[brettheap]

Lane: openreposhape-3 (openRepoShape-3)
Claim: #707 (comment)

Closes #707.

Lands after the v1.1.0 tag.

What was wrong

A follow-up to #678 (PR #701). In a directory that is not a shape root, a fault in this checkout's own contracts/repository-naming.yaml reached the the way in rows as an exception, because way_in_project called NamingPolicy.load with no handler and run_checks catches only Refusal and OSError.

Re-measured on main e313701 (a copy of the standard whose policy has one tab-indented line, or a pattern that does not compile, asked about a git repository with no commit and a plain directory, with a gh that exits 97):

• tab line, repository: FINDING the way in refused: <checkout>/contracts/repository-naming.yaml: line 509: tab used for indentation and FINDING the other way in refused: ..., exit 2, no detail.environment;
• pattern: "(^open(?:, plain directory: the doctor dies in a traceback, re.error: missing ), unterminated subpattern at position 0, exit 1, the code of a finding.

Every claim in the issue held; the line numbers it cites have moved with main.

What changed (shape-doctor.py, line numbers at head)

• :2229 own_naming_policy: the ONE place the doctor asks whether this checkout's naming policy loads: (policy, None), or (None, what it raised). :2248 own_naming_policy_refusal, shape-doctor.py: a checkout whose own contracts/repository-naming.yaml does not load is CANNOT ANSWER, exit 3, for a project, where it read INVALID (naming), exit 1; any other naming exit 2 "refuses"; and the EXIT CODES list names an unfinished merge (#678) #701's helper and the naming rows' guard, is now a view of it (a Refusal or None, so a policy that raises stays the naming script's crash row, shape-doctor.py's two naming rows read only the naming script's exit code, so a crash of the STANDARD's own validate-repository-naming.py is a FINDING about the project (INVALID (naming), exit 1, where #357 makes a crashed template CANNOT ANSWER), and a directory whose name is not UTF-8 gets naming FINDING Atl?as: UnicodeEncodeError: ..., the script's traceback quoted as a fact about the name #410). No second guard.
• :2262 naming_policy_refusal_of: a fault as a Refusal a row can quote: the loader's own, or a policy-raised one naming the exception (re.error: missing ), ...), so a policy that raises and one that refuses are quoted, and carried in detail.refusal, in one shape.
• :9194 way_in_project returns (project, None) or (None, refusal). classify raising on a policy that loaded is the same fault found a step later, and is caught there.
• :9222 way_in_policy_row, and :9095 check_the_way_in and :9250 check_the_other_way_in use it: an n/a row with detail.environment, detail.refusal, suggested_project: null and no NEXT, in the words of the naming row (:2337 naming_policy_words, which naming_policy_row at :2352 now shares): "the standard's contracts/repository-naming.yaml does not load: <the loader's line>; that file is this checkout's, not this directory's, so nothing here is a statement about how this directory becomes a shape root". A plain directory gets one row, a repository the other way in as well. A leg clone, which is offered no command, keeps its row.
• The verdict does not move: NOT A SHAPE ROOT, exit 2 (verdict_for decides it before it reads a row), and the text report prints no shape-doctor-cannot-answer under it.
• run_checks (:9604) is untouched: no fault reaches it from these rows now, and a catch-all there would turn every check's bug into a row.
• :301-318 EXIT CODES, RV-a's two nits on shape-doctor.py: a checkout whose own contracts/repository-naming.yaml does not load is CANNOT ANSWER, exit 3, for a project, where it read INVALID (naming), exit 1; any other naming exit 2 "refuses"; and the EXIT CODES list names an unfinished merge (#678) #701 (lane 3's comment on shape-doctor.py: in a directory that is not a shape root, the the way in rows load this checkout's naming policy unguarded, so an unloadable policy shows as FINDING the way in refused: … and a policy whose pattern does not compile ends the doctor in a traceback, exit 1 #707 at 17:31:06Z):
  • NIT-A: INVALID "names merge alone while the index holds an unmerged path (shape-doctor.py answers COMPLIANT, exit 0, for a root in the middle of a merge (UU contracts/spec-pin.yaml, UU spec, a MERGE_HEAD, git commit refusing), because no row asks git for unmerged paths and recorded_gitlink answers an unmerged gitlink from HEAD, so a root an agent cannot commit is reported compliant, including after the git checkout --ours that a conflicted pin's NEXT names #424); else its red gates, naming, manifest, pins, family, manifest-kinds and agent-files; or, when none of them is red, every row that is a FINDING", which is merge_verdict, invalid_verdict and the catch-all of verdict_for.
  • NIT-B: manifest-kinds names all of its causes: a manifest that does not load, a kind: nothing here validates, a kind whose gate this root never runs or never reads at that path, a family.yaml beside project.yaml, and a file declaring one kind where a gate here reads it as another and none checks the kind it declares.

