Home

# pdschema Validate pandas DataFrames against column contracts. Types, nullability, and per-cell checks — no cleaning or transforms.

Column Contracts Declare dtype, nullability, and per-column validators in a `Schema`.

Class-Based Schemas Define columns as class attributes — inherit, compose, reuse.

Function Validation `@pdfunction` checks DataFrame inputs and outputs at call time.

Extensible Validators Subclass `Validator` or pass any `callable` that returns a bool.

Try it

pip install pdschema

import pandas as pd

from pdschema import Column, IsNonEmptyString, IsPositive, Range, Schema

df = pd.DataFrame({
    "name": ["Alice", "Bob", "Charlie"],
    "age": [25, 30, 35],
    "score": [85.5, 92.0, 78.5],
})

schema = Schema([
    Column("name", str, nullable=False, validators=[IsNonEmptyString()]),
    Column("age", int, validators=[IsPositive()]),
    Column("score", float, validators=[Range(0, 100)]),
])

schema.validate(df)

Guides

Guide	What you learn
Quickstart	Define a schema, validate a DataFrame, read errors
Validators	Built-in checks and writing your own
Functions	Validate inputs and outputs with @pdfunction