openrefine-github-action/.github/workflows/openrefine.yml

name: example # available as environment variable $GITHUB_WORKFLOW

on:
  workflow_dispatch: # allows you to run this workflow manually from the Actions tab

jobs:
  main:
    runs-on: ubuntu-20.04
    steps:
    - uses: actions/checkout@v2 # checks-out your repository under $GITHUB_WORKSPACE, so your job can access it
    - name: install OpenRefine and openrefine-client
      run: |
        mkdir -p output .openrefine .openrefine/data
        cd .openrefine
        wget -q -O openrefine.tar.gz https://github.com/OpenRefine/OpenRefine/releases/download/3.5.0/openrefine-linux-3.5.0.tar.gz
        tar -xzf openrefine.tar.gz --strip 1
        sed -i '$ a JAVA_OPTIONS=-Drefine.headless=true' refine.ini # do not try to open OpenRefine in browser
        sed -i 's/#REFINE_AUTOSAVE_PERIOD=60/REFINE_AUTOSAVE_PERIOD=1440/' refine.ini # set autosave period from 5 minutes to 25 hours
        wget -q -O client https://github.com/opencultureconsulting/openrefine-client/releases/download/v0.3.10/openrefine-client_0-3-10_linux
        chmod +x client
        cd -
    - name: start OpenRefine
      run: |
        .openrefine/refine -m 5120M -v warn -d data >> .openrefine/log.txt 2>&1 &
        timeout 30s bash -c "until wget -q -O - -o /dev/null http://localhost:3333 | cat | grep -q -o OpenRefine; do sleep 1; done"
    - name: import
      run: |
        .openrefine/client \
        --create "$(readlink -m input/schriftstellerinnen.csv)" \
        --encoding UTF-8 \
        --projectName "$GITHUB_WORKFLOW" \
        > >(tee -a .openrefine/log.txt) 2>&1
    - name: transform
      run: |
        .openrefine/client "$GITHUB_WORKFLOW" \
        --apply "$(readlink -m config/history.json)" \
        > >(tee -a .openrefine/log.txt) 2>&1
    - name: export
      run: |
        .openrefine/client "$GITHUB_WORKFLOW" \
        --output "$(readlink -m output/schriftstellerinnen.tsv)" \
        > >(tee -a .openrefine/log.txt) 2>&1
    - name: stop OpenRefine
      if: always()
      run: |
        PID=$(lsof -t -i:3333)
        echo "used $(($(ps --no-headers -o rss -p "$PID") / 1024)) MB RAM" > >(tee -a .openrefine/log.txt) 2>&1
        echo "used $(ps --no-headers -o cputime -p "$PID") CPU time" > >(tee -a .openrefine/log.txt) 2>&1
        kill $PID; while ps -p $PID > /dev/null; do sleep 1; done
    - name: archive OpenRefine projects
      if: always()
      run: for p in .openrefine/data/*/; do tar cfz .openrefine/data/"${p:17:13}.openrefine.tar.gz" -C $p .; done
    - uses: actions/upload-artifact@v2
      if: always()
      with:
        name: OpenRefine project(s)
        path: .openrefine/data/*.openrefine.tar.gz
        retention-days: 5
    - name: check logfile for any warnings
      if: always()
      run: |
        if grep -i 'exception\|error' .openrefine/log.txt
          then echo 1>&2 "log contains warnings!"; echo; cat .openrefine/log.txt; exit 1;
          else echo "log seems to be ok"; echo; cat .openrefine/log.txt
        fi
    - name: commit and push if output changed
      run: |-
        git config user.name "Automated"
        git config user.email "actions@users.noreply.github.com"
        git add -A
        git commit -m "latest change: $(date -u)" || exit 0
        git push