hack/select-e2e.sh and docs/agents/e2e-testing.md both explain the unit/e2e bats split by quoting a count: BATS_UNIT_FILES "keeps 60 files". Counted on current main, hack/*.bats is 94 files with three hack/e2e-*.bats among them, so the filter keeps 91.
The number appears twice in each file, first as "the 60 files it keeps" and again as "those 60 files" in the script and "never executes one of the 60" in the doc. Four sites.
Correcting it to 91 buys about one pull request. The count was accurate when it was written, nothing checks it, and it goes stale the next time anyone adds a bats file under hack/. Both places quote BATS_UNIT_FILES := $(filter-out hack/e2e-%.bats,$(wildcard hack/*.bats)) directly next to the claim, and that expression is what defines the set; the cardinality on any particular day adds nothing a reader needs.
Dropping the number from all four sites leaves both explanations intact and removes the only part of them that can rot.
hack/select-e2e.shanddocs/agents/e2e-testing.mdboth explain the unit/e2e bats split by quoting a count:BATS_UNIT_FILES"keeps 60 files". Counted on current main,hack/*.batsis 94 files with threehack/e2e-*.batsamong them, so the filter keeps 91.The number appears twice in each file, first as "the 60 files it keeps" and again as "those 60 files" in the script and "never executes one of the 60" in the doc. Four sites.
Correcting it to 91 buys about one pull request. The count was accurate when it was written, nothing checks it, and it goes stale the next time anyone adds a bats file under
hack/. Both places quoteBATS_UNIT_FILES := $(filter-out hack/e2e-%.bats,$(wildcard hack/*.bats))directly next to the claim, and that expression is what defines the set; the cardinality on any particular day adds nothing a reader needs.Dropping the number from all four sites leaves both explanations intact and removes the only part of them that can rot.