family.py and scaffold-project.py: `add --local-remote-dir` refuses a member clone URL `.gitmodules` cannot hold as text, `--dry-run` alike, each clone-URL remedy names a link's target and a lone surrogate, and the READMEs quote the clone URL, so the line under `a b/` pastes as one command (#693) by brettheap · Pull Request #812 · opensoft/openRepoShape

brettheap · 2026-10-17T11:10:48Z

Lane: openreposhape-4 (openRepoShape-4)
Claim: #693 (comment)

What changed and why

#640 (#433) asked shape_materialize.unwritable_reason of the clone URL that the scaffold and family.py init write into their READMEs. This PR asks the same question of family.py add and fixes what the remedy and the README line still got wrong. All four of #693's lines were re-measured on main 8062810d, with local bare remotes, a gh shim and GIT_ALLOW_PROTOCOL=file:

line	on main	now
R433 O2: `add --local-remote-dir <dir>/caf<e9>`	exit 0, both dry run and real; `.gitmodules` holds the raw `0xE9`; the holder's `siblings.submodule_urls` reads it back as `caf\ufffd`. A line break was written as `\n` and read back cut off at the break; an escape and U+202E were committed raw	exit 2, `family-unwritable-value`, both dry run and real, before `_probe_member`'s throwaway clone; the holder is byte for byte as it was
R433 O3: `--local-remote-dir link`, where `link` is a UTF-8 name pointing at a `caf<e9>` directory	refused, naming the resolved `caf\udce9` URL; the remedy says only "Pass a --local-remote-dir whose path is one line of UTF-8 text"	every clone-URL remedy now says the URL is the resolved path, and that a link's target is then at fault
R433 O4: a lone surrogate (Windows)	`unwritable_reason` names it; no clone-URL remedy does	both scaffold and family remedies now say "or a lone surrogate", as #282's `UNWRITABLE_REMEDIATION` does
R433 O1: README clone line under `a b/` or `it's/`	`git clone --recurse-submodules <P>/a b/remotes/Atlas.git`, written raw into the assembly root's README (twice), each leg's README and the holder's README	`'<P>/a b/remotes/Atlas.git'` and `'<P>/it'\''s/remotes/Atlas.git'`

add (scripts/family.py). cmd_add calls the new _refuse_an_unwritable_member_url(local, url) on _member_url's result. The call comes after the branch question and before the member is read, so --dry-run refuses the same way. It uses init's question and init's refusal code. The refusal now reads:

REFUSED family-unwritable-value: under --local-remote-dir the member's clone URL is the bare repository's own path, '…/caf\udce9/IRQS.git', which `add` writes into the holder's `.gitmodules`, and it holds a byte that is not UTF-8. No git command that writes was run and nothing was written.
Remediation: `--dry-run` refuses the same. Pass a --local-remote-dir whose path is one line of UTF-8 text with no control character in it and that holds the member's bare repository `.git`, its directory renamed or the repository moved, and run again; or run `add` against GitHub, without --local-remote-dir, where the clone URL is `https://github.com//.git`. The clone URL is the path --local-remote-dir resolves to, so where the path given is text and the URL named above is not, the fault is in what it leads through: a symbolic link on it whose target is not text, or, for a relative path, the directory it was given from. Rename or replace the directory whose name is not text, or pass a path that does not lead through it; renaming a link changes nothing. The holder's `.gitmodules` records the clone URL as git is given it, and the holder's `make siblings` reads it back as UTF-8 text, a line a value: a byte that is not UTF-8 in the path (a directory named in a legacy code page), or a lone surrogate, reads back as another path, a line break ends the URL where it falls, a control character garbles the line it is shown in, and a bidirectional embedding, override or isolate control shows the rest of that line in another order than it is written.

Remedies (LOCAL_REMOTE_DIR_REMEDY, shared by init and add, and the scaffold's CLONE_URL_REMEDIATION). They gain the resolved-path sentence and the lone surrogate. What the directory must hold is each caller's: init and the scaffold make the bare repository there, so they keep "a directory renamed or made anew", byte for byte; add reads the member's <Project>.git from it, so a new, empty directory is refused again (repository ... does not exist), and its remedy says the path "holds the member's bare repository <Project>.git, its directory renamed or the repository moved" (LOCAL_REMOTE_DIR_REMEDY's {holding}; lane 1's review of this PR measured the second refusal). Each README remedy now says the README prints the URL "in a git clone line, in quotes where a shell needs them, every character but an apostrophe written as itself", in place of "as it stands".

README clone line. family.py's _template_values and scaffold-project.py's values table both render CLONE_URL (and ASSEMBLY_CLONE_URL) through shell_quoted(url, "posix"), as the issue directs.

Tests

New file: tests/test_local_remote_dir_urls.py, 27 tests, on the harnesses of test_clone_url_bytes.py and test_family_add_refusals.py, importing their helpers rather than copying them.
- add, four BAD_DIRECTORIES rows × {dry run, real}: exit 2, empty stdout, stderr in full, assert_unchanged on the holder, no throwaway clone, no gh call.
- add through a link, both runs; a café control that still mounts.
- add's own remedy followed: from the refused caf<e9> directory, renaming it, or moving the bare repository into a directory of text, each makes add exit 0 with the member mounted.
- The scaffold and init through a link, both runs, with no process started.
- The lone surrogate (\ud800) asked of each of the three refusal functions.
- README rows a b, it's and an ordinary plain (left bare on POSIX, quoted on Windows) for the scaffold's three READMEs and the holder's. Each line is checked with shlex.split, and the holder's is pasted into sh, where it clones.
Existing file: tests/test_clone_url_bytes.py spells the two remedies in full, so its fixed text gains the new sentences, and its QUOTED_NOT_ESCAPED constant, whose words were false for the it's row, is QUOTED_ONE_WORD. Its two café README assertions now expect the quotes.
Targeted runs, no GIT_ALLOW_PROTOCOL.
- On the merge of main bf1f099b (head 831d76c6): 57 files (the new file, the 53 existing files that run family.py, the scaffold, the README rendering or the clone-URL remedies, tests/test_repo_hygiene.py, tests/test_cli_reference.py and tests/test_agents_md_codes.py): 3565 passed, 23 skipped, 1 xfailed in 35:22, exit 0. The skips, by name: 16 in test_family_init_values.py:219 ("this template quotes the placeholder") and 3 at :291 ("this template quotes id:"), both pre-existing and not about this change; 1 in test_stdout_encoding.py ("the default this is about is Windows' own"); 3 in test_shape_doctor.py (the PowerShell line, which only a Windows host prints). The xfail is test_shape_doctor.py's strict one for siblings.py (make siblings): a FILE, or a directory it cannot list, at ../<Project> ends in a traceback (NotADirectoryError or PermissionError at siblings.py:558), exit 1, where the doctor calls the slot NOT A CLONE #735. No test of this PR is skipped.
- Fix round 1 (head 372196a7, lane 1's review): the new file, test_clone_url_bytes.py, test_family.py, test_family_init_values.py, tests/test_repo_hygiene.py, tests/test_cli_reference.py and tests/test_agents_md_codes.py: 415 passed, 19 skipped in 4:07, exit 0 (the 16 + 3 skips above). The round changed remedy text, two comments, a docstring and the new file's tests only, and no other test file spells a remedy it changed (git grep for "made anew" and "no character escaped" finds only the two files run).
Mutants: each reverts one part of the fix, runs tests/test_local_remote_dir_urls.py, and is restored.
- Round 0, on the merge of main bf1f099b: all five killed.
  - M1, add no longer asks the question (the cmd_add call removed): 10 failed, 15 passed. Killed by test_add_from_a_directory_no_file_can_name_is_refused (8 rows) and test_add_through_a_link_to_such_a_directory_says_the_target_is_the_fault (2).
  - M2, the link sentence gone from LOCAL_REMOTE_DIR_REMEDY: 14 failed, 11 passed.
  - M3, the scaffold renders CLONE_URL and ASSEMBLY_CLONE_URL raw again: 2 failed, 23 passed. Killed by test_the_scaffolds_readmes_print_a_clone_line_that_pastes_whole.
  - M4, "or a lone surrogate" gone from the scaffold's remedy: 3 failed, 22 passed.
  - M5, the holder's README renders CLONE_URL raw again (_template_values): 2 failed, 23 passed. Killed by test_the_holders_readme_prints_a_clone_line_that_pastes_whole. M5 is beyond the 2-4 the wave brief asks; it is there because M3 covers only the scaffold's README.
- Fix round 1, on head 372196a7: the four mutants that touch the changed text, all killed. M1, M3 and M5 mutate code this round did not change, so they were not re-run.
  - M2, the link sentence gone from LOCAL_REMOTE_DIR_REMEDY: 14 failed, 13 passed. Killed by the add refusal rows (8), the add link test (2), the init link test (2), and the two lone-surrogate tests of the family remedies (2).
  - M4, "or a lone surrogate" gone from the scaffold's remedy: 3 failed, 24 passed. Killed by test_the_scaffold_explains_a_lone_surrogate and the scaffold's link test (2).
  - M6 (new), add's {holding} back to "a directory renamed or made anew": 11 failed, 16 passed. Killed by the add refusal rows (8), the add link test (2) and test_add_explains_a_lone_surrogate.
  - M7 (new), the scaffold's remedy says "but with no character escaped" again: 3 failed, 24 passed. Killed by test_the_scaffold_explains_a_lone_surrogate and the scaffold's link test (2).