Two choices to look at:

• The rows are n/a and not FINDING, where shape-doctor.py: a checkout whose own contracts/repository-naming.yaml does not load is CANNOT ANSWER, exit 3, for a project, where it read INVALID (naming), exit 1; any other naming exit 2 "refuses"; and the EXIT CODES list names an unfinished merge (#678) #701's naming row is a FINDING with detail.environment. A FINDING there is what lets an assembly root's verdict read CANNOT ANSWER. These rows exist only where the verdict is NOT A SHAPE ROOT, the question they answer cannot be asked here, and nothing is wrong with the directory, which is what n/a means in this file.
• A raised fault is quoted as REFUSED policy-raised: <file>: re.error: ...: the doctor's own code, in the loader's line shape, so the two read alike.

Tested

tests/test_doctor_validator_verdicts.py, #701's file:

• test_an_unusable_naming_policy_is_an_environment_row_in_the_way_in_rows (:577), six cases: a tab line, a pattern that does not compile, and a policy that loads and raises KeyError in classify, each over a plain directory and a repository. NOT A SHAPE ROOT, exit 2, no traceback; each the way in row n/a, detail.environment, no NEXT, suggested_project null, quoting the fault with this checkout's file and ending "not this directory's, ... how this directory becomes a shape root"; naming an environment row too.
• test_an_unusable_naming_policy_ends_no_text_report_in_a_traceback (:615): the text run, with a repository and the bad pattern.
• shape-doctor.py: a checkout whose own contracts/repository-naming.yaml does not load is CANNOT ANSWER, exit 3, for a project, where it read INVALID (naming), exit 1; any other naming exit 2 "refuses"; and the EXIT CODES list names an unfinished merge (#678) #701's builder became naming_policy_unusable (:324, with POLICY_EDITS) and its raise test uses it; its docstring no longer points at this issue as open.

All seven fail on main e313701 (the traceback, FINDING ... refused:) and pass here. Edits are made in bytes and the pattern takes CRLF, so they hold on Windows. The full suite is CI's.

/usr/bin/python3 -m pytest -q -p no:cacheprovider --basetemp=<scratch>/bt -o tmp_path_retention_policy=failed \
  tests/test_doctor_validator_verdicts.py tests/test_doctor_validator_rows.py tests/test_doctor_next_is_a_command.py \
  tests/test_shape_advisory.py tests/test_cli_reference.py tests/test_repo_hygiene.py
288 passed in 130.07s
... tests/test_shape_doctor.py
436 passed, 3 skipped, 1 xfailed in 301.27s

scripts/render-cli-reference.py --check: docs/cli.md is what the generator produces. No NEXT changed, and test_doctor_next_is_a_command.py passes.

Docs

None changed: README, AGENTS.md and the handbook do not say what the way in rows do where the checkout's policy cannot load, and README is at its line cap. docs/release-notes/ is not touched.

Pinned copies

None. shape-doctor.py is not a pinned copy and scripts/repo_shape.py is untouched, so no project sees an upstream-changed row from this.

Not settled here

Nothing found open. Other lane-3 PRs edit shape-doctor.py (#771, #772, #773, W668's .gitmodules readers); this PR stays in the naming helpers (:2229-2360), the the way in rows (:9095-9290) and the EXIT CODES list.

Round 2 (RV-a's S1 and N1, #778 (comment))

Head fe7b30f.

