import pandas as pd

import numpy as np

def validate_dataset(file_path, expected_columns=None, data_type_checks=None, min_value_checks=None, max_value_checks=None, missing_value_threshold=0.1):

"""

Validates a dataset for common errors.

Args:

file_path (str): Path to the dataset file (CSV or Excel).

expected_columns (list, optional): List of expected column names. Defaults to None.

data_type_checks (dict, optional): Dictionary specifying data type checks for columns.

e.g., {'column_name': 'int64', 'another_column': 'float64'}

min_value_checks (dict, optional): Dictionary specifying minimum value checks for columns.

e.g., {'column_name': 0}

max_value_checks (dict, optional): Dictionary specifying maximum value checks for columns.

e.g., {'column_name': 100}

missing_value_threshold (float, optional): Acceptable percentage of missing values. Defaults to 0.1.

Returns:

dict: A dictionary containing a list of validation errors. Returns an empty dict if no errors found.

"""

errors = []

try:

if file_path.endswith('.csv'):

df = pd.read_csv(file_path)

elif file_path.endswith('.xlsx') or file_path.endswith('.xls'):

df = pd.read_excel(file_path)

else:

errors.append(f"Error: Unsupported file format: {file_path}")

return errors # Exit early if file type is not supported

except FileNotFoundError:

errors.append(f"Error: File not found: {file_path}")

return errors

except Exception as e:

errors.append(f"Error: Could not read file: {e}")

return errors

# Check for missing columns

if expected_columns:

missing_cols = [col for col in expected_columns if col not in df.columns]

if missing_cols:

errors.append(f"Missing columns: {missing_cols}")

# Check for data types

if data_type_checks:

for col, expected_type in data_type_checks.items():

if col not in df.columns:

errors.append(f"Error: Column '{col}' not found in dataset.")

continue # Skip to the next check

try:

df[col] = df[col].astype(expected_type)

except ValueError:

errors.append(f"Error: Column '{col}' has incorrect data type. Expected {expected_type}, got {df[col].dtype}")

except Exception as e:

errors.append(f"Error: Could not convert column '{col}' to type {expected_type}: {e}")

# Check for min/max values

if min_value_checks:

for col, min_val in min_value_checks.items():

if col not in df.columns:

errors.append(f"Error: Column '{col}' not found in dataset.")

continue

if (df[col] < min_val).any():

errors.append(f"Error: Column '{col}' contains values below the minimum of {min_val}")

if max_value_checks:

for col, max_val in max_value_checks.items():

if col not in df.columns:

errors.append(f"Error: Column '{col}' not found in dataset.")

continue

if (df[col] > max_val).any():

errors.append(f"Error: Column '{col}' contains values above the maximum of {max_val}")

# Check for missing values

total_rows = len(df)

missing_count = df.isnull().sum().sum()

missing_percentage = (missing_count / total_rows) if total_rows > 0 else 0

if missing_percentage > missing_value_threshold:

errors.append(f"Warning: Missing values exceed threshold ({missing_percentage:.2%}). Total missing values: {missing_count}")

return errors