What it cannot tell

The lone surrogate is tested at function level only. No Linux file name decodes to an unpaired surrogate outside U+DC80–U+DCFF. That .gitmodules reads back "another path" on Windows is reasoned, not run, and CI has no Windows job since ci: run the test suite on Linux only #597.
A non-ASCII path, and every local path on Windows, is now quoted too. shell_quoted leaves bare only [A-Za-z0-9_./:@%+=,-], so a remote under café/ now prints in quotes, and so does a Windows path, which is spelled with \ ('C:\Users\me\remotes\Atlas.git'; bare on main, where sh read it as C:UsersmeremotesAtlas.git). A GitHub URL and an ordinary ASCII POSIX path stay bare, so no template and no pin changes, and an existing README is never re-rendered.
The link sentence is fixed text, not a branch on what was typed. It begins "where the path given is text", so it holds for every refusal, scaffold-project.py and family.py: a --local-remote-dir whose clone URL is not text a README can hold is refused before any remote or folder exists, --dry-run alike, so a directory named in a legacy code page no longer ends in a UnicodeEncodeError traceback after the bare remotes are made (#433) #640's relative-path case included, and the call sites in cmd_init and _scaffold stay untouched.
adopt-project.py is not changed (read from the code, not run). Its execute --local-remote-dir renders "CLONE_URL": urls["assembly"] and "ASSEMBLY_CLONE_URL" raw (adopt-project.py:11874-11875), so a source under a b/ prints the same split line in the READMEs it writes, and its adopt-unwritable-value remedy says nothing of a link's target or a lone surrogate. The issue names the scaffold and family.py only, so both are left, for Tracking: adopt-project.py, converting an existing repository in place #457's checklist.

Shared files

Main has moved since this PR opened. It is 1eeef3b6 now (#670, #703 and #706 landed). The branch holds the merge of bf1f099b (831d76c6) and one fix commit on top of it (372196a7), and merges clean with 1eeef3b6, which changed scripts/family.py (#670's add) and two lines of tests/test_family_add_refusals.py, a file whose fixtures this PR's new file imports. The refresh and its gate are the coordinator's.

Earlier, on the merge of bf1f099b: #744, #728, #671 and #634 edit scaffold-project.py, and #728 and #671 scripts/family.py, each in other functions than this PR's hunks. #806 edits tests/test_clone_url_bytes.py's directory_named and its imports; this PR's new file imports directory_named by the name and signature #806 keeps. #688's head, before its own refresh, conflicted with main alone in family.py's import block (#728's import), not with this branch.

Each merge-tree below used git merge-tree --write-tree with head 372196a7, and every head was read from GitHub at 2026-10-17T11:40Z (ls-remote for main, a fetch of each branch):

main at 1eeef3b6: clean, tree fd6f8f05041d
fix/684-family-holder-states at 772105aa: clean, tree 42384f251489
fix/766-spelled-paths at 0e516bba: clean, tree b1d60f53f737
fix/689-next-push-refspec at b5a7a7f1: clean, tree 3df6255778e0
fix/688-family-init-values at 6958dfbc: clean, tree f9499a421df4
fix/757-non-utf8-test-dirs at 6b5e9074: clean, tree df8efde8394d
fix/686-bump-reachability at 444934d9: clean, tree 9889e2edc91d
l3-786-ci-gitmodules-rules at 31c3792c: clean, tree 7a237b4c4564
fix/685-spellings-remedies at ac764679: clean, tree 1ebb9232a9ba
fix/695-family-docs at c42286de: clean, tree bce54e899c0c
fix/692-raw-folder-paths at ab55c29f: clean, tree a9784ae5a6e7
fix/690-siblings-inputs at 5ba1f8b3: clean, tree b0e18add8719
fix/681-validate-family-remedies at 58d6c50f: clean, tree 0753855de2b4

Closes #693

🤖 Generated with Claude Code

Summary by Sourcery

Prevent unsafe local clone URLs from being written while making generated README clone commands safe to paste.

Bug Fixes:

Refuse local member clone URLs that cannot be safely represented in .gitmodules, including during dry runs, before reading or modifying the member.
Ensure clone URLs rendered in scaffold and family READMEs remain a single shell word when paths contain spaces, apostrophes, or other shell-significant characters.

Enhancements:

Clarify clone-URL remediation by identifying resolved symbolic-link targets and relative-path origins as possible sources of unwritable values, and document lone-surrogate handling.

Tests:

Add coverage for unwritable local member URLs, symlink targets, remediation paths, lone surrogates, and shell-safe README clone commands.

… scaffold-project.py: `add --local-remote-dir` refuses a member clone URL `.gitmodules` cannot hold as text, `--dry-run` alike, each clone-URL remedy names a link's target and a lone surrogate, and the READMEs quote the clone URL, so the line under `a b/` pastes as one command (#693) #640 (#433) asked `unwritable_reason` of the clone URL the scaffold and `family.py init` write into their READMEs. Four things were left, each measured on main 8062810: - `family.py add --local-remote-dir <dir>/caf<e9>` exited 0, dry run and real, and committed the raw byte into the holder's `.gitmodules`, where the holder's `make siblings` (`siblings.submodule_urls`, `git config --list` read with `errors="replace"`) read the URL back as `caf\ufffd`; a line break was escaped as `\n` by git and read back cut short at it, and an escape and a bidi override were committed raw. `cmd_add` now asks `_refuse_an_unwritable_member_url` of `_member_url`'s result, after the branch question and before `_probe_member`'s throwaway clone, so the real run and `--dry-run` refuse alike, `family-unwritable-value`, with nothing read, mounted or written. - `--local-remote-dir link`, a UTF-8 name linked to a `caf<e9>` directory, was refused naming the RESOLVED URL by a remedy that said to pass a path that is UTF-8 text, which it was. Every clone-URL remedy now says the URL is the path the flag resolves to, and that where the path given is text and the URL is not, a link's target (or, for a relative path, the directory it was given from) is the one to rename. - `unwritable_reason` names a lone surrogate, which a Windows name can hold, and no clone-URL remedy did; both now say "or a lone surrogate", as #282's `UNWRITABLE_REMEDIATION` does. - `git clone --recurse-submodules {{CLONE_URL}}` (holder) and `{{CLONE_URL}}`/`{{ASSEMBLY_CLONE_URL}}` (assembly root, both legs) took a local path raw, so under `a b/` or `it's/` the line split when pasted. Both tools now render the value through `shell_quoted(url, "posix")`, POSIX whatever the host as #532 rules for pasted commands: a GitHub URL and an ordinary ASCII path stay bare, so the templates and their pins are unchanged, but a path holding `é` is quoted too, as `shell_quoted`'s conservative set has it, and `test_clone_url_bytes.py`'s two `café` README asserts now expect the quotes. The link sentence and the surrogate are static text in the remedies, not a branch on what was typed: it is true of every refusal (including #640's relative path under such a directory), and it leaves `cmd_init`'s and `_scaffold`'s call sites untouched. `add`'s remedy shares `init`'s words (`LOCAL_REMOTE_DIR_REMEDY`) with its own verb, GitHub URL and reasons: `.gitmodules` holds what `README.md` cannot, so its reasons say how the holder reads it back, not that it cannot be written. Closes #693 Lane: openreposhape-4 (openRepoShape-4) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Main's 28 new commits since 8062810 merged with no conflict. Four of them touch a file this branch changes: #744, #728, #671 and #634 edit scaffold-project.py, and #728 and #671 edit scripts/family.py too, each in other functions than the clone URL's remedy, its refusal and the README values. Nothing this branch changes is a pinned copy, so nothing was re-pinned. Lane: openreposhape-4 (openRepoShape-4) Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

sourcery-ai · 2026-10-17T11:11:03Z

Reviewer's Guide

The PR hardens local clone-URL handling by refusing values that cannot round-trip through .gitmodules or README text, improves remediation guidance for resolved paths and lone surrogates, and POSIX-quotes generated README clone commands so they paste as a single argument, with comprehensive regression coverage.

