antlr-rust-runtime 0.3.0

name: Copy/Paste Detection

on:
  pull_request:
    branches:
      - main
    paths:
      - "**/*.rs"
      - "!target/**"
      - ".github/workflows/cpd.yml"

permissions:
  contents: read
  pull-requests: write

concurrency:
  group: cpd-${{ github.event.pull_request.number || github.run_id }}
  cancel-in-progress: true

env:
  PMD_VERSION: 7.24.0
  CPD_TOKENS: "100"
  COMMENT_MARKER: <!-- cpd-report -->

jobs:
  cpd:
    name: Copy/Paste Detection
    runs-on: ubuntu-latest

    steps:
      - name: Checkout
        uses: actions/checkout@v6
        with:
          fetch-depth: 0
          persist-credentials: false

      - name: Determine changed Rust files
        id: changed
        shell: bash
        run: |
          set -euo pipefail
          BASE_SHA=$(git merge-base "origin/${GITHUB_BASE_REF}" HEAD)

          git diff --name-only --diff-filter=d "$BASE_SHA" HEAD -- "*.rs" \
            | grep -Ev "^target/" \
            > changed-files.txt || true

          COUNT=$(wc -l < changed-files.txt | tr -d " ")
          echo "count=$COUNT" >> "$GITHUB_OUTPUT"
          echo "Changed Rust files ($COUNT):"
          cat changed-files.txt || true

      - name: Setup Java
        if: steps.changed.outputs.count != '0'
        uses: actions/setup-java@v5
        with:
          distribution: temurin
          java-version: "21"

      - name: Setup PMD
        if: steps.changed.outputs.count != '0'
        shell: bash
        run: |
          set -euo pipefail
          curl -fL "https://github.com/pmd/pmd/releases/download/pmd_releases%2F${PMD_VERSION}/pmd-dist-${PMD_VERSION}-bin.zip" -o pmd.zip
          unzip -q pmd.zip
          rm pmd.zip

      - name: Run CPD
        id: cpd
        shell: bash
        run: |
          set -uo pipefail

          if [ "${{ steps.changed.outputs.count }}" = "0" ]; then
            : > cpd-report.md
            echo "duplications=0" >> "$GITHUB_OUTPUT"
            echo "No changed Rust files."
            exit 0
          fi

          # PMD CPD exit codes:
          #   0 - no duplications
          #   4 - duplications found
          #   5 - recoverable errors, for example a file failed to lex
          set +e
          "pmd-bin-${PMD_VERSION}/bin/pmd" cpd \
            --language rust \
            --minimum-tokens "${CPD_TOKENS}" \
            --file-list changed-files.txt \
            --format markdown \
            > cpd-report.md 2> cpd-stderr.log
          STATUS=$?
          set -e

          if [ "$STATUS" -ne 0 ] && [ "$STATUS" -ne 4 ] && [ "$STATUS" -ne 5 ]; then
            echo "PMD CPD errored (status $STATUS):"
            cat cpd-stderr.log
            exit "$STATUS"
          fi

          if [ -s cpd-stderr.log ]; then
            echo "=== PMD stderr ==="
            cat cpd-stderr.log
          fi

          sed -i "s|${GITHUB_WORKSPACE}/||g" cpd-report.md

          awk '
            BEGIN { open = 0 }
            /^```$/ {
              if (open == 0) { print "```rust"; open = 1 }
              else           { print "```";     open = 0 }
              next
            }
            { print }
          ' cpd-report.md > cpd-report.tagged.md
          mv cpd-report.tagged.md cpd-report.md

          DUP_COUNT=$(grep -c "^Found a " cpd-report.md || true)
          echo "duplications=${DUP_COUNT:-0}" >> "$GITHUB_OUTPUT"

          echo "=== Report ==="
          cat cpd-report.md

      - name: Build comment body
        shell: bash
        env:
          CHANGED_COUNT: ${{ steps.changed.outputs.count }}
          DUP_COUNT: ${{ steps.cpd.outputs.duplications }}
        run: |
          set -euo pipefail
          {
            echo "${COMMENT_MARKER}"
            echo "## Copy/Paste Detection"
            echo ""
            if [ "${DUP_COUNT:-0}" = "0" ]; then
              echo "No duplications found in ${CHANGED_COUNT} changed Rust file(s) (threshold: ${CPD_TOKENS} tokens)."
            else
              echo "Found **${DUP_COUNT}** duplication(s) across ${CHANGED_COUNT} changed Rust file(s) (threshold: ${CPD_TOKENS} tokens)."
              echo ""
              echo "<details>"
              echo "<summary>Show duplications</summary>"
              echo ""
              cat cpd-report.md
              echo ""
              echo "</details>"
            fi
          } > comment-body.md

          SIZE=$(wc -c < comment-body.md)
          if [ "$SIZE" -gt 60000 ]; then
            head -c 60000 comment-body.md > comment-body.trunc.md
            printf "\n\n_(report truncated; full output in workflow logs)_\n" >> comment-body.trunc.md
            mv comment-body.trunc.md comment-body.md
          fi

          cat comment-body.md

      - name: Post sticky PR comment
        if: github.event.pull_request.head.repo.full_name == github.repository
        env:
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          PR_NUMBER: ${{ github.event.pull_request.number }}
        shell: bash
        run: |
          set -euo pipefail
          BODY=$(cat comment-body.md)

          COMMENT_ID=$(gh api "repos/${{ github.repository }}/issues/${PR_NUMBER}/comments" \
            --paginate -q ".[] | select(.body | startswith(\"${COMMENT_MARKER}\")) | .id" | head -1)

          if [ -n "$COMMENT_ID" ]; then
            gh api "repos/${{ github.repository }}/issues/comments/${COMMENT_ID}" \
              -X PATCH -f body="$BODY"
          else
            gh pr comment "$PR_NUMBER" --body "$BODY"
          fi