models (1,000 posts): FastVal 1.27x faster (0.0009s vs 0.0012s)
• Large Payloads (100,000 users): FastVal 2.29x faster (0.0610s vs 0.1397s)

Run python benchmark.py to see live results.

Tests

Run pytest tests/