Merge pull request #157 from microbiomedata/61-add-link-checker

Implement GHA workflow that scans the website file tree for broken links

.github/workflows/assemble-website.yml

@@ -3,6 +3,11 @@
 name: Assemble website
 
 on:
+  # Run this workflow whenever someone opens or adds commits to a PR whose base branch is `main`.
+  # Reference: https://docs.github.com/en/actions/writing-workflows/choosing-when-your-workflow-runs/events-that-trigger-workflows#pull_request
+  pull_request: { branches: [ main ] }
+  # Allow this workflow to be run manually (e.g., via the "Actions" tab on GitHub).
+  # Reference: https://docs.github.com/en/actions/writing-workflows/choosing-when-your-workflow-runs/events-that-trigger-workflows#workflow_dispatch
   workflow_dispatch: { }
   # Allow this workflow to be called by other workflows.
   # Reference: https://docs.github.com/en/actions/using-workflows/reusing-workflows
@@ -61,3 +66,10 @@ jobs:
           #       anyway, here, as a reminder for people that will be implementing workflows that consume the same
           #       artifact; e.g., spell checkers and link checkers.
           name: github-pages
+
+  # Use existing workflow(s) to check the file tree for broken links.
+  check-links:
+    name: Check links
+    uses: ./.github/workflows/check-links.yml
+    needs:
+      - assemble

.github/workflows/check-links.yml

@@ -0,0 +1,47 @@
+# This GitHub Actions workflow finds links in HTML files and creates a GitHub Issue if any of them doesn't work.
+# Reference: https://docs.github.com/en/actions/writing-workflows/workflow-syntax-for-github-actions
+name: Check links
+
+on:
+  # Allow this workflow to be called by other workflows.
+  # Reference: https://docs.github.com/en/actions/using-workflows/reusing-workflows
+  workflow_call: { }
+
+jobs:
+  check-links:
+    name: Check links
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write # required for peter-evans/create-issue-from-file
+    steps:
+      - name: Get website file tree
+        uses: actions/download-artifact@v4  # docs: https://github.com/actions/download-artifact
+        with:
+          name: github-pages
+          path: _downloads
+      - name: Un-tar the archive
+        run: |
+          pwd
+          mkdir -p _build/html
+          tar -xvf _downloads/artifact.tar -C _build/html
+          ls -lR _build/html
+      - name: Use Lychee to check for broken links
+        # This step will populate `steps.lychee.outputs.exit_code` with the exit code returned by lychee.
+        # Reference: https://github.com/lycheeverse/lychee-action
+        id: lychee
+        uses: lycheeverse/lychee-action@v2
+        with:
+          # Note: The `--format` option refers to the format of the report, not the documents being scanned.
+          # Reference: https://github.com/lycheeverse/lychee#commandline-parameters
+          args: --base '_build/html' --verbose --no-progress --format markdown '_build/html/**/*.html'
+          debug: true
+          output: ./lychee/out.md
+          fail: false
+      - name: Create GitHub Issue listing broken links
+        # This step will only run if both (a) lychee returned a non-zero exit code and (b) we are on the `main` branch.
+        # Reference: https://docs.github.com/en/actions/learn-github-actions/variables#using-the-env-context-to-access-environment-variable-values
+        if: steps.lychee.outputs.exit_code != 0 && github.ref == 'refs/heads/main'
+        uses: peter-evans/create-issue-from-file@v5
+        with:
+          title: Website file tree contains broken links
+          content-filepath: ./lychee/out.md

.github/workflows/compile-home-docs.yml

@@ -3,10 +3,8 @@
 name: Compile home documentation into HTML
 
 on:
-  push: { branches: [ main ] }
-  # Run this workflow whenever someone opens or adds commits to a PR whose base branch is `main`.
-  # Reference: https://docs.github.com/en/actions/writing-workflows/choosing-when-your-workflow-runs/events-that-trigger-workflows#pull_request
-  pull_request: { branches: [main] }
+  # Allow this workflow to be run manually (e.g., via the "Actions" tab on GitHub).
+  # Reference: https://docs.github.com/en/actions/writing-workflows/choosing-when-your-workflow-runs/events-that-trigger-workflows#workflow_dispatch
   workflow_dispatch: { }
   # Allow this workflow to be called by other workflows.
   # Reference: https://docs.github.com/en/actions/using-workflows/reusing-workflows

.github/workflows/deploy-to-gh-pages.yml

@@ -3,7 +3,11 @@
 name: Deploy to GitHub Pages
 
 on:
+  # Run this workflow whenever someone pushes commits or tags to the `main` branch.
+  # Reference: https://docs.github.com/en/actions/writing-workflows/choosing-when-your-workflow-runs/events-that-trigger-workflows#push
   push: { branches: [ main ] }
+  # Allow this workflow to be run manually (e.g., via the "Actions" tab on GitHub).
+  # Reference: https://docs.github.com/en/actions/writing-workflows/choosing-when-your-workflow-runs/events-that-trigger-workflows#workflow_dispatch
   workflow_dispatch: { }
 
 # Reference: https://docs.github.com/en/actions/using-jobs/using-concurrency