cleanlab--cleanlab/.github/workflows/gh-pages.yaml

name: GitHub Pages

on:
  push:
    branches:
      - "**"

  release:
    types: [published]

  workflow_dispatch:

jobs:
  deploy:
    runs-on: ubuntu-24.04
    concurrency:
      group: ${{ github.workflow }}-${{ github.ref }}
    steps:
      - uses: actions/checkout@v6
        with:
          fetch-depth: 0

      - name: Install Pandoc
        # pin Pandoc version
        run: |
          cd /tmp
          wget https://github.com/jgm/pandoc/releases/download/2.19.2/pandoc-2.19.2-linux-amd64.tar.gz
          sudo tar xzvf pandoc-2.19.2-linux-amd64.tar.gz --strip-components 1 -C /usr/local

      - name: Setup Python
        uses: actions/setup-python@v6
        with:
          python-version: "3.11"

      - name: Setup Node
        uses: actions/setup-node@v6
        with:
          node-version: "22"
  
      - name: Install FFmpeg
        run: |
          sudo apt-get update
          sudo apt-get install -y ffmpeg

      - name: Upgrade pip
        run: python3 -m pip install --upgrade pip

      # Caching removed to resolve disk space issues
      # Heavy ML packages (TensorFlow, PyTorch, HuggingFace) will be downloaded fresh each run
      # but this avoids the 3.7GB cache that causes "No space left on device" errors

      - name: Install cleanlab
        run: pip install -e ".[all]"

      - name: Install dependencies
        run: python3 -m pip install -r docs/requirements.txt

      - name: Install katex
        run: npm install -g katex

      - name: Find and replace "%pip install cleanlab" with master branch in .ipynb files
        if: ${{ github.ref_type == 'branch' }}
        uses: jacobtomlinson/gha-find-replace@v2
        with:
          find: "%pip install cleanlab  ${{ '#' }} for colab"
          replace: "%pip install git+https://github.com/${{ github.repository }}.git@${{ github.sha }}"
          include: "docs/source**.ipynb"

      - name: Find and replace "%pip install cleanlab" with release tag in .ipynb files
        if: ${{ github.ref_type == 'tag' }}
        uses: jacobtomlinson/gha-find-replace@v2
        with:
          find: "%pip install cleanlab  ${{ '#' }} for colab"
          replace: "%pip install cleanlab==${{ github.ref_name }}"
          include: "docs/source**.ipynb"

      - name: Add and Commit changes to .ipynb files
        run: |
          git config --global user.email "actions@github.com"
          git config --global user.name "GitHub Actions"
          git add -A && git commit -m "Change cleanlab ver in .ipynb files"

      - name: Change tag to the latest commit
        if: ${{ github.ref_type == 'tag' }}
        run: |
          git tag -f $GITHUB_REF_NAME

      - name: Run Sphinx with sphinx-multiversion wrapper - branch trigger
        if: ${{ github.ref_type == 'branch' }}
        env:
          TF_CPP_MIN_LOG_LEVEL: 3
          HF_HOME: /tmp/hf_cache
          HF_HUB_DISABLE_SYMLINKS_WARNING: 1
          TRANSFORMERS_CACHE: /tmp/transformers
          TORCH_HOME: /tmp/torch
        run: sphinx-multiversion docs/source cleanlab-docs -D smv_branch_whitelist=${{ github.ref_name }} -D smv_tag_whitelist=None

      - name: Run Sphinx with sphinx-multiversion wrapper - tag trigger
        if: ${{ github.ref_type == 'tag' }}
        env:
          TF_CPP_MIN_LOG_LEVEL: 3
          HF_HOME: /tmp/hf_cache
          HF_HUB_DISABLE_SYMLINKS_WARNING: 1
          TRANSFORMERS_CACHE: /tmp/transformers
          TORCH_HOME: /tmp/torch
        run: sphinx-multiversion docs/source cleanlab-docs -D smv_branch_whitelist=None -D smv_tag_whitelist=${{ github.ref_name }}

