Skip to content

chore: refresh threats/fort_status goldens against current fixture image - #97

Merged
alexanderolvera merged 1 commit into
mainfrom
chore/refresh-fixture-goldens
Jul 23, 2026
Merged

chore: refresh threats/fort_status goldens against current fixture image#97
alexanderolvera merged 1 commit into
mainfrom
chore/refresh-fixture-goldens

Conversation

@alexanderolvera

Copy link
Copy Markdown
Owner

Summary

  • The committed threats/fort_status goldens were authored 2026-07-21 against an image whose fixture still had a 5-strong great-danger group ("Monster Of Vomit", DEMON_19) active. The currently-built df-headless:53.15 fixture no longer has that group — active_hostiles settled from 7 to 2, same in-game date.
  • Confirmed this is genuine, reproducible fixture-image drift, not per-boot nondeterminism: three fresh vanilla containers built from the same image produced byte-identical threats output, and a from-scratch build off a clean, isolated worktree checkout reproduces the identical diff.
  • No other tool's golden changed. military.json in particular was double-checked to confirm it still matches the currently-shipped mcp_military.lua exactly (ruling out cross-branch contamination noticed mid-investigation).

Per docs/VERIFY.md's own doctrine, a fixture-image bump churning goldens is expected and should land as a reviewable diff — this is that diff.

Test plan

  • npm run typecheck / npm run lint clean
  • verify:t2 passes cleanly against the current fixture image after the update

The committed goldens (last authored 2026-07-21) still showed a
5-strong great-danger group ("Monster Of Vomit", DEMON_19) that no
longer exists in the currently-built df-headless:53.15 fixture image's
world state -- active_hostiles has settled from 7 to 2 (the two goblin
thieves) with no change to in-game date. Verified this is a genuine,
reproducible fixture-state fact, not per-boot nondeterminism: three
fresh vanilla containers built from the same image produced
byte-identical threats output, and re-running against a container
built from a clean, isolated checkout (no contamination from any other
branch) reproduces the same diff. No other tool's golden changed;
military.json in particular was double-checked to confirm it still
matches the currently-shipped mcp_military.lua exactly.

Per VERIFY.md's own doctrine, a fixture-image bump churning goldens is
expected and should be a reviewable diff -- this is that diff.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Copilot AI review requested due to automatic review settings July 23, 2026 03:11

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Refreshes fixture-based golden snapshots to match the current deterministic df-headless:53.15 image.

Changes:

  • Removes the absent five-creature threat group.
  • Updates hostile totals from seven to two.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File Description
test/golden/threats.json Reflects zero active and two contained hostiles.
test/golden/fort_status.json Updates the hostile alert count to two.

@alexanderolvera
alexanderolvera merged commit e3eedea into main Jul 23, 2026
2 checks passed
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