app/models/conversation.py

# ============================================================================
# backend/app/models/conversation.py
# ============================================================================



"""
Conversation Models for MongoDB

Handles conversation persistence with:
- Auto-generated titles from first message
- Message metadata (policy actions, retrieval stats)
- Archive/unarchive support
- Search indexing ready
"""

from datetime import datetime
from typing import List, Optional, Dict, Any, Annotated
from pydantic import BaseModel, Field, ConfigDict, BeforeValidator
from bson import ObjectId


# ============================================================================
# CUSTOM TYPES
# ============================================================================

def validate_object_id(v: Any) -> ObjectId:
    """Validator function for ObjectId"""
    if isinstance(v, ObjectId):
        return v
    if isinstance(v, str):
        if ObjectId.is_valid(v):
            return ObjectId(v)
    raise ValueError("Invalid ObjectId")


# Annotated type for PyObjectId compatible with Pydantic v2
PyObjectId = Annotated[ObjectId, BeforeValidator(validate_object_id)]


# ============================================================================
# MESSAGE MODEL
# ============================================================================

class Message(BaseModel):
    """
    Single message in a conversation.
    
    Contains:
    - User/assistant content
    - Metadata from RAG pipeline (policy action, retrieval stats)
    - Timestamp
    """
    
    role: str = Field(..., description="Role: 'user' or 'assistant'")
    content: str = Field(..., description="Message content")
    timestamp: datetime = Field(default_factory=datetime.utcnow)
    
    # Metadata from RAG pipeline (only for assistant messages)
    metadata: Optional[Dict[str, Any]] = Field(
        default=None,
        description="RAG metadata: policy_action, confidence, docs_retrieved, etc."
    )
    
    model_config = ConfigDict(
        json_encoders={
            datetime: lambda v: v.isoformat()
        },
        json_schema_extra={
            "example": {
                "role": "user",
                "content": "What is my account balance?",
                "timestamp": "2024-01-15T10:30:00",
                "metadata": None
            }
        }
    )


# ============================================================================
# CONVERSATION MODEL (MongoDB Document)
# ============================================================================

class Conversation(BaseModel):
    """
    Full conversation document stored in MongoDB.
    
    Features:
    - Auto-generated title from first user message
    - Message history with metadata
    - Archive/active status
    - User association
    - Search-ready structure
    """
    
    id: Optional[PyObjectId] = Field(alias="_id", default=None)
    user_id: str = Field(..., description="User ID who owns this conversation")
    title: str = Field(..., description="Conversation title (auto-generated or custom)")
    
    messages: List[Message] = Field(
        default_factory=list,
        description="List of messages in chronological order"
    )
    
    # Status flags
    is_archived: bool = Field(default=False, description="Is conversation archived?")
    is_deleted: bool = Field(default=False, description="Soft delete flag")
    
    # Timestamps
    created_at: datetime = Field(default_factory=datetime.utcnow)
    updated_at: datetime = Field(default_factory=datetime.utcnow)
    last_message_at: Optional[datetime] = Field(default=None)
    
    # Metadata
    message_count: int = Field(default=0, description="Total messages (excluding deleted)")
    
    model_config = ConfigDict(
        populate_by_name=True,
        arbitrary_types_allowed=True,
        json_encoders={
            ObjectId: str,
            datetime: lambda v: v.isoformat(),
        },
        json_schema_extra={
            "example": {
                "user_id": "user_123",
                "title": "Account Balance Inquiry",
                "messages": [
                    {
                        "role": "user",
                        "content": "What is my account balance?",
                        "timestamp": "2024-01-15T10:30:00"
                    },
                    {
                        "role": "assistant",
                        "content": "Your current account balance is...",
                        "timestamp": "2024-01-15T10:30:05",
                        "metadata": {
                            "policy_action": "FETCH",
                            "confidence": 0.95,
                            "documents_retrieved": 3
                        }
                    }
                ],
                "is_archived": False,
                "created_at": "2024-01-15T10:30:00",
                "updated_at": "2024-01-15T10:30:05",
                "message_count": 2
            }
        }
    )


# ============================================================================
# REQUEST/RESPONSE MODELS (for API)
# ============================================================================

class CreateConversationRequest(BaseModel):
    """Request body for creating a new conversation"""
    
    title: Optional[str] = Field(
        default=None,
        description="Optional custom title. If not provided, will be auto-generated from first message",
        max_length=100
    )
    first_message: Optional[str] = Field(
        default=None,
        description="Optional first user message to start the conversation",
        max_length=1000
    )
    
