119 lines
4.1 KiB
YAML
119 lines
4.1 KiB
YAML
name: Sync search - PR
|
|
|
|
# **What it does**: This does what `sync-sarch-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:
|
|
pull_request:
|
|
paths:
|
|
- 'src/search/**'
|
|
- 'package*.json'
|
|
# Ultimately, for debugging this workflow itself
|
|
- .github/workflows/sync-search-pr.yml
|
|
|
|
permissions:
|
|
contents: read
|
|
|
|
# This allows a subsequently queued workflow run to interrupt previous runs
|
|
concurrency:
|
|
group: '${{ github.workflow }} @ ${{ github.event.pull_request.head.label || github.head_ref || github.ref }}'
|
|
cancel-in-progress: true
|
|
|
|
env:
|
|
# Yes, it's hardcoded but it makes all the steps look exactly the same
|
|
# as they do in `sync-search-elasticsearch.yml` where it uses
|
|
# that `${{ env.ELASTICSEARCH_URL }}`
|
|
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: ${{ fromJSON('["ubuntu-latest", "ubuntu-20.04-xl"]')[github.repository == 'github/docs-internal'] }}
|
|
if: github.repository == 'github/docs-internal' || github.repository == 'github/docs'
|
|
steps:
|
|
- name: Check out repo
|
|
uses: actions/checkout@3df4ab11eba7bda6032a0b82a6bb43b11571feac # v4.0.0
|
|
|
|
- name: Clone docs-internal.popular-pages
|
|
uses: actions/checkout@3df4ab11eba7bda6032a0b82a6bb43b11571feac # v4.0.0
|
|
with:
|
|
repository: github/docs-internal.popular-pages
|
|
# This works because user `docs-bot` has read access to that private repo.
|
|
token: ${{ secrets.DOCS_BOT_PAT_READPUBLICKEY }}
|
|
path: popular-pages
|
|
|
|
- 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 sync-search-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
|
|
|
|
# The sync-search-index recognizes this env var if you don't
|
|
# use the `--popular-pags <PATH>` option.
|
|
POPULAR_PAGES_JSON: popular-pages/records/popular-pages.json
|
|
|
|
run: |
|
|
mkdir /tmp/records
|
|
npm run sync-search-indices -- /tmp/records \
|
|
--language en \
|
|
--version dotcom
|
|
|
|
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: |
|
|
./src/search/scripts/index-elasticsearch.js /tmp/records \
|
|
--language en \
|
|
--version dotcom
|
|
|
|
- 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
|
|
|
|
- uses: ./.github/actions/slack-alert
|
|
with:
|
|
slack_channel_id: ${{ secrets.DOCS_ALERTS_SLACK_CHANNEL_ID }}
|
|
slack_token: ${{ secrets.SLACK_DOCS_BOT_TOKEN }}
|