[brettheap]

Lane: openreposhape-1 (openRepoShape-1)

A follow-up to #883, the v1.1.1 release commit 5017830, as lane 3 placed it. It changes docs/release-notes/v1.1.1.md only.

• RV883 round 5 found docs/release-notes/v1.1.1.md: the v1.1.1 release notes, for the issues landed so far #883 READY at 02095a54 (docs/release-notes/v1.1.1.md: the v1.1.1 release notes, for the issues landed so far #883 issuecomment-6250558821), and held two SHOULDs and five NITs for the next refresh.
• Brett Heap's word at 2026-10-18T14:09:01Z moved the family root's validate-family.py and bootstrap.py: with the work tree's .gitmodules gone, three remedies undo each other or refuse; a git mv'd checkout is never named; bootstrap says ok beside git's exit 128 #879 to v1.2.0, so docs/release-notes/v1.1.1.md: the v1.1.1 release notes, for the issues landed so far #883 landed at 02095a54 with no further refresh. The held points wait here, for after the tag (docs/release-notes/v1.1.1.md: the v1.1.1 release notes, for the issues landed so far #883 issuecomment-6250568065; docs/release-notes/v1.0.0.md: the v1.0.0 release notes, a draft for Brett to amend before he tags, saying what 1.0 contains and promises, how to pin it by commit, what changed on the way grouped by the five tracking areas, and every known limit as the input that triggers it and what the person sees (#472) #473 issuecomment-6250902576).

What

The seven held points:

1. shape-doctor.py: the the way in rows say a checkout's own contracts/repository-naming.yaml that does not load, or raises, is that checkout's, an n/a environment row, where a tab line was FINDING the way in refused and a pattern that does not compile ended the doctor in a traceback; and the EXIT CODES list says what INVALID and manifest-kinds name (#707) #778's leg-clone sentence. It now says what runs at v1.1.0 and 5017830 show:
   • a leg clone whose own AGENTS.md says so keeps its leg row and its Triad row over each fault;
   • at v1.1.0, over a tab line its leg row was FINDING the way in refused, and over a pattern that does not compile, or a policy that raised on its name, the doctor ended in a traceback, exit 1;
   • a leg known only by its name is read through that policy. Over a tab line or a pattern that does not compile, it gets a repository's two n/a rows and a Triad row that calls it a single repository.
2. "Is", not "stays", exit 2. In the pattern case the run was exit 1 in v1.1.0.
3. update-shape.py: the git show <target>:<source> a remedy prints is shell-quoted, so a source path with a blank or a ' pastes; the missing-name refusal says the validators pass only where none of them meets a line; and its readers skip an if TYPE_CHECKING:, if False: or if 0: body and a statement after a raise, as Python does (#675) #802's clause is dropped. v1.1.0's _blocks_inside already read a try's handlers, else and finally outside its guard; update-shape.py: the git show <target>:<source> a remedy prints is shell-quoted, so a source path with a blank or a ' pastes; the missing-name refusal says the validators pass only where none of them meets a line; and its readers skip an if TYPE_CHECKING:, if False: or if 0: body and a statement after a raise, as Python does (#675) #802 only added test rows.
4. "What 1.1.1 fixes" names "or cannot be used", and the other way in.
5. bootstrap.py: a validator's exit 2 ends bootstrap REFUSED, exit 2, an unreadable upstream path, cache or pin is no answer and a reftable leg's cherry-pick or revert is read from its ref store, so no traceback or ended pick follows (#764) #815's leftover tracebacks were measured on Python 3.12 (its runs used 3.12.3). 3.13 is read, not run.
6. family.py, adopt-project.py and the four tools' --org: init asks GitHub's limits on the topic, the --org and the holder's name before any repository exists, check finds what execute refuses from one list, and an --org GitHub gives no organisation (., a trailing hyphen, --) is refused where GitHub is asked (#863) #871 lets _ through for any owner, an organisation's name included (GITHUB_OWNER_CHARACTERS in scripts/shape_materialize.py).
7. tests: six skips on a failed git call ask the system first, so a failed git call fails the test and an encoding that cannot spell U+202E skips the bidi test (#853, #854) #869 asks the disk only about a name that is not UTF-8.
8. update-shape.py: the git show <target>:<source> a remedy prints is shell-quoted, so a source path with a blank or a ' pastes; the missing-name refusal says the validators pass only where none of them meets a line; and its readers skip an if TYPE_CHECKING:, if False: or if 0: body and a statement after a raise, as Python does (#675) #802's readers. At v1.1.0 the import reader already skipped an if TYPE_CHECKING: body, and the binding reader did not. The sentence now says the readers skip if False:, if 0: and if None: bodies and a statement after an unconditional raise, and the binding reader an if TYPE_CHECKING: body too.
9. A known limit names openRepoShape, setup.sh and install.sh: under the C locale a $OPENREPOSHAPE_REPO URL or path, or a ref, with a letter past ASCII is refused as "a control character", so --version, --install and --preflight refuse what v1.1.0 ran #940, at lane 3's ask (docs/release-notes/v1.0.0.md: the v1.0.0 release notes, a draft for Brett to amend before he tags, saying what 1.0 contains and promises, how to pin it by commit, what changed on the way grouped by the five tracking areas, and every known limit as the input that triggers it and what the person sees (#472) #473 issuecomment-6252055476). Under the C or POSIX locale, the bash openRepoShape and setup.sh's two guards refuse as "a control character" a URL, path or ref with a letter past ASCII. That is a regression from v1.1.0 in the launcher, through openRepoShape: $OPENREPOSHAPE_REPO and $OPENREPOSHAPE_REF are checked once, before --version answers and before any fetch, so a control character or a character a terminal draws as another in either is refused, never printed or sent #872 (openRepoShape: --version and two fetch refusals print $OPENREPOSHAPE_REPO and $OPENREPOSHAPE_REF with no check, so an escape or a bidi override in either reaches the terminal as typed, and a scaffold run sends its fetch with them #866), and a UTF-8 locale takes such a URL or path. It is tracked as openRepoShape, setup.sh and install.sh: under the C locale a $OPENREPOSHAPE_REPO URL or path, or a ref, with a letter past ASCII is refused as "a control character", so --version, --install and --preflight refuse what v1.1.0 ran #940, for v1.2.0.

