0.42.0

🐛 Bug Fixes

• A failing set_up_before_script/set_up now fails every test in the file with the hook's error, keeps the totals consistent, and lets the suite continue — a strict test file can no longer abort the whole run mid-suite (#836)
• Run-mode flags (--parallel, --simple, --strict, --retry, --seed, report paths, …) no longer leak through the environment into nested bashunit runs, so a script under test that calls bashunit gets default behavior; BASHUNIT_*=… configuration is unchanged (#834, #837)
• install.sh fails fast: a failed download, clone, build or copy aborts with a clear error instead of reporting success, and a failed beta clone no longer runs build.sh in the caller's directory (#840)
• ./build.sh --verify exits non-zero when the built binary fails the suite, and the verification run no longer crashes mid-suite from tests resolving repo paths against the build folder (#834)
• ./bashunit bench works again from a repository checkout (the dev entrypoint never sourced src/benchmark.sh) (#834)
• Snapshot placeholders (::ignore::) now work on systems without perl; multi-line placeholders still need perl (#823)
• Runs no longer leak a scratch directory under $TMPDIR/bashunit/run/ — it is removed on exit, including --version/--help, subcommands and Ctrl-C (#811)
• bashunit::helper::get_function_line_number no longer disables extdebug for its caller (#808)
• --test-timeout no longer intermittently reports a fast test as timed out; the watchdog signals by pid and skips a test that already completed

✨ Improvements

• watch subcommand no longer fails when neither inotifywait nor fswatch is installed — it falls back to pure-shell polling (interval via BASHUNIT_WATCH_INTERVAL, default 2s) instead of exiting (#779)
• Shell tab-completion scripts for bash and zsh under completions/ (subcommands, test flags with value hints, assertion names), kept in sync by an anti-drift CI test (#778)
• --rerun-failed (BASHUNIT_RERUN_FAILED) replays only the previously failing tests, recorded in .bashunit/last-failed; composes with --filter/--tag/--parallel and falls back to the full suite when empty (add .bashunit/ to .gitignore) (#776)
• Optional nightly coverage.yml workflow publishes a shields.io coverage badge from --coverage over the unit suite; schedule/manual only, never gates merges (#754)
• --jobs auto / -j auto caps parallel concurrency at the CPU core count (portable across Linux/macOS/BSD); the default stays unlimited (#766)

🛠️ Changes

• build.sh hardened: runs under set -euo pipefail, derives the embed list from the entrypoint's source order (single source of truth), guards against duplicate embeds and missing doc markers, drops eval, and gates every build behind bash -n (#834)
• bashunit doc no longer forks an echo | sed pipe per line of the assertion docs: a single awk pass prints the same bytes in ~50ms instead of ~5s (#832)
• Multi-file runs are no longer quadratic in file count; bashunit's own 63-file unit suite: ~64s -> ~22s sequential, ~26s -> ~7s parallel (#829)
• Major performance work with no behaviour change (near fork-free hot paths, cached snapshot/--tag scans, single-pass failure rendering). On bash 3.2: 100x10 assert_equals ~1.50s -> ~0.76s, 500 snapshot assertions ~7.5s -> ~3.0s, 100 tagged tests ~2.92s -> ~0.68s, acceptance suite ~61s -> ~17s (#761-#764, #772-#775, #798, #801-#807, #809, #810, #813, #817)
• Per-test timing now defaults to auto (BASHUNIT_SHOW_EXECUTION_TIME=true|false|auto): shown only when the clock is fork-free, avoiding perl forks on bash 3.2; --profile/--verbose/reports still measure (see adrs/adr-008-auto-skip-per-test-timing.md) (#765)
• assert_equals/assert_same failures with multiline values now render a git word-diff below the header (requires git, opt out with BASHUNIT_NO_DIFF=true, respects --no-color); machine reports keep the raw values (#777)

👥 Contributors

• @Chemaclass

Checksum

SHA256: a0e39761363d8b6876059cd5927cd4bed1b578be616c5490a8bf4102284a308c

Full Changelog: 0.41.0...0.42.0

0.41.0

✨ Improvements

• --retry <n> / BASHUNIT_RETRY re-runs a failed test up to N times; passes if any attempt passes, annotates retried tests, and works with --parallel and --stop-on-failure (#737)
• --random-order with --seed <n> / BASHUNIT_SEED randomizes test file and function order to surface inter-test coupling; prints the seed for reproducible replay and works with --parallel (#738)
• --shard <index>/<total> runs a deterministic, non-overlapping subset of the test files to split a suite across parallel CI runners; composes with --parallel (#739)
• --report-tap <file> writes a TAP v13 report to a file (complements the streaming --output tap) (#740)
• --report-json <file> writes results as JSON (summary counts + per-test records); no jq dependency (#741)
• assert_file_permissions <mode> <file> asserts a file's octal permission mode; portable across GNU/BSD stat (#742)
• assert_array_length <n> <array> asserts an array has exactly n elements (#743)
• assert_within_delta <expected> <actual> <delta> asserts a number is within a tolerance; supports floats (#744)

