server.py

from flask import Flask, render_template, request
from flask_socketio import SocketIO, emit
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

app = Flask(__name__)
socketio = SocketIO(app, cors_allowed_origins="*")

code_content = {"text": "", "language": "python"}

@app.route("/")
def index():
    return render_template("index.html")

@socketio.on("code_update")
def handle_code_update(data):
    code_content["text"] = data["code"]
    emit("update_code", data, broadcast=True)

@socketio.on("language_change")
def handle_language_change(data):
    code_content["language"] = data["language"]
    emit("update_language", data, broadcast=True)

MODEL_NAME = "bigcode/santacoder"
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, torch_dtype=torch.float16, device_map="auto")

@app.route("/debug", methods=["POST"])
def debug_code():
    """AI-assisted debugging using SantaCoder (lightweight model)."""

    user_code = request.json.get("code")
    language = request.json.get("language")

    print("Received Code for Debugging")  

    if tokenizer.pad_token is None:
        tokenizer.pad_token = tokenizer.eos_token

    prompt = f"Analyze the following {language} code for syntax errors and optimizations:\n\n" \
             f"```{language}\n{user_code}\n```\n\nProvide structured suggestions."

    device = "cuda" if torch.cuda.is_available() else "cpu"
    inputs = tokenizer(prompt, return_tensors="pt", padding=True, truncation=True).to(device)

    
    model.config.pad_token_id = tokenizer.pad_token_id
    stopping_criteria = torch.tensor([tokenizer.eos_token_id], device=device)

    print("Generating Debugging Suggestions...")  

    try:
        output = model.generate(
            **inputs,
            max_length=256,
            do_sample=True,
            top_k=50,
            top_p=0.95,
            temperature=0.7,
            num_return_sequences=1,
            eos_token_id=tokenizer.eos_token_id
        )
        
        print("Model Output Generated")  

        suggestion = tokenizer.decode(output[0], skip_special_tokens=True).strip()
        
        if not suggestion or suggestion == prompt:
            suggestion = "No issues detected, or AI couldn't analyze the code properly."
        
        return {"suggestion": suggestion}
    
    except Exception as e:
        print("Error in AI Debugging:", str(e))  
        return {"suggestion": "AI debugging failed due to an error."}

if __name__ == "__main__":
    socketio.run(app, debug=True, host="0.0.0.0", port=5000)