Also: the draft-time lines in the sections marked "remove at tag time" now say main 2d372bb, where #778 landed file for file as its reviewed head 8259ea1. Those are the header comment, "Not landed at drafting time", the open list's header and the shape-copy range.

• "Not landed at drafting time" names docs/release-notes/v1.1.1.md: the v1.1.1 release notes, for the issues landed so far #883, which landed as the release commit 5017830.
• The the family root's validate-family.py and bootstrap.py: with the work tree's .gitmodules gone, three remedies undo each other or refuse; a git mv'd checkout is never named; bootstrap says ok beside git's exit 128 #879 item says it moved to v1.2.0 at 2026-10-18T14:10:11Z.
• The release body never carries those sections.

Review

Lane 1's independent reviewer r893 (opus), who is not the writer (#893 issuecomment-6251576393):

• NOT READY at b795ef74.
  • One MUST: the leg-clone sentence was false for a leg known only by its name, for the Triad row over a tab line, and for the raising policy.
  • Two NITs: the readers' sentence (:430) and the draft lines.
• READY at ca49f125. Each changed sentence was checked against runs at v1.1.0 and 5017830, and there is no new finding. The writer narrowed the reviewer's wording for :430, which the runs bore out.
• 6cdb5551 adds the openRepoShape, setup.sh and install.sh: under the C locale a $OPENREPOSHAPE_REPO URL or path, or a ref, with a letter past ASCII is refused as "a control character", so --version, --install and --preflight refuse what v1.1.0 ran #940 limit (item 9). Lane 1's reviewer r893b (sonnet), not the writer, found it READY (docs/release-notes/v1.1.1.md: the points RV883 round 5 held for after the release commit, #778's leg-clone sentence and #802's readers sentence, each checked against runs at v1.1.0 and main (#883 follow-up) #893 issuecomment-6253065635). It checked each clause against openRepoShape, setup.sh and install.sh: under the C locale a $OPENREPOSHAPE_REPO URL or path, or a ref, with a letter past ASCII is refused as "a control character", so --version, --install and --preflight refuse what v1.1.0 ran #940's body and measured the launcher at 64464d31 and v1.1.0 under C, POSIX and C.UTF-8. Its NITs are left as written.

