Message-ID: <gh-pgjdbc-pgjdbc-3812-c3312122810@github.com>
From: "sehrope (@sehrope)" <noreply+sehrope@github.com>
To: "pgjdbc/pgjdbc" <noreply+pgjdbc-pgjdbc@github.com>
Date: Fri, 19 Sep 2025 13:04:37 +0000
Subject: Re: [pgjdbc/pgjdbc] issue #3812: Release is releasing a SNAPSHOT version
In-Reply-To: <gh-pgjdbc-pgjdbc-3812@github.com>
References: <gh-pgjdbc-pgjdbc-3812@github.com>
List-Id: <gh-pgjdbc-pgjdbc.github.com>
X-GitHub-Author-Login: sehrope
X-GitHub-Comment-Id: 3312122810
X-GitHub-Comment-Type: issue_comment
X-GitHub-Issue: 3812
X-GitHub-Repo: pgjdbc/pgjdbc
X-GitHub-Type: comment
X-GitHub-Url: https://github.com/pgjdbc/pgjdbc/issues/3812#issuecomment-3312122810
Content-Type: text/plain; charset=utf-8

I don't consider it fixed until I see a release done end-to-end without any manual intervention with the action successful. The CI shows failure so that clearly has not happened yet.

And it sounds like we either need to backpatch whatever was fixed to the older branches. Or update how things run so that it uses the latest version of the workflow in master but publishes from a version built from a separate branch. That feels like it'd have less maintenance in the future (only worry about keeping it in sync in master) but I'm not sure if we'd need any special handling for publishing back-patch releases.

Are you going to back patch the release fixes?

When this is all settled, I actually want to run the whole thing again end to end. We don't have to wait for any meaningful code changes. Version numbers are cheap so we can burn as many as we want to ensure this thing is working.