From ea87febd2bdf0aebf603532be9448e6435f1fea9 Mon Sep 17 00:00:00 2001
From: Hyejin Yoon <0327jane@gmail.com>
Date: Fri, 6 Oct 2023 14:36:32 +0900
Subject: [PATCH] fix: add retry for fetch_url (#8958)

---
 docs-website/download_historical_versions.py | 34 ++++++++++++++------
 1 file changed, 24 insertions(+), 10 deletions(-)

diff --git a/docs-website/download_historical_versions.py b/docs-website/download_historical_versions.py
index 83157edc1972c..53ee9cf1e63ef 100644
--- a/docs-website/download_historical_versions.py
+++ b/docs-website/download_historical_versions.py
@@ -1,6 +1,7 @@
 import json
 import os
 import tarfile
+import time
 import urllib.request
 
 repo_url = "https://api.github.com/repos/datahub-project/static-assets"
@@ -16,17 +17,30 @@ def download_file(url, destination):
                 f.write(chunk)
 
 
-def fetch_urls(repo_url: str, folder_path: str, file_format: str):
+def fetch_urls(
+    repo_url: str, folder_path: str, file_format: str, max_retries=3, retry_delay=5
+):
     api_url = f"{repo_url}/contents/{folder_path}"
-    response = urllib.request.urlopen(api_url)
-    data = response.read().decode("utf-8")
-    urls = [
-        file["download_url"]
-        for file in json.loads(data)
-        if file["name"].endswith(file_format)
-    ]
-    print(urls)
-    return urls
+    for attempt in range(max_retries + 1):
+        try:
+            response = urllib.request.urlopen(api_url)
+            if response.status == 403 or (500 <= response.status < 600):
+                raise Exception(f"HTTP Error {response.status}: {response.reason}")
+            data = response.read().decode("utf-8")
+            urls = [
+                file["download_url"]
+                for file in json.loads(data)
+                if file["name"].endswith(file_format)
+            ]
+            print(urls)
+            return urls
+        except Exception as e:
+            if attempt < max_retries:
+                print(f"Attempt {attempt + 1}/{max_retries}: {e}")
+                time.sleep(retry_delay)
+            else:
+                print(f"Max retries reached. Unable to fetch data.")
+                raise
 
 
 def extract_tar_file(destination_path):