From 03b2b06e90f34e67fbda1172eab2c7de6dc8246a Mon Sep 17 00:00:00 2001
From: Thomas Desrosiers <681004+thomasdesr@users.noreply.github.com>
Date: Tue, 8 Mar 2022 16:21:13 -0500
Subject: [PATCH] fix: Update time grain expressions for Spark >= 3.x (#18690)

* Fix the time grain expressions for Spark >= 2.3.0

Spark removed date format string 'u' in Spark 3.0. Switch to using date_trunc which has been around since 2.3

* Review: Pull out time_grain_expressoins into its own thing
---
 superset/db_engine_specs/databricks.py | 25 +++++++++++++++++++++----
 1 file changed, 21 insertions(+), 4 deletions(-)

diff --git a/superset/db_engine_specs/databricks.py b/superset/db_engine_specs/databricks.py
index 50ea59bae..f5f46bf49 100644
--- a/superset/db_engine_specs/databricks.py
+++ b/superset/db_engine_specs/databricks.py
@@ -21,6 +21,24 @@ from typing import Any, Dict, Optional
 from superset.db_engine_specs.base import BaseEngineSpec
 from superset.db_engine_specs.hive import HiveEngineSpec
 
+time_grain_expressions = {
+    None: "{col}",
+    "PT1S": "date_trunc('second', {col})",
+    "PT1M": "date_trunc('minute', {col})",
+    "PT1H": "date_trunc('hour', {col})",
+    "P1D": "date_trunc('day', {col})",
+    "P1W": "date_trunc('week', {col})",
+    "P1M": "date_trunc('month', {col})",
+    "P3M": "date_trunc('quarter', {col})",
+    "P1Y": "date_trunc('year', {col})",
+    "P1W/1970-01-03T00:00:00Z": (
+        "date_trunc('week', {col} + interval '1 day') + interval '5 days'"
+    ),
+    "1969-12-28T00:00:00Z/P1W": (
+        "date_trunc('week', {col} + interval '1 day') - interval '1 day'"
+    ),
+}
+
 
 class DatabricksHiveEngineSpec(HiveEngineSpec):
     engine = "databricks"
@@ -28,16 +46,15 @@ class DatabricksHiveEngineSpec(HiveEngineSpec):
     driver = "pyhive"
     _show_functions_column = "function"
 
+    _time_grain_expressions = time_grain_expressions
+
 
 class DatabricksODBCEngineSpec(BaseEngineSpec):
     engine = "databricks"
     engine_name = "Databricks SQL Endpoint"
     driver = "pyodbc"
 
-    # the syntax for the ODBC engine is identical to the Hive one, so
-    # we can reuse the expressions from `HiveEngineSpec`
-    # pylint: disable=protected-access
-    _time_grain_expressions = HiveEngineSpec._time_grain_expressions
+    _time_grain_expressions = time_grain_expressions
 
     @classmethod
     def convert_dttm(