chat_agent/services/chat_agent.py

"""

Core Chat Agent Service



This module provides the main ChatAgent class that orchestrates message processing,

language context management, chat history, and LLM interactions for the multi-language

chat agent system.

"""

import logging
from typing import Dict, Any, Optional, Generator, List
from datetime import datetime

from .groq_client import GroqClient, ChatMessage, LanguageContext
from .language_context import LanguageContextManager
from .session_manager import SessionManager, SessionNotFoundError, SessionExpiredError
from .chat_history import ChatHistoryManager, ChatHistoryError
from .programming_assistance import ProgrammingAssistanceService, AssistanceType
from ..models.message import Message
from ..models.chat_session import ChatSession
from ..utils.error_handler import ChatAgentError, ErrorCategory, ErrorSeverity, get_error_handler, error_handler_decorator
from ..utils.logging_config import get_logger, get_performance_logger

logger = get_logger('chat_agent')
performance_logger = get_performance_logger('chat_agent')


# Remove legacy ChatAgentError class - using the one from error_handler


class ChatAgent:
    """

    Core chat agent service that orchestrates message processing workflow.

    

    Handles language context, history retrieval, LLM calls, and response streaming

    for the multi-language programming assistant chat system.

    """
    
    def __init__(self, groq_client: GroqClient, language_context_manager: LanguageContextManager,

                 session_manager: SessionManager, chat_history_manager: ChatHistoryManager,

                 programming_assistance_service: ProgrammingAssistanceService = None):
        """

        Initialize the chat agent with required service dependencies.

        

        Args:

            groq_client: Groq LangChain client for LLM interactions

            language_context_manager: Manager for programming language contexts

            session_manager: Manager for chat sessions

            chat_history_manager: Manager for chat history storage and retrieval

            programming_assistance_service: Service for specialized programming assistance

        """
        self.groq_client = groq_client
        self.language_context_manager = language_context_manager
        self.session_manager = session_manager
        self.chat_history_manager = chat_history_manager
        self.programming_assistance_service = programming_assistance_service or ProgrammingAssistanceService()
        
        # Initialize error handler
        self.error_handler = get_error_handler()
        
        logger.info("ChatAgent initialized successfully", extra={
            'components': ['groq_client', 'language_context_manager', 'session_manager', 'chat_history_manager', 'programming_assistance_service'],
            'error_handling': 'enabled'
        })
    
    @error_handler_decorator(get_error_handler(), return_fallback=False)
    def process_message(self, session_id: str, message: str, 

                       language: Optional[str] = None) -> Dict[str, Any]:
        """

        Process a user message through the complete chat workflow.

        

        This method handles:

        1. Session validation and activity updates

        2. Language context management

        3. Chat history retrieval

        4. LLM response generation

        5. Message and response storage

        

        Args:

            session_id: Unique session identifier

            message: User's input message

            language: Optional language override for this message

            

        Returns:

            Dict containing response and metadata

            

        Raises:

            ChatAgentError: For various processing errors

        """
        start_time = datetime.utcnow()
        
        # 1. Validate session and update activity
        session = self._validate_and_update_session(session_id)
        
        # 2. Handle language context
        current_language = self._handle_language_context(session_id, language, session)
        
        # 3. Store user message
        user_message = self._store_user_message(session_id, message, current_language)
        
        # 4. Retrieve chat history for context
        chat_history = self._get_chat_context(session_id)
        
        # 5. Generate LLM response
        response_content, response_metadata = self._generate_response(
            message, chat_history, current_language
        )
        
        # 6. Store assistant response
        assistant_message = self._store_assistant_message(
            session_id, response_content, current_language, response_metadata
        )
        
        # 7. Update session message count
        self.session_manager.increment_message_count(session_id)
        
        # Log performance
        processing_time = (datetime.utcnow() - start_time).total_seconds()
        performance_logger.log_operation(
            operation="process_message",
            duration=processing_time,
            context={
                'session_id': session_id,
                'language': current_language,
                'message_length': len(message),
                'history_size': len(chat_history)
            }
        )
        
        return {
            'response': response_content,
            'language': current_language,
            'session_id': session_id,
            'message_id': assistant_message.id,
            'metadata': response_metadata,
            'processing_time': processing_time,
            'timestamp': datetime.utcnow().isoformat()
        }
    
    def switch_language(self, session_id: str, language: str) -> Dict[str, Any]:
        """

        Switch programming language context for a session while maintaining chat continuity.

