Adding Query Validation Workflow

Author: ntanwar-sumo

.github/workflows/validate-queries.yml

@@ -0,0 +1,24 @@
+name: Validate SumoLogic Queries
+on: [pull_request, push]
+
+jobs:
+  validate-sql:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: "3.10"
+
+      - name: Install dependencies
+        run: |
+          pip install requests pyyaml
+
+      - name: Run validation
+        env:
+          SUMO_LOGIC_ACCESS_ID: ${{ secrets.SUMO_LOGIC_ACCESS_ID }}
+          SUMO_LOGIC_ACCESS_KEY: ${{ secrets.SUMO_LOGIC_ACCESS_KEY }}
+        run: |
+          python scripts/validate_queries.py

scripts/sumologic_client.py

@@ -0,0 +1,46 @@
+import os
+import requests
+from datetime import datetime, timedelta
+
+class SumoLogicClient:
+    def __init__(self):
+        self.base_url = "https://api.sumologic.com/api/v1"
+        self.session = requests.Session()
+        self.session.auth = (
+            os.getenv("SUMO_LOGIC_ACCESS_ID"),
+            os.getenv("SUMO_LOGIC_ACCESS_KEY")
+        )
+
+    def test_query(self, query):
+        """Execute a query in Sumo Logic and check for results"""
+        job_id = self._create_search_job(query)
+        status = self._wait_for_job(job_id)
+        return self._check_results(job_id) if status == "DONE GATHERING RESULTS" else False
+
+    def _create_search_job(self, query):
+        end_time = datetime.utcnow()
+        start_time = end_time - timedelta(hours=24)
+        payload = {
+            "query": query,
+            "from": start_time.isoformat() + "Z",
+            "to": end_time.isoformat() + "Z",
+            "timeZone": "UTC"
+        }
+        response = self.session.post(f"{self.base_url}/search/jobs", json=payload)
+        response.raise_for_status()
+        return response.json()["id"]
+
+    def _wait_for_job(self, job_id, max_attempts=10):
+        for _ in range(max_attempts):
+            response = self.session.get(f"{self.base_url}/search/jobs/{job_id}")
+            response.raise_for_status()
+            status = response.json()["state"]
+            if status in ["DONE GATHERING RESULTS", "CANCELLED"]:
+                return status
+            time.sleep(3)
+        return "TIMEOUT"
+
+    def _check_results(self, job_id):
+        response = self.session.get(f"{self.base_url}/search/jobs/{job_id}/messages")
+        response.raise_for_status()
+        return len(response.json()["messages"]) > 0

scripts/validate_queries.sh

@@ -0,0 +1,41 @@
+#!/usr/bin/env python3
+import re
+import sys
+from pathlib import Path
+from sumologic_client import SumoLogicClient
+
+def find_sql_blocks_in_pr():
+    """Detect changed SQL blocks without file modifications"""
+    changed_files = sys.argv[1:] if len(sys.argv) > 1 else [
+        str(p) for p in Path("docs").rglob("*.md")
+        if "search-query-language" in str(p)
+    ]
+
+    sql_blocks = []
+    for file in changed_files:
+        content = Path(file).read_text()
+        sql_blocks.extend([
+            (file, sql.strip())
+            for sql in re.findall(r'```sql\n(.*?)```', content, re.DOTALL)
+        ])
+    return sql_blocks
+
+def validate_queries():
+    client = SumoLogicClient()
+    failed = False
+
+    for file, query in find_sql_blocks_in_pr():
+        print(f"Validating SQL in {file}...")
+        try:
+            if not client.test_query(query):
+                print(f"::error file={file},title=Query Validation Failed::Query returned no results")
+                failed = True
+        except Exception as e:
+            print(f"::error file={file},title=Query Execution Failed::{str(e)}")
+            failed = True
+
+    if failed:
+        sys.exit(1)
+
+if __name__ == "__main__":
+    validate_queries()