pr-review-collect.yml

# Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.
# SPDX-License-Identifier: Apache-2.0
#
# Workflow A: Collect PR data (unprivileged).
# Triggered by pull_request (no access to secrets). Gathers diff and metadata,
# uploads as artifact for the privileged pr-review-run.yml workflow.

name: "AI PR Review - Collect"

on:
  pull_request:
    types: [opened, synchronize]
    paths:
      - "aws-cli/**"
      - "cpp/example_code/**"
      - "dotnetv3/**"
      - "dotnetv4/**"
      - "gov2/**"
      - "javav2/example_code/**"
      - "javav2/usecases/**"
      - "javascriptv3/example_code/**"
      - "kotlin/services/**"
      - "kotlin/usecases/**"
      - "php/cross_service/**"
      - "php/example_code/**"
      - "python/cross_service/**"
      - "python/example_code/**"
      - "ruby/cross_service_examples/**"
      - "ruby/example_code/**"
      - "rustv1/cross_service/**"
      - "rustv1/examples/**"
      - "scenarios/**"
      - "swift/example_code/**"

permissions:
  contents: read

jobs:
  collect:
    runs-on: ubuntu-latest
    steps:
      - name: Collect PR metadata
        env:
          GH_TOKEN: ${{ github.token }}
          PR_NUMBER: ${{ github.event.pull_request.number }}
          PR_TITLE: ${{ github.event.pull_request.title }}
          PR_BODY: ${{ github.event.pull_request.body }}
          PR_HEAD_SHA: ${{ github.event.pull_request.head.sha }}
          PR_EVENT_ACTION: ${{ github.event.action }}
        run: |
          mkdir -p /tmp/pr-data

          # Save metadata
          cat <<EOF > /tmp/pr-data/metadata.json
          {
            "pr_number": "${PR_NUMBER}",
            "pr_title": $(echo "$PR_TITLE" | jq -Rs .),
            "pr_body": $(echo "$PR_BODY" | jq -Rs .),
            "head_sha": "${PR_HEAD_SHA}",
            "event_action": "${PR_EVENT_ACTION}",
            "repository": "${{ github.repository }}"
          }
          EOF

          # Get file list
          gh api repos/${{ github.repository }}/pulls/${PR_NUMBER}/files \
            --paginate --jq '.[].filename' > /tmp/pr-data/pr_files.txt

          # Get unified diff (with file headers for inline comment mapping)
          gh api repos/${{ github.repository }}/pulls/${PR_NUMBER} \
            --header "Accept: application/vnd.github.diff" > /tmp/pr-data/pr_diff.txt

          # Fetch full file contents for code files
          mkdir -p /tmp/pr-data/full_files
          gh api repos/${{ github.repository }}/pulls/${PR_NUMBER}/files \
            --paginate --jq '.[] | select(.status != "removed") | .filename' | while read -r file; do
            case "$file" in
              *.py|*.java|*.js|*.ts|*.cs|*.go|*.rs|*.swift|*.rb|*.php|*.cpp|*.h|*.kt|*.sh)
                echo "Fetching: $file"
                gh api "repos/${{ github.repository }}/contents/${file}?ref=${PR_HEAD_SHA}" \
                  --jq '.content' | base64 -d > "/tmp/pr-data/full_files/$(basename "$file")" 2>/dev/null || true
                ;;
            esac
          done

          # Get previous review comments if this is a synchronize event
          if [ "${PR_EVENT_ACTION}" = "synchronize" ]; then
            gh api repos/${{ github.repository }}/issues/${PR_NUMBER}/comments \
              --paginate --jq '.[] | select(.body | startswith("## 🤖 AI Code Example Review")) | .body' \
              > /tmp/pr-data/previous_reviews.txt || true

            gh api repos/${{ github.repository }}/pulls/${PR_NUMBER}/comments \
              --paginate --jq '.[] | select(.body | contains("🤖") or .pull_request_review_id) | {path: .path, line: .line, body: .body}' \
              > /tmp/pr-data/previous_inline_comments.json || true
          fi

      - name: Upload PR data artifact
        uses: actions/upload-artifact@v4
        with:
          name: pr-review-data
          path: /tmp/pr-data/
          retention-days: 1