        

        Args:

            session_id: Unique session identifier

            language: New programming language to switch to

            

        Returns:

            Dict containing switch confirmation and context info

            

        Raises:

            ChatAgentError: If language switch fails

        """
        try:
            # 1. Validate session
            session = self._validate_and_update_session(session_id)
            
            # 2. Validate and set new language
            if not self.language_context_manager.validate_language(language):
                raise ChatAgentError(f"Unsupported language: {language}")
            
            # Get previous language for context
            previous_language = self.language_context_manager.get_language(session_id)
            
            # 3. Update language context
            success = self.language_context_manager.set_language(session_id, language)
            if not success:
                raise ChatAgentError(f"Failed to set language to {language}")
            
            # 4. Update session language
            self.session_manager.set_session_language(session_id, language)
            
            # 5. Store language switch message for continuity
            switch_message = f"Language context switched from {previous_language} to {language}. " \
                           f"I'm now ready to help you with {language} programming!"
            
            self._store_assistant_message(
                session_id, switch_message, language, 
                {'type': 'language_switch', 'previous_language': previous_language}
            )
            
            logger.info(f"Language switched from {previous_language} to {language} for session {session_id}")
            
            return {
                'success': True,
                'previous_language': previous_language,
                'new_language': language,
                'session_id': session_id,
                'message': switch_message,
                'timestamp': datetime.utcnow().isoformat()
            }
            
        except (SessionNotFoundError, SessionExpiredError) as e:
            logger.error(f"Session error switching language: {e}")
            raise ChatAgentError(f"Session error: {e}")
        except Exception as e:
            logger.error(f"Unexpected error switching language: {e}")
            raise ChatAgentError(f"Language switch failed: {e}")  
  
    def stream_response(self, session_id: str, message: str, 

                       language: Optional[str] = None) -> Generator[Dict[str, Any], None, None]:
        """

        Generate streaming response for real-time chat experience.

        

        Args:

            session_id: Unique session identifier

            message: User's input message

            language: Optional language override for this message

            

        Yields:

            Dict containing response chunks and metadata

            

        Raises:

            ChatAgentError: For various processing errors

        """
        try:
            # 1. Validate session and update activity
            session = self._validate_and_update_session(session_id)
            
            # 2. Handle language context
            current_language = self._handle_language_context(session_id, language, session)
            
            # 3. Store user message
            user_message = self._store_user_message(session_id, message, current_language)
            
            # 4. Retrieve chat history for context
            chat_history = self._get_chat_context(session_id)
            
            # 5. Create language context for streaming
            language_context = LanguageContext(
                language=current_language,
                prompt_template=self.language_context_manager.get_language_prompt_template(current_language),
                syntax_highlighting=current_language
            )
            
            # 6. Stream response from Groq
            response_chunks = []
            start_time = datetime.utcnow()
            
            yield {
                'type': 'start',
                'session_id': session_id,
                'language': current_language,
                'timestamp': start_time.isoformat()
            }
            
            for chunk in self.groq_client.stream_response(message, chat_history, language_context):
                response_chunks.append(chunk)
                yield {
                    'type': 'chunk',
                    'content': chunk,
                    'session_id': session_id,
                    'timestamp': datetime.utcnow().isoformat()
                }
            
            # 7. Store complete response
            complete_response = ''.join(response_chunks)
            end_time = datetime.utcnow()
            
            response_metadata = {
                'streaming': True,
                'chunks_count': len(response_chunks),
                'processing_time': (end_time - start_time).total_seconds()
            }
            
            assistant_message = self._store_assistant_message(
                session_id, complete_response, current_language, response_metadata
            )
            