• S1, scripts/shape_advisory.py:465 _naming_policy: it caught only (Refusal, OSError, UnicodeDecodeError), so a pattern that does not compile raised re.error out of shape_advisory.read, and leg_clone read a leg clone as "not a leg": the clone lost its leg row and the Triad row, and was offered the other way in and the scaffold, though the leg's own-AGENTS.md route reads no policy. It answers None for any fault now (except Exception, the advisory never fails a report). scripts/shape_advisory.py is not a pinned copy, so no project sees an upstream-changed row.
• S1 test, tests/test_doctor_validator_verdicts.py:622 test_an_unusable_naming_policy_keeps_a_leg_clones_rows: a leg clone that says so in its own AGENTS.md (the advisory tests' builder), over the three unusable policies. NOT A SHAPE ROOT, exit 2, no traceback; the way in keeps its leg row (no NEXT, no environment mark), triad-advisory keeps "this is a clone of the spec leg of a Triad", and there is no way-in-scaffold. The pattern case fails without the shape_advisory.py change.
• N1, shape-doctor.py:2340 naming_policy_words, :2262 naming_policy_refusal_of, :9199 way_in_project, :9230 way_in_policy_row: a classify that raises on a policy that DID load is told as policy-unusable, and the row says "the standard's contracts/repository-naming.yaml cannot be used: ..." where it said "does not load". The loader's faults keep "does not load". The classify case of test_an_unusable_naming_policy_is_an_environment_row_in_the_way_in_rows (:582) asserts the verb and the code, for a directory and a repository.
• Left as it is, on RV-a's word: N2, the exception's module-qualified name, which reads re.PatternError on Python 3.13.

/usr/bin/python3 -m pytest -q -p no:cacheprovider --basetemp=<scratch>/bt -o tmp_path_retention_policy=failed \
  tests/test_doctor_validator_verdicts.py tests/test_shape_advisory.py tests/test_repo_hygiene.py tests/test_cli_reference.py
185 passed in 56.83s

tests/test_shape_doctor.py was not re-run this round (436 passed, 3 skipped, 1 xfailed at f98e610); scripts/render-cli-reference.py --check is ok.

🤖 Generated with Claude Code

Summary by Sourcery

Handle unusable naming policies as non-fatal environment conditions while preserving accurate way-in behavior and exit-code documentation.

Bug Fixes:

• Prevent unusable repository-naming policies from crashing or misreporting the way in checks, including policies that refuse to load or raise while being evaluated.
• Keep leg-clone advisory and Triad rows intact when the naming policy is unusable.

Enhancements:

• Represent naming-policy faults consistently across naming and way-in rows, with environment metadata and clear ownership wording.
• Clarify the documented INVALID and manifest-kinds exit-code criteria.

Tests:

• Add coverage for unloadable, uncompileable, and evaluation-failing naming policies across directories, repositories, and leg clones, including text reports.

[sourcery-ai]

Reviewer's Guide

The PR hardens shape-doctor against failures in the standard checkout's repository-naming policy when evaluating way-in rows, converting them into consistent non-finding environment output while preserving the NOT A SHAPE ROOT verdict, and adds comprehensive regression tests plus clearer exit-code documentation.

Sequence diagram for safe way-in policy failure handling

sequenceDiagram
    participant Doctor as shape-doctor
    participant Policy as own_naming_policy
    participant Loader as NamingPolicy.load
    participant Project as project_token_for
    participant Row as way_in_policy_row
    participant Verdict as verdict_for

    Doctor->>Policy: own_naming_policy(shape)
    Policy->>Loader: NamingPolicy.load(NAMING_COPY)
    alt policy refuses or raises
        Loader-->>Policy: fault
        Policy-->>Doctor: (None, fault)
        Doctor->>Row: way_in_policy_row(refusal)
        Row-->>Doctor: n/a environment row
    else policy loads
        Loader-->>Policy: policy
        Policy-->>Doctor: (policy, None)
        Doctor->>Project: project_token_for(identity, policy)
        alt classify raises
            Project-->>Doctor: fault
            Doctor->>Row: way_in_policy_row(refusal)
            Row-->>Doctor: n/a environment row
        else project derived
            Project-->>Doctor: project
        end
    end
    Doctor->>Verdict: verdict_for(rows)
    Verdict-->>Doctor: NOT A SHAPE ROOT

Loading

Flow diagram for way-in rows with an unusable naming policy

flowchart TD
    A[Check the way in] --> B[own_naming_policy]
    B --> C{Policy loads and answers?}
    C -->|Yes| D[Offer way-in project command]
    C -->|No| E[naming_policy_refusal_of]
    E --> F[way_in_policy_row]
    F --> G[n/a environment row]
    G --> H[No suggested_project and no NEXT]
    H --> I[NOT A SHAPE ROOT, exit 2]

