Tests

test: add ShouldProcess and comment-based-help coverage sweeps, and fix the gaps they find #157

Workflow file for this run

.github/workflows/cross-platform-tests.yml at 0a81619

	name: Tests

	# Runs the Pester suite across every OS/PowerShell-edition combination the module
	# claims to support (PureStorageFlashBladePowerShell.psd1: PowerShellVersion = '5.1'),
	# so platform-specific bugs (e.g. a .NET marshaling difference between Windows and
	# Linux/macOS) are caught on every push/PR instead of only surfacing much later on
	# whichever CI job happens to run on a non-Windows runner.
	#
	# Two separate jobs rather than one shell-matrixed job: the `shell:` key on a step does
	# NOT have access to the `matrix` context (unlike `run:`, `env:`, or a job's own `name:`/
	# `runs-on:`) -- confirmed via `gh workflow run`: "Unrecognized named-value: 'matrix'.
	# Located at position 1 within expression: matrix.shell". In a workflow, `shell:` must be
	# a literal. (Inside a composite action `shell:` does accept `${{ inputs.* }}` -- but
	# still not `matrix`/`env` -- which is how .github/actions/install-test-modules serves
	# both editions from one implementation.)
	#
	# workflow_call: publish-to-gallery.yml calls this workflow as its test gate, so a
	# release only ships after passing on all 4 OS/PowerShell-edition combinations below --
	# not just whichever single platform a standalone Pester step would happen to run on.

	# `push` is filtered to main; `pull_request` is not. An unfiltered `push` matched every branch
	# commit, and a commit on a branch with an open PR ALSO matches `pull_request` -- so one push
	# ran the whole matrix twice, 10 jobs for one commit, with Windows pwsh alone near 17 minutes
	# on each side.
	#
	# Branch commits keep their coverage: `pull_request` fires on every push to a branch with an
	# open PR (the `synchronize` event). The only case that loses a run is a branch commit pushed
	# BEFORE its PR exists -- and a `pull_request` run is the better of the two anyway, because it
	# tests the MERGE commit rather than the branch tip, i.e. the tree that would actually ship.
	on:
	push:
	branches: [main]
	pull_request:
	workflow_dispatch: {}
	workflow_call: {}

	# Read-only: this workflow checks out, restores/saves the spec cache, moves artifacts between
	# jobs and runs Pester. It never writes to the repository. Declared rather than inherited
	# because without a permissions block the GITHUB_TOKEN gets whatever the repo's
	# default_workflow_permissions setting happens to be -- a setting an admin can flip to write
	# with no change to any file here. Cache and artifact actions are unaffected either way: they
	# authenticate with ACTIONS_RUNTIME_TOKEN, not this token.
	permissions:
	contents: read

	jobs:
	# Issue #63: tools/specs/ is a ~50MB cache of raw OpenAPI specs, gitignored because it is a
	# build input rather than source, so on a bare runner it does not exist. Every tooling test
	# gated on its presence skipped gracefully while the job reported success -- roughly 23% of
	# the suite, including the absolute-path regression guards from PR #62, invisible in the run
	# summary. This job materialises the cache once per run and hands it to every test leg.
	#
	# The cache key prefix is deliberately shared with update-api-capability-map.yml, so a warm
	# cache written by either workflow serves both. Update-PfbApiSpecs.ps1 skips any version
	# already on disk, so a cache hit means only newly-published versions get fetched instead of
	# the full ~29-version history.
	#
	# An artifact rather than a per-leg cache restore: it fetches from the published spec index
	# at most once per run instead of up to four times, and it is immune to cache key/branch
	# scoping differences across the matrix. 29 JSON files, ~50MB raw, which compress well.
	prepare-specs:
	name: Prepare API spec cache
	runs-on: ubuntu-latest
	steps:
	- name: Checkout
	uses: actions/checkout@v7

	- name: Restore cached spec files
	uses: actions/cache/restore@v6
	with:
	path: tools/specs
	key: pfb-specs-${{ github.run_id }}
	restore-keys: \|
	pfb-specs-

	- name: Fetch any missing REST API spec versions
	shell: pwsh
	run: ./tools/Update-PfbApiSpecs.ps1

	# The whole point of this job. A silent no-op here would put us straight back to the
	# green-but-empty runs issue #63 is about, so an empty result is a hard failure.
	- name: Assert the spec set is non-empty
	shell: pwsh
	run: ./scripts/Assert-PfbSpecCache.ps1

	- name: Save spec cache
	uses: actions/cache/save@v6
	if: always()
	with:
	path: tools/specs
	key: pfb-specs-${{ github.run_id }}

	- name: Publish specs to the test jobs
	uses: actions/upload-artifact@v5
	with:
	name: pfb-specs
	path: tools/specs
	retention-days: 1

	test-pwsh:
	name: Test (${{ matrix.os }}, pwsh)
	runs-on: ${{ matrix.os }}
	needs: prepare-specs
	strategy:
	fail-fast: false
	matrix:
	os: [ubuntu-latest, windows-latest, macos-latest]
	steps:
	- name: Checkout
	uses: actions/checkout@v7

	# Issue #63: without this, every tools/specs-gated test skips and the job still passes.
	- name: Download API spec cache
	uses: actions/download-artifact@v5
	with:
	name: pfb-specs
	path: tools/specs

	# Pinned + cached; see .github/actions/install-test-modules/action.yml for why
	# (including why Posh-SSH is needed at all).
	- name: Install test dependencies
	uses: ./.github/actions/install-test-modules
	with:
	shell: pwsh

	# Suite invocation + the issue-#63 coverage gate live in scripts/Invoke-PfbCiPester.ps1
	# rather than inline here: this same block was copy-pasted in three places across two
	# workflows, and all three needed the same change. A script is also lintable, diffable
	# and runnable outside Actions, which YAML-embedded PowerShell is not.
	- name: Run Pester tests
	shell: pwsh
	run: ./scripts/Invoke-PfbCiPester.ps1 -Edition pwsh7

	test-windows-powershell-5-1:
	name: Test (windows-latest, Windows PowerShell 5.1)
	runs-on: windows-latest
	needs: prepare-specs
	steps:
	- name: Checkout
	uses: actions/checkout@v7

	# Wired here too, for symmetry with the pwsh job. The tools/specs-gated Describes are
	# additionally PS7-gated so they still skip on this leg, but a future spec-dependent
	# test that does NOT carry the PS7 guard then works with no workflow change.
	- name: Download API spec cache
	uses: actions/download-artifact@v5
	with:
	name: pfb-specs
	path: tools/specs

	# Same pinned modules and same cache entry as the pwsh job on this OS -- the saved
	# files are edition-independent (Pester 6.0.1 declares PowerShellVersion = '5.1').
	# `shell:` is passed explicitly because a composite action has no default shell and
	# cannot read `matrix`.
	- name: Install test dependencies
	uses: ./.github/actions/install-test-modules
	with:
	shell: powershell

	# `shell:` stays per-leg and must remain a literal (see the header note about the
	# matrix context) -- it selects the interpreter. -Edition selects only which
	# Tests/coverage-baseline.psd1 block applies, since the two editions legitimately
	# differ by ~200 skips.
	- name: Run Pester tests
	shell: powershell
	run: ./scripts/Invoke-PfbCiPester.ps1 -Edition winps51

Navigation Menu

test: add ShouldProcess and comment-based-help coverage sweeps, and fix the gaps they find #157

Workflow file

test: add ShouldProcess and comment-based-help coverage sweeps, and fix the gaps they find #157

Uh oh!

Workflow file for this run