            # 8. Update session message count
            self.session_manager.increment_message_count(session_id)
            
            yield {
                'type': 'complete',
                'session_id': session_id,
                'message_id': assistant_message.id,
                'total_chunks': len(response_chunks),
                'processing_time': response_metadata['processing_time'],
                'timestamp': end_time.isoformat()
            }
            
        except (SessionNotFoundError, SessionExpiredError) as e:
            logger.error(f"Session error in streaming: {e}")
            yield {
                'type': 'error',
                'error': f"Session error: {e}",
                'session_id': session_id,
                'timestamp': datetime.utcnow().isoformat()
            }
        except Exception as e:
            logger.error(f"Unexpected error in streaming: {e}")
            yield {
                'type': 'error',
                'error': f"Processing failed: {e}",
                'session_id': session_id,
                'timestamp': datetime.utcnow().isoformat()
            }  
  
    def get_chat_history(self, session_id: str, limit: int = 10) -> List[Dict[str, Any]]:
        """

        Retrieve recent conversation history for a session.

        

        Args:

            session_id: Unique session identifier

            limit: Maximum number of messages to retrieve

            

        Returns:

            List of message dictionaries with formatted data

            

        Raises:

            ChatAgentError: If history retrieval fails

        """
        try:
            # Validate session
            self._validate_and_update_session(session_id)
            
            # Get recent messages
            messages = self.chat_history_manager.get_recent_history(session_id, limit)
            
            # Format messages for response
            formatted_messages = []
            for message in messages:
                formatted_messages.append({
                    'id': message.id,
                    'role': message.role,
                    'content': message.content,
                    'language': message.language,
                    'timestamp': message.timestamp.isoformat(),
                    'metadata': message.message_metadata
                })
            
            return formatted_messages
            
        except (SessionNotFoundError, SessionExpiredError) as e:
            logger.error(f"Session error getting history: {e}")
            raise ChatAgentError(f"Session error: {e}")
        except ChatHistoryError as e:
            logger.error(f"Chat history error: {e}")
            raise ChatAgentError(f"History error: {e}")
        except Exception as e:
            logger.error(f"Unexpected error getting history: {e}")
            raise ChatAgentError(f"Failed to get history: {e}")
    
    def get_session_info(self, session_id: str) -> Dict[str, Any]:
        """

        Get comprehensive session information including context and statistics.

        

        Args:

            session_id: Unique session identifier

            

        Returns:

            Dict containing session info, language context, and statistics

            

        Raises:

            ChatAgentError: If session info retrieval fails

        """
        try:
            # Get session
            session = self._validate_and_update_session(session_id)
            
            # Get language context
            language_context = self.language_context_manager.get_session_context_info(session_id)
            
            # Get message count
            message_count = self.chat_history_manager.get_message_count(session_id)
            
            # Get cache stats
            cache_stats = self.chat_history_manager.get_cache_stats(session_id)
            
            return {
                'session': {
                    'id': session.id,
                    'user_id': session.user_id,
                    'language': session.language,
                    'created_at': session.created_at.isoformat(),
                    'last_active': session.last_active.isoformat(),
                    'message_count': session.message_count,
                    'is_active': session.is_active,
                    'metadata': session.session_metadata
                },
                'language_context': language_context,
                'statistics': {
                    'total_messages': message_count,
                    'session_message_count': session.message_count,
                    'cache_stats': cache_stats
                },
                'supported_languages': list(self.language_context_manager.get_supported_languages())
            }
            
        except (SessionNotFoundError, SessionExpiredError) as e:
            logger.error(f"Session error getting info: {e}")
            raise ChatAgentError(f"Session error: {e}")
        except Exception as e:
            logger.error(f"Unexpected error getting session info: {e}")
            raise ChatAgentError(f"Failed to get session info: {e}")
    
    def process_programming_assistance(self, session_id: str, message: str, 

                                     code: str = None, error_message: str = None,

                                     assistance_type: AssistanceType = None) -> Dict[str, Any]:
        """

        Process a programming assistance request with specialized handling.

