fix basic function

remove client for session and using global variable to pass result
also adding unit test goal

engineered_chatgpt_prompts.py

@@ -7,7 +7,7 @@
 import asyncio
 import os
 import sys
-import traceback
+
 import openai
 from dotenv import load_dotenv, find_dotenv
 from PyQt5.QtWidgets import (QApplication,  # pylint: disable=no-name-in-module
@@ -21,10 +21,9 @@
 
 _ = load_dotenv(find_dotenv())  # read local .env file
 
-client = openai.AsyncOpenAI(
-    api_key=os.getenv('OPENAI_API_KEY'),
-    organization=os.getenv('OPENAI_ORGANIZATION'),
-)
+openai.api_key = os.getenv('OPENAI_API_KEY')
+openai.organization = os.getenv('OPENAI_ORGANIZATION')
+response = ""
 
 print("OpenAI version:", openai.__version__)
 
@@ -36,41 +35,26 @@ async def get_completion(prompt,
     method to query openai API
     """
     messages = [{"role": "user", "content": prompt}]
-    chat = None
-    try:
-        # chat = openai.ChatCompletion.create(
-        chat = await client.chat.completions.create(
+    chat = openai.ChatCompletion.create(
             model=model,
             messages=messages,
             temperature=0,
-            stream=True,
+            # stream=True,
             # this is the randomness degree of the model's output
         )
 
-    except openai.APIConnectionError as e:
-        print("The server could not be reached")
-        print(e.__cause__)  # an underlying Exception, likely raised within http
-    except openai.RateLimitError as e:
-        print("A 429 status code was received; we should back off a bit.")
-    except openai.APIStatusError as e:
-        print("Another non-200-range status code was received")
-        print(e.status_code)
-        print(e.response)
-        return None
-
-    response = ""
-    async for part in chat:
-        response += part.choices[0].delta.content or ""
-        sys.stdout.write(f"\r{response}>")
-        sys.stdout.flush()
-        # print(response)
+
+    global response
+    response = chat.choices[0].message["content"]
+    sys.stdout.write(f"\r{response}>")
+    sys.stdout.flush()
     return response
 
 
 class EngineeredChatgptPrompts(
     QWidget):  # pylint: disable=too-many-instance-attributes
     """
-    class to hold widgets and preocess method of main application
+    class to hold widgets and process method of main application
     """
 
     def __init__(self):
@@ -147,6 +131,8 @@ def process_text(self):
                            f"process the following text with specified goal"
                            f"(delimited by triple backticks): ```{input_text}```")
         asyncio.run(get_completion(complete_prompt))
+        global response
+        self.output_text.setPlainText(response)
 
     def load_goal(self):
         """ open a dialog inspecting text files on file system """