WweiL commented on code in PR #41540:
URL: https://github.com/apache/spark/pull/41540#discussion_r1230161386
##########
python/pyspark/sql/streaming/listener.py:
##########
@@ -155,11 +156,31 @@ class QueryStartedEvent:
This API is evolving.
"""
- def __init__(self, jevent: JavaObject) -> None:
- self._id: uuid.UUID = uuid.UUID(jevent.id().toString())
- self._runId: uuid.UUID = uuid.UUID(jevent.runId().toString())
- self._name: Optional[str] = jevent.name()
- self._timestamp: str = jevent.timestamp()
+ def __init__(
+ self, id: uuid.UUID, runId: uuid.UUID, name: Optional[str], timestamp:
str
+ ) -> None:
+ self._id: uuid.UUID = id
+ self._runId: uuid.UUID = runId
+ self._name: Optional[str] = name
+ self._timestamp: str = timestamp
+
+ @classmethod
+ def fromJObject(cls, jevent: JavaObject) -> "QueryStartedEvent":
+ return cls(
+ id=uuid.UUID(jevent.id().toString()),
+ runId=uuid.UUID(jevent.runId().toString()),
+ name=jevent.name(),
+ timestamp=jevent.timestamp(),
+ )
+
+ @classmethod
+ def fromJson(cls, j: Dict[str, Any]) -> "QueryStartedEvent":
Review Comment:
I think we need to still let users to use `import
pyspark.sql.streaming.Query<xxx>Event` in Connect. If we put the code in
connect folder they might need to do `import
pyspark.sql.connect.streaming.Query<xxx>Event`.
A way I can think of doing this is to refactor the code to make the event
classes abstract here:
```
class QueryStartedEvent(ABC):
@property
@abstractmethod
def id(self) -> uuid.UUID:
...
```
Do you think we should do that?
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