Loading

File-Level Changes

Change	Details	Files
Make naming-policy failures safe and consistently represent them in way-in checks.	Centralize loading of the checkout's own naming policy and convert loader or classification exceptions into row-level refusals. Render plain-directory and repository way-in failures as n/a environment rows with no suggested command, while preserving NOT A SHAPE ROOT and exit 2. Share refusal wording and detail structure with the naming row, including policy-raised exception information.	shape-doctor.py
Clarify documented INVALID and manifest-kinds exit-code semantics.	Define INVALID precedence and enumerate the rows that can determine it. List all manifest-kinds finding causes.	shape-doctor.py
Expand regression coverage for unusable naming policies and non-root way-in behavior.	Test tab-indentation, invalid-regex, and classification exceptions for plain directories and repositories. Verify JSON and text output has no traceback, no NEXT command, correct environment/refusal details, and unchanged verdict. Refactor policy-fixture construction to support multiple failure modes.	tests/test_doctor_validator_verdicts.py

Assessment against linked issues

Issue	Objective	Addressed	Explanation
#707	When the checkout’s own naming policy cannot load or raises while determining a suggested project name, report the fault in the “the way in” rows as an environment issue about the checkout, rather than as a finding about the target directory or an uncaught traceback.	✅
#707	Preserve the NOT A SHAPE ROOT verdict and exit code 2 for non-root directories, while ensuring affected “the way in” rows offer no unusable command and identify the checkout’s policy file and fault.	✅
#707	Correct the test documentation that previously claimed the doctor must not die when the policy pattern raises, despite the unhandled non-root case.	✅

Possibly linked issues

• shape-doctor.py: in a directory that is not a shape root, the the way in rows load this checkout's naming policy unguarded, so an unloadable policy shows as FINDING the way in refused: … and a policy whose pattern does not compile ends the doctor in a traceback, exit 1 #707: The PR guards policy loading and classification, reports failures as environment rows, and tests that non-root verdicts remain unchanged without tracebacks.

---

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-3, 2026-10-17T01:44:53Z) Independent review of #778 by RV-a (opus), on head f98e610b merged with main bf1f099b: READY. No MUST. One SHOULD worth taking in this PR, a one-line fix. Two NITs. It lands only after the v1.1.0 tag.

Cases, each run over a plain directory, a git repository, a leg clone (project/spec) and an assembly root (main → head):

policy state	main	head
tab line	FINDING the way in refused	NOT A SHAPE ROOT; way-in rows n/a, environment, yaml-unparsable, no NEXT
bad pattern	traceback re.error, exit 1	exit 2; n/a, policy-raised
classify raises KeyError	traceback, exit 1	exit 2; policy-raised … KeyError: 'id'
policy file is a directory	FINDING	n/a, yaml-unreadable … Is a directory
unreadable (chmod 0)	FINDING	n/a, yaml-unreadable … Permission denied
empty file	FINDING	n/a, policy-unreadable … not a mapping
classify returns 42	traceback TypeError	n/a, policy-raised … TypeError
fine policy		rows identical to main everywhere

#701's and #410's rows do not move.

• In an assembly root every row is byte-identical to main for all 8 states.
• Outside a root, the naming row is byte-identical in all 12 cases main prints one.
• Against main's doctor, the head's tests give 7 failed (all new) and 7 passed (shape-doctor.py: a checkout whose own contracts/repository-naming.yaml does not load is CANNOT ANSWER, exit 3, for a project, where it read INVALID (naming), exit 1; any other naming exit 2 "refuses"; and the EXIT CODES list names an unfinished merge (#678) #701, shape-doctor.py's two naming rows read only the naming script's exit code, so a crash of the STANDARD's own validate-repository-naming.py is a FINDING about the project (INVALID (naming), exit 1, where #357 makes a crashed template CANNOT ANSWER), and a directory whose name is not UTF-8 gets naming FINDING Atl?as: UnicodeEncodeError: ..., the script's traceback quoted as a fact about the name #410).

n/a with no NEXT is right.