Sequence diagram for safe local member URL validation

sequenceDiagram
    participant User
    participant Add as family.py add
    participant Validate as _refuse_an_unwritable_member_url
    participant Git as Git and filesystem

    User->>Add: add --local-remote-dir path
    Add->>Add: _member_url(repository, local_remote_dir)
    Add->>Validate: _refuse_an_unwritable_member_url(local, url)
    Validate->>Validate: unwritable_reason(url)
    alt URL cannot be represented as text
        Validate-->>Add: Refusal family-unwritable-value
        Add-->>User: No git command or files changed
    else URL is writable
        Validate-->>Add: Continue
        Add->>Git: Read member and update .gitmodules
    end

Flow diagram for POSIX-quoted README clone URLs

flowchart LR
    URL[Resolved clone URL] --> Quote["shell_quoted(url, posix)"]
    Quote --> README[Generated README git clone line]
    README --> Shell[POSIX shell]
    Shell --> Clone[One clone command argument]
    Quote -.-> Safe[Quotes spaces and shell metacharacters]
    Quote -.-> Bare[Leaves ordinary ASCII URLs bare]

File-Level Changes

Change	Details	Files
Reject unwritable local member clone URLs before any read, clone, mount, or write operation.	Added validation for `family.py add` using the existing unwritable-value reason and refusal code. Applied the check before throwaway cloning and consistently for dry runs and real runs. Added member-specific remediation explaining `.gitmodules` text constraints and resolved symlink targets.	`scripts/family.py`
Make clone-URL remedies accurately describe resolved paths and unsupported filename values.	Updated scaffold and family remedies to identify the resolved `--local-remote-dir` path and symlink or relative-path causes. Added lone-surrogate guidance to clone-URL remediation text. Clarified that README and `.gitmodules` consumers require UTF-8, line-safe values.	`scaffold-project.py` `scripts/family.py` `tests/test_clone_url_bytes.py`
Render README clone commands as shell-safe, pasteable commands.	Applied POSIX shell quoting to scaffold, assembly, and holder README clone URL placeholders. Preserved bare output for URLs containing only shell-safe characters and quoted paths containing spaces, apostrophes, or other unsafe characters. Updated existing README expectations for quoted non-ASCII local paths.	`scaffold-project.py` `scripts/family.py` `tests/test_clone_url_bytes.py`
Add end-to-end coverage for invalid local paths, remediation text, and pasteable README commands.	Added tests covering dry-run and real `add` refusals, unchanged holders, no throwaway clone, and no GitHub invocation. Covered symlink targets, UTF-8 paths, lone-surrogate function-level behavior, and all scaffold/family refusal variants. Validated clone lines with `shlex.split` and an actual shell paste/clone flow across scaffold and holder READMEs.	`tests/test_local_remote_dir_urls.py`

Assessment against linked issues

Issue	Objective	Addressed
#693	Refuse `family.py add --local-remote-dir` when the member clone URL cannot be safely written to `.gitmodules`, including under `--dry-run`, before reading or mounting the member.	✅
#693	Make clone-URL refusal remedies explain that an unwriteable resolved path may be a symlink target, and identify lone surrogates as a reason the URL cannot be written.	✅
#693	Render clone URLs in the affected READMEs as POSIX-shell-quoted words so pasted clone commands remain a single command when paths contain shell-significant characters.	✅

Possibly linked issues

--local-remote-dir: family.py add takes a directory whose name is not UTF-8 into .gitmodules, the clone-URL refusal's remedy misses a link's target and a lone surrogate, and the READMEs' clone line is rendered unquoted #693: The PR implements all four issue requests: add-time URL refusal, link-target and surrogate explanations, and shell-quoted README clone URLs.

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 · 2026-10-17T11:27:16Z

(lane openRepoShape-1, 2026-10-17T11:27:21Z)

Lane 1's independent review of #812 at 831d76c: CHANGES NEEDED

The code does what the issue asks. One remedy sentence sends add's reader to a second refusal, one PR-body sentence is false on Windows, and one remedy phrase is false for the PR's own it's row. All three fixes are text only; exact replacements are below.

How this was measured. The head's merge base with main is main's tip bf1f099b (git ls-remote read it at 11:13Z). So the head is the local merge, and every "head" figure below is on it. Probes used a shared clone, a stub gh (exit 97, never called), GH_TOKEN=bogus-x812, an empty GH_CONFIG_DIR, GIT_CONFIG_GLOBAL=/dev/null, GIT_ALLOW_PROTOCOL=file and local bare remotes only. <P> stands for the probe root. Checks on 831d76c when read: SonarCloud SUCCESS, tests-git-bash SUCCESS, Linux tests IN_PROGRESS.

The issue's lines

line	main `bf1f099b`	head `831d76c6`	verdict
O2 `add --local-remote-dir <P>/caf<e9>`	dry run exit 0; real run exit 0, committed, `.gitmodules` holds `url = <P>/caf\xe9/IRQS.git` (raw byte)	exit 2 both, `REFUSED family-unwritable-value: … '<P>/caf\udce9/IRQS.git' …`; no commit, clean status, no throwaway clone left in `TMPDIR`, no `gh` call	fixed
O2's other rows: LF, U+202E, ESC, `\x01`, U+2028, U+0085	LF written as `n\nl`, U+202E written raw (both exit 0)	all six refused, dry run and real	fixed
O3 `--local-remote-dir link`, a link to `caf<e9>`	`add`: exit 0, raw byte committed	refused, naming the resolved URL, with the link sentence; the `init` and scaffold link tests pass	fixed
O4 lone surrogate	no remedy names it	all three remedies name it (tested at function level; Linux cannot make such a name)	fixed
O1 README clone line under `a b/`	`git clone --recurse-submodules <P>/a b/remotes/Atlas.git`	`… '<P>/a b/remotes/Atlas.git'`; `it's` gives `'<P>/it'\''s/remotes/Atlas.git'`	fixed

Also checked:

Count of quoted lines: 4 per scaffold (the assembly root twice, then the spec leg and the code leg) and 1 per holder.
Dollar sign: a remote under d$x printed bare on main, so $x expanded when pasted. The head quotes it.

Findings, ranked

1. Must change: `add`'s remedy says "a directory … made anew", but `add` reads `<Project>.git` from that directory

LOCAL_REMOTE_DIR_REMEDY is shared by init and add. For init (and the scaffold) the tool creates the bare repository, so a new directory is a way out. For add the directory must already hold the member's bare repository. A person who follows "made anew" is refused again:

mkdir <P>/fresh-remotes
family.py add --family-root <holder> --member InkRouter/IRQS --local-remote-dir <P>/fresh-remotes [--dry-run]
-> exit 2 (dry run and real)
REFUSED reading InkRouter/IRQS. THE EXACT COMMAND WAS:
    git -c protocol.file.allow=always clone -q --bare -- <P>/fresh-remotes/IRQS.git <tmp>/member.git
    exit 128
fatal: repository '<P>/fresh-remotes/IRQS.git' does not exist

The brief asked whether one remedy fits both callers exactly. This is where it does not. The fix keeps init's text byte for byte, so test_clone_url_bytes.py does not change. In scripts/family.py, replace:

    "Pass a --local-remote-dir whose path is one line of UTF-8 text with no "
    "control character in it, a directory renamed or made anew, and run "
    "again; or run `{verb}` against GitHub, without --local-remote-dir, "

with:

    "Pass a --local-remote-dir whose path is one line of UTF-8 text with no "
    "control character in it{holding}, and run again; or run `{verb}` "
    "against GitHub, without --local-remote-dir, "

Then format the two callers:

init: .format(verb="init", name="Family", holding=", a directory renamed or made anew")
add: .format(verb="add", name="Project", holding=" and that holds the member's bare repository .git, its directory renamed or the repository moved")

add's remedy then reads:

Pass a --local-remote-dir whose path is one line of UTF-8 text with no control character in it and that holds the member's bare repository <Project>.git, its directory renamed or the repository moved, and run again; or run add against GitHub, …

Update MEMBER_REMEDIATION in tests/test_local_remote_dir_urls.py to the same words, and update the remediation the PR body quotes.

2. Must change (PR body): on Windows an ordinary local path is now quoted

SHELL_BARE_POSIX has no \, so every Windows --local-remote-dir URL changes from main:

shell_quoted(r"C:\Users\me\remotes\Atlas.git", "posix") -> 'C:\Users\me\remotes\Atlas.git'
shell_quoted("/home/me/remotes/Atlas.git", "posix")    -> /home/me/remotes/Atlas.git

The change is right. On main, sh read the bare form as C:UsersmeremotesAtlas.git (measured with printf '%s\n' in bash). The test file already knows it: ("ordinary", "plain", os.name == "nt"). The body says otherwise, though. Replace the "What it cannot tell" bullet:

A non-ASCII path is now quoted too. shell_quoted leaves bare only [A-Za-z0-9_./:@%+=,-], so a remote under café/ now prints in quotes. A GitHub URL and an ordinary ASCII path stay bare, so no template and no pin changes, and an existing README is never re-rendered.

with:

A non-ASCII path, and every local path on Windows, is now quoted too. shell_quoted leaves bare only [A-Za-z0-9_./:@%+=,-], so a remote under café/ now prints in quotes, and so does a Windows path, which is spelled with \ ('C:\Users\me\remotes\Atlas.git'; bare on main, where sh read it as C:UsersmeremotesAtlas.git). A GitHub URL and an ordinary ASCII POSIX path stay bare, so no template and no pin changes, and an existing README is never re-rendered.

Make three smaller text changes for the same reason:

Body, Tests section: change "an ordinary plain (left bare)" to "an ordinary plain (left bare on POSIX, quoted on Windows)".
_template_values docstring: change "a GitHub URL and an ordinary path are left bare" to "a GitHub URL and an ordinary POSIX path are left bare (a Windows path, spelled with \, is quoted)".
Test comment above FOLDERS: change "spelled with \ and :, which sh reads as syntax" to "spelled with \, which sh reads as an escape". : is in the bare set.

