Added validation configuration for EPA_airqualityindex

scripts/us_epa/airqualityindex/manifest.json

@@ -20,10 +20,11 @@
                 }
             ],
             "cron_schedule": "15 22 10 1,7 *",
+            "validation_config_file": "validation_config.json",
             "resource_limits": {
                 "cpu": 32,
                 "memory": 128,
-                "disk": 512
+                "disk": 300
             }
         }
     ]

scripts/us_epa/airqualityindex/validation_config.json

@@ -0,0 +1,28 @@
+{
+    "schema_version": "1.0",
+    "rules": [
+        {
+            "rule_id": "check_deleted_records_percent",
+            "description": "Strictly enforce historical deletion average threshold of 0.1%",
+            "validator": "DELETED_RECORDS_PERCENT",
+            "params": {
+                "threshold": 0.1
+            }
+        },
+        {
+            "rule_id": "check_max_date_consistent",
+            "description": "Checks if the MaxDate is the same for all StatVars.",
+            "validator": "MAX_DATE_CONSISTENT",
+            "params": {}
+        },
+        {
+            "rule_id": "check_max_date_freshness",
+            "description": "Verifies that the dataset's maximum observation date meets freshness requirements (within 1 year of current year)",
+            "validator": "SQL_VALIDATOR",
+            "params": {
+                "query": "SELECT MAX(CAST(LEFT(CAST(MaxDate AS VARCHAR), 4) AS INT)) AS max_year FROM stats",
+                "condition": "max_year >= (EXTRACT(YEAR FROM CURRENT_DATE) - 1)"
+            }
+        }
+    ]
+}