src/gaia/config/loader.py

"""
Configuration loader for GAIA agent.

This module provides a hierarchical configuration system for the GAIA agent,
supporting configuration from multiple sources with proper precedence:
1. Default values
2. Configuration files
3. Environment variables
4. Runtime overrides

It includes a Configuration class for typed access to configuration values
and validation of configuration.
"""

import os
import json
import yaml
import logging
from typing import Dict, Any, Optional, List, Union, TypeVar, Generic, Type, cast
from pathlib import Path
from copy import deepcopy

from src.gaia.config.default import DEFAULT_CONFIG
from src.gaia.config.env import load_env_config

# Type variable for configuration value types
T = TypeVar('T')

logger = logging.getLogger("gaia.config")

class ConfigurationError(Exception):
    """Exception raised for configuration errors."""
    pass

class Configuration:
    """
    Configuration class providing structured access to GAIA configuration.
    
    This class wraps a configuration dictionary and provides typed access to
    configuration values with dot notation, default values, and validation.
    """
    
    def __init__(self, config_dict: Dict[str, Any]):
        """
        Initialize the Configuration with the provided dictionary.
        
        Args:
            config_dict: The configuration dictionary
        """
        self._config = deepcopy(config_dict)
    
    def get(self, path: str, default: Optional[T] = None) -> Union[T, Any]:
        """
        Get a configuration value by path with optional default.
        
        Args:
            path: Dot-separated path to the configuration value
            default: Default value if the path doesn't exist
            
        Returns:
            The configuration value or default
        """
        parts = path.split('.')
        current = self._config
        
        try:
            for part in parts:
                current = current[part]
            return current
        except (KeyError, TypeError):
            return default
    
    def get_typed(self, path: str, value_type: Type[T], default: Optional[T] = None) -> T:
        """
        Get a configuration value with type checking.
        
        Args:
            path: Dot-separated path to the configuration value
            value_type: Expected type of the value
            default: Default value if the path doesn't exist
            
        Returns:
            The configuration value with the specified type
            
        Raises:
            ConfigurationError: If the value doesn't match the expected type
        """
        value = self.get(path, default)
        
        if value is None:
            if default is None:
                raise ConfigurationError(f"Required configuration value '{path}' is missing")
            return default
        
        if not isinstance(value, value_type):
            # Attempt type conversion for simple types
            try:
                if value_type is bool and isinstance(value, str):
                    if value.lower() in ('true', 'yes', '1', 'on'):
                        return cast(T, True)
                    if value.lower() in ('false', 'no', '0', 'off'):
                        return cast(T, False)
                
                if value_type in (int, float, str):
                    return cast(T, value_type(value))
            except (ValueError, TypeError):
                pass
                
            raise ConfigurationError(
                f"Configuration value '{path}' has incorrect type: "
                f"expected {value_type.__name__}, got {type(value).__name__}"
            )
        
        return cast(T, value)
    
    def set(self, path: str, value: Any) -> None:
        """
        Set a configuration value by path.
        
        Args:
            path: Dot-separated path to the configuration value
            value: The value to set
        """
        parts = path.split('.')
        current = self._config
        
        # Navigate to the parent of the final key
        for part in parts[:-1]:
            if part not in current:
                current[part] = {}
            current = current[part]
        
        # Set the value at the final key
        current[parts[-1]] = value
    
    def update(self, config_dict: Dict[str, Any]) -> None:
        """
        Update configuration with values from another dictionary.
        
        This performs a deep update, preserving nested structure.
        
        Args:
            config_dict: Dictionary with configuration values to update
        """
        self._deep_update(self._config, config_dict)
    
    def _deep_update(self, target: Dict[str, Any], source: Dict[str, Any]) -> None:
        """
        Recursively update a dictionary with values from another dictionary.
        
        Args:
            target: Target dictionary to update
            source: Source dictionary with values to copy
        """
        for key, value in source.items():
            if isinstance(value, dict) and key in target and isinstance(target[key], dict):
                # Recursively update nested dictionaries
                self._deep_update(target[key], value)
            else:
                # Direct assignment for non-dict values or new keys
                target[key] = value
    
    def to_dict(self) -> Dict[str, Any]:
        """
        Convert the configuration to a dictionary.
        
        Returns:
            A deep copy of the configuration dictionary
        """
        return deepcopy(self._config)
    
    def validate(self, schema: Dict[str, Any]) -> List[str]:
        """
        Validate the configuration against a schema.
        
        Args:
            schema: Dictionary defining the expected structure and types
            
        Returns:
            List of validation errors, empty if validation succeeds
        """
        errors = []
        self._validate_recursive(self._config, schema, "", errors)
        return errors
    
    def _validate_recursive(
        self, config: Dict[str, Any], schema: Dict[str, Any], 
        path_prefix: str, errors: List[str]
    ) -> None:
        """
        Recursively validate configuration against schema.
        