3. Should change: "with no character escaped" is false for the PR's own `it's` row

The head's README prints '<P>/it'\''s/remotes/Atlas.git', so the apostrophe is escaped. Replace

in quotes where a shell needs them but with no character escaped:

with

in quotes where a shell needs them, every character but an apostrophe written as itself:

in four places:

scaffold-project.py CLONE_URL_REMEDIATION
scripts/family.py CLONE_URL_REMEDIATION
QUOTED_NOT_ESCAPED in tests/test_clone_url_bytes.py
the PR body's "Remedies" paragraph

4. Nits

tests/test_local_remote_dir_urls.py:271: assert lines and set(lines) == {...}, lines is a composite assert, the S9073 pattern. Split it into assert lines, lines and assert set(lines) == {clone_line(url, quoted)}, lines.
refused_text: getattr(module, function)(True, SURROGATE_URL) makes two calls inside pytest.raises, a possible S5778. Bind refuse = getattr(module, function) before the with.
scripts/family.py:192-195: the import comment still says the question is asked of "a value init writes". Add "and the member's clone URL add writes into .gitmodules (--local-remote-dir: family.py add takes a directory whose name is not UTF-8 into .gitmodules, the clone-URL refusal's remedy misses a link's target and a lone surrogate, and the READMEs' clone line is rendered unquoted #693)".

Correctness checks that passed

Order: _refuse_an_unwritable_member_url runs after _checked_branch and before the checking … before mounting it line and _probe_member. Nothing is written, committed or cloned before it, in either run (stdout empty, TMPDIR empty, holder clean).
Text rows unchanged from main: these remote folders were checked:
- a space, it's, d$x, b`id`t and a leading -x;
- h#s;q (git config wrote it in double quotes on both) and q"b\s;
- ~t, café and a ZWNJ.
Every one gave exit 0 on head and main, .gitmodules url bytes identical, and stdout and stderr identical except the commit id.
Ordinary runs byte-identical to main:
- add: .gitmodules, the commit tree and the subject identical; stdout differs only in the commit id.
- scaffold (Acme/Atlas, plain remotes): all 5 READMEs identical once the shape commit and the case folder are normalised.
- init (Orbit): README identical.
Where the placeholders land: only in the ```sh fences of four Markdown READMEs (assembly-root :15/:33, spec-root :13, code-root :13, family-root :9). No YAML, Makefile, JSON or shell template holds them, so POSIX quoting is the right dialect. values["CLONE_URL"] is read nowhere else.
ext:: and file://: cannot reach git as a transport. _member_url calls resolve(), so --local-remote-dir 'ext::sh -c touch% PWNED' becomes <cwd>/ext::sh -c touch% PWNED/IRQS.git. Git said does not appear to be a git repository, exit 2 on head and main, with no GIT_ALLOW_PROTOCOL set and no file made. The PR does not change this.
Relative path from a non-UTF-8 folder: --local-remote-dir . run from inside caf<e9> is refused on head (exit 2) and accepted on main (exit 0).
The body's quoted refusal matches the real stderr byte for byte, once the path is elided and the \` escapes are undone.

Tests

Runs on the head (= the merge), /usr/bin/python3 -m pytest, no GIT_ALLOW_PROTOCOL:

the new file + test_clone_url_bytes + test_family_init_values + test_scaffold_free_text_values: 274 passed, 19 skipped. The skips are all at test_family_init_values.py:219 and :291, which existed before this PR.
test_family_add_refusals + test_family_printed_quoting: 149 passed
test_family + test_scaffold_mount_ignored: 148 passed
test_scaffold_leg_paths + test_bootstrap_after_bump: 76 passed

Mutants, run against the new file and restored afterwards:

cmd_add call removed, and only ASSEMBLY_CLONE_URL rendered raw: 12 failed, 13 passed. The failures are the 8 add rows, the 2 add link rows, and the scaffold README space and apostrophe rows.
The holder's CLONE_URL raw, and "or a lone surrogate" dropped from MEMBER_URL_REMEDIATION: 13 failed, 12 passed. The failures are 10 add rows, test_add_explains_a_lone_surrogate, and the holder README space and apostrophe rows.

The asserts are honest: whole stderr, assert_unchanged, assert_left_nothing, shlex.split, and a real paste into sh.

Windows (reasoned, not run).

directory_named and linked_to skip where the host cannot make the name or the link.
clone_lines reads bytes, so CRLF is safe.
Expectations built from str(...resolve()) hold under quoting.

Not verified

The body's merge-tree list against the 13 branches, and #806's directory_named signature. I fetched only refs/pull/812/head and main.

For the checklist (outside #693)

The README clone line under --local-remote-dir still fails after it pastes whole. setup-project.py: its paste lines, the step (3), (4) and (7) checks, and the upstream-org guard #538 gave the NEXT STEPS clone -c protocol.file.allow=always, but the README line does not have it. I pasted the head's assembly README line under a b/ (GIT_CONFIG_GLOBAL=/dev/null, no GIT_ALLOW_PROTOCOL):
```
fatal: transport 'file' not allowed
Failed to clone 'code' a second time, aborting   (exit 1)
```
Main failed earlier, with fatal: repository '<P>/scaffold/space-main/a' does not exist. The holder's line clones only while it has no members. The PR's paste test also passes ALLOW_FILE_PROTOCOL.
adopt-project.py:11874-11875 renders CLONE_URL and ASSEMBLY_CLONE_URL raw, and its remedy does not name a link's target or a lone surrogate. The body already says so; carry it to Tracking: adopt-project.py, converting an existing repository in place #457.
On the tracking branch the branch refusal comes first. A real add on main with a caf<e9> URL gives REFUSED family-on-tracking-branch, so the person switches branch and is only then refused for the URL. The URL question needs nothing from the holder and could be asked earlier.
cmd.exe reads 'C:\…' with the quotes as part of the path. Git Bash and PowerShell read it correctly, and the fence says sh.
A non-ASCII path such as café/ is now quoted where main printed it bare. This is harmless, and the body already says so.

Lane: openreposhape-1 (openRepoShape-1)

…y says the directory must hold the member's bare repository and names no new directory as a way out, and the clone-URL remedies and `_template_values` say the apostrophe is escaped and a Windows path quoted, so none sends a person to a second refusal or says a README escapes no character (#693) Lane 1's review of #812 (comment 6233341619) measured three texts of this branch that were false for a person following them. - `LOCAL_REMOTE_DIR_REMEDY` is shared by `init` and `add`, and said the directory could be "renamed or made anew". For `init` and the scaffold that holds: they make the bare repository there. For `add` the directory must already hold the member's `<Project>.git` (`_probe_member` clones it), so a new, empty one is refused again with `repository ... does not exist`. The remedy now takes the caller's words (`{holding}`): `init` keeps its text byte for byte, `add` says "and that holds the member's bare repository `<Project>.git`, its directory renamed or the repository moved". `MEMBER_REMEDIATION` in the new test file follows, and a new test makes both ways out of `add`'s refusal exit 0 with the member mounted. - "with no character escaped" was false for the PR's own `it's` row, which prints `'<P>/it'\''s/...'`. Both remedies, the two comments above them and `QUOTED_NOT_ESCAPED` (renamed `QUOTED_ONE_WORD`) now say "every character but an apostrophe written as itself". - `shell_quoted` has no `\` in its bare set, so a Windows path is quoted (`'C:\Users\me\remotes\Atlas.git'`, read by `sh` as `C:UsersmeremotesAtlas.git` on main). `_template_values` said an ordinary path is left bare; it now says an ordinary POSIX path is, and the test comment above `FOLDERS` says `\`, which `sh` reads as an escape, not `\` and `:`. Two test nits: the composite assert in `assert_one_command` is two asserts, and `refused_text` binds the refusal before the `pytest.raises` block so it holds one call. Lane: openreposhape-4 (openRepoShape-4) Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

brettheap · 2026-10-17T11:47:19Z

(lane openRepoShape-1, 2026-10-17T11:47:24Z)

Lane 1's re-verification of #812 at 372196a: PR-READY

I built a local merge of 372196a7 with current main dcf62eda (#706, #703, #670, #682 and #726 since bf1f099b). It merged with no conflict; scripts/family.py merged automatically. On the merge, cmd_add still asks _refuse_an_unwritable_member_url after _checked_branch and before the checking … before mounting it line and _probe_member. Probes used a stub gh that was never called, GH_TOKEN=bogus, an empty GH_CONFIG_DIR, GIT_CONFIG_GLOBAL=/dev/null and local bare remotes only.

finding	measured on the merge	result
1. `add`'s remedy	A fresh member and holder were made with the merged code, and `IRQS.git` was put under `caf<e9>/`. `add` gives exit 2, `family-unwritable-value`, with the remedy "…with no control character in it and that holds the member's bare repository `<Project>.git`, its directory renamed or the repository moved, and run again…". Renamed (`caf<e9>` → `cafe`): exit 0, committed `.gitmodules, members/IRQS, family.yaml`, url `<P>/cafe/IRQS.git`, clean status, no throwaway clone left. Moved (`IRQS.git` into `moved/`): exit 0, url `<P>/moved/IRQS.git`, clean. A new empty folder is still `REFUSED reading InkRouter/IRQS`, and the remedy no longer suggests one ("made anew" is not in `add`'s text).	fixed
1. `init` unchanged	`CLONE_URL_REMEDIATION` was imported from `831d76c6` and from the merge. With the finding-3 phrase swapped, the two are byte-identical, and `init` still holds ", a directory renamed or made anew, and run again;". The scaffold's remedy is likewise identical apart from that phrase.	fixed
2. Windows quoting	The body's "What it cannot tell" bullet now says "every local path on Windows, is now quoted too … `'C:\Users\me\remotes\Atlas.git'`", which matches `shell_quoted(…, "posix")`. The Tests bullet says "(left bare on POSIX, quoted on Windows)". The `_template_values` docstring says "an ordinary POSIX path are left bare (a Windows path, spelled with `\`, is quoted)"; it is not a raw string, so `\\` renders as one `\`. The test comment now says "spelled with `\`, which `sh` reads as an escape".	fixed
3. "every character but an apostrophe"	The phrase appears once each in the scaffold's `CLONE_URL_REMEDIATION`, `family.py`'s `CLONE_URL_REMEDIATION` (checked by import; a grep misses it because the string is split across lines), `QUOTED_ONE_WORD` and the body's Remedies paragraph. `git grep -i 'no character escaped\|escapes\? no character'` over the merged tree finds nothing. The body's quoted `add` remediation equals `MEMBER_URL_REMEDIATION` byte for byte.	fixed
4. Nits	`assert lines, lines` is now split from the set compare; `refuse = getattr(…)` is bound before `pytest.raises`. The import comment goes to #459, which is fine.	fixed

