Refactor debug.py for cleaner debugpy handling, enhance logging levels in main.py, and improve error handling in task_list.py; Checkpoint before refactoring summary.py

Author: mrjcleaver

src/ai_chunking.py

@@ -36,7 +36,7 @@ def chunk_messages(prior_messages, recent_channel_messages, max_tokens=30000, as
         logging.debug(f"Number of chunks: {len(chunks)}")
         return chunks
 
-
+# This is an Arun's implementation of chunking messages for OpenAI API.
 def group_messages(messages, model):
     """
     Groups messages into chunks based on the token limit of the model.

src/debug.py

@@ -50,15 +50,11 @@ def _watch():
 
 watch_for_debugger_connects()
 
-if True:
-    if os.getenv("DEBUGPY_WAIT") == "true":
-        logging.info("🪛 debugpy enabled, now waiting_for_client")
-        debugpy.wait_for_client()  # noqa # NEVER BREAKPOINT HERE
-        debugpy.breakpoint()
+if os.getenv("DEBUGPY_WAIT") == "true":
+    logging.info("🪛 debugpy enabled, now waiting_for_client")
+    debugpy.wait_for_client()  # noqa # NEVER BREAKPOINT HERE
+    debugpy.breakpoint()
 
-        logging.info("🪛 client attached")
-        logging.info("Breakpoint should hit here")  # ⛔ Set breakpoint here
+    logging.info("🪛 client attached")
+    logging.info("Breakpoint should hit here")  # ⛔ Set breakpoint here
 
-#if os.getenv("DEBUGPY_WAIT") == "true":
-#    logging.info("🪛 debugpy enabled, now waiting_for_client")
-#    debugpy.wait_for_client()  # noqa

src/main.py

@@ -1,5 +1,20 @@
 import logging
-logging.basicConfig(level=logging.INFO)
+logging.basicConfig(level=logging.DEBUG)
+logging.getLogger('apscheduler').setLevel(logging.INFO)
+logging.getLogger('discord').setLevel(logging.INFO)
+logging.getLogger('werkzeug').setLevel(logging.INFO)
+logging.getLogger('flask').setLevel(logging.INFO)
+logging.getLogger('apscheduler.executors.default').setLevel(logging.INFO)
+logging.getLogger('apscheduler.executors.pool').setLevel(logging.INFO)
+logging.getLogger('apscheduler.scheduler').setLevel(logging.INFO)
+logging.getLogger('apscheduler.triggers.cron').setLevel(logging.INFO)
+logging.getLogger('apscheduler.triggers.interval').setLevel(logging.INFO)
+logging.getLogger('apscheduler.triggers.date').setLevel(logging.INFO)
+logging.getLogger('apscheduler.triggers.combining').setLevel(logging.INFO)
+logging.getLogger('apscheduler.job').setLevel(logging.INFO)
+logging.getLogger('apscheduler.jobstores.memory').setLevel(logging.INFO)
+logging.getLogger('apscheduler.jobstores.sqlalchemy').setLevel(logging.INFO)
+
 
 logging.info("Starting main.py")
 import os
@@ -75,7 +90,8 @@ def run_webhook(bot, summary_func):
 # Run the Webhook Flask app in a thread to avoid blocking the bot
 webhook_server_thread = threading.Thread(target=run_webhook, 
                                          args=(bot,
-                                               summary_for_webhook)
+                                               summary_for_webhook),
+                                         name="webhook_server_thread",      
                                         )
 webhook_server_thread.start()
 

src/task_list.py

@@ -58,6 +58,8 @@ def perform_job(self, payload):
 
 
 ### WEBHOOK RELATED FUNCTIONS
+    # This effectively calls ```summary_for_webhook.py:summary_for_webhook```
+
     def call_webhook_with_payload(self, url, payload):
         logging.debug(f"Sending {payload} to {url}")
         headers = {"Content-Type": "application/json"}
@@ -78,16 +80,19 @@ def perform_job_using_webhook(self, payload):
         else:
             result = response.text
 
+        data = {}
+        
         try:
             data = json.loads(result) 
-        except json.JSONDecodeError:
-            logging.error(f"Error decoding JSON: {result}")
-            
 
-        if data["message"]:
-            message = data["message"]
-        else:
+            if data["message"]:
+                message = data["message"]
+
+        except json.JSONDecodeError:
+            message = "Error decoding JSON response - Chances are, we passed too many tokens."
+            logging.error(message)
             message = result
+        
 
         self.store_job_result(payload, message) #TODO: think about name
         return message