From 8f236a12886295bff9e72e41d69808f07732d339 Mon Sep 17 00:00:00 2001
From: YASH <139299779+Yash-2707@users.noreply.github.com>
Date: Tue, 22 Oct 2024 17:37:07 +0530
Subject: [PATCH] Update reset_indexes.py

Error Handling: Add more specific error handling to make it easier to debug issues.
Configuration Management: Use environment variables or a configuration file for settings like DOCUMENT_INDEX_NAME and DOCUMENT_ID_ENDPOINT.
Logging: Improve logging to include more details about the operations.
Retry Mechanism: Add a retry mechanism for network requests to handle transient errors.
Testing: Add unit tests for the functions to ensure they work as expected
---
 backend/scripts/reset_indexes.py | 48 +++++++++++++++++++++++++-------
 1 file changed, 38 insertions(+), 10 deletions(-)

diff --git a/backend/scripts/reset_indexes.py b/backend/scripts/reset_indexes.py
index 4ec8d9bf3128..a8c22bf1f31b 100644
--- a/backend/scripts/reset_indexes.py
+++ b/backend/scripts/reset_indexes.py
@@ -1,9 +1,12 @@
 # This file is purely for development use, not included in any builds
 import os
 import sys
-
+import logging
 import requests
 
+from requests.exceptions import RequestException
+from time import sleep
+
 # makes it so `PYTHONPATH=.` is not required when running this script
 parent_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 sys.path.append(parent_dir)
@@ -14,23 +17,48 @@ from danswer.utils.logger import setup_logger  # noqa: E402
 
 logger = setup_logger()
 
-
 def wipe_vespa_index() -> None:
+    """
+    Wipes the Vespa index by deleting all documents.
+    """
     continuation = None
     should_continue = True
+    retries = 3
+
     while should_continue:
         params = {"selection": "true", "cluster": DOCUMENT_INDEX_NAME}
         if continuation:
-            params = {**params, "continuation": continuation}
-        response = requests.delete(DOCUMENT_ID_ENDPOINT, params=params)
-        response.raise_for_status()
+            params["continuation"] = continuation
 
-        response_json = response.json()
-        print(response_json)
+        for attempt in range(retries):
+            try:
+                response = requests.delete(DOCUMENT_ID_ENDPOINT, params=params)
+                response.raise_for_status()
 
-        continuation = response_json.get("continuation")
-        should_continue = bool(continuation)
+                response_json = response.json()
+                logger.info(f"Response: {response_json}")
 
+                continuation = response_json.get("continuation")
+                should_continue = bool(continuation)
+                break  # Exit the retry loop if the request is successful
+
+            except RequestException as e:
+                logger.error(f"Request failed: {e}")
+                sleep(2 ** attempt)  # Exponential backoff
+        else:
+            logger.error("Max retries exceeded. Exiting.")
+            sys.exit(1)
+
+def main():
+    """
+    Main function to execute the script.
+    """
+    try:
+        wipe_vespa_index()
+        logger.info("Vespa index wiped successfully.")
+    except Exception as e:
+        logger.error(f"An error occurred: {e}")
+        sys.exit(1)
 
 if __name__ == "__main__":
-    wipe_vespa_index()
+    main()