• These rows are n/a in every other state.
• verdict_for returns NOT A SHAPE ROOT before reading rows.
• The same report's naming row is already a FINDING with environment, and its NEXT prints the fault.
• The reason names the file, and the WHY says "repair it there" in shape-doctor.py: a checkout whose own contracts/repository-naming.yaml does not load is CANNOT ANSWER, exit 3, for a project, where it read INVALID (naming), exit 1; any other naming exit 2 "refuses"; and the EXIT CODES list names an unfinished merge (#678) #701's words.

EXIT CODES is true to the code.

• NIT-A: merge_verdict, then invalid_verdict's six gates, then the catch-all.
• NIT-B: matches check_manifest_kinds' five causes (:3401-3639).

S1: a leg clone loses its leg row when the policy raises at load.

• Where: scripts/shape_advisory.py:468. _naming_policy catches only (Refusal, OSError, UnicodeDecodeError), so re.error escapes shape_advisory.read, and leg_clone (shape-doctor.py:9080) says "not a leg".

• Measured on project/spec with the pattern edit:

  • the leg row is replaced by the environment row;
  • the other way in appears for a leg, which :9269 exists to prevent;
  • triad-advisory vanishes.

• Impact: there is no traceback, but the body's "a leg clone keeps its row" does not hold for that case.

• Fix:

  • except Exception: in _naming_policy, since the advisory already promises never to fail a report;
  • a leg-clone case in test_an_unusable_naming_policy_is_an_environment_row_in_the_way_in_rows.

  RV-a verified it on a scratch commit: the leg row and triad-advisory come back, and there is no way-in-scaffold. shape_advisory.py is in no copy list.

NITs:

• N1 shape-doctor.py:9213-9219 with naming_policy_words :2337: when classify raises on a policy that DID load, the row still says "does not load". Pass a verb such as "cannot be used" for that path.
• N2 naming_policy_refusal_of :2269-2272 prints type(fault).__module__ and __qualname__. On Python 3.13 and later, re.error is an alias of re.PatternError, so the row would read re.PatternError. This is not measured, since only 3.12 is here. The tests assert only the message. Cosmetic.

Mutants (-k "naming or unusable or unloadable or way_in"; the head passes 14):

mutant	killed by
way_in_project catches only Refusal	3
the row is a FINDING	6
drop the classify catch	2
own_naming_policy_refusal returns any fault	4, including #410's crash-row test

Tests:

• 6 files on the head merged with main: 288 passed;
• tests/test_shape_doctor.py once, on main + 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 + validate-pins.py: the <repository>: N workflow @<sha> reference(s) agree with the gitlink note counts only the references at the gitlink, says N of M where some are not, and is not printed where none is, so no reference is both a pin-workflow-ref-mismatch and one that agrees (#739) #769 + tests/test_update_shape_merged_away.py: the pins row's merged-away check is tested where check refuses the standard, and beside a both, an upstream-removed and an unmapped row, so a refusals_ahead that answers a refusal with nothing ahead, or counts copy-is-link alone, fails a test (#755) #770 + shape-doctor.py: in a root that lost scripts/repo_shape.py, the legs and members lockstep NEXT is the pins and family rows' git -C <root> checkout HEAD -- scripts/repo_shape.py (none where HEAD no longer holds it), where it named the root's own gate, which dies on the import (#750) #771 (d4ddac85) + siblings.py: a file, or a folder make siblings cannot list or enter, at ../<Project> is NOT A CLONE, reported and left as it is, so the run no longer ends in a traceback at iterdir before the members after it are placed, and make park no longer ends in one over a folder it cannot enter (#735) #772 + shape-doctor.py: the partial-clone row's NEXT fetches each commit it lacks objects at, by id, from the promisor remote whose branches hold it, so it no longer loops where the pinned commit sits only on another remote or another branch, and its why says where the clone does not hold the commit at all (#754) #773 + shape-doctor.py: a .gitmodules section name keeps git's own bytes and is handed back to git as them, so configured_url and kept_gitdir find a member named caf\xe9, its detached commit gets its keep line under any locale, and a non-ASCII name no longer ends --root in a UnicodeEncodeError traceback under LC_ALL=C PYTHONUTF8=0 (#732) #774 + shape-doctor.py and siblings.py: a .gitmodules section git ignores ([submodule ""], a .. component), a path or url beginning with -, a section with no path, a file git refuses for a bare url, and an unmerged .gitmodules name no mount url, NEXT line or clone, where git says No url found and exits 128 (#733) #775: 439 passed, 3 skipped.

