Create schema_submission.yml

Author: Sulstice

.github/workflows/schema_submission.yml

@@ -0,0 +1,157 @@
+name: Schema Submission
+
+on:
+  issues:
+    types: [opened]
+
+jobs:
+  ingest:
+    # Only run if issue title starts with "schema-submission:"
+    if: startsWith(github.event.issue.title, 'schema-submission:')
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+      issues: write
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install dependencies
+        run: pip install -r requirements.txt
+
+      - name: Extract schema name from issue title
+        id: meta
+        run: |
+          TITLE="${{ github.event.issue.title }}"
+          # title format: "schema-submission:my-schema-name"
+          SCHEMA_NAME="${TITLE#schema-submission:}"
+          SCHEMA_NAME=$(echo "$SCHEMA_NAME" | tr -s ' ' | xargs | tr ' ' '-' | tr '[:upper:]' '[:lower:]')
+          echo "schema_name=$SCHEMA_NAME" >> $GITHUB_OUTPUT
+
+      - name: Extract YAML from issue body
+        id: extract
+        env:
+          ISSUE_BODY: ${{ github.event.issue.body }}
+        run: |
+          python3 - << 'PYEOF'
+          import os, re, sys
+
+          body = os.environ["ISSUE_BODY"]
+
+          # Accept YAML fenced in ```yaml ... ``` or bare (no fence)
+          fenced = re.search(r'```ya?ml\s*\n(.*?)```', body, re.DOTALL)
+          if fenced:
+              yaml_content = fenced.group(1)
+          else:
+              yaml_content = body.strip()
+
+          # Basic sanity check — must look like LinkML
+          if "classes:" not in yaml_content:
+              print("ERROR: issue body does not contain a valid LinkML schema (missing 'classes:' key).")
+              sys.exit(1)
+
+          schema_name = os.environ.get("SCHEMA_NAME", "unknown")
+          out_path = f"schemas/{schema_name}.yml"
+          os.makedirs("schemas", exist_ok=True)
+          with open(out_path, "w") as f:
+              f.write(yaml_content)
+          print(f"Wrote {out_path}")
+          PYEOF
+        env:
+          SCHEMA_NAME: ${{ steps.meta.outputs.schema_name }}
+
+      - name: Validate YAML parses as LinkML
+        run: |
+          python3 - << 'PYEOF'
+          import yaml, sys
+          name = "${{ steps.meta.outputs.schema_name }}"
+          path = f"schemas/{name}.yml"
+          with open(path) as f:
+              data = yaml.safe_load(f)
+          required = ["classes", "slots"]
+          missing = [k for k in required if k not in data]
+          if missing:
+              print(f"ERROR: schema missing required keys: {missing}")
+              sys.exit(1)
+          print(f"Valid LinkML schema: {len(data.get('classes', {}))} classes, {len(data.get('slots', {}))} slots")
+          PYEOF
+
+      - name: Ingest schema into registry
+        run: |
+          python neuro_registry/ingest_linkml.py \
+            --file schemas/${{ steps.meta.outputs.schema_name }}.yml \
+            --db registry.lbug
+
+      - name: Run alignment
+        run: |
+          python neuro_registry/align.py \
+            --source ${{ steps.meta.outputs.schema_name }} \
+            --db registry.lbug
+
+      - name: Export registry snapshot to JSON
+        run: |
+          python neuro_registry/export_json.py \
+            --db registry.lbug \
+            --out data/registry.json
+
+      - name: Commit schema file + updated data snapshot
+        run: |
+          git config user.name  "github-actions[bot]"
+          git config user.email "github-actions[bot]@users.noreply.github.com"
+          git add schemas/${{ steps.meta.outputs.schema_name }}.yml data/registry.json
+          git diff --cached --quiet || git commit -m \
+            "feat(schema): ingest ${{ steps.meta.outputs.schema_name }} from issue #${{ github.event.issue.number }}"
+          git push
+
+      - name: Comment on issue — success
+        if: success()
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const name = "${{ steps.meta.outputs.schema_name }}";
+            const pagesUrl = `https://${context.repo.owner}.github.io/${context.repo.repo}`;
+            await github.rest.issues.createComment({
+              owner: context.repo.owner,
+              repo:  context.repo.repo,
+              issue_number: context.issue.number,
+              body: [
+                `✅ **Schema \`${name}\` ingested successfully.**`,
+                ``,
+                `- Classes and properties have been added to the registry.`,
+                `- Alignment against existing schemas has been computed.`,
+                `- Registry snapshot updated at \`data/registry.json\`.`,
+                ``,
+                `Browse the updated schema at: ${pagesUrl}`,
+              ].join("\n"),
+            });
+            await github.rest.issues.update({
+              owner: context.repo.owner,
+              repo:  context.repo.repo,
+              issue_number: context.issue.number,
+              state: "closed",
+            });
+
+      - name: Comment on issue — failure
+        if: failure()
+        uses: actions/github-script@v7
+        with:
+          script: |
+            await github.rest.issues.createComment({
+              owner: context.repo.owner,
+              repo:  context.repo.repo,
+              issue_number: context.issue.number,
+              body: [
+                `❌ **Schema ingestion failed.**`,
+                ``,
+                `Please check that your issue body contains a valid LinkML YAML schema`,
+                `with both \`classes:\` and \`slots:\` keys, optionally fenced in \`\`\`yaml ... \`\`\`.`,
+                ``,
+                `[View the workflow run for details](https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }})`,
+              ].join("\n"),
+            });