        

        Args:

            session_id: Unique session identifier

            message: User's message/question

            code: Optional code to analyze

            error_message: Optional error message to debug

            assistance_type: Optional specific type of assistance needed

            

        Returns:

            Dict containing specialized assistance response

            

        Raises:

            ChatAgentError: For various processing errors

        """
        try:
            # 1. Validate session and update activity
            session = self._validate_and_update_session(session_id)
            current_language = self.language_context_manager.get_language(session_id)
            
            # 2. Detect assistance type if not provided
            if not assistance_type:
                assistance_type = self.programming_assistance_service.detect_assistance_type(message, code)
            
            # 3. Get specialized prompt template
            context = {
                'beginner_mode': 'beginner' in message.lower() or 'new to' in message.lower(),
                'code_provided': bool(code),
                'error_provided': bool(error_message)
            }
            
            specialized_prompt = self.programming_assistance_service.get_assistance_prompt_template(
                assistance_type, current_language, context
            )
            
            # 4. Perform analysis based on assistance type
            analysis_result = None
            if assistance_type in [AssistanceType.CODE_EXPLANATION, AssistanceType.CODE_REVIEW] and code:
                analysis_result = self.programming_assistance_service.analyze_code(code, current_language)
            elif assistance_type in [AssistanceType.ERROR_ANALYSIS, AssistanceType.DEBUGGING] and error_message:
                analysis_result = self.programming_assistance_service.analyze_error(
                    error_message, code, current_language
                )
            elif assistance_type == AssistanceType.BEGINNER_HELP:
                # Extract topic from message for beginner explanations
                topic = self._extract_topic_from_message(message)
                analysis_result = self.programming_assistance_service.generate_beginner_explanation(
                    topic, current_language, code
                )
            
            # 5. Build enhanced message with analysis
            enhanced_message = self._build_enhanced_message(
                message, code, error_message, analysis_result, assistance_type
            )
            
            # 6. Store user message with assistance metadata
            user_message = self._store_user_message(
                session_id, message, current_language, {
                    'assistance_type': assistance_type.value,
                    'code_provided': bool(code),
                    'error_provided': bool(error_message)
                }
            )
            
            # 7. Get chat history for context
            chat_history = self._get_chat_context(session_id)
            
            # 8. Create specialized language context
            language_context = LanguageContext(
                language=current_language,
                prompt_template=specialized_prompt,
                syntax_highlighting=current_language
            )
            
            # 9. Generate response with specialized context
            response_content, response_metadata = self._generate_response(
                enhanced_message, chat_history, current_language, language_context
            )
            
            # 10. Format response if analysis was performed
            if analysis_result and assistance_type != AssistanceType.BEGINNER_HELP:
                formatted_response = self.programming_assistance_service.format_assistance_response(
                    assistance_type, analysis_result, current_language
                )
                response_content = f"{formatted_response}\n\n---\n\n{response_content}"
            
            # 11. Store assistant response
            assistant_message = self._store_assistant_message(
                session_id, response_content, current_language, {
                    **response_metadata,
                    'assistance_type': assistance_type.value,
                    'analysis_performed': bool(analysis_result)
                }
            )
            
            # 12. Update session message count
            self.session_manager.increment_message_count(session_id)
            
            return {
                'response': response_content,
                'assistance_type': assistance_type.value,
                'language': current_language,
                'session_id': session_id,
                'message_id': assistant_message.id,
                'analysis_result': analysis_result,
                'metadata': response_metadata,
                'timestamp': datetime.utcnow().isoformat()
            }
            
        except (SessionNotFoundError, SessionExpiredError) as e:
            logger.error(f"Session error in programming assistance: {e}")
            raise ChatAgentError(f"Session error: {e}")
        except Exception as e:
            logger.error(f"Unexpected error in programming assistance: {e}")
            raise ChatAgentError(f"Programming assistance failed: {e}")
    
    def explain_code(self, session_id: str, code: str, question: str = None) -> Dict[str, Any]:
        """

        Provide detailed code explanation.

