framehints 1.0.1

No project description provided

framehints

Simple library (decorator & function) for validating Pandas Dataframes structure

Installation

pip install git+https://github.com/lutrarutra/framehints

Usage

from dataclasses import dataclass
from datetime import datetime
from pathlib import Path
import numpy as np
import pandas as pd

import framehints as fh

@dataclass
class CustomClass():
    field1: int
    field2: str

schema = {
    "name": str,
    "value": float,
    "date": datetime | None,
    "path": Path | str,
    "data": np.ndarray | list | None,
    "custom": CustomClass,
    "column_that_doesnt_exists": int  # This column is optional, so it won't raise an error if it's missing
}

# Usage with function decorator
@fh.pdvalidate({"df": schema}, optional_columns={"df": ["column_that_doesnt_exists"]})
def your_function(df: pd.DataFrame):
    return df.shape


df = pd.DataFrame({
    'name': ['Alice', 'Bob', 'Charlie'],
    'value': [1.0, 2.5, 2.2],
    'date': [datetime.now(), datetime.now(), None],
    "path": [Path("/tmp"), Path("/var"), "/home"],
    "data": [np.zeros([1, 2]), None, [1,2,3]],
    "custom": [CustomClass(1, "a"), CustomClass(2, "b"), CustomClass(2, "b")]
})

try:
    your_function(df)
    print("1. DataFrame is valid.")
except TypeError as e:
    print(e)


# or usage with function
try:
    df = fh.validate_df(df, schema, optional_columns=["column_that_doesnt_exists"])
    print("2. DataFrame is valid.")
except TypeError as e:
    print(e)