🐛 Bug Fixes

• watch subcommand failed with bashunit::watch::run: command not found in the released binary because src/watch.sh was missing from the build; it is now bundled (#735)

🛠️ Changes

• Faster test execution by removing subprocess forks from hot paths (no behaviour change)

👥 Contributors

• @Chemaclass

Checksum

SHA256: 146c9b1f5462633d40c377ab0548bbd1a720ce365f49ef6942621192b4d15f79

Full Changelog: 0.40.0...0.41.0

0.40.0

✨ Improvements

• --test-timeout <seconds> flag and BASHUNIT_TEST_TIMEOUT env var to abort a test that runs longer than N seconds, mark it failed and continue. Disabled by default; no external timeout needed, works on Bash 3.2+ (#721)

🐛 Bug Fixes

• A non-zero test exit no longer leaks into later tests in the same file (per-test exit code is now reset)
• Coverage now counts backslash line-continuation lines as covered (#722)
• Spying/mocking the printf builtin no longer breaks coverage collection: the buffer is flushed with builtin printf (#724)

🛠️ Changes

• URLs now point to the new primary domain bashunit.com (old bashunit.typeddevs.com redirects)
• Docs site deploys to GitHub Pages on the bashunit.com custom domain (deploy-gh-pages.yml)

Removed

• Weekly-downloads chart on the docs homepage (non-portable data source)

👥 Contributors

• @Chemaclass
• @truffle-dev

Checksum

SHA256: 0ee0474803b6e88e7dfa4f4c2486ea8f8e53fd8324134a9fe604ec3df8b5e72c

Full Changelog: 0.39.1...0.40.0

0.39.1

🐛 Bug Fixes

• bashunit upgrade resolved the floating v0 Action tag as the latest version, breaking upgrades; it now only considers exact-version tags

👥 Contributors

• @Chemaclass

Checksum

SHA256: 0cb0153ebcf7198371051332b8e8fce36c47b514eef9bb362148f404837c2e82

Full Changelog: 0.39.0...0.39.1

0.39.0

✨ Improvements

• GitHub Action args input: when set, runs bashunit <args> after installing, so a workflow can install and run the suite in a single step
• Floating major tag for the GitHub Action: the release process now force-moves v0 to each release, so workflows can pin TypedDevs/bashunit@v0 to track the latest release within a major (#700)
• bashunit init now scaffolds a .github/workflows/tests.yml CI workflow using the official action (existing files are left untouched) (#702)
• --report-junit <file> flag as an alias of --log-junit, for naming parity with --report-html (#705)

🛠️ Changes

• The GitHub Action's version input now defaults to the version pinned at the action ref instead of latest, so pinning the action by SHA pins a visible bashunit version
• install.sh now verifies the release checksum by default (set BASHUNIT_VERIFY_CHECKSUM=false to opt out); it soft-skips with a warning when a checksum asset is unavailable unless verification was explicitly requested (#703)
• --log-gha annotations now include the failing test's line (::error file=…,line=…), so they pin to the exact line in a pull request's "Files changed" tab (#704)

👥 Contributors

• @Chemaclass
• @RevanProdigalKnight

Checksum

SHA256: 37c228533b5e565147b4daaadfb2a1fe5227aa268de7dc41dd967fa85cab288f

Full Changelog: 0.38.0...0.39.0

0.38.0

✨ Improvements

• Official TypedDevs/bashunit GitHub Action: composite install action, pinnable by commit SHA for immutable installs (version, directory, add-to-path, verify-checksum inputs; path, version outputs) (#695)
• install.sh sha256 checksum verification via BASHUNIT_VERIFY_CHECKSUM=true, validating the download against the release checksum asset (#695)

🐛 Bug Fixes

• install.sh fails loudly (non-zero exit, no stub binary) on a failed download and retries transient failures, instead of silently reporting success (#695)
• install.sh creates nested target directories (mkdir -p) (#695)

👥 Contributors

• @Chemaclass

Checksum

SHA256: b1364d4874e61bc8cee4d9dea15844badebdbcc855d4d5b5439c9f4ae8b2204c

Full Changelog: 0.37.0...0.38.0

0.37.0

✨ Improvements

• --profile reports the slowest tests (BASHUNIT_PROFILE_COUNT, default 10) (#678)
• Readable snapshot diff when git is unavailable (#679)
• Failure output shows the test file:line (#680)
• .bashunitrc project config file (#681)
• Killed tests report the cause: timeout, SIGINT, SIGKILL/OOM, SIGTERM (#683)

🐛 Bug Fixes

• --strict skips Bash 3.0's broken pipefail (pipefail only on Bash >= 3.1)
• bashunit watch forwards --filter and other flags correctly (#682)
• learn and coverage use mktemp -d for temp directories
• parallel::cleanup refuses to rm -rf outside */bashunit/parallel/*

Internal

• Global-slot return pattern for hot-path helpers; _BASHUNIT_SPY_* namespace (#674)
• Windows acceptance CI: matrix shards now cover every test file (bashunit_test.sh, coverage_*, install, mock were silently skipped) and job labels show only the shard name