        

        Args:

            session_id: Unique session identifier

            code: Code to explain

            question: Optional specific question about the code

            

        Returns:

            Dict containing code explanation response

        """
        message = question or "Please explain this code:"
        return self.process_programming_assistance(
            session_id, message, code=code, assistance_type=AssistanceType.CODE_EXPLANATION
        )
    
    def debug_code(self, session_id: str, code: str, error_message: str, 

                   description: str = None) -> Dict[str, Any]:
        """

        Provide debugging assistance for code with errors.

        

        Args:

            session_id: Unique session identifier

            code: Code that has errors

            error_message: Error message received

            description: Optional description of the problem

            

        Returns:

            Dict containing debugging assistance response

        """
        message = description or "I'm getting an error with this code. Can you help me debug it?"
        return self.process_programming_assistance(
            session_id, message, code=code, error_message=error_message, 
            assistance_type=AssistanceType.DEBUGGING
        )
    
    def analyze_error(self, session_id: str, error_message: str, 

                     context: str = None) -> Dict[str, Any]:
        """

        Analyze and explain an error message.

        

        Args:

            session_id: Unique session identifier

            error_message: Error message to analyze

            context: Optional context about when the error occurred

            

        Returns:

            Dict containing error analysis response

        """
        message = context or "I got this error and don't understand what it means:"
        return self.process_programming_assistance(
            session_id, message, error_message=error_message, 
            assistance_type=AssistanceType.ERROR_ANALYSIS
        )
    
    def review_code(self, session_id: str, code: str, focus_areas: List[str] = None) -> Dict[str, Any]:
        """

        Provide code review and improvement suggestions.

        

        Args:

            session_id: Unique session identifier

            code: Code to review

            focus_areas: Optional list of specific areas to focus on

            

        Returns:

            Dict containing code review response

        """
        focus_text = f" Please focus on: {', '.join(focus_areas)}" if focus_areas else ""
        message = f"Please review this code and provide feedback.{focus_text}"
        return self.process_programming_assistance(
            session_id, message, code=code, assistance_type=AssistanceType.CODE_REVIEW
        )
    
    def get_beginner_help(self, session_id: str, topic: str, 

                         specific_question: str = None) -> Dict[str, Any]:
        """

        Provide beginner-friendly help on programming topics.

        

        Args:

            session_id: Unique session identifier

            topic: Programming topic or concept

            specific_question: Optional specific question about the topic

            

        Returns:

            Dict containing beginner-friendly explanation

        """
        message = specific_question or f"I'm new to programming. Can you explain {topic} in simple terms?"
        return self.process_programming_assistance(
            session_id, message, assistance_type=AssistanceType.BEGINNER_HELP
        ) 
   
    # Private helper methods
    
    def _validate_and_update_session(self, session_id: str) -> ChatSession:
        """Validate session exists and update activity."""
        session = self.session_manager.get_session(session_id)
        self.session_manager.update_session_activity(session_id)
        return session
    
    def _handle_language_context(self, session_id: str, language: Optional[str], 

                                session: ChatSession) -> str:
        """Handle language context for the session."""
        if language:
            # Validate and set new language if provided
            if not self.language_context_manager.validate_language(language):
                logger.warning(f"Invalid language {language}, using session default")
                return self.language_context_manager.get_language(session_id)
            
            # Set language context
            self.language_context_manager.set_language(session_id, language)
            
            # Update session language if different
            if session.language != language:
                self.session_manager.set_session_language(session_id, language)
            
            return language
        else:
            # Use existing session language
            return self.language_context_manager.get_language(session_id)
    

    
    def _store_assistant_message(self, session_id: str, content: str, language: str, 

                                metadata: Optional[Dict[str, Any]] = None) -> Message:
        """Store assistant message in chat history."""
        return self.chat_history_manager.store_message(
            session_id=session_id,
            role='assistant',
            content=content,
            language=language,
            message_metadata=metadata
        )
    
