ci(core-ref): pin the core on main, now that the cutover has landed #27

wtclaude · 2026-09-01T12:48:22-05:00

wtclaude commented

2026-09-01 12:48:22 -05:00

A leftover from cutover step 4. ci/core-ref.json still names website edge — a branch deleted by the cutover — because #25 moved the pin there mid-window to unbreak frozen-manifest, with its own note saying it reverts to a main sha afterwards.

ref      52eac24d…  →  66bb3b9a…
refName  edge @ MODULE_API 1.9.0, engagement Phase 11b (website#179)
      →  main @ MODULE_API 1.9.0, the engagement cutover (website#180)

Nothing regenerates, and that is checkable rather than asserted

The file's own instruction is "bump it, regenerate routes.manifest.json, and commit both together". There is nothing to pair with here: website's tree at 66bb3b9a (main, the cutover merge) and at 52eac24d (the edge head it merged) are the same tree, e7a7240 — main was zero commits ahead, so the merge carried edge's tree unchanged. frozen-manifest clones a different commit and reads identical bytes.

So this is documentary, which is what the pin is for: it tells the next person which core this module was last proved against, named by a ref they can still resolve. A sha on a deleted branch still clones — that is exactly why a wrong label here would have sat unnoticed.

AI-assisted: written with Claude Code.

🤖 Generated with Claude Code

A leftover from cutover step 4. `ci/core-ref.json` still names website `edge` — a branch deleted by the cutover — because #25 moved the pin there mid-window to unbreak `frozen-manifest`, with its own note saying it reverts to a `main` sha afterwards. ``` ref 52eac24d… → 66bb3b9a… refName edge @ MODULE_API 1.9.0, engagement Phase 11b (website#179) → main @ MODULE_API 1.9.0, the engagement cutover (website#180) ``` ### Nothing regenerates, and that is checkable rather than asserted The file's own instruction is *"bump it, regenerate `routes.manifest.json`, and commit both together"*. There is nothing to pair with here: website's tree at **66bb3b9a** (`main`, the cutover merge) and at **52eac24d** (the `edge` head it merged) are the **same tree**, `e7a7240` — `main` was zero commits ahead, so the merge carried edge's tree unchanged. `frozen-manifest` clones a different commit and reads identical bytes. So this is documentary, which is what the pin is for: it tells the next person which core this module was last proved against, named by a ref they can still resolve. A sha on a deleted branch still clones — that is exactly why a wrong label here would have sat unnoticed. - [x] AI-assisted: written with Claude Code. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

wtclaude added 1 commit 2026-09-01 12:48:24 -05:00

ci(core-ref): pin the core on main, now that the cutover has landed

PR Checks / client-build (pull_request) Successful in 33s

Details

PR Checks / frozen-manifest (pull_request) Successful in 47s

Details

PR Checks / server-tests (pull_request) Successful in 8m24s

Details

c679944181

Module-uo#25 moved this pin onto website `edge` (52eac24) to unbreak
frozen-manifest during the engagement window, with its own note saying it
reverts to a `main` sha at the cutover. The cutover is step 4 of 7, merged as
#26, and website#180 landed the same code on `main` -- so the pin now names a
branch that no longer exists.

No regeneration, and the reason is checkable rather than asserted: website's
tree at 66bb3b9a (main, the cutover merge) and at 52eac24d (the edge head it
merged) are the SAME tree, e7a7240. `main` was zero commits ahead, so the merge
carried edge's tree unchanged. The frozen-manifest job clones a different commit
and reads identical bytes; routes.manifest.json cannot move.

What changes is what a reader learns from the file: which core this module was
last proved against, named by a ref they can still resolve.

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

whitlocktech merged commit 144242fe8f into main

2026-09-01 13:03:22 -05:00

whitlocktech deleted branch ci/core-ref-main

2026-09-01 13:03:23 -05:00

whitlocktech referenced this issue from a commit

2026-09-01 13:03:23 -05:00

Merge pull request 'ci(core-ref): pin the core on `main`, now that the cutover has landed' (#27) from ci/core-ref-main into main

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: RunicGateway/Module-uo#27

ci(core-ref): pin the core on main, now that the cutover has landed #27

Nothing regenerates, and that is checkable rather than asserted

ci(core-ref): pin the core on `main`, now that the cutover has landed #27