Adopt the four remaining upstream CHECK_EPOCH hardening items from
sneak/prompts #26, closing the gap #85 left open deliberately.
Fail closed on a missing value. Each check stage now asserts
`[ -n "$CHECK_EPOCH" ] || exit 1` before running anything. An unset ARG
is an empty string and an empty string is a stable cache key, so the
second and every later bare `docker build .` on an unchanged tree
replayed all three check layers, executed nothing, and still exited 0 --
and `docker build .` is the command REPO_POLICIES.md names verbatim as a
thing that must be green, so the documented command was precisely the
one that lied. Failed steps are never cached, which is what makes the
guard fire on every invocation rather than once.
Expand the epoch into each check command rather than leaving it a bare
declaration, so the cache miss does not depend on BuildKit's
unreferenced-ARG handling staying as it is, and so the value appears in
the build log where a reader can see the layer was keyed fresh.
Make the epoch unique per invocation rather than per second:
`epoch="$(date +%s%N)$$"`. `%N` alone is not enough, since busybox drops
it silently and exits 0, handing back second granularity with no
warning; `$$` differs between concurrent invocations regardless. The
bare-assignment form is kept on purpose -- inlined into an argument, a
failing substitution does not abort under `set -eu` and would yield an
empty constant epoch, restoring the exact false green this prevents.
Pass the same fresh value from script/docker. It is not the CI gate, but
local builds are almost always warm, so it was the likelier fooling in
practice, and two entrypoints disagreeing about whether the tree is
green is worse than either being wrong alone.
The ARG placement from #85 is unchanged, below apk add, COPY go.mod
go.sum and go mod download, so dependency layers still cache and the
build is not cold. Verified by negative control rather than inspection;
measurements are recorded once, in the PR verification comment.
.golangci.yml, the lint-stage FROM line and its digest, script/lint,
REPO_POLICIES.md and .gitea/workflows/check.yml are untouched.
script/cibuild was a bare `docker build .`. On an unchanged tree Docker
served the check RUN layers from cache, so make fmt-check, make lint and
make test never executed - and the build still exited 0. Measured at
221ms with zero ok lines and every check layer CACHED, against 162s for a
real run. CI showed the same signature: 6 second "successes" on main.
An ARG CHECK_EPOCH now sits immediately above the check RUNs in both
stages - each stage declares its own, since ARG scope is per-stage - and
script/cibuild passes a fresh value per invocation. Dependency and module
layers sit above the ARG and still cache, so this does not make every
build cold.
The epoch is assigned before the build rather than inlined into the
--build-arg. Under `set -eu` a command substitution that fails inside an
argument does not abort the script: CHECK_EPOCH would become an empty
string, an empty string is a constant, a constant CHECK_EPOCH restores
the cached false green, and the guard would silently disarm itself while
still exiting 0. As a bare assignment, set -e catches a failing date and
no build starts.
The README and Dockerfile state the guarantee conditionally. It holds per
build context and CHECK_EPOCH value, and depends on script/cibuild
passing a fresh one - a bare `docker build .` with no --build-arg still
replays the check layers from the second consecutive run onward. That
residual gap is tracked in #91 along with the remaining upstream
hardening.
Verification is recorded once, in the PR's verification comment, rather
than restated with differing numbers in three places.
