Jobergum/document v1 id formatter (#626) #333
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: pyvespa documentation search feed | |
| on: | |
| push: | |
| branches: [ master ] | |
| env: | |
| DATA_PLANE_PUBLIC_KEY : ${{ secrets.VESPA_TEAM_DATA_PLANE_PUBLIC_CERT }} | |
| DATA_PLANE_PRIVATE_KEY : ${{ secrets.VESPA_TEAM_DATA_PLANE_PRIVATE_KEY }} | |
| VESPA_CLI_DATA_PLANE_CERT : ${{ secrets.VESPA_TEAM_VESPA_CLI_DATA_PLANE_CERT }} | |
| VESPA_CLI_DATA_PLANE_KEY : ${{ secrets.VESPA_TEAM_VESPA_CLI_DATA_PLANE_KEY }} | |
| jobs: | |
| build: | |
| runs-on: ubuntu-latest | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - uses: ruby/setup-ruby@v1 | |
| with: | |
| ruby-version: 3.1 | |
| bundler-cache: true | |
| - name: Install Sphinx | |
| run: | | |
| sudo apt-get install -y pandoc | |
| python3 -m pip install --upgrade pip | |
| python3 -m pip install -r docs/sphinx/source/requirements-feed.txt | |
| - name: Build site | |
| run: | | |
| sphinx-build -E -D nbsphinx_allow_errors=True -b html docs/sphinx/source docs/sphinx/build | |
| find docs/sphinx/build/ -name \*.html | while read f; do (echo -e "---\n---\n"; cat ${f})>${f}.new; mv ${f}.new ${f}; done | |
| # Strip jekyll liquid macros | |
| sed -i.orig 's/{%/{ %/g; s/%}/% }/g; s/{{/{ {/g; s/}}/} }/g' docs/sphinx/build/*.html ; rm docs/sphinx/build/*.orig | |
| bundle exec jekyll build -s docs/sphinx/build/ -p _plugins-vespafeed --config _config.yml | |
| - name: Setup Python | |
| uses: actions/setup-python@v4 | |
| with: | |
| python-version: '3.x' | |
| - name: Install dependencies | |
| run: | | |
| pip3 install PyYAML spacy mmh3 requests html5lib beautifulsoup4 markdownify tiktoken | |
| - name: Get Vespa CLI - update to later versions as needed | |
| run: | | |
| # Workaround: Rename the "vespa" directory in pyvespa, conflicts with the cli install below | |
| # This directory is not needed for the feeding anyway | |
| mv vespa vespa.org | |
| apt update && apt -y install curl | |
| curl -SsLo vespa-cli.tar.gz https://github.com/vespa-engine/vespa/releases/download/v8.209.11/vespa-cli_8.209.11_linux_amd64.tar.gz | |
| tar -xvf vespa-cli.tar.gz && ln -fs vespa-cli_8.209.11_linux_amd64/bin/vespa | |
| - name: Feed site | |
| run: | | |
| # The python scripts below uses the Vespa CLI for feeding / data access. | |
| # See https://docs.vespa.ai/en/vespa-cli.html. | |
| # The environment variables below have credentials for endpoint access - | |
| # use the key/cert files in .vespa and paste their content into GitHub Secrets. | |
| # VESPA_CLI_API_KEY must be set and empty as below. | |
| export VESPA_CLI_DATA_PLANE_CERT | |
| export VESPA_CLI_DATA_PLANE_KEY | |
| export VESPA_CLI_API_KEY= | |
| ./feed_to_vespa.py _config.yml | |
| - name: Feed paragraphs site | |
| run: | | |
| export VESPA_CLI_DATA_PLANE_CERT | |
| export VESPA_CLI_DATA_PLANE_KEY | |
| export VESPA_CLI_API_KEY= | |
| ./feed-split.py pyvespa_index.json https://pyvespa.readthedocs.io/en/latest questions.jsonl | |
| ./feed_to_vespa.py _paragraphs_config.yml | |
| - name: Feed suggestions | |
| run: | | |
| export VESPA_CLI_DATA_PLANE_CERT | |
| export VESPA_CLI_DATA_PLANE_KEY | |
| export VESPA_CLI_API_KEY= | |
| ./feed_to_vespa.py _suggestions_config.yml |