reporter

Node's own spec reporter with the per-evaluation coverage summary held back, so the merged table the runner prints is the only coverage verdict a run shows.

Usage

Name the entry with --test-reporter, the way any custom reporter is named. The runner behind runProjectTests passes it by file URL and sends it to stdout, with Node's lcov reporter writing the raw report to a file beside it:

node --import @hyperfrontend/testing/hooks --test \
  --experimental-test-coverage \
  --test-reporter=@hyperfrontend/testing/reporter \
  --test-reporter-destination=stdout \
  'src/**/*.spec.ts'

The output is exactly what Node's spec reporter prints, test by test and suite by suite, minus the coverage report that would otherwise follow the summary.

Design

Node reports each evaluation of a module as a file of its own. A suite that calls jest.resetModules therefore produces one coverage entry per generation, and a summary that counts every line of the module once per evaluation while crediting only the evaluation that ran it. The runner prints a merged table instead, built from the lcov report by mergeLcov and renderCoverageTable, and printing both would show two different answers for the same run.

The reporter is an async generator that wraps an instance of Node's spec reporter: every event from the runner's stream is forwarded to it except test:coverage, and whatever the spec reporter renders is yielded as it comes. Nothing about pass, fail, skip, todo or diagnostic output is changed.