      - name: Debug temporary cache usage (not saved)
        run: |
          echo "=== Temporary Cache Usage ==="
          echo "These caches are created in /tmp and cleaned up after build:"
          echo "  /tmp/hf_cache: $(du -sh /tmp/hf_cache 2>/dev/null | cut -f1 || echo 'not found')"
          echo "  /tmp/transformers: $(du -sh /tmp/transformers 2>/dev/null | cut -f1 || echo 'not found')"
          echo "  /tmp/torch: $(du -sh /tmp/torch 2>/dev/null | cut -f1 || echo 'not found')"
          echo "Pip cache directory: $(pip cache dir)"
          echo "Pip cache size: $(du -sh $(pip cache dir) 2>/dev/null | cut -f1 || echo 'not found')"
          echo "========================"

      - name: Get latest stable release
        id: stable_release
        continue-on-error: true
        uses: pozetroninc/github-action-get-latest-release@master
        with:
          owner: ${{ github.repository_owner }}
          repo: cleanlab
          excludes: prerelease, draft

      - name: Create latest_release.txt file
        if: ${{ steps.stable_release.outcome == 'success' }}
        run: echo ${{ steps.stable_release.outputs.release }} > cleanlab-docs/latest_release.txt

      - name: Find and replace "placeholder_version_number" in versioning.js
        if: ${{ steps.stable_release.outcome == 'success' }}
        uses: jacobtomlinson/gha-find-replace@v2
        with:
          find: "placeholder_version_number"
          replace: ${{ steps.stable_release.outputs.release }}
          include: "docs/source/_templates/versioning.js"

      - name: Find and replace "placeholder_commit_hash" in versioning.js
        uses: jacobtomlinson/gha-find-replace@v2
        with:
          find: "placeholder_commit_hash"
          replace: ${{ github.sha }}
          include: "docs/source/_templates/versioning.js"

      - name: Copy versioning JS file to cleanlab-docs/
        run: |
          cp docs/source/_templates/versioning.js cleanlab-docs/versioning.js

      - name: Find and replace "stable_url" in redirect-to-stable.html
        if: ${{ github.ref_type == 'tag' && steps.stable_release.outcome == 'success' }}
        uses: jacobtomlinson/gha-find-replace@v2
        with:
          find: "stable_url"
          replace: ${{ env.DOCS_SITE_URL }}stable/index.html
          include: "docs/source/_templates/redirect-to-stable.html"

      - name: Copy redirecting HTML file to cleanlab-docs/
        if: ${{ github.ref_type == 'tag' }}
        run: cp docs/source/_templates/redirect-to-stable.html cleanlab-docs/index.html

      - name: Update OpenGraph URLs in HTML files
        # This code fixes the URLs in the HTML files with OpenGraph tags.
        # It uses a regular expression to capture two parts of the URL:
        #   $1: The first part of the URL, including the protocol and domain.
        #   $2: The rest of the URL, including the path and query string.
        # The replacement string combines the captured parts with the ref name.
        # This ensures that the ref name is inserted into the URL correctly.
        env:
          REF_URL_SEGMENT: ${{ (github.ref_type == 'tag' && steps.stable_release.outcome == 'success') && 'stable' || github.ref_name }}
        uses: jacobtomlinson/gha-find-replace@v2
        with:
          find: '(<meta property="og:url" content="https?://[^/]+)/([^"]+")'
          replace: '$1/${{ env.REF_URL_SEGMENT }}/$2'
          include: "cleanlab-docs/**/*.html"
          regex: true
      
      - name: Deploy
        if: ${{ (github.ref == 'refs/heads/master') || (github.ref_type == 'tag') }}
        uses: peaceiris/actions-gh-pages@v4
        with:
          deploy_key: ${{ secrets.ACTIONS_DEPLOY_KEY }}
          external_repository: ${{ github.repository_owner }}/cleanlab-docs
          publish_branch: master
          publish_dir: cleanlab-docs
          keep_files: true
          exclude_assets: ""

      - uses: actions/upload-artifact@v5
        with:
          name: docs-html
          path: cleanlab-docs
          retention-days: 1 # don't need this except in link checking step below

  links:
    needs: deploy
    runs-on: ubuntu-24.04
    steps:
      - uses: actions/download-artifact@v5
        with:
          name: docs-html
          path: cleanlab-docs
      - uses: anishathalye/proof-html@v1
        with:
          directory: ./cleanlab-docs
          external_only: true
          url_ignore_re: |
            ^https://pradyunsg.me/
            ^https://keras.io/