    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "title": "Savings Account Help",
                "first_message": "How do I open a savings account?"
            }
        }
    )


class AddMessageRequest(BaseModel):
    """Request body for adding a message to conversation"""
    
    message: str = Field(..., description="User message to add")
    
    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "message": "What are the interest rates?"
            }
        }
    )


class UpdateConversationRequest(BaseModel):
    """Request body for updating conversation properties"""
    
    title: Optional[str] = Field(default=None, description="New title")
    is_archived: Optional[bool] = Field(default=None, description="Archive status")
    
    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "title": "Fixed Deposit Rates Discussion"
            }
        }
    )


class ConversationResponse(BaseModel):
    """Response model for single conversation"""
    
    id: str = Field(..., description="Conversation ID")
    user_id: str
    title: str
    messages: List[Message]
    is_archived: bool
    created_at: datetime
    updated_at: datetime
    last_message_at: Optional[datetime]
    message_count: int
    
    model_config = ConfigDict(
        json_encoders={
            datetime: lambda v: v.isoformat()
        }
    )


class ConversationListResponse(BaseModel):
    """Response model for list of conversations (without full messages)"""
    
    id: str
    user_id: str
    title: str
    preview: str = Field(..., description="Last message preview (first 100 chars)")
    is_archived: bool
    created_at: datetime
    updated_at: datetime
    last_message_at: Optional[datetime]
    message_count: int
    
    model_config = ConfigDict(
        json_encoders={
            datetime: lambda v: v.isoformat()
        },
        json_schema_extra={
            "example": {
                "id": "507f1f77bcf86cd799439011",
                "user_id": "user_123",
                "title": "Account Balance Inquiry",
                "preview": "What is my current account balance?",
                "is_archived": False,
                "created_at": "2024-01-15T10:30:00",
                "updated_at": "2024-01-15T10:35:00",
                "last_message_at": "2024-01-15T10:35:00",
                "message_count": 6
            }
        }
    )


class ConversationListResult(BaseModel):
    """Paginated list of conversations"""
    
    conversations: List[ConversationListResponse]
    total: int = Field(..., description="Total conversations matching filter")
    page: int = Field(default=1, description="Current page number")
    page_size: int = Field(default=20, description="Items per page")
    has_more: bool = Field(..., description="Are there more pages?")
    
    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "conversations": [],
                "total": 42,
                "page": 1,
                "page_size": 20,
                "has_more": True
            }
        }
    )

# ============================================================================
# UPDATE: backend/app/models/conversation.py
# ADD THIS TO Message CLASS (line ~40)
# ============================================================================

class Message(BaseModel):
    """
    Single message in a conversation.
    
    Contains:
    - User/assistant content
    - Metadata from RAG pipeline (policy action, retrieval stats)
    - User reaction (👍 👎)
    - Timestamp
    """
    
    role: str = Field(..., description="Role: 'user' or 'assistant'")
    content: str = Field(..., description="Message content")
    timestamp: datetime = Field(default_factory=datetime.utcnow)
    
    # Metadata from RAG pipeline (only for assistant messages)
    metadata: Optional[Dict[str, Any]] = Field(
        default=None,
        description="RAG metadata: policy_action, confidence, docs_retrieved, etc."
    )
    
    # ========================================================================
    # 🆕 NEW: User reaction to message
    # ========================================================================
    reaction: Optional[str] = Field(
        default=None,
        description="User reaction: 'like', 'dislike', or None",
        pattern="^(like|dislike)$"  # Only allow these values
    )
    
    model_config = ConfigDict(
        json_encoders={
            datetime: lambda v: v.isoformat()
        },
        json_schema_extra={
            "example": {
                "role": "assistant",
                "content": "Your account balance is $1,234.56",
                "timestamp": "2024-01-15T10:30:00",
                "metadata": {
                    "policy_action": "FETCH",
                    "confidence": 0.95
                },
                "reaction": "like"  # 👍
            }
        }
    )


# ============================================================================
# 🆕 NEW REQUEST MODEL
# ============================================================================

class ReactToMessageRequest(BaseModel):
    """Request body for reacting to a message"""
    
    reaction: str = Field(
        ...,
        description="Reaction type: 'like' or 'dislike'",
        pattern="^(like|dislike)$"
    )
    
    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "reaction": "like"
            }
        }
    )