implement retry strategy for remote calls · docling-project/docling#2465

(2 comments) (0 reactions) (1 assignee)Python (59,751 stars) (4,140 forks)batch import

enhancementgood first issue

説明

Hi, when for example sending images for description to remote API, it would be beneficial to add a retry strategy in the underlying requests calls, i.e.:

import requests

from requests.adapters import HTTPAdapter, Retry

s = requests.Session()

retries = Retry(total=5,
                backoff_factor=0.1,
                status_forcelist=[ 500, 502, 503, 504 ])

s.mount('http://', HTTPAdapter(max_retries=retries))

s.get('http://httpstat.us/500')

This way we could solve various errors that might happen during these calls, such as:

"Requests to the ChatCompletions_Create Operation under Azure OpenAI API have exceeded token rate limit of your current AIServices S0 pricing tier. Please retry after 11 seconds.

コントリビューターガイド

技術スタック: python
領域: backend
Issue 種別: feature
難度: 2
推定時間: 1-3 hours
活動状況: blocked
明確さ: clear
前提条件: Pythonrequests library
初心者向け度: 60
調査方針: The issue requests adding a retry strategy to remote HTTP calls in docling/utils/api image request.py. The current code uses `requests`; a session with retry adapter should be implemented as shown in the issue description. The retry should handle 500 level errors and rate limits. No linked PRs or maintainer comments are present, so the implementation can follow the provided snippet.