Checks

At ca49f125 and again at 6cdb5551:

• the three test files that read the release notes give 148 passed;
• the tag-time cut grep gives 0 hits;
• git diff --check is clean, and there is no character outside those v1.1.0's notes use.

At b795ef74, the file was byte-identical to the one checked at 97b9ab5d on 02095a54, and 5017830 holds the same notes file as 02095a54:

• the three test files that read the release notes (tests/test_repo_hygiene.py, tests/test_release_workflow.py, tests/test_workflow_concurrency.py) gave 148 passed;
• the tag-time simulation grep prints nothing;
• there is no character outside ASCII and the three v1.1.0's notes use;
• git diff --check is clean.

Milestone

v1.2.0, to land after the v1.1.1 tag, when main opens for v1.2.0 work. Lane 3, the release owner, may place it otherwise. It touches the v1.1.1 notes only, so it needs no back-port: the published release body is cut from the tag's copy.

🤖 Generated with Claude Code

Summary by Sourcery

Refresh the v1.1.1 release notes with post-release verification results, clarified behavior, and the newly documented #940 locale limitation.

Enhancements:

• Update the v1.1.1 release notes with verified behavior for leg clones, naming policies, readers, locale handling, encoding checks, Python-version coverage, and owner-name validation.
• Document the known non-ASCII locale regression tracked in openRepoShape, setup.sh and install.sh: under the C locale a $OPENREPOSHAPE_REPO URL or path, or a ref, with a letter past ASCII is refused as "a control character", so --version, --install and --preflight refuse what v1.1.0 ran #940 for v1.2.0.
• Refresh release-note draft metadata, milestone status, release-commit references, and tag-time cleanup sections.

Documentation:

• Revise the v1.1.1 release notes to reflect post-release verification and clarify the documented behavior and limitations of the release.

[sourcery-ai]

Reviewer's Guide

This documentation-only follow-up updates docs/release-notes/v1.1.1.md after the v1.1.1 release commit, incorporating seven held RV883 corrections and synchronizing draft-time provenance, milestone, and shape-copy references with main at 2d372bb; no production code or published release body is changed.

File-Level Changes

Change	Details	Files
Reconcile the release-note provenance and draft-time references with the post-release state of main.	Update the round-history header to record the held-point refresh at main 2d372bb and shape-doctor.py: the the way in rows say a checkout's own contracts/repository-naming.yaml that does not load, or raises, is that checkout's, an n/a environment row, where a tab line was FINDING the way in refused and a pattern that does not compile ended the doctor in a traceback; and the EXIT CODES list says what INVALID and manifest-kinds name (#707) #778’s reviewed head. Remove shape-doctor.py: the the way in rows say a checkout's own contracts/repository-naming.yaml that does not load, or raises, is that checkout's, an n/a environment row, where a tab line was FINDING the way in refused and a pattern that does not compile ended the doctor in a traceback; and the EXIT CODES list says what INVALID and manifest-kinds name (#707) #778 from the not-yet-landed list and update milestone and shape-copy references to 2d372bb. Keep tag-time placeholder sections consistent with the release-commit workflow.	docs/release-notes/v1.1.1.md
Incorporate the seven RV883 round-5 corrections into the v1.1.1 behavioral notes.	Clarify the leg-clone failure behavior and remove the inaccurate claim that the doctor offers the other way in. Correct the exit-status wording for pattern failures and retain exit 2 for the final verdict. Remove the misleading update-shape.py: the git show <target>:<source> a remedy prints is shell-quoted, so a source path with a blank or a ' pastes; the missing-name refusal says the validators pass only where none of them meets a line; and its readers skip an if TYPE_CHECKING:, if False: or if 0: body and a statement after a raise, as Python does (#675) #802 handler else/finally claim. Describe unavailable naming policies and the other way in behavior more precisely. Limit bootstrap.py: a validator's exit 2 ends bootstrap REFUSED, exit 2, an unreadable upstream path, cache or pin is no answer and a reftable leg's cherry-pick or revert is read from its ref store, so no traceback or ended pick follows (#764) #815’s traceback measurement to Python 3.12 and state that 3.13 was not run. Clarify that _ is accepted for any owner name and that disk probing applies only to non-UTF-8 names.	docs/release-notes/v1.1.1.md

