Request Body & Validation

JustAPI validates request bodies against schemas defined with Pydantic models or the built-in Schema class. Validation runs in Rust by default, keeping the GIL released.

Using Pydantic Models

from justapi import JustAPIApp
from pydantic import BaseModel

app = JustAPIApp()


class Item(BaseModel):
    name: str
    description: str | None = None
    price: float
    tax: float | None = None


@app.post("/items/")
def create_item(request, item: Item):
    item_dict = item.model_dump()
    if item.tax:
        item_dict["price_with_tax"] = item.price + item.tax
    return item_dict

curl -X POST http://127.0.0.1:8000/items/ \
  -H "Content-Type: application/json" \
  -d '{"name": "Laptop", "price": 999.99, "tax": 50.0}'
# Output: {"name":"Laptop","description":null,"price":999.99,"tax":50.0,"price_with_tax":1049.99}

Validation Errors

Invalid data returns 422 with details:

curl -X POST http://127.0.0.1:8000/items/ \
  -H "Content-Type: application/json" \
  -d '{"name": "Laptop", "price": "free"}'
# Output: {"detail":"validation error for price: invalid type"}

Using `body_schema` for Rust-Side Validation

For maximum performance, pass a schema explicitly via body_schema. Validation runs entirely in Rust before Python is called:

from justapi import JustAPIApp, Schema
from pydantic import Field


class ProductCreate(Schema):
    name: str = Field(..., min_length=1, max_length=100)
    price: float = Field(..., gt=0, le=100000)
    quantity: int = Field(0, ge=0)


@app.post("/products/", body_schema=ProductCreate)
def create_product(request):
    data = request.json()
    return {"status": "created", "product": data}

The Native Fast Path (`native=True`)

For the absolute fastest path (724,000+ RPS), use native=True. The entire handler runs in Rust — no Python bytecode execution:

@app.post("/fast-items/", body_schema=ProductCreate, native=True)
def fast_create(request):
    return {"status": "ok"}

Performance Comparison

Approach	RPS	p50 Latency
Python handler + Pydantic	~60,000	0.8 ms
`body_schema` (Rust validates)	~280,000	0.2 ms
`native=True` (full Rust)	~724,000	0.07 ms

Using JustAPI's Native `Schema` Class

The built-in Schema class supports field constraints with Field():

from justapi import JustAPIApp, Schema
from pydantic import Field


class UserCreate(Schema):
    username: str = Field(..., min_length=3, max_length=50, regex=r"^\w+$")
    email: str = Field(..., format="email")
    age: int = Field(..., ge=18, le=120)
    role: str = Field("user", enum=["user", "admin", "moderator"])

Supported Field Constraints

Constraint	Type	Description
`gt`	number	Greater than
`ge`	number	Greater than or equal
`lt`	number	Less than
`le`	number	Less than or equal
`min_length`	int	Minimum string length
`max_length`	int	Maximum string length
`regex`	str	Regex pattern match
`format`	str	Named format (`email`, `uri`, `uuid`, `date`, `date-time`, `hostname`, `ipv4`)
`enum`	list	Allowed values
`default`	any	Default value
`description`	str	Field description for OpenAPI

Nested Models

class Address(Schema):
    street: str
    city: str
    zip_code: str = Field(..., regex=r"^\d{5}(-\d{4})?$")


class Customer(Schema):
    name: str
    address: Address
    tags: list[str]

The nested Address schema is lifted into $defs and referenced via $ref in the generated JSON Schema. Validation runs entirely in Rust.

Request Body Size Limits

By default, request bodies are limited to 50 MiB. You can configure this:

app.run("0.0.0.0:8000", max_body_size=1024 * 1024)  # 1 MiB cap

Bodies exceeding the limit receive 413 Payload Too Large immediately.

Next Steps

Dependency Injection — Access request components with Depends
Error Handling — Custom validation error responses
Native Fast Path — Deep dive into Rust-native execution

 Prev: Query Parameters Next: Error Handling 