handle multiple PRs with schema changes

.github/workflows/update-community-prs.yaml

@@ -0,0 +1,100 @@
+name: Update Community PRs After Merge
+
+on:
+  push:
+    branches: [main]
+    paths:
+      - 'community/**'
+      - 'schemas/community_submission.*.schema.json'
+
+permissions:
+  contents: write
+  pull-requests: write
+
+jobs:
+  update-open-prs:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          token: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Setup Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install dependencies
+        run: pip install jsonschema
+
+      - name: Get current schema version
+        id: schema
+        run: |
+          # Find the latest schema version on main
+          latest=$(ls schemas/community_submission.v*.schema.json 2>/dev/null | sed 's/.*\.v\([0-9]*\)\.schema\.json/\1/' | sort -n | tail -1)
+          echo "latest_version=${latest:-1}" >> "$GITHUB_OUTPUT"
+          echo "Latest schema version: ${latest:-1}"
+
+      - name: Find and update open community PRs
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          LATEST_SCHEMA_VERSION: ${{ steps.schema.outputs.latest_version }}
+        run: |
+          # Get list of open community PRs
+          prs=$(gh pr list --label community --state open --json number,headRefName --jq '.[] | "\(.number) \(.headRefName)"')
+          
+          if [ -z "$prs" ]; then
+            echo "No open community PRs found"
+            exit 0
+          fi
+          
+          echo "$prs" | while read pr_number branch_name; do
+            echo "Processing PR #$pr_number (branch: $branch_name)"
+            
+            # Checkout PR branch
+            git fetch origin "$branch_name"
+            git checkout "$branch_name"
+            
+            # Check if this PR has a schema file that needs updating
+            pr_schema=$(ls schemas/community_submission.v*.schema.json 2>/dev/null | sed 's/.*\.v\([0-9]*\)\.schema\.json/\1/' | sort -n | tail -1)
+            
+            if [ "$pr_schema" -le "$LATEST_SCHEMA_VERSION" ] 2>/dev/null; then
+              echo "  PR schema version ($pr_schema) <= main version ($LATEST_SCHEMA_VERSION)"
+              
+              # Merge main into PR branch
+              git config user.name "github-actions[bot]"
+              git config user.email "github-actions[bot]@users.noreply.github.com"
+              
+              if git merge origin/main -m "Merge main to update schema baseline"; then
+                # Regenerate schema for this PR's submission
+                python -m src.contributions.regenerate_pr_schema || true
+                
+                # If there are changes, commit and push
+                if [ -n "$(git status --porcelain schemas/)" ]; then
+                  new_version=$((LATEST_SCHEMA_VERSION + 1))
+                  git add schemas/
+                  git commit -m "Update schema to v${new_version} (rebased on main)"
+                  git push origin "$branch_name"
+                  echo "  Updated PR #$pr_number with new schema"
+                else
+                  git push origin "$branch_name"
+                  echo "  Merged main into PR #$pr_number"
+                fi
+              else
+                echo "  Merge conflict in PR #$pr_number, adding comment"
+                gh pr comment "$pr_number" --body "⚠️ **Merge Conflict**
+
+Another community submission was merged and this PR has conflicts.
+
+A maintainer may need to:
+1. Close this PR
+2. Remove the \`approved\` label from the original issue
+3. Re-add the \`approved\` label to regenerate the PR"
+                git merge --abort
+              fi
+            fi
+            
+            git checkout main
+          done

src/contributions/regenerate_pr_schema.py

@@ -0,0 +1,68 @@
+#!/usr/bin/env python3
+"""
+Regenerate schema for a PR branch after main has been merged in.
+This script looks at the submission files in this branch and generates
+an updated schema version if new tags were introduced.
+
+Usage: python -m src.contributions.regenerate_pr_schema
+"""
+
+import json
+import sys
+from pathlib import Path
+
+# Add parent to path for imports when running as script
+sys.path.insert(0, str(Path(__file__).parent.parent.parent))
+
+from src.contributions.read_community_data import read_all_submissions, build_tag_type_registry
+from src.contributions.update_schema import (
+    get_existing_tag_definitions,
+    check_for_new_tags,
+    generate_new_schema,
+)
+from src.contributions.schema import get_latest_schema_version, load_schema, SCHEMAS_DIR
+
+
+def main():
+    """Main entry point."""
+    # Get current schema version and load it
+    current_version = get_latest_schema_version()
+    current_schema = load_schema()
+    
+    # Get existing tag definitions from schema
+    existing_tags = get_existing_tag_definitions(current_schema)
+    
+    # Read all submissions (including ones from this PR branch)
+    submissions = read_all_submissions()
+    
+    if not submissions:
+        print("No submissions found")
+        return
+    
+    # Build tag registry from all submissions
+    tag_registry = build_tag_type_registry(submissions)
+    
+    # Check for new tags not in the current schema
+    new_tags = check_for_new_tags(tag_registry, current_schema)
+    
+    if new_tags:
+        # Generate new schema version
+        new_version = current_version + 1
+        print(f"Found new tags: {new_tags}")
+        print(f"Generating schema v{new_version}")
+        
+        # Generate new schema with updated tag definitions
+        new_schema = generate_new_schema(current_schema, tag_registry, new_version)
+        
+        # Write new schema version
+        new_schema_path = SCHEMAS_DIR / f"community_submission.v{new_version}.schema.json"
+        with open(new_schema_path, 'w') as f:
+            json.dump(new_schema, f, indent=2)
+        
+        print(f"Created {new_schema_path}")
+    else:
+        print("No new tags found, schema is up to date")
+
+
+if __name__ == "__main__":
+    main()