[GitHub] [airflow] uranusjr commented on a change in pull request #15794: Make SparkSqlHook use Connection

[GitBox <gi...@apache.org>]

uranusjr commented on a change in pull request #15794:
URL: https://github.com/apache/airflow/pull/15794#discussion_r630953999



##########
File path: airflow/providers/apache/spark/hooks/spark_sql.py
##########
@@ -72,13 +77,33 @@ def __init__(
         executor_memory: Optional[str] = None,
         keytab: Optional[str] = None,
         principal: Optional[str] = None,
-        master: str = 'yarn',
+        master: Optional[str] = None,
         name: str = 'default-name',
         num_executors: Optional[int] = None,
         verbose: bool = True,
-        yarn_queue: str = 'default',
+        yarn_queue: Optional[str] = None,
     ) -> None:
         super().__init__()
+
+        try:
+            conn: "Optional[Connection]" = self.get_connection(conn_id)
+        except AirflowNotFoundException:
+            conn = None
+            options = {}
+        else:
+            options = conn.extra_dejson
+
+        # Set arguments to values set in Connection if not explicitly provided.
+        if master is None:
+            if conn is None:
+                master = "yarn"
+            elif conn.port:
+                master = f"{conn.host}:{conn.port}"
+            else:
+                master = conn.host
+        if yarn_queue is None:
+            yarn_queue = options.get("queue", "default")
+
         self._sql = sql
         self._conf = conf
         self._conn = self.get_connection(conn_id)

Review comment:
       Why? I think most hooks accept an ID instead of the connection itself.




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org