Add an environment variable to switch on/off sort columns

redash/models/__init__.py

@@ -216,7 +216,7 @@ def get_schema(self, refresh=False):
             schema = query_runner.get_schema(get_stats=refresh)
 
             try:
-                out_schema = self._sort_schema(schema)
+                out_schema = self._sort_schema(schema, settings.SCHEMAS_SORT_COLUMNS)
             except Exception:
                 logging.exception("Error sorting schema columns for data_source {}".format(self.id))
                 out_schema = schema
@@ -226,11 +226,17 @@ def get_schema(self, refresh=False):
 
         return out_schema
 
-    def _sort_schema(self, schema):
-        return [
-            {"name": i["name"], "columns": sorted(i["columns"], key=lambda x: x["name"] if isinstance(x, dict) else x)}
-            for i in sorted(schema, key=lambda x: x["name"])
-        ]
+    def _sort_schema(self, schema, sort_columns=True):
+        if sort_columns:
+            return [
+                {
+                    "name": i["name"],
+                    "columns": sorted(i["columns"], key=lambda x: x["name"] if isinstance(x, dict) else x),
+                }
+                for i in sorted(schema, key=lambda x: x["name"])
+            ]
+        else:
+            return sorted(schema, key=lambda x: x["name"])
 
     @property
     def _schema_key(self):

redash/settings/__init__.py

@@ -49,6 +49,7 @@
 # default set query results expired ttl 86400 seconds
 QUERY_RESULTS_EXPIRED_TTL = int(os.environ.get("REDASH_QUERY_RESULTS_EXPIRED_TTL", "86400"))
 
+SCHEMAS_SORT_COLUMNS = parse_boolean(os.environ.get("REDASH_SCHEMAS_SORT_COLUMNS", "true"))
 SCHEMAS_REFRESH_SCHEDULE = int(os.environ.get("REDASH_SCHEMAS_REFRESH_SCHEDULE", 30))
 
 AUTH_TYPE = os.environ.get("REDASH_AUTH_TYPE", "api_key")

tests/models/test_data_sources.py

@@ -58,7 +58,28 @@ def test_schema_sorter(self):
             {"name": "zoo", "columns": ["is_cow", "is_snake", "is_zebra"]},
         ]
 
-        real_output = self.factory.data_source._sort_schema(input_data)
+        real_output = self.factory.data_source._sort_schema(input_data, sort_columns=True)
+
+        self.assertEqual(real_output, expected_output)
+
+    def test_schema_sorter_columns_unsorted(self):
+        input_data = [
+            {"name": "zoo", "columns": ["is_zebra", "is_snake", "is_cow"]},
+            {
+                "name": "all_terain_vehicle",
+                "columns": ["has_wheels", "has_engine", "has_all_wheel_drive"],
+            },
+        ]
+
+        expected_output = [
+            {
+                "name": "all_terain_vehicle",
+                "columns": ["has_wheels", "has_engine", "has_all_wheel_drive"],
+            },
+            {"name": "zoo", "columns": ["is_zebra", "is_snake", "is_cow"]},
+        ]
+
+        real_output = self.factory.data_source._sort_schema(input_data, sort_columns=False)
 
         self.assertEqual(real_output, expected_output)