from fastapi import FastAPI, HTTPException
from fastapi import FastAPI
from fastapi.staticfiles import StaticFiles
from fastapi.responses import FileResponse
import requests
from transformers import pipeline, AutoModelForSequenceClassification, AutoTokenizer
import torch
from pydantic import BaseModel

app = FastAPI()

# Load model directly from Hugging Face Hub
model_name = "SandboxBhh/sentiment-thai-text-model"

try:
    device = 0 if torch.cuda.is_available() else -1  # Check if GPU is available
    # Ensure correct indentation here
    reloaded_pipe = pipeline(
        "text-classification",
        model=model_name,
        tokenizer=model_name,
        device=device,

    )

except Exception as e:
    print(f"Error loading model: {e}")
    reloaded_pipe = None

class TextInput(BaseModel):
    text: str

def send_line_notification(message, line_token, result, score, sentiment):
    url = "https://notify-api.line.me/api/notify"
    headers = {"Authorization": f"Bearer {line_token}"}
    
    # Create a detailed message including result, score, and sentiment
    detailed_message = (f"{message}\n\n"
                        f"Result: {json.dumps(result, ensure_ascii=False)}\n"
                        f"Score: {score:.4f}\n"
                        f"Sentiment: {sentiment}")
    
    data = {"message": detailed_message}
    response = requests.post(url, headers=headers, data=data)
    return response.status_code

def split_message(message, max_length=1000):
    return [message[i:i+max_length] for i in range(0, len(message), max_length)]

# Use environment variable for LINE token
line_token = "C9r65PpEvIvOJSK2xMhgl53WvmOhhnKEOuQq7DsiVJT"

@app.post("/classify-text")
async def classify_text(input: TextInput):
    if reloaded_pipe is None:
        raise HTTPException(status_code=500, detail="Model not loaded")
    try:
        result = reloaded_pipe(input.text)
        sentiment = result[0]['label'].lower()  
        score = result[0]['score']
        
        if sentiment == 'neg':
            message = f"[แจ้งเตือน CSI-Negetive]: \nข้อความ: {input.text} \n csi score: {score:.2f} \n ความคิดเห็นเชิงลบ"
            message_parts = split_message(message)
            notification_status = []
            for i, part in enumerate(message_parts):
                # status = send_line_notification(part, line_token)
                status = 200
                if status == 200:
                    notification_status.append(f"ส่งการแจ้งเตือนส่วนที่ {i+1}/{len(message_parts)} ผ่าน LINE สำเร็จ")
                else:
                    notification_status.append(f"การส่งการแจ้งเตือนส่วนที่ {i+1}/{len(message_parts)} ผ่าน LINE ล้มเหลว")
            return {
                "result": result, 
                "sentiment": "negative",
                "score": score,
                "message": "Negative sentiment detected and notification sent to LINE.",
                "formatted_message": message,
                "notification_status": notification_status
            }
        else:
            message = f"[ไม่พบความคิดเห็นเชิงลบ]: ข้อความ: {input.text} \ncsi score: {score:.2f}"
            return {
                "result": result, 
                "sentiment": "positive" if sentiment == 'pos' else "neutral",
                "score": score,
                "message": "Sentiment is not negative. No notification sent.",
                "formatted_message": message
            }
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))


app.mount("/", StaticFiles(directory="static", html=True), name="static")

@app.get("/")
def index() -> FileResponse:
    return FileResponse(path="/app/static/index.html", media_type="text/html")