feat: api key protection, better err handling, vercel setup

.gitignore

@@ -1 +1,2 @@
-/venv/**
+/venv/**
+.flaskenv

main.py

@@ -1,9 +1,14 @@
 from flask import Flask, request, Response, jsonify
 import g4f
 import sys
+import os
+from dotenv import load_dotenv
+load_dotenv()
 
 app = Flask(__name__)
 
+API_KEY = os.environ.get('API_KEY')
+
 def generate_response(model, llm, content):
  response_generator = g4f.ChatCompletion.create(
  model=model,
@@ -19,11 +24,19 @@ def generate_response(model, llm, content):
 def chat_completion():
  data = request.get_json()
 
- if 'content' not in data or 'provider' not in data:
- return jsonify({"error": "Missing 'content' or 'provider' in the request"}), 400
+ required_params = {'content', 'provider', 'api_key'}
+ missing_params = required_params - set(data.keys())
+
+ if missing_params:
+ error_msg = f"Missing parameters: {', '.join(missing_params)}"
+ return jsonify({"error": error_msg}), 400
 
  content = data['content']
  pname = data['provider']
+ api_key = data['api_key']
+
+ if not api_key == API_KEY:
+ return jsonify({"error": "Invalid API key"}), 401
 
  try:
  llm = getattr(getattr(getattr(sys.modules[__name__], "g4f"), "Provider"), pname)
@@ -32,7 +45,7 @@ def chat_completion():
 
  model = "gpt-3.5-turbo"
 
- return Response(generate_response(model, llm, content), mimetype="text/event-stream")
+ return Response(generate_response(model, llm, content), content_type='text/plain', mimetype='text/event-stream')
 
 @app.route('/working_providers', methods=['GET'])
 def working_providers():

requirements.txt

@@ -50,6 +50,7 @@ pydantic_core==2.14.5
 PyExecJS==1.5.1
 pyjsparser==2.7.1
 PySocks==1.7.1
+python-dotenv==1.0.0
 requests==2.31.0
 selenium==4.16.0
 six==1.16.0

vercel.json

@@ -0,0 +1,15 @@
+{
+ "version": 2,
+ "builds": [
+ {
+ "src": "./main.py",
+ "use": "@vercel/python"
+ }
+ ],
+ "routes": [
+ {
+ "src": "/(.*)",
+ "dest": "/"
+ }
+ ]
+}