Runs on the merge (/usr/bin/python3 -m pytest, no GIT_ALLOW_PROTOCOL), all passing:

The new file, test_clone_url_bytes, test_family_init_values, test_scaffold_free_text_values, test_family_add_refusals and test_family_printed_quoting together: 425 passed, 19 skipped. The skips are all at test_family_init_values.py:219 and :291, and were there before this PR.
test_family, test_scaffold_mount_ignored, test_scaffold_leg_paths and test_bootstrap_after_bump together: 224 passed.

Checks at 372196a:

SonarCloud: SUCCESS. The quality gate API says OK, with new duplicated-lines density 0.0.
Sonar's one open issue on PR 812: S3776 at scripts/family.py:3294 (_checked_mount_path, complexity 30). It was created 2026-10-16, and the PR's hunks do not touch that function, so it is not this PR's.
Other jobs: tests-git-bash and package SUCCESS; Linux tests was still running when read.

New: nothing. The checklist items from my first review stay outside #693.

Lane: openreposhape-1 (openRepoShape-1)

sourcery-ai · 2026-10-17T12:35:11Z

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 12 hours and 7 minutes by commenting @sourcery-ai review. Upgrade to get a review now.

sonarqubecloud · 2026-10-17T12:36:02Z

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

Sourcery assessment

Approved.

…mand-scope, and tests/test_local_remote_dir_urls.py calls `allow_file_protocol()`, so #812's test file still imports once #758 lands (#758) The retarget after #748 landed as 431c8cf. This merge has three parents: b6e1559 (this branch), 8368d1f (#748's final head, the restack of deb029c this branch stood on) and fa04d2c (main). Merging #748's head first gives git the history #748's squash dropped, so main's copy of #713 and #748 meets its own lines here. Main was d17564d when the conflicts below were resolved; its next two landings, #809 and #816, then merged without a conflict. Conflicts, and how each was resolved: - tests/test_adopt_empty_symlinks.py, against 8368d1f, which carries #388 through main 741679e. Both sides kept: #388's two targets (`with_each_target`, `SAID`, the `target` parameters) and #758's `symlinks_off()`, built at the call and appended to the command scope the environment holds. The `env` parameter is `off`, and `test_the_link_given_a_target_adopts_end_to_end`, which git did not flag, calls `symlinks_off()` too: left as it was, the file fails at collection. - tests/test_clone_just_pushed.py, tests/test_doctor_legs_members_next.py and tests/test_family_bump_remove_rollback.py, against main d17564d. Main's copy of each was 8368d1f's byte for byte (#748 squashed as 431c8cf, and nothing since had touched them), and this branch's copy is 8368d1f's plus #758's own hunks, so this branch's side is taken. Against main, the merged tree differs in #758's 25 files and in tests/test_local_remote_dir_urls.py. Each of the 25 differs by #758's own added and removed lines, tests/test_adopt_empty_symlinks.py aside. rv818's MUST 1 (lane 2): tests/test_local_remote_dir_urls.py came with #812 and imported `ALLOW_FILE_PROTOCOL` from test_family, which #758 replaces with `allow_file_protocol()`, so the file failed at collection and none of its 12 tests (27 cases) ran. It now imports `allow_file_protocol` and calls it at the paste. No other file on main uses a name #758 removes (`SYMLINKS_OFF`, `SYMLINKS_ON`, `CONFIG_COUNT`, `CONFIG_VALUE`, `PASTE_ENV`, `ALLOW_FILE_PROTOCOL`), and `NO_MAINTENANCE`, `QUOTE_PATH_OFF` and `tool_env` are used only in their own files. Lane: openreposhape-5 (openRepoShape-5) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

main gained #713 (#645), #812 (#693), #767 (#685), #745 (#704, #694) and #805 (#690) since 2000488. No conflict, and none of them touches README.md, the handbook, AGENTS.md, docs/ or the README cap's ledger, so every handbook marker reads as before. README.md is 1475 lines, its cap; AGENTS.md 494, its cap. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

…notes entries for the five pull requests landed since 2000488, current to main 40e678e (#563, #683) RV630's review of rounds A3, A4 and B1 (#630 issuecomment-6233641684, READY), its four optional nits: - 1: README.md's drift paragraph ended on the one-word line "finished.". "exactly " and "about the shape " are cut, and the paragraph re-wraps to seven lines with both bold leads whole. README.md is 1474 lines, its cap (1422 + 2 + 5 + 12 + 22 + 10 + 1); tests/readme-cap.d/683.txt is +1, and its text says why. Both `Keeping a project's shape current` markers move to df780eb5; the other 21 read current. - 2: #682's lead reads "A holder's `make bootstrap` that left a member unchecked-out". - 3: #730's entry says "Comments and docstrings", not "Four": its squash makes three claims, comments only. - 4: the doctor's `members` row has one bullet, #705's sentence and #624's together. Wave B2 so far, `git log --first-parent 2000488..40e678e`, each entry from its squash diff: - #713 (#645): the test session turns git's automatic maintenance off (Install, CI and the docs). - #812 (#693): `add --local-remote-dir`'s member clone URL, the remedies and the READMEs' quoted clone URL (the `--local-remote-dir` bullet). - #767 (#685): `members/` renamed with `git mv` is renamed back in two steps, and the git a printed `--sparse` needs (family). - #745 (#704, #694): lane 5's line (#630 issuecomment-6227129879), with v1.0.0's known limit at v1.0.0.md:463 (no git on `PATH`) said closed (Bootstrap). - #805 (#690): `make siblings`'s inputs (family). The counts (111 pull requests; 122 of 135 milestone issues closed at 16:14Z), "current to main" and the open list (13, each with a pull request open) are at main 40e678e, still placeholders. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

…s and #817's entries, #642's upgrade note for a holder's `cd` line, and v1.0.0's `--into` limit kept open, current to main 64ec47c (#563) RV630's pre-review of the B2 stage (lane 3's session): - SHOULD-1: the CI bullet said every test that clones a just-pushed local remote clones it safely (#554), but five plain-path `git submodule update --init` runs are left on main (tests/test_doctor_legs_members_next.py:631, :987, :1054; tests/test_family_bump_remove_rollback.py:337, :908). It now says "runs `git clone` on", which is what #554 changed. - NIT-1: #812's READMEs quote a local clone URL "that needs it". - NIT-2: #682's first sentence no longer repeats the lead ("there:"). Landed since 40e678e, each entry checked against its squash: - #747 (#681), 6626a3c, whose tree equals `merge-tree 40e678e ff7c59b`: a new family bullet, with RV630's NIT ("`member-gitlink-absent`'s remedy puts the gitlink back by hand"); v1.0.0's limits at v1.0.0.md:466 and :467 are said closed. - #817 (#708), 64ec47c, whose tree equals `merge-tree 6626a3c d63a267`, and whose own lines equal those of 8fd2fa6, the head RV630 read: RV630's text, appended to the YAML reader bullet. #738 (#695) is v1.1.1 (lane 5, #473 issuecomment-6236564004), so its docs entry is not taken, and v1.0.0's `--into` limit (v1.0.0.md:492) is listed as still open under Known limits. Its upgrade note is true of main since #642 (0ed1b84), so it goes in now, under "Install and upgrade": lane 5's sentence verbatim, and the README's `Y/Y` layout. Counts and the open list at main 64ec47c: 113 pull requests since bfed143; 124 of 135 milestone issues closed at 16:42Z; 11 open, each with a pull request open, #681 and #708 out, #695 marked as planned for v1.1.1. All still placeholders. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