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it. You can also reply to a
  review comment with @sourcery-ai issue to create an issue from it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time. You can also comment
  @sourcery-ai title on the pull request to (re-)generate the title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time exactly where you
  want it. You can also comment @sourcery-ai summary on the pull request to
  (re-)generate the summary at any time.
• Generate reviewer's guide: Comment @sourcery-ai guide on the pull
  request to (re-)generate the reviewer's guide at any time.
• Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
  pull request to resolve all Sourcery comments. Useful if you've already
  addressed all the comments and don't want to see them anymore.
• Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
  request to dismiss all existing Sourcery reviews. Especially useful if you
  want to start fresh with a new review - don't forget to comment
  @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[brettheap]

(lane openRepoShape-1, 2026-10-18T15:17:51Z) Independent review of #893, by lane 1's reviewer r893 (opus), who is not its writer: READY at ca49f125.

Round 1, at b795ef74: NOT READY.

• MUST: shape-doctor.py: the the way in rows say a checkout's own contracts/repository-naming.yaml that does not load, or raises, is that checkout's, an n/a environment row, where a tab line was FINDING the way in refused and a pattern that does not compile ended the doctor in a traceback; and the EXIT CODES list says what INVALID and manifest-kinds name (#707) #778's leg-clone sentence (:756-759) was false for part of what it covered. r893 ran shape-doctor.py --root <target> --json at v1.1.0 and 5017830 over the three faults shape-doctor.py: the the way in rows say a checkout's own contracts/repository-naming.yaml that does not load, or raises, is that checkout's, an n/a environment row, where a tab line was FINDING the way in refused and a pattern that does not compile ended the doctor in a traceback; and the EXIT CODES list says what INVALID and manifest-kinds name (#707) #778's tests use: a tab line, a pattern that does not compile, and a policy that raises.
  • Only a leg clone whose own AGENTS.md says so keeps both rows.
  • A clone known only by its name (Atlas-spec) gets two n/a "does not load" rows, and a Triad row saying it is a single repository.
  • Over a tab line, v1.1.0 kept the Triad row.
  • The raising policy's traceback came only where it raised on the clone's own name.
• NIT, :430. The readers' if TYPE_CHECKING: clause.
• NIT, the draft lines. "This pull request, which lands last" reads as docs/release-notes/v1.1.1.md: the points RV883 round 5 held for after the release commit, #778's leg-clone sentence and #802's readers sentence, each checked against runs at v1.1.0 and main (#883 follow-up) #893, and the the family root's validate-family.py and bootstrap.py: with the work tree's .gitmodules gone, three remedies undo each other or refuse; a git mv'd checkout is never named; bootstrap says ok beside git's exit 128 #879 move needed a time.
• Confirmed: points 2 to 7, shape-doctor.py: the the way in rows say a checkout's own contracts/repository-naming.yaml that does not load, or raises, is that checkout's, an n/a environment row, where a tab line was FINDING the way in refused and a pattern that does not compile ended the doctor in a traceback; and the EXIT CODES list says what INVALID and manifest-kinds name (#707) #778's squash 2d372bb equal to its reviewed head 8259ea1 file for file, every PR and issue number, and 148 passed.

Round 2, at ca49f125: READY, with no new finding.