⚡ Performance

• Faster runtime-error detection: single case glob (#668)
• Cache coverage-enabled flag in hot path (#664)
• Parallel runner uses wait -n on Bash 4.3+ instead of polling (#667)
• Hot-path result helpers return via global slots, dropping a fork per call (#662)
• Drop grep/random_str forks in generate_id and normalize_variable_name (#663)

👥 Contributors

• @Chemaclass

Checksum

SHA256: 18f1e8354213001b80e37c722b8520ebe26ce10fce11cb20ee471ddc96a21b11

Full Changelog: 0.36.0...0.37.0

0.36.0

🐛 Bug Fixes

• bashunit upgrade exits non-zero with a clear error when the download fails (no more false success message)

✨ Improvements

• --show-output displays captured test output on assertion failures (#637)
• npm registry distribution: npm install -g bashunit (#244)
• bashunit::env::supports_color and bashunit::io::clear_screen helpers (#247)
• LCOV reports now include FN, FNDA, FNF and FNH function records, consumed by genhtml, Codecov and Coveralls
• LCOV reports now include BRDA, BRF and BRH branch records for if/elif/else chains and case patterns (see adrs/adr-007-branch-coverage-mvp.md)
• BASHUNIT_COVERAGE_SHOW_FUNCTIONS=true adds a per-function coverage block to the text report
• BASHUNIT_COVERAGE_SHOW_UNCOVERED=true adds an "Uncovered Lines" block to the text report, with consecutive line numbers compressed into ranges

🛠️ Changes

• Docs moved into their own npm workspace under docs/ (use cd docs && npm ci or make docs/install)
• Pre-commit hook skips the test suite when no shell files are staged
• ANSI escapes route through _BASHUNIT_COLOR_* constants (#247)

⚡ Performance

• Faster coverage report generation: single-pass file scan, native bash regex, cached file stats (#636)

Internal

• bashunit::runner::run_test split into nine named helpers (no behavior change)
• release.sh handles docs/package.json as a first-class release file
• Structural test guards root package.json against regaining dependencies or scripts

👥 Contributors

@Chemaclass

Checksum

SHA256: 29bffc3d492296c859640f7425a66f7ae0549e667e5265e3e84859a828edbf64

Full Changelog: 0.35.0...0.36.0

0.35.0

✨ Improvements

• bashunit::spy accepts an optional exit code or custom implementation function (#600)
• Assert functions accept an optional trailing label to override the failure title (#77)
• --fail-on-risky flag and BASHUNIT_FAIL_ON_RISKY env var treat no-assertion tests as failures (#115)
• --log-gha <file> flag and BASHUNIT_LOG_GHA env var emit GitHub Actions workflow commands so failed, risky and incomplete tests show up as inline PR annotations (#280)
• assert_exec accepts --stdin, --stdout-contains, --stdout-not-contains, --stderr-contains and --stderr-not-contains flags to test interactive prompt commands and substring output (#301)

🛠️ Changes

• Parallel test execution is now enabled on Alpine Linux (#370)

🐛 Bug Fixes

• Dim/faint labels now render as gray (SGR 90) so keywords like Expected and Tests: stay colored in GitHub Actions logs (#323)
• Syntax error in a test file now fails the suite instead of passing silently (#220)
• --stop-on-failure now stops on runtime errors such as command not found or illegal option (#383)
• Spying on echo or printf no longer hangs via infinite recursion (#607)
• LCOV and HTML coverage reports no longer produce empty output under set -e (#618)
• clock::now handles EPOCHREALTIME values that use a comma decimal separator
• Invalid .env.example coverage threshold entry; CI now copies .env.example to .env so config parse errors are caught
• Coverage no longer counts case patterns with trailing comments (e.g. *thing) # note) or loop terminators with redirections/pipes (e.g. done < file, done <<<"$var", done | sort) as executable lines (#634)
• assert_true and assert_false now report empty strings as assertion failures instead of trying to execute them

👥 Contributors

• @Chemaclass
• @SauronBot
• @antonio-gg-dev
• @objctp

Checksum

SHA256: bfe9f69bda77034234a38f26182cc34e1f7648d846dab57a513a14cf91977544

Full Changelog: 0.34.1...0.35.0

0.34.1

✨ Improvements

• Add risky test detection for tests with zero assertions (shown as warning, does not fail)

🐛 Bug Fixes

• Fix source of non-existent file in set_up() silently passing all tests (#611)
• Fix set_up running before strict mode — unbound variables in hooks now detected with --strict
• Fix source failure in tear_down(), set_up_before_script(), and tear_down_after_script() silently passing
• Add missing runtime error patterns: ambiguous redirect, integer expression expected, too many arguments, value too great, not a valid identifier, unexpected EOF

👥 Contributors

• @Chemaclass

Checksum

SHA256: b620423b0fb201ff1c8d2c04735061cdea4237be84007348218a7e8d8522d8d0

Full Changelog: 0.34.0...0.34.1