frankenbit/release-toolkit

Fork 0

docs(integration): "Replace `v0.62.1`" is an ungraded number that forwards to the section #1401 emptied #1415

New issue

Closed

opened 2026-09-07 01:35:51 +02:00 by bosun · 1 comment

bosun commented

2026-09-07 01:35:51 +02:00

Owner

docs/integration.md tells an adopter to "Replace v0.62.1 with the toolkit tag you want" and points at the README Status section for the answer — a number nothing grades, in a sentence that forwards the reader to a section which, after #1401, no longer carries one.

Measured on main `8d207d74`

docs/integration.md:863
  "Replace `v0.62.1` with the toolkit tag you want to pin to (the latest released
   tag -- see [README Status](../README.md#status) ...)"

carries no `@`      -> readme-pin-check does not grade it
MUTATION  Replace `v0.62.1` -> `v0.11.1`   gate rc=0, SILENT
IS machine-set      -> internal/prep/doc_version_refs.go:95, replaceTagRE

Same shape as the **Latest: vX.Y.Z** line #1401 removed: machine-SET, not machine-KEPT. prep advances it at cut time; nothing catches a hand-edit, and nothing notices when it goes stale between cuts.

And the cross-reference makes it worse than its sibling

⚠️ It forwards the reader to README Status for the authoritative version — and that section is where the decision on #1401 removed the descriptive number. So the sentence sends a reader to a section that deliberately no longer answers the question the sentence promised it would.

Scope

Not #1411, deliberately — that tracker is about a SETTER losing its subject; this is about a CLAIM in a different file with different reviewers. And not a widening of readme-pin-check: the rule is a number may appear iff something grades it, and the remedy here may be to remove the number rather than to grade it.

AC

docs/integration.md:863 either carries a number something grades, or carries none
The cross-reference points at a section that answers what the sentence says it answers
If replaceTagRE ends up with no subject, that is decided rather than left inert -- the same judgement #1411 records for latestLineRE

Anchor

@shipwright, measured on origin/main while deciding #1411, requested as a sibling rather than folded into it. Mutation re-run by @bosun. Related: #1401, #1411.

docs/integration.md tells an adopter to "Replace `v0.62.1` with the toolkit tag you want" and points at the README Status section for the answer — a number nothing grades, in a sentence that forwards the reader to a section which, after #1401, no longer carries one. ## Measured on main `8d207d74` ``` docs/integration.md:863 "Replace `v0.62.1` with the toolkit tag you want to pin to (the latest released tag -- see [README Status](../README.md#status) ...)" carries no `@` -> readme-pin-check does not grade it MUTATION Replace `v0.62.1` -> `v0.11.1` gate rc=0, SILENT IS machine-set -> internal/prep/doc_version_refs.go:95, replaceTagRE ``` **Same shape as the `**Latest: vX.Y.Z**` line `#1401` removed: machine-SET, not machine-KEPT.** *prep advances it at cut time; nothing catches a hand-edit, and nothing notices when it goes stale between cuts.* ## And the cross-reference makes it worse than its sibling ⚠️ **It forwards the reader to `README Status` for the authoritative version — and that section is where the decision on `#1401` removed the descriptive number.** *So the sentence sends a reader to a section that deliberately no longer answers the question the sentence promised it would.* ## Scope **Not `#1411`, deliberately** — that tracker is about a SETTER losing its subject; this is about a CLAIM in a different file with different reviewers. **And not a widening of `readme-pin-check`**: the rule is *a number may appear iff something grades it*, and the remedy here may be to remove the number rather than to grade it. ## AC - [x] `docs/integration.md:863` either carries a number something grades, or carries none - [x] The cross-reference points at a section that answers what the sentence says it answers - [x] If `replaceTagRE` ends up with no subject, that is decided rather than left inert -- the same judgement `#1411` records for `latestLineRE` ## Anchor @shipwright, measured on `origin/main` while deciding `#1411`, requested as a sibling rather than folded into it. Mutation re-run by @bosun. Related: `#1401`, `#1411`.

bosun added the

priority/medium

size/S

kind/docs

labels

2026-09-07 01:35:51 +02:00

shipwright was assigned by bosun

2026-09-07 01:35:51 +02:00

bosun referenced this issue

2026-09-07 02:39:02 +02:00

docs(rule): "a number may appear iff something grades it" must exempt numbers that name the past #1423

shipwright referenced this issue from a commit

2026-09-07 02:42:34 +02:00

fix(docs): the integration guide names no ungraded version

shipwright referenced this issue

2026-09-07 02:42:54 +02:00

fix(docs): the integration guide names no ungraded version (#1415) #1424

bosun referenced this issue

2026-09-07 02:43:58 +02:00

chore(prep): UpdateDocVersionRefs writes nothing on every call now that both its subjects are gone #1425

bosun referenced this issue from a commit

2026-09-07 08:45:07 +02:00

fix(docs): the integration guide names no ungraded version

release-bot referenced this issue

2026-09-07 08:45:59 +02:00

chore(release): v0.62.3 #1429

bosun commented

2026-09-07 08:46:01 +02:00

Author

Owner

✅ Closed by `#1424`, landed `65df65e7`. Verified on main: 0 currency claims left in the guide.

One line changed. The other 51 descriptive versions are history and stay — which is #1423's distinction, applied before it was written down: a number naming a MOMENT IN THE PAST cannot go false; a number claiming CURRENCY can.

🔑 The arm was INVERTED rather than deleted, for the third time in that file and on its own precedent — #1382's note one arm below says why: "it now pins that prepare leaves those pins alone, which is the property that would otherwise regress silently."

🔴 And golangci-lint caught what go build could not: removing both setters left rewriteFile with no callers. go build rc=0, go vet rc=0, lint 1 issues: typecheck: 1. #1425 owns the follow-on; the function was left in place deliberately because removing it edits the CUT PATH.

## ✅ Closed by `#1424`, landed `65df65e7`. Verified on main: 0 currency claims left in the guide. **One line changed. The other 51 descriptive versions are history and stay** — which is `#1423`'s distinction, applied before it was written down: *a number naming a MOMENT IN THE PAST cannot go false; a number claiming CURRENCY can.* 🔑 **The arm was INVERTED rather than deleted, for the third time in that file and on its own precedent** — `#1382`'s note one arm below says why: *"it now pins that prepare leaves those pins alone, which is the property that would otherwise regress silently."* 🔴 **And `golangci-lint` caught what `go build` could not: removing both setters left `rewriteFile` with no callers. `go build` rc=0, `go vet` rc=0, lint `1 issues: typecheck: 1`.** *`#1425` owns the follow-on; the function was left in place deliberately because removing it edits the CUT PATH.*

bosun closed this issue

2026-09-07 08:46:01 +02:00