name: ci-cpd
on:
  push:
jobs:
  build:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v3
    - uses: ./.github/actions/cpd
      with:
        language: python

name: cpd
inputs:
  pmd_version:
    default: '6.52.0'
  minimum_tokens:
    default: '100'
  fail_on_violation:
    default: true
  language:
    required: true
runs:
  using: "composite"
  steps:
  - uses: actions/setup-java@v3
    with:
      distribution: 'temurin'
      java-version: '11'
  - uses: actions/setup-python@v4
    with:
      python-version: '3.10'
  - name: setup-pmd
    run: |
      export PMD_VERSION=${{ inputs.pmd_version }}
      cd $HOME
      wget -q https://github.com/pmd/pmd/releases/download/pmd_releases%2F$PMD_VERSION/pmd-bin-$PMD_VERSION.zip
      unzip -q pmd-bin-$PMD_VERSION.zip
      echo "$HOME/pmd-bin-$PMD_VERSION/bin" >> $GITHUB_PATH      
    shell: bash
  - name: run-pmd-cpd
    run: |
      run.sh cpd --dir . --minimum-tokens ${{ inputs.minimum_tokens }} --language ${{ inputs.language }} --format xml --fail-on-violation ${{ inputs.fail_on_violation }} | python -c '
      import xml.etree.ElementTree as ET
      import sys
      root = ET.fromstring(sys.stdin.read())
      report_each_file = [
          {
              "path": f.attrib["path"],
              "total": int(f.attrib["totalNumberOfTokens"]),
              "copy": len(set(sum([
                  list(range(int(d.attrib["begintoken"]),int(d.attrib["endtoken"])))
                  for d in root.findall("duplication/file")
                  if d.attrib["path"] == f.attrib["path"]
              ], [])))
          }
          for f in root.findall("file")
          if int(f.attrib["totalNumberOfTokens"]) > 0
      ]
      total_token = sum([f["total"] for f in report_each_file])
      dup_token = sum([f["copy"] for f in report_each_file])

      print("CopyRate: {}".format(float(dup_token)/total_token))
      print("CopyTokenSize: {}".format(dup_token))
      print("TotalTokenSize: {}".format(total_token))
      for f in report_each_file:
          if f["copy"] > 0:
              print("{}: {} (copy_tokens={}, total_tokens={})".format(
                  f["path"], float(f["copy"])/f["total"], f["copy"], f["total"]
              ))
      '      
    shell: bash