    def _get_chat_context(self, session_id: str) -> List[ChatMessage]:
        """Get recent chat history formatted for LLM context."""
        messages = self.chat_history_manager.get_recent_history(session_id)
        
        chat_messages = []
        for message in messages:
            chat_messages.append(ChatMessage(
                role=message.role,
                content=message.content,
                language=message.language,
                timestamp=message.timestamp.isoformat()
            ))
        
        return chat_messages
    
    def _generate_response(self, message: str, chat_history: List[ChatMessage], 

                          language: str, language_context: LanguageContext = None) -> tuple[str, Dict[str, Any]]:
        """Generate response using Groq LLM with context."""
        start_time = datetime.utcnow()
        
        # Create language context if not provided
        if not language_context:
            language_context = LanguageContext(
                language=language,
                prompt_template=self.language_context_manager.get_language_prompt_template(language),
                syntax_highlighting=language
            )
        
        # Generate response
        response = self.groq_client.generate_response(
            prompt=message,
            chat_history=chat_history,
            language_context=language_context
        )
        
        end_time = datetime.utcnow()
        
        # Create response metadata
        metadata = {
            'processing_time': (end_time - start_time).total_seconds(),
            'language': language,
            'context_messages': len(chat_history),
            'model_info': self.groq_client.get_model_info()
        }
        
        return response, metadata
    
    def _store_user_message(self, session_id: str, content: str, language: str, 

                           metadata: Optional[Dict[str, Any]] = None) -> Message:
        """Store user message in chat history with optional metadata."""
        return self.chat_history_manager.store_message(
            session_id=session_id,
            role='user',
            content=content,
            language=language,
            message_metadata=metadata
        )
    
    def _extract_topic_from_message(self, message: str) -> str:
        """Extract programming topic from user message."""
        # Simple keyword extraction - could be enhanced with NLP
        common_topics = [
            'variables', 'functions', 'loops', 'conditionals', 'classes', 'objects',
            'arrays', 'lists', 'dictionaries', 'strings', 'integers', 'floats',
            'inheritance', 'polymorphism', 'encapsulation', 'recursion', 'algorithms',
            'data structures', 'debugging', 'testing', 'modules', 'packages'
        ]
        
        message_lower = message.lower()
        for topic in common_topics:
            if topic in message_lower:
                return topic
        
        # If no specific topic found, extract potential topic from question words
        words = message_lower.split()
        for i, word in enumerate(words):
            if word in ['what', 'how', 'explain', 'understand'] and i + 1 < len(words):
                # Return the next few words as potential topic
                return ' '.join(words[i+1:i+3])
        
        return 'programming concepts'
    
    def _build_enhanced_message(self, message: str, code: str = None, 

                               error_message: str = None, analysis_result: Any = None,

                               assistance_type: AssistanceType = None) -> str:
        """Build enhanced message with code and analysis context."""
        enhanced_parts = [message]
        
        if code:
            enhanced_parts.append(f"\n\nCode to analyze:\n```\n{code}\n```")
        
        if error_message:
            enhanced_parts.append(f"\n\nError message:\n```\n{error_message}\n```")
        
        if analysis_result and assistance_type == AssistanceType.BEGINNER_HELP:
            # For beginner help, the analysis_result is already the formatted explanation
            return analysis_result
        
        return '\n'.join(enhanced_parts)


def create_chat_agent(groq_client: GroqClient, language_context_manager: LanguageContextManager,

                     session_manager: SessionManager, chat_history_manager: ChatHistoryManager,

                     programming_assistance_service: ProgrammingAssistanceService = None) -> ChatAgent:
    """

    Factory function to create a ChatAgent instance.

    

    Args:

        groq_client: Groq LangChain client for LLM interactions

        language_context_manager: Manager for programming language contexts

        session_manager: Manager for chat sessions

        chat_history_manager: Manager for chat history storage and retrieval

        programming_assistance_service: Service for specialized programming assistance

        

    Returns:

        ChatAgent: Configured chat agent instance

    """
    return ChatAgent(groq_client, language_context_manager, session_manager, chat_history_manager, programming_assistance_service)