from fastapi import FastAPI, HTTPException
from fastapi.middleware.cors import CORSMiddleware
from pydantic import BaseModel
from typing import Optional
import os

app = FastAPI(
    title="Innovatieplatform AI Service",
    description="AI service providing chat, summarization and semantic search for the Innovatieplatform.",
    version="0.1.0",
)

# CORS — allow requests from the Laravel app and local development
app.add_middleware(
    CORSMiddleware,
    allow_origins=[
        "http://laravel-app",
        "http://nginx",
        "http://localhost",
        "http://localhost:80",
        os.getenv("LARAVEL_APP_URL", "http://localhost"),
    ],
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)


# --- Request/Response models ---

class ChatRequest(BaseModel):
    message: str
    project_id: Optional[int] = None
    conversation_history: Optional[list] = []


class ChatResponse(BaseModel):
    reply: str
    project_id: Optional[int] = None


class SummarizeRequest(BaseModel):
    content: str
    project_id: Optional[int] = None
    summary_type: Optional[str] = "general"


class SummarizeResponse(BaseModel):
    summary: str
    project_id: Optional[int] = None


class SearchRequest(BaseModel):
    query: str
    project_id: Optional[int] = None
    limit: Optional[int] = 10


class SearchResult(BaseModel):
    id: int
    content: str
    score: float
    metadata: Optional[dict] = {}


class SearchResponse(BaseModel):
    results: list[SearchResult]
    query: str


# --- Endpoints ---

@app.get("/health")
async def health_check():
    """Health check endpoint used by Docker and monitoring."""
    return {"status": "ok", "service": "ai-service"}


@app.post("/api/chat", response_model=ChatResponse)
async def chat(request: ChatRequest):
    """
    Handle a chat message, optionally scoped to a project.
    Placeholder — wire up LangGraph + Anthropic in the full implementation.
    """
    # TODO: integrate LangGraph agent with Anthropic Claude
    reply = (
        f"[AI placeholder] Received: '{request.message}'. "
        "Full AI integration pending."
    )
    return ChatResponse(reply=reply, project_id=request.project_id)


@app.post("/api/summarize", response_model=SummarizeResponse)
async def summarize(request: SummarizeRequest):
    """
    Summarize content for a given project.
    Placeholder — wire up Anthropic in the full implementation.
    """
    # TODO: integrate Anthropic Claude summarization
    summary = (
        f"[AI placeholder] Summary of {len(request.content)} characters "
        f"(type: {request.summary_type}). Full AI integration pending."
    )
    return SummarizeResponse(summary=summary, project_id=request.project_id)


@app.post("/api/search", response_model=SearchResponse)
async def search(request: SearchRequest):
    """
    Semantic search using pgvector embeddings.
    Placeholder — wire up pgvector + embeddings in the full implementation.
    """
    # TODO: integrate pgvector similarity search with embeddings
    return SearchResponse(
        results=[],
        query=request.query,
    )