openrefine-task-runner/Taskfile.yml

version: '3'

silent: true

env:
  OPENREFINE_MEMORY: 5120M
  OPENREFINE_PORT: 3333
  OPENREFINE_APPDIR:
    sh: readlink -m .openrefine
  OPENREFINE_TMPDIR:
    sh: mktemp -d

tasks:
  default:
    desc: run tasks start, import, transform, export, stats, check, kill and cleanup
    cmds:
      - defer: { task: cleanup } # will run even when one of the following commands fail
      - task: start
      - defer: { task: kill } # will run before cleanup
      - defer: { task: check } # will run before kill
      - defer: { task: stats } # will run before check
      - task: import
      - task: transform
      - task: export
    sources:
      - Taskfile.yml
      - input/**
      - config/**
    generates:
      - output/**

  install:
    desc: install OpenRefine and openrefine-client into subdirectory ${OPENREFINE_APPDIR}
    cmds:
      - mkdir -p "${OPENREFINE_APPDIR}"
      - | # install OpenRefine into subdirectory ${OPENREFINE_APPDIR}
        wget --no-verbose -O openrefine.tar.gz https://github.com/OpenRefine/OpenRefine/releases/download/3.5.2/openrefine-linux-3.5.2.tar.gz
        tar -xzf openrefine.tar.gz -C "${OPENREFINE_APPDIR}" --strip 1 && rm openrefine.tar.gz
      - | # optimize OpenRefine for batch processing
        sed -i 's/cd `dirname $0`/cd "$(dirname "$0")"/' "${OPENREFINE_APPDIR}/refine" # fix path issue in OpenRefine startup file
        sed -i '$ a JAVA_OPTIONS=-Drefine.headless=true' "${OPENREFINE_APPDIR}/refine.ini" # do not try to open OpenRefine in browser
        sed -i 's/#REFINE_AUTOSAVE_PERIOD=60/REFINE_AUTOSAVE_PERIOD=1440/' "${OPENREFINE_APPDIR}/refine.ini" # set autosave period from 5 minutes to 25 hours
      - | # install openrefine-client into subdirectory ${OPENREFINE_APPDIR}
        wget --no-verbose -O "${OPENREFINE_APPDIR}/client" https://github.com/opencultureconsulting/openrefine-client/releases/download/v0.3.10/openrefine-client_0-3-10_linux
        chmod +x "${OPENREFINE_APPDIR}/client"

  start:
      - | # requirement OpenRefine
        if [ ! -f "${OPENREFINE_APPDIR}/refine" ]; then
          echo 1>&2 "OpenRefine missing; try task install"; exit 1
        fi
      - | # launch OpenRefine with specific data directory and redirect its output to a log file
        "${OPENREFINE_APPDIR}/refine" -v warn -p "$OPENREFINE_PORT" -m "$OPENREFINE_MEMORY" -d "${OPENREFINE_TMPDIR}" >> "${OPENREFINE_TMPDIR}/log.txt" 2>&1 &
      - | # wait until OpenRefine API is available
        timeout 30s bash -c "until wget -q -O - -o /dev/null http://localhost:${OPENREFINE_PORT} | cat | grep -q -o OpenRefine; do sleep 1; done"

  import:
      - | # import (requires absolute path)
        "${OPENREFINE_APPDIR}/client" \
        --create "$(readlink -m input/duplicates.csv)" \
        --projectName myproject \
        > >(tee -a "${OPENREFINE_TMPDIR}/log.txt") 2>&1

  transform:
      - | # apply undo/redo history
        for f in config/*.json; do
          "${OPENREFINE_APPDIR}/client" myproject --apply "$f" \
          > >(tee -a "${OPENREFINE_TMPDIR}/log.txt") 2>&1
        done

  export:
      - mkdir -p output
      - | # export to TSV
        "${OPENREFINE_APPDIR}/client" myproject \
        --output "$(readlink -m output/deduped.tsv)" \
        > >(tee -a "${OPENREFINE_TMPDIR}/log.txt") 2>&1

  stats:
      - | # print RAM and CPU usage
        PID="$(lsof -t -i:${OPENREFINE_PORT})"
        echo "used $(($(ps --no-headers -o rss -p "$PID") / 1024)) MB RAM" \
        > >(tee -a "${OPENREFINE_TMPDIR}/log.txt") 2>&1
        echo "used $(ps --no-headers -o cputime -p "$PID") CPU time" \
        > >(tee -a "${OPENREFINE_TMPDIR}/log.txt") 2>&1

  check:
      - | # check log file for any warnings
        if grep -i 'exception\|error' "${OPENREFINE_TMPDIR}/log.txt"
          then echo 1>&2 "log contains warnings!"; echo; cat "${OPENREFINE_TMPDIR}/log.txt"; exit 1
        fi

  stop:
      - | # shut down OpenRefine gracefully
        PID="$(lsof -t -i:${OPENREFINE_PORT})"
        kill $PID; while ps -p $PID > /dev/null; do sleep 1; done

  kill:
      - | # shut down OpenRefine immediately to save time
        PID="$(lsof -t -i:${OPENREFINE_PORT})"
        kill -9 $PID

  cleanup: rm -rf "${OPENREFINE_TMPDIR}"

  git:
    desc: commit and push if something changed
    cmds:
      - git add -A
      - git commit -m "latest change $(date -u)" || exit 0
      - git push