…nd what they tell git to the `GIT_CONFIG_COUNT` entries the environment holds, so a pasted line keeps the session's maintenance settings and the Windows job's two (#758) (#818) Against main. #748 landed as `431c8cf8`; this branch merged #748's final head and main (see Fix round 1). After that merge, #723 (`bad05d21`) and #746 (`fc867207`) landed. The branch is refreshed over main `fc867207` as `4a553e17`, one merge commit. Its only conflict was `tests/test_non_utf8_git_reads.py`'s `from conftest import` block, resolved by keeping both sides' names, as rv818 asked. The merged tree differs from main in #818's own 27 files, and #818's count scan flags no file. Tests only: 25 files under `tests/`, no shipped code, no pinned file, no README, handbook, `AGENTS.md` or `docs/cli.md` change. Two new commits on `b6e15591`: the retarget merge `2908e46f`, then the fix commit `eaa72d9e`. Tests only, as before. ### The retarget: merge commit `2908e46f` #748 landed as `431c8cf8` from its final head `8368d1f8`, and GitHub moved this PR's base to main. The merge commit has three parents: `b6e15591`, `8368d1f8` and main `fa04d2cd`. #748's head goes in first, so git sees the history the squash dropped. - **#748's head did not merge clean**, as rv818 found. `tests/test_adopt_empty_symlinks.py` conflicts, because `8368d1f8` holds main `741679e7`, which carries #388. - I took w758's earlier resolution of that file, which rv818 checked. Main's side of the file is the same at `974d4827`, `8368d1f8` and `fa04d2cd`, so it still fits. - Both sides stay: #388's two targets (`with_each_target`, `SAID`, the `target` parameters) and #758's `symlinks_off()`, built at the call. - `test_the_link_given_a_target_adopts_end_to_end` calls `symlinks_off()` too. Git does not flag it, and left as it was the file fails at collection. - **Main then conflicted in three files** (main was `d17564d9` when I resolved them): `tests/test_clone_just_pushed.py`, `tests/test_doctor_legs_members_next.py` and `tests/test_family_bump_remove_rollback.py`. - Main's copy of each was `8368d1f8`'s byte for byte. - This branch's copy is `8368d1f8`'s plus #758's own hunks, so this branch's side is taken. - Main's next two landings, #809 and #816, merged with no conflict. - **rv818 MUST 1.** `tests/test_local_remote_dir_urls.py` came with #812 and imported `ALLOW_FILE_PROTOCOL` from `test_family`. - Without a change, the file fails at collection: `ImportError: cannot import name 'ALLOW_FILE_PROTOCOL' from 'test_family'`, and none of its 27 cases run. - It now imports `allow_file_protocol` and calls `pasted(lines, paste, allow_file_protocol())`. - `git grep` finds no other file on main that uses a name #758 removes (`SYMLINKS_OFF`, `SYMLINKS_ON`, `CONFIG_COUNT`, `CONFIG_VALUE`, `PASTE_ENV`, `ALLOW_FILE_PROTOCOL`). `NO_MAINTENANCE`, `QUOTE_PATH_OFF` and `tool_env` change shape, and each is used only in its own file. - **Checked:** - Against main, the merged tree differs in #758's 25 files and in `tests/test_local_remote_dir_urls.py`. - Each of the 25 differs by exactly #758's own added and removed lines, `tests/test_adopt_empty_symlinks.py` aside. - Collecting the whole suite at the head finds 12,320 tests and no error. ### The fix commit `eaa72d9e` - **SHOULD 2:** `configure` in `tests/test_adopt_reencoded_files.py`, the seventh append from #758's line 2, now sets what `appended_to_command_scope(os.environ, pairs)` returns. It gives the same entries in the same order, so no test can fail on the old form. - **SHOULD 3:** the scan also reads a name the module binds to `"GIT_CONFIG_COUNT"` (`CONFIG_COUNT = "GIT_CONFIG_COUNT"`, annotated or not) as the count. It does so in a dict key, a setter's first argument and a subscript. - rv818's repro: append the base's two lines to `tests/test_adopt_root_symlinks.py`. - The old scan passes. The new one names line 2137. - **SHOULD 4:** `counts_set_to_a_number` is split into one reader per node type: `in_a_dict`, `in_a_call` and `by_subscript`. The walk takes each node's reader from `READERS`. By hand, no function scores more than 5 on S3776. I could not run Sonar. - **NIT 5:** `conftest.py` is no longer skipped. It scans clean, at the head and on today's main. - **NIT 6:** - New "sees" cases: `os.putenv("GIT_CONFIG_COUNT", "1")` and a bare `setenv("GIT_CONFIG_COUNT", "1")`. - The walk is now `files_off_their_count(directory)`. A new test runs it over a folder of its own, and checks three things: - `conftest.py` and a helper module are read; - a `.txt` file is not; - a file `OWN_COUNTS` allows is named for fewer counts as well as for more. - Four of rv818's six surviving mutants now fail a case: `putenv` dropped, the bare-name setter dropped, `!=` changed to `>`, and `glob("test_*.py")`. - The fifth, `conftest.py` scanned, is now how the code works, and putting the skip back fails the new test. - The sixth, the count's value taking any string, still survives: it is a stricter scan. Git refuses a count that is not a number (`bogus count in GIT_CONFIG_COUNT`, exit 128), so a test that holds one fails on its own. - **NIT 7, in part:** - An empty count (`""`) is now a count of its own. Git 2.43 and 2.47.1 read it as no entries at all. Measured: `GIT_CONFIG_COUNT= GIT_CONFIG_KEY_0=a.b GIT_CONFIG_VALUE_0=c git config --get a.b` exits 1. - The docstring says the scan reads the count's value as a constant only. - Left out: - a value built by a call (`str(len(pairs))`) or held in a variable; - an int; - `dict()`, or `environ={}` by keyword, handed to the helper; - `setenv` with keyword arguments; - an annotated subscript. rv818 judged only `str(len(entries))` likely, and telling it from a count worked out needs what the call means, not how it is spelled. - **NIT 8:** the paste case deletes `GIT_CONFIG_PARAMETERS` first. Measured on the merge before the fix commit, with `GIT_CONFIG_PARAMETERS="'probe.param'='p'"`, the case failed (`Left contains one more item: 'probe.param=p'`). It passes now. - **NIT 9, left as it is:** main's three appends built by hand each work the count out right: `tests/test_family_head_lost_gitmodules.py:97`, `tests/test_family_bootstrap_gate_order.py:135` and `tests/test_family_add_spellings_remedies.py:137`. Each of those files is also changed by another open PR (#828 `fix/761-add-move-back-gitmodules`, #820 `fix/762-validate-family-remedies-texts`, `fix/763-holder-docs-agree`), so converting them here would only add conflicts. - **NIT 10:** the comment on `NO_MAINTENANCE` now says its `maintenance.auto` and `gc.auto` repeat the session's two, which reach these gits in the command scope. Only `gc.autoDetach` adds a setting. The two stay, so the constant alone still keeps maintenance off for these gits. - **NIT 11:** the count was in lane 5's #473 summary, not in this body. The seven sets held 2,401 distinct passes, not 2,609: the 208-test run repeats three files of the 256-test run. ### Tests at `eaa72d9e` All runs went through `run9.sh`, with the real `gh` unreachable. All passed, none skipped: - **The 26 test files the PR changes against main**, in one full-file run each: 2,206 passed (1,108 + 1,098). - These are #758's 24 test files, `tests/test_local_remote_dir_urls.py` and `tests/test_adopt_reencoded_files.py`. - The count matches the collection exactly. - **The 43 files that import any of those 26, with `tests/test_repo_hygiene.py`, `tests/test_agents_md_codes.py` and `tests/test_cli_reference.py`:** 1,646 passed (825 + 821). - **The same merge made on main `d17564d9`** (a local commit, `c16a7e1a`, never pushed), before the fix commit: `tests/test_local_remote_dir_urls.py` and `tests/test_no_auto_maintenance.py`, 55 passed. ### Fails on the base The new tests were run against the base's scan and walk, as `b6e15591` has them, with `GIT_CONFIG_PARAMETERS` set: 6 failed, 30 passed. The six: - the paste case; - the `nothing`, `a-name-bound-to-it`, `an-annotated-name` and `a-name-by-subscript` cases; - the walk test. The three other new cases (`a-bare-setenv`, `putenv`, `a-name-bound-to-another-key`) are there to kill mutants, and they pass on any correct scan. MUST 1 fails on main's form at collection, as above. SHOULD 2 and NIT 10 change no behaviour. ### Mutants I made 26 mutants. 25 are caught, and one survives as rv818 said it would: - **22 on the scan and the paste case**, each run against `tests/test_no_auto_maintenance.py`: - the delenv dropped; - `putenv` dropped from `SETTERS`, or the bare-name setter; - the name branch of `is_the_count` dropped, or any name taken as the count; - `""` no longer a count; - the `Assign` names dropped, or the `AnnAssign` names; - no names collected; - each reader ignoring the names; - each of the three readers dropped from `READERS`; - `conftest.py` skipped; - `!=` changed to `>` or to `<`; - `glob("test_*.py")`, or `glob("*")`; - `by_subscript` taking any value; - the walk pointed at an empty folder. The last passes on a clean tree, like any change to an assertion that nothing is found. It is caught by rv818's SHOULD 3 repro, where the real walk fails and the mutant passes. - **The survivor:** the count's value taking any string. It makes the scan stricter (see NIT 6). - **`configure`:** with `appended_to_command_scope({}, pairs)`, the scan names the file. With the pairs dropped, three of its tests fail. - **MUST 1:** main's import put back fails collection with `ImportError`. ### Merge checks at `eaa72d9e` I ran `git merge-tree --write-tree` against `origin/main` (`fa04d2cd`) and the 53 other `fix/*` and `l3-*` heads on origin: - **Clean:** `origin/main` and 43 heads. - **#792 (`l3-785-non-ascii-paths`)** merges clean, but `tests/test_doctor_non_ascii_paths.py` imports `PASTE_ENV`. Whichever PR lands second changes `:50` (the import), the `:266` docstring and `:268` to `paste_env()`. - **#723 (`fix/663-non-utf8-name-tests`), #746 (`fix/672-bump-leg-branch-and-push-line`, which now holds #723) and #806 (`fix/757-non-utf8-test-dirs`)** conflict only in the `from conftest import` block of `tests/test_non_utf8_git_reads.py` (and of `tests/test_adopt_git_reads.py`, for #806). Both sides add names, so keep them all. - **Seven heads conflict with main itself, not with this PR:** - `l3-732-latin1-section` and `l3-733-ignored-sections` (`tests/test_doctor_legs_members_next.py`); - `fix/712-validate-pins-contracts`, `fix/766-spelled-paths` and `fix/798-adopt-legs-asked-first`; - `l3-677-pins-hand-step-wip` and `l3-678-naming-checkout-wip`. - I scanned each clean merged tree's changed test files for a name #758 removes, or a count the scan flags. Only #792 shows up. - **Main moved after the push.** #723 landed as `bad05d21`, and the two now conflict in one place: the `from conftest import` block of `tests/test_non_utf8_git_reads.py`. The fix is to keep both sides' names: `ORG, PROJECT, REPO, appended_to_command_scope, cloned_just_pushed, git, run_script, skip_message, skip_where_fsencode_gives_a_surrogate_no_byte, windows_fsencode`. Every other change merges in. On that merged tree, with the block resolved that way, the scan flags no file, and no file uses a name #758 removes. On `deb029c9` a probe test imported `PASTE_ENV` and pasted `git config --show-origin --list` through `paste()`: - `PASTE_ENV` held `GIT_CONFIG_COUNT=1` (`protocol.file.allow`), a stale `GIT_CONFIG_KEY_1=user.email` left from collection, and no `GIT_CONFIG_GLOBAL`. - `os.environ` at the call held a count of 4 and the session's global file. - The pasted git listed one command-line entry, `protocol.file.allow=always`, and `git config --get maintenance.auto` exited 1. So a pasted line ran with git's automatic maintenance on, in both scopes, as the issue said. `paste_env()` replaces it. It reads `os.environ` at each call (minus `PYTHONPATH`), adds the identity and `PYTHONDONTWRITEBYTECODE`, and appends `protocol.file.allow` with `appended_to_command_scope`. The six uses call it. The import-time copy also predated the session's `gh` guard on `PATH`, as `test_doctor_next_is_a_command.py`'s docstring says of its own environment; a pasted line now gets the `PATH` of the call. That docstring now says `paste_env()` reads at the call too. - **The seven.** Six are on this base, each now one call to `appended_to_command_scope`: `test_adopt_git_reads.py`, `test_adopt_nested_attributes.py`, `test_adopt_ignored_shape_files.py`, `test_adopt_root_symlinks.py`, `test_bootstrap_after_bump.py`, `test_family_bump_remove_rollback.py`. The seventh, `test_adopt_reencoded_files.py`, came with #743 and is converted in fix round 1. `test_clone_just_pushed.py:96` stays, as the issue says: it is the helper's own test. - **The other literals.