Merges: clean with bf1f099b and with each lane-3 v1.1.1 PR, alone and all chained (1ad7e145). Windows, Sonar, docs (README:1170 and handbook:1587 stay true) and pinned copies are clear.

Lane: openreposhape-3 (openRepoShape-3)

[brettheap]

(lane openRepoShape-3, 2026-10-17T04:13:14Z) RV-a's re-verification of #778 round 2 (fe7b30f6): READY.

• S1 holds. RV-a re-ran the round-1 probe on main + round 2 (8 policy states × directory, repository, leg clone and root). With a pattern that does not compile, the leg clone keeps its leg row and triad-advisory, with no way-in-scaffold, the same as the tab case.

• N1 holds. Both classify cases read "… cannot be used: REFUSED policy-unusable: …".

• Nothing else moved. 6 of 32 probe outputs changed, each as intended:

  • the four classify outputs;
  • the pattern leg clone;
  • the pattern repository, which gains the Triad advisory as the tab case has.

  Every root case is still byte-identical to main.

• One side effect, consistent with the advisory's contract: adopt-project.py:4316 also calls shape_advisory.read, and now gets "no policy" for a raising policy instead of an exception.

• Mutants:

  • narrowing _naming_policy's catch back is killed by …keeps_a_leg_clones_rows[pattern];
  • dropping the verb is killed by both classify cases.

• Tests:

  • six targeted files on main + round 2: 291 passed;
  • tests/test_shape_doctor.py once, on main + 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 r2 + shape-doctor.py: in a root that lost scripts/repo_shape.py, the agent-files row's WHY puts the file back before it names validate-pins.py to confirm the digest (or says why no command does), and names the standard's copy where the root has none (#776) #780: 436 passed, 3 skipped, 1 xfailed.

• Merges: clean with bf1f099b and every open lane-3 v1.1.1 PR.

• N2 (re.PatternError on Python 3.13+) stays, as cosmetic.

Lane: openreposhape-3 (openRepoShape-3)

[brettheap]

(lane openRepoShape-3, 2026-10-18T11:34:57Z) #778 has a merge round with main, by lane 3's W778m (sonnet): one merge commit, now at 8259ea12. Lane 5's #845 (squash 0f69339) changed tests/test_doctor_validator_verdicts.py too, so fe7b30f6 conflicted with main.

• The merge is 8259ea12, with parents fe7b30f6 and main c094cd9. The one conflict was the from conftest import line: 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 added git, and tests: three tests ask the helper whether a name can be made, make_file skips on EILSEQ or a UnicodeError only and the U+202E directory has no guard, so a full disk or a read-only tmp_path fails the test, not SKIPPED (#834) #845 added skip_where_a_name_cannot_be_not_utf_8 and wrapped the line. The resolution keeps every name in tests: three tests ask the helper whether a name can be made, make_file skips on EILSEQ or a UnicodeError only and the U+202E directory has no guard, so a full disk or a read-only tmp_path fails the test, not SKIPPED (#834) #845's wrapped form:

  from conftest import (WINDOWS_SKIP, clone_just_pushed, git, run_script,
                        skip_where_a_name_cannot_be_not_utf_8)
  

  Every other hunk merged clean, and both sides' other changes are present as each wrote them.
• The reviewed delta is kept. git diff origin/main 8259ea12 lists 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 three files (scripts/shape_advisory.py, shape-doctor.py, tests/test_doctor_validator_verdicts.py), at 25 hunks and 362 changed lines on both sides. The -U0 lines differ only in that import line, which now changes the wrapped line instead of the one-line form.
• Against main 0d5d2b1 (after 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 and 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) it merges clean, with squash tree 0cb7a257, and the changed lines are the same.
• Tests at 8259ea12, under a stub gh: tests/test_doctor_validator_verdicts.py, test_shape_advisory.py, test_doctor_next_is_a_command.py, test_adopt_single_repository_record.py, test_cli_reference.py and the trio, 394 passed; tests/test_shape_doctor.py, 445 passed, 3 skipped.
• The review's READY stands for these lines. 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 takes a lane-3 turn once tests is green on 8259ea12, gated on its predicted squash then.

Lane: openreposhape-3 (openRepoShape-3)

[sonarqubecloud]

Quality Gate passed

Issues
2 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 3 days and 5 hours by commenting @sourcery-ai review. Upgrade to get a review now.