#10018 ci: Re-roll Docker images on a schedule against the latest Python base image Pull request opened by edgarrmondragon Description The weekly scheduled run of this workflow already built and scanned Docker images, but never actually pushed them anywhere — the

push:

condition was

false

for

schedule

events, and the registry was always

<http://ghcr.io|ghcr.io>

for non-release triggers. This change makes the weekly schedule perform a full re-roll of the latest stable release onto the current Python base images and push the result to Docker Hub. Specifically: •

build

job: on

schedule

, resolves the highest semver stable release tag (using

sort -V

over all non-prerelease tags, guarding against backport releases overwriting a newer version), then checks out that tag before building the wheel •

build_docker

job: routes

schedule

events to

<http://docker.io|docker.io>

(same as

release: published

), enables

push:

for

schedule

, and runs the

get-latest-version

step on

schedule

so the

latest

-tag anti-backport guard applies to re-rolls as well • Release Marker: extended to

schedule

events so re-rolled images are correctly identified as production builds •

pypi_release

: unchanged — still gated strictly on

release: published

Checklist • I have read the contribution guide Was generative AI tooling used to co-author this PR? • Yes (please specify the tool below) Related Issues • References https://github.com/meltano/meltano/discussions/10008 Summary by Sourcery Enable the scheduled CI workflow to fully rebuild and publish Docker images for the latest stable release using the current Python base image. CI: • Allow the build job on scheduled runs to resolve and check out the latest stable release tag before building artifacts. • Mark scheduled builds as production via the existing release marker mechanism. • Route scheduled Docker builds to Docker Hub, enable image pushing on scheduled events, and reuse the latest-version logic for these pushes. meltano/meltano