link-checker.yml

name: Link Checker

on:
  workflow_dispatch:
  schedule:
    - cron: "2 0 * * *"

  # START Temporary for testing.
  pull_request:
    branches: [main]
  push:
    branches: ["link-checker-workflow-configuration"]
  # END Temporary for testing.

defaults:
  run:
    # Specify to ensure "pipefail and errexit" are set.
    # Ref: https://docs.github.com/en/actions/writing-workflows/workflow-syntax-for-github-actions#defaultsrunshell
    shell: bash

jobs:
  link-checker-documentation:
    runs-on: ubuntu-latest

    steps:
      - uses: actions/checkout@v4

      - name: Setup Ruby
        uses: ruby/setup-ruby@v1
        with:
          ruby-version: 3.1
          bundler-cache: true

      - name: check links
        env:
          LANG: "C.UTF-8"
        run: |
          bundle exec jekyll build
          #
          # Remove the redirect-files before link-check
          find _site/en _site/documentation -name \*.html | \
            xargs grep -l "Click here if you are not redirected." | xargs rm
          #
          # htmlproofer does not check links inside <code>-elements
          find _site -name \*.html | xargs sed -i.orig 's/<code[^>]*>//g; s/<\/code>//g; s/<pre[^>]*>//g; s/<\/pre>//g;'
          find _site -name \*.orig | xargs rm
          #
          bundle exec htmlproofer \
            --assume-extension .html \
            --no-enforce-https \
            --no-check-external-hash \
            --allow-missing-href \
            --ignore-files '/playground/index.html/' \
            --ignore-urls '\
              /localhost:8080/,\
              /docs.vespa.ai/playground/,\
              /javadoc.io.*#/,\
              /readthedocs.io.*#/,\
              /linux.die.net/,\
              /arxiv.org/,\
              /hub.docker.com/r/,\
              /platform.openai.com/' \
            --typhoeus '{"connecttimeout": 10, "timeout": 30, "accept_encoding": "zstd,br,gzip,deflate"}' \
            --hydra '{"max_concurrency": 1}' \
            --swap-urls '(https\://github.com.*/master/.*)#.*:\1,(https\://github.com.*/main/.*)#.*:\1' \
            _site