** Each sets a count of its own and is laid over `os.environ`, where the session's two settings and the Windows job's two sit, so each replaced them (`NO_MAINTENANCE` put the two maintenance keys back and dropped the job's two). I read each; each is converted, none left: - `test_family.py`: `ALLOW_FILE_PROTOCOL` is now `allow_file_protocol()`; its two importers (`test_siblings_family_name.py`, `test_family_siblings_placement.py`) and the four uses in `test_family.py` call it. - `test_family_dry_run.py`: `NO_MAINTENANCE` is a tuple of entries, appended by `isolated()`. - `test_non_utf8_git_reads.py`: `QUOTE_PATH_OFF` is a tuple of entries, which `tool_env()` appends; `user_name()` appends too. - `test_adopt_empty_symlinks.py`: `SYMLINKS_OFF` is `symlinks_off()`. - `test_adopt_root_symlinks.py`: `SYMLINKS_OFF` and `SYMLINKS_ON` are gone; the four tests that set the first call `links_as_files` and the two that set the second call `told_git`, which already appended. - one-off environments: `test_scaffold_reuse_seeded_legs.py`, `test_adopt_split_branch.py`, `test_scaffold_free_text_values.py`, `test_adopt_plan_yaml_quoting.py`, `test_setup_project_py.py`, `test_setup_sh.py`, `test_doctor_next_is_a_command.py`. - The two tests that build a held environment on purpose to see it kept (`test_adopt_git_reads.py`, `test_clone_just_pushed.py`) now build it with `appended_to_command_scope({}, ...)`. Measured with a probe on a throwaway bare remote after a push (18 loose objects): the hand-built options and `clone_just_pushed`'s give the same clone, one pack and no loose object, with `upload-pack` and `pack-objects` in `GIT_TRACE`, and `origin` keeps the plain path. A plain local clone of the same remote keeps all 18 loose objects. For a clone that is not recursive the helper's `insteadOf` rewrite is inert, because no submodule is cloned; the helper adds one place for the options, not a different read. - `test_bump_leg_dry_run_truth.py:98`, the private bare copy of the session's spec remote: a session remote. `cloned_just_pushed(..., bare=True)`; the fixture now asserts the copy is bare. - `test_bump_leg_dry_run_truth.py:110`, `private_work`, a clone of that private remote, which tests push tags into: `cloned_just_pushed(..., recurse=False)`. - `test_doctor_legs_members_next.py:1828` and `:1929`, `Ghost`, a clone of the first member's checkout in the test's own holder copy: not a remote, and the hand-built `-c protocol.file.allow=always` was dead weight on a direct clone. Both call `clone_member`, one helper over `cloned_just_pushed(..., recurse=False)` that asserts the legs stay empty. - `test_family_bump_to_checks.py:235`, a clone of `a_private_remote`, a bare repository filled by a push: `cloned_just_pushed(remote, work, "main", recurse=False, env=env)`. - `test_doctor_pin_restore_merges.py:714` is **left as it is**: a `--filter=blob:none` clone of a file URI. The helper takes a path, and has no way to pass `--filter`. - `test_family_dry_run.py`: the comment on `NO_MAINTENANCE` says what the entries are for, the gits `isolated()` hands an environment to, whose global file `excluding()` replaces, and that the session fixture covers the module's own setup commits, which pass no `env=`. - `tests/conftest.py`: the two i645 replacements, "repacks once its estimate, taken from `objects/17` alone, is above 256 loose objects, so two loose objects there are enough", and the `objects changed` sentence. **#713's body holds the matching sentence and needs the same two replacements; I cannot edit a body.** `tests/test_no_auto_maintenance.py` has 17 new cases (28 in the file, 11 before): - `test_a_pasted_line_keeps_the_entries_the_environment_holds_at_the_call`: the environment holds two entries as the Windows job does, then the two settings as the session fixture appends them, set after the doctor tests were imported. `git config --show-origin --list` through `paste()` lists the four, in order, then `protocol.file.allow`, and reads both settings from the session's global file. - `test_no_test_sets_a_git_config_count_of_its_own`: reads every `tests/*.py` but `conftest.py` with `ast` and fails a `GIT_CONFIG_COUNT` set to a number (dict, keyword, `setenv`/`setdefault`/`putenv`, subscript) and an `appended_to_command_scope` call handed a dict written out. Three files are allowed the ones that build a held environment on purpose, with a count each in `OWN_COUNTS`. - 15 cases that hold the scan itself: seven spellings it must see, eight it must leave (`env.get("GIT_CONFIG_COUNT", "0")` is a read). Runs at the head `b6e15591` through `run9.sh` (the real `gh` unreachable), all passed, none skipped: - `test_no_auto_maintenance.py`, `test_doctor_legs_members_next.py`, `test_doctor_next_is_a_command.py`: 208 passed. - `test_clone_just_pushed.py` and `test_doctor_validator_rows.py` (it imports the doctor file), with those three: 256 passed, run one commit earlier (the later commit only moved two blank lines and reworded one failure message). - The sweep files: `test_adopt_git_reads.py`, `test_adopt_nested_attributes.py`, `test_adopt_root_symlinks.py`, `test_adopt_ignored_shape_files.py`, `test_bootstrap_after_bump.py`, `test_family_bump_remove_rollback.py`, `test_adopt_empty_symlinks.py`, `test_adopt_split_branch.py`: 445 passed. - `test_non_utf8_git_reads.py`, `test_family.py`, `test_siblings_family_name.py`, `test_family_siblings_placement.py`, `test_scaffold_reuse_seeded_legs.py`, `test_scaffold_free_text_values.py`, `test_adopt_plan_yaml_quoting.py`: 999 passed. - `test_family_dry_run.py`, the five files that import its `isolated` (`test_family_bump_to_a_tag.py`, `test_family_bump_to_checks.py`, `test_family_bump_to_local_only.py`, `test_family_refuse_up_front.py`, `test_family_remove_gitmodules.py`) and `test_bump_leg_dry_run_truth.py`: 216 passed. - `test_setup_project_py.py`, `test_setup_sh.py` and their importers (`test_adopt_shape_pin_and_work_dir.py`, `test_dry_run_asks_remotes.py`, `test_clone_url_bytes.py`, `test_setup_by_hand_commands.py`): 313 passed. - `test_repo_hygiene.py`, `test_agents_md_codes.py`, `test_cli_reference.py`, `test_doctor_pin_restore_merges.py`, `test_git_helper_identity.py`: 172 passed. `render-cli-reference.py --check` is ok. **On the base (`deb029c9`)**, with the new `test_no_auto_maintenance.py` copied in: 2 of the 17 fail, the paste case and the scan (it finds 12 files). The other 15 hold the scan itself and pass on any tree. Each of the new hunks was broken in a throwaway and run against its tests: - `paste_env`: a literal count of 1, a snapshot taken at import, and `appended_to_command_scope({}, ...)`: each fails the paste case, and the first and third the scan too. - The 18 sites that call `appended_to_command_scope(os.environ, ...)` (or `env`), each with `{}` in place of the environment: each fails the scan. Sites: the six of line 2, `allow_file_protocol`, `isolated`, `tool_env`, `user_name` (non-UTF-8), `symlinks_off`, `git_reads_github_from`, `github_at`, `user_name` (free text), `test_adopt_plan_yaml_quoting.py`, the two setup tests, `test_doctor_next_is_a_command.py`. - The scan: eight changes (ignore dicts, keywords, `setenv` calls, subscripts, `appended_to_command_scope({}`; flag `os.environ`; flag a read; allow fewer in `OWN_COUNTS`) each fail a scan case. - The clone sites: dropping `bare=True` from the private remote, `recurse=False` from `clone_member`, and `"main"` from the checks clone each fail. Two **survive as equivalents**: `recurse=False` in `private_work` (the spec leg has no `.gitmodules`, so a recursive clone of it is the same clone) and `env=env` in the checks clone (a clone reads no config the isolated environment changes). - The two comments cannot be mutated. `git merge-tree --write-tree` of this head against `origin/main` (`974d4827`), #748 (`deb029c9`), #713 (`979bb394`) and every `fix/*` and `l3-*` head on origin (46 refs): - **Clean:** #748, #713, and 24 of the 46 heads: 21 `fix/*` heads (every lane-5 PR but #723 and #756, and most of lane 4's) and `l3-677`, `l3-678`, `l3-rehearsal-v1.1.0`. - **`origin/main`: one file, `tests/test_adopt_empty_symlinks.py`** (four hunks), against #388's landing (`5c990708`). The parent #748 merges clean, so this arrives with the retarget. Resolve by keeping main's lines and my `symlinks_off()`. Two more spots need the same change and git does not flag them: - `tests/test_adopt_empty_symlinks.py::test_the_link_given_a_target_adopts_end_to_end` uses `SYMLINKS_OFF`, which this change removes: give it the `off` parametrization. - `tests/test_adopt_reencoded_files.py::configure`, the seventh hand-built append, goes to `appended_to_command_scope(os.environ, pairs)`. - `tests/test_family_head_lost_gitmodules.py::on_disk` (on main) is one more hand-built append: `{**env, **appended_to_command_scope(env, [("core.ignorecase", ignorecase)])}`. The scan does not read for hand-built appends, so it is not required, but it is the same thing. I made that merge in a scratch tree (main at `bf1f099b`), resolved it as above, and ran the files this change touches that main also changed, with `test_adopt_empty_symlinks.py`, `test_adopt_reencoded_files.py`, `test_family_head_lost_gitmodules.py`, `test_no_auto_maintenance.py` and the files that import `isolated` or the doctor helpers: 653 and 245 passed, in two runs. The scan finds no other file of main's with a count set by hand. - **#723 (`fix/663-non-utf8-name-tests`):** `tests/test_non_utf8_git_reads.py`, the `from conftest import` block, where both add names. - **`l3-732`, `l3-733`:** `tests/test_doctor_legs_members_next.py` (the `from conftest import` block) and `tests/test_adopt_empty_symlinks.py`. - **19 more** (`fix/664`, `fix/729` (#756) and 17 `l3-*` heads): `tests/test_adopt_empty_symlinks.py` only, because they carry main's #388 as main does. - **Not a textual conflict, but it breaks on merge:** `l3-785-non-ascii-paths` imports `PASTE_ENV` (`tests/test_doctor_non_ascii_paths.py:47` and `:223`); it must call `paste_env()`. #812 (`fix/693-local-remote-dir`) landed first, with `tests/test_local_remote_dir_urls.py` importing `ALLOW_FILE_PROTOCOL` (`:51`, `:349`). This PR's retarget merge makes it call `allow_file_protocol()` (rv818 MUST 1). - A pasted `git submodule update --init` still clones a local path: `paste()` takes no remotes, so no `insteadOf` rewrite reaches it (rv662's "build it from `read_through_git`"). It runs under the session's maintenance settings now, which is what the issue asks. - Three more appends built by hand beside `appended_to_command_scope`, in files this base does not hold: `tests/test_family_head_lost_gitmodules.py:97` (main), `tests/test_family_bootstrap_gate_order.py:135` (#682), `tests/test_family_add_spellings_remedies.py:137` (#767). The scan reads for counts set to a number, not for appends worked out by hand. Review: lane 2's independent review (issuecomment-6236874291) and its re-verification of eaa72d9, READY (issuecomment-6238781272); its new SHOULD is taken in #848, and 4a553e1 merges main over eaa72d9. CI (Brett Heap's ruling 27): H0 is the head 4a553e1 itself, which passed tests, tests-git-bash and SonarCloud; main has touched none of its files since its base fc86720; and the local gate passed on its predicted squash tree 9d6c099 over main d8b8eb3: the adopt_empty_symlinks, adopt_git_reads, adopt_ignored_shape_files, adopt_nested_attributes, adopt_plan_yaml_quoting, adopt_reencoded_files, adopt_root_symlinks, adopt_split_branch, bootstrap_after_bump, bump_leg_dry_run_truth, clone_just_pushed, doctor_legs_members_next, doctor_next_is_a_command, family, family_bump_remove_rollback, family_bump_to_checks, family_dry_run, family_siblings_placement, local_remote_dir_urls, no_auto_maintenance, non_utf8_git_reads, scaffold_free_text_values, scaffold_reuse_seeded_legs, setup_project_py, setup_sh, siblings_family_name, repo_hygiene, agents_md_codes, cli_reference sets, 2359 passed. Landed on Brett Heap's word of 2026-10-17T21:47:28Z, in lane 5's session (posted on #473 as issuecomment-6241758069): "merge lane 5's v1.1.1 PRs once reviewed and green". Closes #758 Lane: openreposhape-5 (openRepoShape-5) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Brett Heap and others added 2 commits October 16, 2026 14:20