• The leg-clone sentence now says what the runs show, at v1.1.0 and at 5017830. That includes the clone named Atlas, on which the raising policy does raise: KeyError: 'id', exit 1, at v1.1.0.
• :430. The writer found r893's suggested wording too broad, and the runs agree: at v1.1.0 the binding reader read every one of those bodies as bound, and only the import reader skipped if TYPE_CHECKING:. The sentence now says so, as update-shape.py: the git show <target>:<source> a remedy prints is shell-quoted, so a source path with a blank or a ' pastes; the missing-name refusal says the validators pass only where none of them meets a line; and its readers skip an if TYPE_CHECKING:, if False: or if 0: body and a statement after a raise, as Python does (#675) #802's squash does.
• The draft lines. docs/release-notes/v1.1.1.md: the v1.1.1 release notes, for the issues landed so far #883 merged as 5017830 at 14:13:11Z. the family root's validate-family.py and bootstrap.py: with the work tree's .gitmodules gone, three remedies undo each other or refuse; a git mv'd checkout is never named; bootstrap says ok beside git's exit 128 #879 moved to v1.2.0 at 14:10:11Z, by its timeline.
• Checks:
  • 148 passed;
  • the tag-time cut grep gives 0 hits;
  • no character outside those v1.1.0's notes use;
  • git diff --check is clean.

Lane: openreposhape-1 (openRepoShape-1)

[brettheap]

(lane openRepoShape-1, 2026-10-18T17:43:01Z) Independent re-check of #893 at 6cdb5551, by lane 1's reviewer r893b (sonnet), who is not its writer: READY, with no MUST and no SHOULD.

6cdb5551 adds one known limit naming #940, at lane 3's ask (#473 issuecomment-6252055476). r893b checked each clause against #940's body, and measured it with the bash launcher at 64464d31 and at v1.1.0 under C, POSIX and C.UTF-8, with a stub gh and curl:

• True:
  • the locales;
  • which scripts and checks;
  • install.sh's words in both locales;
  • the regression through openRepoShape: $OPENREPOSHAPE_REPO and $OPENREPOSHAPE_REF are checked once, before --version answers and before any fetch, so a control character or a character a terminal draws as another in either is refused, never printed or sent #872 (openRepoShape: --version and two fetch refusals print $OPENREPOSHAPE_REPO and $OPENREPOSHAPE_REF with no check, so an escape or a bidi override in either reaches the terminal as typed, and a scaffold run sends its fetch with them #866);
  • the four runs v1.1.0 ran: --version, --install from a file, --doctor beside a checkout, and --preflight through a real setup.sh;
  • who meets it;
  • "it refuses and does nothing else": with a logging git, setup.sh made one rev-parse and no clone;
  • the way around it, a UTF-8 locale for a URL or path, rightly not claimed for a ref;
  • the milestone.
• Placement: last in the launcher's group of "Known limits", in the section's form. It is the only change: 13 lines, 66 to 75 columns.
• Checks:
  • 148 passed;
  • the tag-time cut grep prints nothing;
  • no character outside those v1.1.0's notes use;
  • git diff --check is clean.
• NITs, left as written:
  • "It is a regression" could be scoped to "In the launcher": v1.1.0's setup.sh refused the same value.
  • A café ref is refused too.
  • The squash message should say a UTF-8 locale takes "a URL or path", not a ref. It will.
• Not false, but now incomplete, for lane 3 to weigh: "What 1.1.1 fixes" (:38-40), "What a script sees" (:207-208, where a line naming openRepoShape, setup.sh and install.sh: under the C locale a $OPENREPOSHAPE_REPO URL or path, or a ref, with a letter past ASCII is refused as "a control character", so --version, --install and --preflight refuse what v1.1.0 ran #940's four runs would help most) and :797-802.

Lane 1 asks for #893's turn after #881 on #473, gated over #881's predicted squash.

Lane: openreposhape-1 (openRepoShape-1)

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[sourcery-ai]

Sorry @brettheap, you've used your own review budget of 250,000 diff characters for the last 7 days.

You can request another review in 1 day and 6 hours by commenting @sourcery-ai review. Upgrade to get a review now.