        Args:
            config: Configuration dictionary to validate
            schema: Schema dictionary to validate against
            path_prefix: Current path prefix for error messages
            errors: List to append validation errors to
        """
        # Check for required fields
        for key, spec in schema.items():
            if isinstance(spec, dict) and "required" in spec and spec["required"]:
                if key not in config:
                    errors.append(f"Missing required field '{path_prefix}.{key}'")
        
        # Validate each field in the config
        for key, value in config.items():
            path = f"{path_prefix}.{key}" if path_prefix else key
            
            if key not in schema:
                # Ignore fields not in schema
                continue
                
            spec = schema[key]
            
            if isinstance(spec, dict):
                if "type" in spec:
                    # Validate type
                    expected_type = spec["type"]
                    if not self._check_type(value, expected_type):
                        errors.append(
                            f"Invalid type for '{path}': expected {expected_type}, "
                            f"got {type(value).__name__}"
                        )
                
                if "values" in spec and isinstance(value, (list, tuple)):
                    # Validate list items
                    for i, item in enumerate(value):
                        if not self._check_type(item, spec["values"]):
                            errors.append(
                                f"Invalid type for item {i} in '{path}': "
                                f"expected {spec['values']}, got {type(item).__name__}"
                            )
                
                if "properties" in spec and isinstance(value, dict):
                    # Recursively validate nested properties
                    self._validate_recursive(value, spec["properties"], path, errors)
    
    def _check_type(self, value: Any, expected_type: str) -> bool:
        """
        Check if a value matches the expected type.
        
        Args:
            value: Value to check
            expected_type: Expected type name as string
            
        Returns:
            True if the value matches the expected type, False otherwise
        """
        if expected_type == "string":
            return isinstance(value, str)
        elif expected_type == "number":
            return isinstance(value, (int, float))
        elif expected_type == "integer":
            return isinstance(value, int)
        elif expected_type == "boolean":
            return isinstance(value, bool)
        elif expected_type == "array":
            return isinstance(value, (list, tuple))
        elif expected_type == "object":
            return isinstance(value, dict)
        elif expected_type == "null":
            return value is None
        elif expected_type == "any":
            return True
        else:
            return False


class ConfigLoader:
    """
    Configuration loader handling multiple configuration sources.
    
    This class loads configuration from multiple sources and combines them
    with proper precedence to create a final configuration.
    """
    
    def __init__(self):
        """Initialize the ConfigLoader."""
        pass
    
    def load_config(self, config_file: Optional[str] = None) -> Configuration:
        """
        Load configuration from all sources.
        
        Args:
            config_file: Optional path to a configuration file
            
        Returns:
            Configuration instance with combined configuration
        """
        # Start with default configuration
        config_dict = deepcopy(DEFAULT_CONFIG)
        
        # Load configuration from file if provided
        if config_file:
            file_config = self._load_config_file(config_file)
            self._deep_update(config_dict, file_config)
        
        # Load configuration from environment variables
        env_config = load_env_config()
        self._deep_update(config_dict, env_config)
        
        # Create and return the Configuration instance
        return Configuration(config_dict)
    
    def _load_config_file(self, config_file: str) -> Dict[str, Any]:
        """
        Load configuration from a file.
        
        Supports JSON and YAML files based on extension.
        
        Args:
            config_file: Path to the configuration file
            
        Returns:
            Dictionary with configuration from the file
            
        Raises:
            ConfigurationError: If the file cannot be loaded
        """
        path = Path(config_file)
        
        if not path.exists():
            raise ConfigurationError(f"Configuration file not found: {config_file}")
        
        try:
            with open(path, 'r') as f:
                if path.suffix.lower() in ('.yaml', '.yml'):
                    return yaml.safe_load(f) or {}
                elif path.suffix.lower() == '.json':
                    return json.load(f)
                else:
                    raise ConfigurationError(
                        f"Unsupported configuration file format: {path.suffix}"
                    )
        except (yaml.YAMLError, json.JSONDecodeError) as e:
            raise ConfigurationError(f"Error parsing configuration file: {str(e)}")
    
    def _deep_update(self, target: Dict[str, Any], source: Dict[str, Any]) -> None:
        """
        Recursively update a dictionary with values from another dictionary.
        
        Args:
            target: Target dictionary to update
            source: Source dictionary with values to copy
        """
        for key, value in source.items():
            if isinstance(value, dict) and key in target and isinstance(target[key], dict):
                # Recursively update nested dictionaries
                self._deep_update(target[key], value)
            else:
                # Direct assignment for non-dict values or new keys
                target[key] = value