From 3ff5074cf5970d1285b4a49d1b7221593f0e8a99 Mon Sep 17 00:00:00 2001
From: sabaimran <narmiabas@gmail.com>
Date: Wed, 5 Jul 2023 14:59:44 -0700
Subject: [PATCH] Log the end-to-end time of generating a streamed response
 from OpenAI

---
 src/khoj/processor/conversation/utils.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/src/khoj/processor/conversation/utils.py b/src/khoj/processor/conversation/utils.py
index cab4ef19..4c3e7619 100644
--- a/src/khoj/processor/conversation/utils.py
+++ b/src/khoj/processor/conversation/utils.py
@@ -2,6 +2,7 @@
 import os
 import logging
 from datetime import datetime
+from time import perf_counter
 from typing import Any
 from threading import Thread
 import json
@@ -38,6 +39,7 @@ class ThreadedGenerator:
         self.compiled_references = compiled_references
         self.completion_func = completion_func
         self.response = ""
+        self.start_time = perf_counter()
 
     def __iter__(self):
         return self
@@ -45,6 +47,8 @@ class ThreadedGenerator:
     def __next__(self):
         item = self.queue.get()
         if item is StopIteration:
+            time_to_response = perf_counter() - self.start_time
+            logger.info(f"Time to stream full response: {time_to_response:.3f}")
             if self.completion_func:
                 # The completion func effective acts as a callback.
                 # It adds the aggregated response to the conversation history. It's constructed in api.py.