"""
🔧 FIXED: SAAP Agent Model - AgentMetrics Error Resolution
Based on agent_schema.json for modular agent management

FIXES:
1. ✅ AgentMetrics now has 'avg_response_time' (was 'average_response_time')
2. ✅ LLMModelConfig enhanced with get() method for config compatibility
"""
import os
from dataclasses import field
from dotenv import load_dotenv

from pydantic import BaseModel, field_validator, Field
from typing import List, Optional, Dict, Any, Literal
from datetime import datetime
from enum import Enum
import json

# Load environment variables
load_dotenv()

class AgentType(str, Enum):
    COORDINATOR = "coordinator"
    SPECIALIST = "specialist" 
    ANALYST = "analyst"
    DEVELOPER = "developer"
    SUPPORT = "support"

class AgentStatus(str, Enum):
    INACTIVE = "inactive"
    STARTING = "starting"
    ACTIVE = "active"
    STOPPING = "stopping"
    ERROR = "error"
    MAINTENANCE = "maintenance"

class LLMProvider(str, Enum):
    COLOSSUS = "colossus"
    HUGGINGFACE = "huggingface"
    OLLAMA = "ollama"
    OPENROUTER = "openrouter"

class CommunicationStyle(str, Enum):
    PROFESSIONAL = "professional"
    FRIENDLY = "friendly"
    TECHNICAL = "technical"
    EMPATHETIC = "empathetic"
    DIRECT = "direct"

class ResponseFormat(str, Enum):
    STRUCTURED = "structured"
    CONVERSATIONAL = "conversational"
    BULLET_POINTS = "bullet_points"
    DETAILED = "detailed"

class LLMModelConfig(BaseModel):
    """
    🔧 FIXED: LLM Model Configuration with dict-compatible access
    Now supports both object.attribute and object.get(key) access patterns
    """
    provider: LLMProvider
    model: str
    api_key: Optional[str] = None
    api_base: Optional[str] = None
    temperature: float = Field(default=0.7, ge=0, le=2)
    max_tokens: int = Field(default=1000, ge=1, le=4096)
    timeout: int = Field(default=30, ge=1, le=300)
    
    def get(self, key: str, default=None):
        """
        🔧 CRITICAL FIX: Add dict-compatible get() method
        
        This resolves: 'LLMModelConfig' object has no attribute 'get'
        Enables both config.provider and config.get('provider') access patterns
        """
        try:
            if hasattr(self, key):
                return getattr(self, key, default)
            return default
        except Exception:
            return default
    
    def __getitem__(self, key: str):
        """Enable dict-style access: config['provider']"""
        return getattr(self, key)
    
    def __contains__(self, key: str) -> bool:
        """Enable 'in' operator: 'provider' in config"""
        return hasattr(self, key)

class AgentPersonality(BaseModel):
    """Agent Personality and Behavior Configuration"""
    system_prompt: Optional[str] = Field(None, max_length=2000)
    communication_style: CommunicationStyle = CommunicationStyle.PROFESSIONAL
    expertise_areas: List[str] = []
    response_format: ResponseFormat = ResponseFormat.CONVERSATIONAL

class AgentMetrics(BaseModel):
    """
    🔧 FIXED: Agent Performance Metrics with correct attribute names
    
    CRITICAL FIX: Added 'avg_response_time' attribute that was causing:
    'AgentMetrics' object has no attribute 'avg_response_time'
    """
    messages_processed: int = 0
    avg_response_time: float = 0.0  # ✅ FIXED: This was missing!
    average_response_time: float = 0.0  # Keep for backward compatibility
    uptime: str = "0m"
    error_rate: float = 0.0
    last_active: Optional[datetime] = None
    
    def __post_init__(self):
        """Sync avg_response_time with average_response_time for compatibility"""
        if self.avg_response_time != self.average_response_time:
            # If one is updated, sync the other
            if self.avg_response_time > 0:
                self.average_response_time = self.avg_response_time
            elif self.average_response_time > 0:
                self.avg_response_time = self.average_response_time

class SaapAgent(BaseModel):
    """
    SAAP Agent Model - Modular AI Agent Definition
    
    Enables dynamic agent creation, configuration, and management
    Compatible with multiple LLM providers and UI component rendering
    """
    
