add link-checking CI

moving over from the fern repo. Planning to schedule for weekly runs and send alerts to myself to resolve issues.

Author: davidkonigsberg

.github/workflows/check-links.yml

@@ -0,0 +1,82 @@
+name: Check Links
+
+on:
+  # TODO: Enable scheduled scanning once standing issues have been resolved
+  #schedule:
+  #- cron: "0 0 * * *"
+  workflow_dispatch:
+jobs:
+  check-links:
+    name: Check links
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        site:
+        - name: Fern Docs
+          url: "buildwithfern.com/learn"
+    steps:
+      - name: Install link checker
+        run: pip install linkchecker
+      - name: Create config
+        run: echo -e "[csv]\nseparator=,\n[filtering]\nignore=\n\texample.com\n\tus.i.posthog.com\n\tc.vialoops.com/CL0\n[output]\nignoreerrors=\n  ^http ^403 Forbidden" > lcconfig
+      - name: Check ${{ matrix.site.name }} Links
+        run: |
+          set +e
+          cat lcconfig
+          linkchecker https://${{ matrix.site.url }} --check-extern --no-status --no-warnings --config lcconfig -F csv/utf-8/link_report.csv
+
+          if [ $? -ne 0 ]; then
+            echo "Bad links found. Please see report."
+          else
+            echo "Check completed. No issues found."
+            exit 0
+          fi
+
+          echo "Scan done, generating summary"
+
+          started=false
+          shouldfail=false
+          while read p; do
+            # skip comments (top and bottom)
+            if [[ $p == \#* ]]; then
+              continue
+            fi
+            # make sure first line after comments is skipped
+            if [[ "$started" = false ]]; then
+              started=true
+              continue
+            fi
+            
+            IFS=',' read -r -a array <<< "$p"
+
+            ret=$(curl -I -s "${array[0]}" -o /dev/null -w "%{http_code}\n")
+            if [[ $ret == 200 ]]; then
+              echo "Site now seems to be working, we should continue here"
+              continue
+            else
+              echo "There is still an issue with ${array[0]}"
+            fi
+
+            shouldfail=true
+
+            if [[ ${array[10]} == '' ]]; then
+              echo "::error::URL: ${array[0]} on page: ${array[1]} is returning a ${array[3]}"
+            else
+              echo "::error::URL: ${array[0]} linked from '${array[10]}' on page: ${array[1]} is returning a ${array[3]}"
+            fi
+
+          done < link_report.csv
+
+          if [[ "$shouldfail" = false ]]; then
+            exit 0
+          fi
+            
+          exit 1
+
+      - name: Upload report
+        if: failure()
+        uses: actions/upload-artifact@v4
+        with:
+          name: Report - ${{ matrix.site.name }}
+          path: link_report.csv