Release 1.0.2 - #705
Merged
Merged
Conversation
Five fixes have been on main since v1.0.1 and none has reached anyone. Four of
them are about the installer and the hook, and that boundary has exactly one
delivery mechanism.
#696 the hook pinned the release it was installed from
#698 install.sh never installed the Codex plugin; install.ps1 did
#700 a failed Codex integration reported healthy and exited 0
#702 documents claiming more than the product does
#704 the PowerShell 5.1 floor documented but not enforced
#700 is the one a user meets first: registering Codex is two steps, and the
installer called the host healthy when the second failed. It said "plugin step
failed" in its detail and exited 0, because ok is computed from the field rather
than the sentence.
dist is unchanged and the canonical digest is identical to before the bump. The
version is read from package.json at runtime rather than compiled in. Rebuilt
through the canonical builder and verified rather than assumed, as with the two
releases before it.
Thirty-eight pins across nine files. The installers and READMEs carry the tag in
URLs that resolve only once the tag exists, so this is one commit and the tag
goes on it.
Limit: an installer-boundary fix reaches nobody until it is released
Blast: system
Undo: costly
Certainty: firm
Provenance: authored
Record-Id: r-rel102
CommitLore — record lintTrailers: clean — 2 commits in Trailer violations fail this check. Active constraints are informational — they are what the repository already decided, not a verdict on this PR. |
The README had install and no upgrade. Every release since v1.0.0 has been about that boundary -- the hook pinned to its release, the plugin cache the installer never touched, a Codex step one platform skipped -- and a user following the README had no reason to run doctor afterwards, which is where all of it surfaces. Two things a release cannot reach, said plainly: hooks already installed in a repository, because the installer cannot know which repositories have them, and sessions already running, because a host loads its runtime once. Neither is a defect and both look like one. English only. The three translations carry no operational guidance at this level and nothing asserts they must, so adding it to all four would be widening scope rather than finishing this. Limit: an upgrade path is documented where the install path is Blast: local Undo: easy Certainty: firm Provenance: authored Record-Id: r-rdupgr
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Five fixes have been on
mainsince v1.0.1 and none has reached anyone.Four are at the installer and hook boundary, and that boundary has exactly one
delivery mechanism. Measured at v1.0.1: re-installing with
main'sinstall.shchanged nothing, because the code being installed is the released binary.
#700 is the one a user meets first. Registering Codex is two steps, and the
installer called the host healthy when the second failed — it said plugin step
failed in its detail and exited 0, because
okis computed from the fieldrather than the sentence. That defect was introduced in #698 while fixing #697,
and found by the closeout review.
The bump
Thirty-eight pins across nine files.
distunchanged, canonical digestidentical —
ca608927. The version is read frompackage.jsonat runtime ratherthan compiled in; rebuilt through the canonical builder and verified rather than
assumed, as with the two releases before it.
The installers and READMEs carry the tag in URLs that resolve only once the tag
exists, so this is one commit and the tag goes on it.
After publishing
docs/RELEASE-GATE.md§6b and §6c apply: read the check runs at the commit themerge creates, and install over v1.0.1 rather than fresh. The specific
readbacks this release needs:
Baseline measured before the release, for comparison: CLI
1.0.1, hook pinnedat
…/v1.0.1/dist/commitlore.mjs,doctor12 ok / 6 warnings / 0 failed.Not in this release
#691 —
install.ps1wires hosts directly as well as delegating. Removing theduplicates needs a Windows machine with agents installed; CI cannot substitute,
because its runners have none and every host reports
notDetectedthere.