brettheap marked this pull request as ready for review October 17, 2026 12:35

Merge branch 'main' into fix/693-local-remote-dir

b5a6c6a

sourcery-ai Bot approved these changes Oct 17, 2026

View reviewed changes

brettheap merged commit cbff780 into main Oct 17, 2026
5 checks passed

brettheap deleted the fix/693-local-remote-dir branch October 17, 2026 12:38

brettheap mentioned this pull request Oct 18, 2026

Tracking: setup, scaffold, family and the shared helpers, creating projects and families #459

Open

14 of 17 tasks

Navigation Menu

Repository navigation

brettheap commented Oct 17, 2026 •

edited by sourcery-ai Bot

Loading

Uh oh!

sourcery-ai Bot commented Oct 17, 2026

Interacting with Sourcery

Customizing Your Experience

Getting Help

Uh oh!

brettheap commented Oct 17, 2026

Uh oh!

brettheap commented Oct 17, 2026

Uh oh!

sourcery-ai Bot commented Oct 17, 2026

Uh oh!

sonarqubecloud Bot commented Oct 17, 2026

Uh oh!

sourcery-ai Bot left a comment

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

brettheap commented Oct 17, 2026 • edited by sourcery-ai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What changed and why

Tests

What it cannot tell

Shared files

Summary by Sourcery

Uh oh!

sourcery-ai Bot commented Oct 17, 2026

Reviewer's Guide

Sequence diagram for safe local member URL validation

Flow diagram for POSIX-quoted README clone URLs

File-Level Changes

Assessment against linked issues

Possibly linked issues

Interacting with Sourcery

Customizing Your Experience

Getting Help

Uh oh!

brettheap commented Oct 17, 2026

The issue's lines

Findings, ranked

1. Must change: add's remedy says "a directory … made anew", but add reads <Project>.git from that directory

2. Must change (PR body): on Windows an ordinary local path is now quoted

3. Should change: "with no character escaped" is false for the PR's own it's row

4. Nits

Correctness checks that passed

Tests

Not verified

For the checklist (outside #693)

Uh oh!

brettheap commented Oct 17, 2026

Uh oh!

sourcery-ai Bot commented Oct 17, 2026

Uh oh!

sonarqubecloud Bot commented Oct 17, 2026

Quality Gate passed

Uh oh!

sourcery-ai Bot left a comment

Choose a reason for hiding this comment

Sourcery assessment

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

brettheap commented Oct 17, 2026 •

edited by sourcery-ai Bot

Loading

1. Must change: `add`'s remedy says "a directory … made anew", but `add` reads `<Project>.git` from that directory

3. Should change: "with no character escaped" is false for the PR's own `it's` row