Implement automatic changelog generation for PRs

Co-authored-by: saulshanabrook <[email protected]>

Author: Copilot

.github/scripts/update_changelog.py

@@ -0,0 +1,96 @@
+#!/usr/bin/env python3
+"""
+Script to automatically update the changelog with PR information.
+"""
+
+import argparse
+import re
+import sys
+from pathlib import Path
+
+
+def find_unreleased_section(lines):
+    """Find the line number where UNRELEASED section starts and ends."""
+    unreleased_start = None
+    content_start = None
+    
+    for i, line in enumerate(lines):
+        if line.strip() == "## UNRELEASED":
+            unreleased_start = i
+            continue
+        
+        if unreleased_start is not None and content_start is None:
+            # Skip empty lines after ## UNRELEASED
+            if line.strip() == "":
+                continue
+            else:
+                content_start = i
+                break
+    
+    return unreleased_start, content_start
+
+
+def update_changelog(changelog_path, pr_number, pr_title, pr_url):
+    """Update the changelog with the new PR entry."""
+    
+    # Read the current changelog
+    with open(changelog_path, 'r', encoding='utf-8') as f:
+        lines = f.readlines()
+    
+    # Find the UNRELEASED section
+    unreleased_start, content_start = find_unreleased_section(lines)
+    
+    if unreleased_start is None:
+        print("ERROR: Could not find '## UNRELEASED' section in changelog")
+        return False
+    
+    if content_start is None:
+        print("ERROR: Could not find content start after UNRELEASED section")
+        return False
+    
+    # Create the new entry
+    new_entry = f"- {pr_title} [#{pr_number}]({pr_url})\n"
+    
+    # Check if this PR entry already exists to avoid duplicates
+    for line in lines[content_start:]:
+        if f"[#{pr_number}]" in line:
+            print(f"Changelog entry for PR #{pr_number} already exists")
+            return False
+        # Stop checking when we reach the next section
+        if line.startswith("## ") and not line.strip() == "## UNRELEASED":
+            break
+    
+    # Insert the new entry at the beginning of the unreleased content
+    lines.insert(content_start, new_entry)
+    
+    # Write the updated changelog
+    with open(changelog_path, 'w', encoding='utf-8') as f:
+        f.writelines(lines)
+    
+    print(f"Added changelog entry for PR #{pr_number}: {pr_title}")
+    return True
+
+
+def main():
+    parser = argparse.ArgumentParser(description='Update changelog with PR information')
+    parser.add_argument('--pr-number', required=True, help='Pull request number')
+    parser.add_argument('--pr-title', required=True, help='Pull request title')
+    parser.add_argument('--pr-url', required=True, help='Pull request URL')
+    parser.add_argument('--changelog-path', default='docs/changelog.md', help='Path to changelog file')
+    
+    args = parser.parse_args()
+    
+    changelog_path = Path(args.changelog_path)
+    
+    if not changelog_path.exists():
+        print(f"ERROR: Changelog file not found: {changelog_path}")
+        sys.exit(1)
+    
+    success = update_changelog(changelog_path, args.pr_number, args.pr_title, args.pr_url)
+    
+    if not success:
+        sys.exit(1)
+
+
+if __name__ == '__main__':
+    main()

.github/workflows/update-changelog.yml

@@ -0,0 +1,53 @@
+name: Update Changelog
+
+on:
+  pull_request:
+    types: [opened, edited]
+
+jobs:
+  update-changelog:
+    # Only run if this is not a PR from a fork to avoid permission issues
+    # and not a commit made by GitHub Action to avoid infinite loops
+    if: github.event.pull_request.head.repo.full_name == github.repository && github.actor != 'github-actions[bot]'
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+      pull-requests: write
+    
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          # Checkout the PR head ref
+          ref: ${{ github.event.pull_request.head.ref }}
+          token: ${{ secrets.GITHUB_TOKEN }}
+      
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.12'
+      
+      - name: Update changelog
+        run: |
+          python .github/scripts/update_changelog.py \
+            --pr-number="${{ github.event.pull_request.number }}" \
+            --pr-title="${{ github.event.pull_request.title }}" \
+            --pr-url="${{ github.event.pull_request.html_url }}"
+      
+      - name: Check for changes
+        id: changes
+        run: |
+          if git diff --quiet docs/changelog.md; then
+            echo "changed=false" >> $GITHUB_OUTPUT
+          else
+            echo "changed=true" >> $GITHUB_OUTPUT
+          fi
+      
+      - name: Commit and push changes
+        if: steps.changes.outputs.changed == 'true'
+        run: |
+          git config --local user.email "[email protected]"
+          git config --local user.name "GitHub Action"
+          git add docs/changelog.md
+          git commit -m "Add changelog entry for PR #${{ github.event.pull_request.number }}"
+          git push

docs/changelog.md

@@ -4,6 +4,7 @@ _This project uses semantic versioning_
 
 ## UNRELEASED
 
+- Automatically Generate Changelog Entries for PRs [#312](https://github.com/egraphs-good/egglog-python/pull/312)
 - Upgrade egglog which includes new backend.
   - Fixes implementation of the Python Object sort to work with objects with dupliating hashes but the same value.
     Also changes the representation to be an index into a list instead of the ID, making egglog programs more deterministic.