import os
import json
from fastapi import FastAPI, Request, HTTPException
from fastapi.responses import StreamingResponse, HTMLResponse
from fastapi.middleware.cors import CORSMiddleware
from pydantic import BaseModel
from typing import List, Optional
from huggingface_hub import AsyncInferenceClient
from dotenv import load_dotenv

load_dotenv()

app = FastAPI(title="Vibe Coding Backend - 100% Free Edition")

# Enable CORS
app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"],
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)

# Initialize Hugging Face Inference Client (100% Free)
# User needs a FREE HF Token from huggingface.co/settings/tokens
HF_TOKEN = os.getenv("HF_TOKEN")
# Model: DeepSeek-Coder-V2-Lite-Instruct (Very powerful and free on HF API)
MODEL_ID = "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct"

client = AsyncInferenceClient(model=MODEL_ID, token=HF_TOKEN)

class Message(BaseModel):
    role: str
    content: str

class ChatRequest(BaseModel):
    messages: List[Message]
    stream: bool = True
    temperature: float = 0.6
    max_tokens: int = 4096

VIBE_SYSTEM_PROMPT = """You are the ultimate 'Vibe Coder'. Your purpose is to build entire, high-end web applications from top to bottom in a single response.
When a user gives you a 'vibe' or a concept:
1. Write 100% complete, functional code (HTML, CSS, JS).
2. Use modern, premium aesthetics: Glassmorphism, deep gradients, silky animations (Framer Motion style), and high-end typography (Inter/Outfit).
3. Do not use placeholders. If you need images, use Unsplash URLs.
4. Structure the code so it's ready to be copied into a single file or a clean project structure.
5. Be bold. Be creative. Make it look like a $100k startup design."""

@app.get("/", response_class=HTMLResponse)
async def root():
    with open("index.html", "r") as f:
        return f.read()

@app.post("/vibe")
async def vibe_coding(request: ChatRequest):
    """
    100% Free Vibe Coding using Hugging Face Serverless API.
    """
    # Inject Vibe System Prompt
    messages = [{"role": "system", "content": VIBE_SYSTEM_PROMPT}]
    for msg in request.messages:
        messages.append({"role": msg.role, "content": msg.content})

    async def stream_generator():
        try:
            # Using HF's free streaming capability
            stream = await client.chat_completion(
                messages=messages,
                max_tokens=request.max_tokens,
                stream=True,
                temperature=request.temperature,
            )
            
            async for chunk in stream:
                content = chunk.choices[0].delta.content
                if content:
                    yield f"data: {json.dumps({'content': content})}\n\n"
        except Exception as e:
            yield f"data: {json.dumps({'error': str(e)})}\n\n"

    return StreamingResponse(stream_generator(), media_type="text/event-stream")

if __name__ == "__main__":
    import uvicorn
    port = int(os.getenv("PORT", 7860))
    uvicorn.run(app, host="0.0.0.0", port=port)