All five gates now live in the check domain. canvas-version produces
byte-identical output to the original on the live tree.
It is the first real consumer of core/process.run. The git calls pass
check=False deliberately: a git failure here is not an error to report but a
signal that there is nothing to compare, since a fresh clone with no remote is
a legitimate state rather than a broken one. The argv-list and missing-binary
guards still apply.
Its two skips are kept distinct from its pass. NO_BASE and DIFF_FAILED exit 0,
as does CLEAN — but only CLEAN means the gate actually looked at something.
Collapsing them would hide a gate that had silently stopped running, which for
this check in particular is the exact failure it exists to prevent.
Found a second rich path while a NameError was rendering as a full-width
box-drawn traceback: typer's pretty-exception handler is a different mechanism
from rich_markup_mode, and setting one does nothing about the other. Same log
pollution T-1259 thought it had closed, arriving through another door and
landing in the worst place — a hook log at the moment something has already
gone wrong. pretty_exceptions_enable=False now on the root and on every domain
built by cli.domain().
test_canvas_version_check.py moves with the code it guards. It had been loading
the extensionless script through a SourceFileLoader and reaching canvas_sources
by sys.path insert, both only because tooling/ was not importable. Second
instance of that debt evaporating on contact. What it asserts is unchanged,
which is the point: diff_has_version_bump was kept pure in the port so its six
properties still hold without constructing git history.
Also restores an import the check router dropped in T-1267 when it moved to
cli.domain() — caught by running the command rather than by reading it.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
project.yaml's version is the Atlas disk cache's only invalidation signal, and
nothing enforced that changing canvas GENERATION also moved it. It broke five
times -- 0.4.2 lake_margin_q, 0.4.3 coast_warp_px, 0.4.4 the extent inversion,
0.4.5 the Global sentinel, 0.4.6 one-course-per-river -- each bumped only after
someone noticed a wrong map. The failure is invisible to its author: it needs a
warm cache to reproduce, so a cold checkout looks fine. T-1239 is the last one,
and it took eight days.
tooling/canvas_sources.py is the path registry; tooling/check-canvas-version
rejects a push that touches those paths without moving project.yaml's version
line. Wired into the pre-push hook, `make check-canvas-version`, and, for the
parsing units, `make test-tooling`.
Verified against real history rather than a synthetic branch: run over
4e503c356 -- the commit that actually caused T-1239 -- the gate rejects and names
the three files. Run over the commits that DID bump (bdea71953, 39f0fd8c5, and
T-1239's own fix), it passes.
The registry is globbed, not hand-listed. step_canvas.rs imports ten sibling
modules and those import more, so a traced closure would be stale within a month,
and stale here is silent. It over-includes on purpose: a false positive costs one
bump and one round of cache misses, a false negative costs another week of a
wrong map -- the ticket's own ruling.
Two deliberate calls worth naming. The registry includes ITSELF, which closes the
narrowing hole: remove a path and change that same path in one push, and the gate
still fires because the registry file is in the set. And there is no override
flag -- it would be reached for exactly when someone is certain their change is
harmless, which is the reasoning behind all five regressions.
Version bumped 0.4.6 -> 0.4.7 with NO canvas-generation change: self-inclusion
means adding the registry trips its own rule. Spent rather than special-cased,
because the first exception is how a rule like this dies.
The units cover the property no branch run can show -- that editing project.yaml's
comment block, which quotes old version NUMBERS directly above the field, is not
a bump -- plus a registry-coverage test naming the files each of the five known
regressions touched, so a future narrowing past them fails loudly.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
jpmschweitzerandClaude Opus 5