src/data/loaders/base.py

"""
Abstract base class for data loaders.
Dependency Inversion Principle: depend on abstractions, not concrete implementations.
"""
from abc import ABC, abstractmethod
from pathlib import Path
from typing import Any, Dict, Optional
import pandas as pd


class BaseDataLoader(ABC):
    """
    Abstract base loader for all data sources.
    Concrete loaders must implement load() method.
    """

    def __init__(self, data_path: Path, **kwargs):
        """
        Initialize loader with data path.

        Args:
            data_path: Path to data file or directory
            **kwargs: Additional loader-specific parameters
        """
        self.data_path = Path(data_path)
        self.kwargs = kwargs
        self._validate_path()

    def _validate_path(self) -> None:
        """Validate that data path exists."""
        if not self.data_path.exists():
            raise FileNotFoundError(f"Data path not found: {self.data_path}")

    @abstractmethod
    def load(self) -> pd.DataFrame:
        """
        Load data and return as DataFrame.
        Must be implemented by concrete loaders.

        Returns:
            DataFrame with loaded data
        """
        pass

    def get_metadata(self) -> Dict[str, Any]:
        """
        Get metadata about the loaded data.
        Can be overridden by subclasses.

        Returns:
            Dictionary with metadata
        """
        return {
            "source": self.__class__.__name__,
            "path": str(self.data_path)
        }