Skip to content
name: Index general search in Elasticsearch on PR
# **What it does**: This does what `index-general-search-elasticsearch.yml` does but
# with a localhost Elasticsearch and only for English.
# **Why we have it**: To test that the script works and the popular pages json is valid.
# **Who does it impact**: Docs engineering
on:
workflow_dispatch:
pull_request:
paths:
- 'src/search/**'
- 'package*.json'
# For debugging this workflow
- .github/workflows/index-general-search-pr.yml
# Make sure we run this if the composite action changes
- .github/actions/setup-elasticsearch/action.yml
permissions:
contents: read
# This allows a subsequently queued workflow run to interrupt previous runs
concurrency:
group: '${{ github.workflow }} @ ${{ github.event.pull_request.head.label || github.head_ref || github.ref }}'
cancel-in-progress: true
env:
ELASTICSEARCH_URL: http://localhost:9200
# Since we'll run in NDOE_ENV=production, we need to be explicit that
# we don't want Hydro configured.
HYDRO_ENDPOINT: ''
HYDRO_SECRET: ''
jobs:
dryRunElasticsearchIndexes:
runs-on: ubuntu-20.04-xl
if: github.repository == 'github/docs-internal'
steps:
- name: Check out repo
uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # v4.1.1
- name: Clone docs-internal-data
uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # v4.1.1
with:
repository: github/docs-internal-data
# This works because user `docs-bot` has read access to that private repo.
token: ${{ secrets.DOCS_BOT_PAT_READPUBLICKEY }}
path: docs-internal-data
- uses: ./.github/actions/setup-elasticsearch
- uses: ./.github/actions/node-npm-setup
- uses: ./.github/actions/cache-nextjs
- name: Build
run: npm run build
- name: Start the server in the background
env:
ENABLE_DEV_LOGGING: false
run: |
npm run general-search-scrape-server > /tmp/stdout.log 2> /tmp/stderr.log &
# first sleep to give it a chance to start
sleep 6
curl --retry-connrefused --retry 4 -I http://localhost:4002/
- if: ${{ failure() }}
name: Debug server outputs on errors
run: |
echo "____STDOUT____"
cat /tmp/stdout.log
echo "____STDERR____"
cat /tmp/stderr.log
- name: Scrape records into a temp directory
env:
# If a reusable, or anything in the `data/*` directory is deleted
# you might get a
#
# RenderError: Can't find the key 'site.data.reusables...' in the scope
#
# But that'll get fixed in the next translation pipeline. For now,
# let's just accept an empty string instead.
THROW_ON_EMPTY: false
DOCS_INTERNAL_DATA: docs-internal-data
run: |
mkdir /tmp/records
npm run general-search-scrape -- /tmp/records \
--language en \
--version fpt
ls -lh /tmp/records
- name: Check that Elasticsearch is accessible
run: |
curl --fail --retry-connrefused --retry 5 -I ${{ env.ELASTICSEARCH_URL }}
- name: Index into Elasticsearch
run: |
npm run index-general-search -- /tmp/records \
--language en \
--version fpt
- name: Check created indexes and aliases
run: |
curl --fail --retry-connrefused --retry 5 ${{ env.ELASTICSEARCH_URL }}/_cat/indices?v
curl --fail --retry-connrefused --retry 5 ${{ env.ELASTICSEARCH_URL }}/_cat/indices?v