Answers #44: the canonical Go make test target omitted -count=1, so Go replayed cached successful results and the target could exit 0 having executed no test. Every repository that copied it inherited that false green.
The change adds -count=1 to both go test invocations in the example in REPO_POLICIES.md, with a short paragraph saying why, and records the step in TODO.md. It defeats only the test-result cache, not the build cache, so it costs the suite's runtime and no recompilation. It is independent of the Docker layer cache that #26 addresses.
Rebased onto current main; the check is green. Repositories pick it up the next time each vendors the canonical files.
Answers https://git.eeqj.de/sneak/prompts/issues/44: the canonical Go `make test` target omitted `-count=1`, so Go replayed cached successful results and the target could exit 0 having executed no test. Every repository that copied it inherited that false green.
The change adds `-count=1` to both `go test` invocations in the example in `REPO_POLICIES.md`, with a short paragraph saying why, and records the step in `TODO.md`. It defeats only the test-result cache, not the build cache, so it costs the suite's runtime and no recompilation. It is independent of the Docker layer cache that https://git.eeqj.de/sneak/prompts/issues/26 addresses.
Rebased onto current `main`; the check is green. Repositories pick it up the next time each vendors the canonical files.
Model: opus-5 (change); fable-5-1 (this description)
The canonical Go `test` target in `REPO_POLICIES.md` omitted `-count=1`, so
Go replayed cached successful results and the target could exit 0 having
executed no test. Every repo that copied it inherited the false green.
Both invocations get the flag; the rerun needs it so a failure is
reproduced rather than replayed.
Go's test-result cache sits below the Docker layer cache that issue #26
addresses. CHECK_EPOCH forces the RUN make test step to re-execute, but
GOCACHE baked into earlier image layers survives into the re-executed
step, so go test can still do no work. Record that these are two
separate defects so a fix for one is not recorded as covering the other.
Also correct the TODO.md Completed Steps date to the landing date, which
matches the REPO_POLICIES.md last_modified front matter.
Closes#44
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Answers #44: the canonical Go
make testtarget omitted-count=1, so Go replayed cached successful results and the target could exit 0 having executed no test. Every repository that copied it inherited that false green.The change adds
-count=1to bothgo testinvocations in the example inREPO_POLICIES.md, with a short paragraph saying why, and records the step inTODO.md. It defeats only the test-result cache, not the build cache, so it costs the suite's runtime and no recompilation. It is independent of the Docker layer cache that #26 addresses.Rebased onto current
main; the check is green. Repositories pick it up the next time each vendors the canonical files.Model: opus-5 (change); fable-5-1 (this description)
@clawbot unmergeable
clawbot referenced this pull request2026-09-03 20:28:09 +02:00
clawbot referenced this pull request2026-09-03 20:29:13 +02:00
3933e6bdfatoc937cd7a4c