required-regen: the hand-maintained collision refuses under its own remedy (and the traced defect does not exist)

[gunbai-bot]

required-regen: the hand-maintained collision refuses under its own remedy

THE FINDING I WAS SENT TO FIX DOES NOT EXIST, and the confirming dispatch is what
established that. The brief traced HAND_MAINTAINED_STAGE0_DIRS being consulted by one
of three call sites in required_regen_host.rs and predicted a live refusal on eight
hand-maintained basenames. Measured at main 90986d1:

claim_executor --required-regen --source-root dag --source-root src/v2
first_generation_equal=true planned=132 executed=132 declared_divergent=1 [main.rs]

No population refusal. The traced mechanism is not merely unobserved, it is
UNREACHABLE: it requires the emitter to produce a file INTO a hand-maintained
directory, and emitted Rust filenames are flat by construction --
v1.compiler.emit_core_support module_to_filename is split(".") |> join("_") under
rust_source_root(). A three-way disagreement over an uninhabited domain is a code
fact, not a defect. Scanning the last 95 failed witnesses.yml runs for the refusal
string found four instances, all flat generated names on branches that introduced
modules; mod.rs and index.rs appear in none of them, and #8691's own failure is an
ordinary drifted mirror on v1_compiler_emit_rust.rs. The eight-basename population is
retracted by its author.

WHAT IS REAL IS A DIFFERENT MECHANISM WITH A DAMAGING REMEDY. Both compared
populations key on BASENAME and the committed walk enumerates only the top level, so
a module authored with a bare name equal to a file inside cli_run/ or
module_path_index/ emits a colliding basename that lands in emitted_not_committed.
The two remedies attached there both destroy hand-authored code: installing the
mirror overwrites it with generated bytes, deleting it removes it. A refusal whose
every offered remedy is wrong for the class it fires on routes a compliant author
into damage, and the only non-damaging response left is to ignore the wall -- which
is how a correct wall decays into advice, fastest for the authors who take it most
seriously.

So the collision gets its own typed cause and its own remedy (de-collide the module
name), and NOTHING IS EXCLUDED. Teaching is_hand_maintained_path about the directory
list -- the fix the brief proposed -- drops a genuinely generated surface out of the
comparison entirely: a wrong refusal traded for no refusal, the same false green as
making the committed walk descend, entered from the other side. DIRS is therefore
consulted at the seam that CLASSIFIES a refusal, never at one that excludes from it.

ZERO EXPOSURE, in the carrier and in those words. Zero collisions in the corpus, no
upstream guard found that would refuse such a module name (the nearest candidates in
gunbc.stage0_rust_source_lifecycle_scaffold are scoped to top-level stage0 paths and
cannot see a subdirectory file), and the state is reachable by one .dag module. This
is a wall for a state nothing currently inhabits, not the repair of a live defect --
recorded that way so the class is not read as closed.

EVIDENCE. witness_hand_maintained_collision_refuses_under_its_own_remedy asserts both
halves in one run: the collision refuses under MirrorMissingShadowsHandMaintainedSource
naming its directory, AND a genuine orphan in the same run still refuses under
MirrorMissingForEmittedSurface, with neither path appearing under the other's cause --
a change that silenced the collision by exclusion would take the orphan with it and go
green on a wall that no longer stands. Executed remotely at this tree: three witnesses
PASS; re-running the same witness with the collision re-classified as a plain orphan
(the pre-change behavior) FAILS.

Co-Authored-By: Claude Opus 5 (1M context) noreply@anthropic.com

[gunbai-bot]

V1 FREEZE ADMISSION, stated rather than assumed. This touches src/v1/stage0/src/required_regen_host.rs, which is inside the semantics-frozen v1 seed. The admission test is PURPOSE (gunbc.v1_maintenance_standing v1_seed_standing): a change is admitted when it serves the v2 self-host program. This one does. required-regen is the mechanism that keeps the committed stage0 mirror honest against a fresh emit, and that mirror IS the self-host path — a refusal on that path whose only offered remedies damage hand-authored source is a defect in the instrument the self-host program is steered by. The change adds no capability to the seed and no row to any of its growth surfaces: one refusal cause splits out of an existing one, and the seed's behaviour on every population it already handled is byte-identical.

It is worth recording that the admission is not independent — the session that authorized this build is the one that reviewed it. So the reasoning is written down here to be checked rather than taken.

Last-write-wins finding is fixed (552b8be), and the carrier changed rather than the host working around it: hand_maintained_homes is now a List<String>, because with a single-home field the host has no correct choice available and insert last-write-wins is the only shape there is. mod.rs under both cli_run/ and module_path_index/ now refuses naming both. Homes are sorted host-side so the refusal text is a function of the collision, not of the roster's authoring order. The witness carries a two-home row asserting the exact rendering, and the genuine-orphan separation half is unchanged and still holds; both PASS, executed remotely at this tree.

— sent from quick-heron-894