From df4141ec28368507fbf71cb89e8ab334522e3fc9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Fran=C3=A7oise=20Conil?= <francoise.conil@insa-lyon.fr> Date: Thu, 4 Jul 2024 12:05:02 +0200 Subject: [PATCH] Naive curl requests to retrieve all issues I made a first request to get the number of pages then several curl requests. Unfortunately there were duplicates issues. There may be some headers to catch and reuse but I do not know which ones. GitHub strongly insists on using their CLI (gh) to acccess the API. --- curl/get-all-pypi-issues_curl.sh | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) create mode 100755 curl/get-all-pypi-issues_curl.sh diff --git a/curl/get-all-pypi-issues_curl.sh b/curl/get-all-pypi-issues_curl.sh new file mode 100755 index 0000000..7fd6c44 --- /dev/null +++ b/curl/get-all-pypi-issues_curl.sh @@ -0,0 +1,17 @@ +#!/bin/bash + +# https://blog.exploratory.io/analyzing-issue-data-with-github-rest-api-63945017dedc +# (state = "all", per_page = 100, page = i +# -o pypi-issues_31-05-2024.${page}.json \ + +# Ou plus simplement dans la doc de l'API GitHub +#Â https://docs.github.com/en/rest/issues/issues?apiVersion=2022-11-28#list-repository-issues + +seq -w 1 42 | while read page +do + curl -s -L -H "Accept: application/vnd.github+json" \ + -H "Authorization: Bearer ${MY_PYPI_ISSUES_TOKEN}" \ + -H "X-GitHub-Api-Version: 2022-11-28" \ + -o pypi_issues.${page}.json \ + "https://api.github.com/repos/pypi/support/issues?state=all&per_page=100&page=${page}" +done -- GitLab