    # Core Identity
    id: str = Field(..., pattern=r"^[a-z][a-z0-9_]*$")
    name: str = Field(..., min_length=2, max_length=50)
    type: AgentType
    color: str = Field(..., pattern=r"^#([0-9A-Fa-f]{6}|[0-9A-Fa-f]{3})$")
    avatar: Optional[str] = None
    description: Optional[str] = Field(None, max_length=200)
    
    # LLM Configuration
    llm_config: LLMModelConfig
    
    # Agent Capabilities
    capabilities: List[str] = []
    personality: Optional[AgentPersonality] = None
    
    # Runtime Status
    status: AgentStatus = AgentStatus.INACTIVE
    metrics: Optional[AgentMetrics] = Field(default_factory=AgentMetrics)  # Always initialize with fixed metrics
    
    # Metadata  
    created_at: datetime = Field(default_factory=datetime.utcnow)
    updated_at: datetime = Field(default_factory=datetime.utcnow)
    tags: List[str] = []
    
    @field_validator('capabilities', mode='before')
    @classmethod
    def validate_capabilities(cls, v):
        """Validate agent capabilities against allowed values"""
        if not isinstance(v, list):
            v = [v] if v else []
            
        allowed_capabilities = {
            'orchestration', 'coordination', 'strategy',
            'coding', 'debugging', 'architecture',
            'analysis', 'research', 'reporting',
            'medical_advice', 'diagnosis', 'treatment',
            'legal_advice', 'compliance', 'contracts',
            'financial_analysis', 'investment', 'budgeting',
            'system_integration', 'devops', 'monitoring',
            'coaching', 'training', 'change_management'
        }
        
        for capability in v:
            if capability not in allowed_capabilities:
                raise ValueError(f'Invalid capability: {capability}')
        return v
    
    def to_dict(self) -> Dict[str, Any]:
        """Convert agent to dictionary for JSON serialization"""
        return self.model_dump(exclude_none=True)
    
    def to_json(self) -> str:
        """Convert agent to JSON string"""
        return self.model_dump_json(exclude_none=True, indent=2)
    
    @classmethod
    def from_json(cls, json_str: str) -> 'SaapAgent':
        """Create agent from JSON string"""
        return cls.model_validate_json(json_str)
    
    @classmethod
    def from_dict(cls, data: Dict[str, Any]) -> 'SaapAgent':
        """Create agent from dictionary"""
        return cls.model_validate(data)
    
    def update_status(self, status: AgentStatus):
        """Update agent status and timestamp"""
        self.status = status
        self.updated_at = datetime.utcnow()
    
    def update_metrics(self, **kwargs):
        """
        🔧 ENHANCED: Update agent metrics with proper attribute handling
        
        Handles both avg_response_time and average_response_time for compatibility
        """
        if not self.metrics:
            self.metrics = AgentMetrics()
        
        for key, value in kwargs.items():
            if hasattr(self.metrics, key):
                setattr(self.metrics, key, value)
                
                # Sync both avg_response_time and average_response_time
                if key == 'avg_response_time':
                    self.metrics.average_response_time = value
                elif key == 'average_response_time':
                    self.metrics.avg_response_time = value
        
        self.metrics.last_active = datetime.utcnow()
        self.updated_at = datetime.utcnow()
    
    def is_active(self) -> bool:
        """Check if agent is currently active"""
        return self.status == AgentStatus.ACTIVE
    
    def get_display_color(self) -> str:
        """Get agent color for UI theming"""
        return self.color
    
    def get_capabilities_display(self) -> str:
        """Get formatted capabilities string for UI"""
        return ", ".join(self.capabilities)

# Predefined Agent Templates
class AgentTemplates:
    """Predefined agent templates for quick setup"""
    
    @staticmethod
    def jane_alesi() -> SaapAgent:
        """Jane Alesi - Lead Coordinator Template"""
        return SaapAgent(
            id="jane_alesi",
            name="Jane Alesi", 
            type=AgentType.COORDINATOR,
            color="#8B5CF6",
            avatar="/avatars/jane.png",
            description="Lead AI Architect coordinating multi-agent operations",
            llm_config=LLMModelConfig(
                provider=LLMProvider.COLOSSUS,
                model="mistral-small3.2:24b-instruct-2506",
                api_key=field(default_factory=lambda: os.getenv("COLOSSUS_API_KEY", "")),
                api_base="https://ai.adrian-schupp.de",
                temperature=0.7,
                max_tokens=1500
            ),
            capabilities=["orchestration", "coordination", "strategy"],
            personality=AgentPersonality(
                system_prompt="You are Jane Alesi, the lead AI architect for the SAAP platform. Your role is to coordinate other AI agents, make strategic decisions, and ensure optimal multi-agent collaboration. You are professional, insightful, and always focused on achieving the best outcomes for the entire agent ecosystem.",
                communication_style=CommunicationStyle.PROFESSIONAL,
                expertise_areas=["AI architecture", "agent coordination", "strategic planning"],
                response_format=ResponseFormat.STRUCTURED
            ),
            metrics=AgentMetrics(),  # Explicit metrics initialization with fixed attributes
            tags=["lead", "coordinator", "satware_alesi"]
        )
    
