Skip to content

fix(podman): the guidance pins a tag that exists, and the test agrees - #64

Merged
LMPrado-DZ23 merged 1 commit into
release/v3.8.55from
fix/podman-tag-after-release
Sep 20, 2026
Merged

LMPrado-DZ23 merged 1 commit into
release/v3.8.55from
fix/podman-tag-after-release

Conversation

@LMPrado-DZ23

Copy link
Copy Markdown
Owner

My regression from #63, and the test was right to catch it.

#63 rewrote the docs from :next to :latest. tests/unit/podman-machine-guidance-8497.test.ts asserts the tag the podman guidance names, so it went red on the base — proven by the agent working on #56, who ran it on the tip either side of that merge: 7/7 before, 6/7 after.

It also exposed something #63 left half-done: I changed contrib/podman/README.md but not contrib/podman/omniroute.container next to it, so the Quadlet unit still said :next while the README said :latest. The guidance disagreed with itself.

Why the original pin was right, and why it stopped being right

The test carries its own reasoning, from Fase 8:

:next until a version is released — :latest is only promoted by a SemVer publish, so pinning it here documented a pull that returns "manifest unknown".

Correct when written. v3.8.54 shipped on 2026-09-19:

docker manifest inspect ghcr.io/lmprado-dz23/omniroute:latest  -> present
docker manifest inspect ghcr.io/lmprado-dz23/omniroute:3.8.49  -> absent

The premise expired. This is the same class of staleness #63 fixed one layer up — a correct reason, recorded in a comment, that quietly became false when the world changed. So the comment is rewritten to state the new reason and the evidence, rather than leaving behind a justification that no longer holds.

Change

File Was Is
contrib/podman/omniroute.container Image=…/omniroute:next Image=…/omniroute:latest
the two assertions in podman-machine-guidance-8497.test.ts :next :latest

The test keeps its teeth: it still pins the tag the guidance names, so the next person who changes one and not the other gets the same failure I did.

Gate Result
tests/unit/podman-machine-guidance-8497.test.ts 7 pass / 0 fail
eslint --max-warnings=0 exit 0
prettier --check clean
check:doc-links PASS

🤖 Generated with Claude Code

#63 rewrote the docs from `:next` to `:latest` and broke
`podman-machine-guidance-8497.test.ts`, which asserts the tag the podman
guidance names. My regression, and the test was right to catch it: the
Quadlet unit still said `:next` while the README next to it said
`:latest`, so the guidance disagreed with itself.

The test's own comment records why `:next` was pinned in Fase 8:

  `:latest` is only promoted by a SemVer publish, so pinning it here
  documented a pull that returns "manifest unknown"

That was true when written. v3.8.54 shipped on 2026-09-19 and `:latest`
resolves:

  docker manifest inspect ghcr.io/lmprado-dz23/omniroute:latest  -> present
  docker manifest inspect ghcr.io/lmprado-dz23/omniroute:3.8.49  -> absent

So the premise expired — the same class of staleness #63 fixed, one layer
down. `contrib/podman/omniroute.container` now pins `:latest` like the
README, and both assertions follow, with the comment rewritten to say why
rather than leaving a reason that is no longer true.

  tests 7 / pass 7 / fail 0
  eslint exit 0, prettier clean, [doc-links] PASS

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
@coderabbitai

coderabbitai Bot commented Sep 20, 2026

Copy link
Copy Markdown

Important

  • 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Advanced

Run ID: 715e37af-cb64-4cc8-bcbd-eab6601d7084


Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

@LMPrado-DZ23
LMPrado-DZ23 merged commit 5f52c9b into release/v3.8.55 Sep 20, 2026
15 checks passed
LMPrado-DZ23 added a commit that referenced this pull request Sep 20, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants