siblings.py (make siblings): the printed mv goes astray over an existing .holder/.folder or a dangling link, a lone surrogate or NUL in project: is a traceback, a url with NEL or U+2028 is cut, two warnings print paths raw #690

Goal: make siblings either clones each member where it says, or stops before cloning with a move that runs as typed, and no manifest or disk text ends it in a traceback or a forged line.

Filed by lane openRepoShape-5 on Brett Heap's word of 2026-10-16T12:17:02Z ("#459's family lines (~38)"), from checklist lines on #459 (comment 6192725913, lane 5's #578 lines; comment 6214512824, R454 O1 and O2; comment 6214631091, R432 O2). Lane 4 judged R454 O1/O2 and R432 O2 OPEN on 95780dd (cluster T11); re-read and probed on d4a7fab.

Pinned copy. templates/family-root/scripts/siblings.py is pinned in every holder (FAMILY_COPIED_VERBATIM, scripts/shape_materialize.py:523), so each holder takes the edit as one upstream-changed row: bundle it with the other pinned-file edits (#639, #648, #634).

Lane 3's #460 lines. Lane 3 now owns #460's lines, R605's six about siblings.py and the family-root README and AGENTS templates among them. Its fix of R605's template line may land first; rebase over it. R432's O2 is split: the siblings.py half is here, the doctor's submodule_config half is lane 3's.

Lines (line numbers at d4a7fab)

• siblings.py and family.py: make siblings asks about the family folder before it clones, clones each member on its tracking branch, and family.py siblings runs the holder's own copy (#543) #578: an existing staging name sends the folder moves astray. parent_folder_warning prints mv/mkdir -p/mv through <Family>.holder (templates/family-root/scripts/siblings.py:1000) or <Family>.folder (:1012) and checks neither for existing. Probed on Linux with <t>/InkRouter.holder already a directory: the three printed lines all exit 0 and leave the holder at .../InkRouter/InkRouter/InkRouter/family.yaml, one level too deep (on a disk that ignores case, with .folder, R578's writer measured InkRouter/inkrouter/InkRouter). The next run stops again with a rename that works. Direction: where the staging path exists (os.path.lexists), pick another name (a numbered one) or stop with a sentence naming it; test both names.
• siblings.py and family.py: make siblings asks about the family folder before it clones, clones each member on its tracking branch, and family.py siblings runs the holder's own copy (#543) #578: a dangling link named after the family breaks the staged move halfway. not wanted.exists() follows the link (siblings.py:994-995), so the staged move is printed; probed: <t>/InkRouter a link to nothing, the holder at <t>/t/InkRouter: mv to <Family>.holder exits 0, then mkdir -p <t>/InkRouter fails File exists, exit 1. Direction: os.path.lexists, and say a dangling link of that name is in the way (move it aside), as _checked_mount_path does for a dangling link at a mount path.
• R454 O1, a real make siblings over a member whose project: holds a lone surrogate or a NUL ends in a traceback. place hands project to git clone through str(target) (siblings.py:577-581). Probed on main with a row project: "IR\ud83dRS": exit 1, UnicodeEncodeError: 'utf-8' codec can't encode character '\ud83d' in position 127: surrogates not allowed, at siblings.py:577. R454 measured the NUL as ValueError: embedded null byte at the same site (not re-run); the dry run prints both spelled. This is assembly-root bootstrap.py: a lone surrogate or NUL in tracking_branch (or a leg's branch:) ends make bootstrap in a subprocess traceback in step (a), and validate-manifest.py says manifest ok, so no tool names it #455's class. Direction: a per-member FINDING before git runs, in the words tracking_branch already uses for a value that does not print (:342-386, isprintable), and the other members still go on.
• R454 O2, parent_folder_warning prints the holder's path and the parent folder RAW on stderr. The prose lines {root} should sit in a plain folder named {family} (siblings.py:974) and would land in {folder} (:978-979) print raw, and {family} too, (spelled, as place does at :569, spells a manifest value; R454 notes the summary line under the warning spells {family}); the first warning line already spells the folder name (!r, :972). Under a parent folder named x<ESC>[31my, stderr holds the raw ESC (stderr's backslashreplace means no traceback). The mv line stays as typed (shlex.join, :1023): quoting a control character in a pasted command is family.py and siblings.py print commands with unquoted paths, so under a folder with a space the member-git-dir-cached remedy's rm -rf would delete the wrong directory, and add's push line fails #486's question. Direction: spell the prose values with spelled(...); leave the mv lines.
• R432 O2, submodule_urls truncates a url holding a UTF-8 line-boundary character. proc.stdout.splitlines() over a text-mode git config -f .gitmodules --list (siblings.py:247) splits on U+0085 (NEL), U+2028, U+2029, U+001C to U+001E, VT and FF, where git uses the whole url: both this and the doctor's submodule_config read <remotes>/caf of caf<NEL>.... Read from the code (str.splitlines), not re-run. Direction: split on "\n" only, or read --list -z and split on NUL; the doctor's half is lane 3's (Tracking: shape-doctor.py, reporting a repository's compliance #460).
• siblings.py and family.py: make siblings asks about the family folder before it clones, clones each member on its tracking branch, and family.py siblings runs the holder's own copy (#543) #578: a tracking_branch: that names only a tag on the member's remote gives a detached clone. place clones with --branch=<name> (siblings.py:580), which git clone takes a tag for, and its line says "on , its tracking branch" (:569); branch_of then reads detached. Read from the code, not run. Direction: where the clone's HEAD is detached though a branch was asked, say the name is not a branch of the remote (a tag), in the placement line.

Overlaps: lane 3's R605 template lines (#460), which may land first. No open PR edits siblings.py; lane 5's #648 edits the family-root bootstrap.py and validate-family.py, not this file. scripts/family.py's siblings subcommand runs this file from the standard's checkout and is untouched.

Mark the six lines → this issue on #459, and tick them there when this closes.

🤖 Generated with Claude Code