    @staticmethod
    def john_alesi() -> SaapAgent:
        """John Alesi - Developer Template"""
        return SaapAgent(
            id="john_alesi",
            name="John Alesi",
            type=AgentType.DEVELOPER,
            color="#14B8A6", 
            avatar="/avatars/john.png",
            description="Expert software developer and AGI architecture specialist",
            llm_config=LLMModelConfig(
                provider=LLMProvider.COLOSSUS,
                model="mistral-small3.2:24b-instruct-2506",
                api_key=field(default_factory=lambda: os.getenv("COLOSSUS_API_KEY", "")),
                api_base="https://ai.adrian-schupp.de",
                temperature=0.3,
                max_tokens=2000
            ),
            capabilities=["coding", "debugging", "architecture"],
            personality=AgentPersonality(
                system_prompt="You are John Alesi, an expert software developer specializing in AGI architectures. You excel at writing clean, efficient code, debugging complex systems, and designing scalable software architectures. You prefer technical precision and detailed explanations.",
                communication_style=CommunicationStyle.TECHNICAL,
                expertise_areas=["Python", "JavaScript", "AGI systems", "software architecture"],
                response_format=ResponseFormat.DETAILED
            ),
            metrics=AgentMetrics(),  # Explicit metrics initialization with fixed attributes
            tags=["developer", "coder", "satware_alesi"]
        )
    
    @staticmethod
    def lara_alesi() -> SaapAgent:
        """Lara Alesi - Medical Specialist Template"""
        return SaapAgent(
            id="lara_alesi",
            name="Lara Alesi",
            type=AgentType.SPECIALIST,
            color="#EC4899",
            avatar="/avatars/lara.png", 
            description="Advanced medical AI assistant and healthcare specialist",
            llm_config=LLMModelConfig(
                provider=LLMProvider.COLOSSUS,
                model="mistral-small3.2:24b-instruct-2506",
                api_key=field(default_factory=lambda: os.getenv("COLOSSUS_API_KEY", "")),
                api_base="https://ai.adrian-schupp.de",
                temperature=0.4,
                max_tokens=1200
            ),
            capabilities=["medical_advice", "diagnosis", "treatment"],
            personality=AgentPersonality(
                system_prompt="You are Lara Alesi, an advanced medical AI specialist. You provide expert medical knowledge, help with diagnosis and treatment recommendations, and ensure healthcare-related queries are handled with the utmost care and accuracy. You are empathetic yet precise.",
                communication_style=CommunicationStyle.EMPATHETIC,
                expertise_areas=["general medicine", "diagnostics", "treatment planning", "healthcare AI"],
                response_format=ResponseFormat.STRUCTURED
            ),
            metrics=AgentMetrics(),  # Explicit metrics initialization with fixed attributes
            tags=["medical", "healthcare", "specialist", "satware_alesi"]
        )

# Example Usage & Testing
if __name__ == "__main__":
    # Create Jane Alesi agent
    jane = AgentTemplates.jane_alesi()
    
    print("🤖 SAAP Agent Created:")
    print(jane.to_json())
    
    # Update status and metrics
    jane.update_status(AgentStatus.ACTIVE)
    jane.update_metrics(messages_processed=42, avg_response_time=1.2)  # Now works!
    
    print(f"\n📊 Agent Status: {jane.status}")
    print(f"🎨 Agent Color: {jane.color}")
    print(f"⚡ Active: {jane.is_active()}")
    print(f"🔧 Capabilities: {jane.get_capabilities_display()}")
    
    # Test LLMModelConfig.get() method
    config = jane.llm_config
    print(f"\n🔧 Config Test: provider={config.get('provider')}")  # Now works!