app.py

from flask import Flask, request, jsonify, render_template_string
from transformers import T5Tokenizer, T5ForConditionalGeneration
from deep_translator import GoogleTranslator
from PIL import Image
from langdetect import detect, DetectorFactory
import torch
import requests
import io
import base64
import os
from dotenv import load_dotenv

# Load environment variables
load_dotenv()

app = Flask(__name__)

# Translation setup
translator = GoogleTranslator(source='auto', target='en')
DetectorFactory.seed = 0

# Hugging Face image generation API
HF_TOKEN = os.getenv("HF_TOKEN")
headers = {"Authorization": f"Bearer {HF_TOKEN}"}
IMAGE_MODEL_API_URL = "https://api-inference.huggingface.co/models/black-forest-labs/FLUX.1-dev"

# Load text generation model
model_id = "google/flan-t5-base"
tokenizer = T5Tokenizer.from_pretrained(model_id, legacy=False)
model = T5ForConditionalGeneration.from_pretrained(model_id).to("cuda" if torch.cuda.is_available() else "cpu")

# Text generation function
def query_text_model(prompt):
    structured_prompt = f"Describe the following in bullet points: {prompt}"
    inputs = tokenizer(structured_prompt, return_tensors="pt").to(model.device)
    outputs = model.generate(**inputs, max_new_tokens=256)
    text = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return [line.strip("-• ") for line in text.split("\n") if line.strip()]

# Format text for HTML display
def format_text_response(prompt, points):
    formatted_response = f"<strong>{prompt}:</strong><ul>"
    for point in points:
        formatted_response += f"<li style='text-indent: 20px;'>{point}</li>"
    formatted_response += "</ul>"
    return formatted_response

# Call Hugging Face API for image generation
def query_image_model(prompt):
    try:
        response = requests.post(IMAGE_MODEL_API_URL, headers=headers, json={"inputs": prompt})
        if response.status_code == 200 and response.headers['Content-Type'].startswith('image/'):
            return response.content
        else:
            raise ValueError("Image generation failed or unexpected content type")
    except requests.exceptions.RequestException as e:
        print("Image generation error:", str(e))
        return None

# Convert image bytes to base64
def image_to_base64(image_bytes):
    image = Image.open(io.BytesIO(image_bytes))
    buffered = io.BytesIO()
    image.save(buffered, format="JPEG")
    return base64.b64encode(buffered.getvalue()).decode("utf-8")

# Load HTML page
@app.route('/')
def index():
    return render_template_string(open("index.html", "r", encoding="utf-8").read())

# Chat endpoint handling text, image, and translation
@app.route('/chat', methods=['POST'])
def chat():
    data = request.json
    user_input = data.get('message', '').strip()
    response_type = data.get('response_type', 'text')

    print(f"[Received] {response_type.upper()} | {user_input}")

    if not user_input:
        return jsonify({"response": "Please enter something first."})

    if response_type == 'image':
        image_bytes = query_image_model(user_input)
        if image_bytes:
            image_base64 = image_to_base64(image_bytes)
            return jsonify({"response": f'<img src="data:image/jpeg;base64,{image_base64}"/>'})
        return jsonify({"response": "Image generation failed."})

    elif response_type == 'translation':
        try:
            translated = translator.translate(user_input)
            return jsonify({"response": translated})
        except Exception as e:
            print("Translation error:", e)
            return jsonify({"response": "Sorry, translation failed."})

    else:  # Default: text
        try:
            text_points = query_text_model(user_input)
            if text_points:
                formatted_response = format_text_response(user_input, text_points)
                return jsonify({"response": formatted_response})
            return jsonify({"response": "No meaningful response generated."})
        except Exception as e:
            print("Text generation error:", e)
            return jsonify({"response": "Something went wrong while generating a response."})

# Run the app
if __name__ == '__main__':
    port = int(os.environ.get("PORT", 7860))
    app.run(host='0.0.0.0', port=port)