rawwar commented on code in PR #44960:
URL: https://github.com/apache/airflow/pull/44960#discussion_r1890228496
##########
providers/src/airflow/providers/databricks/operators/databricks.py:
##########
@@ -1037,17 +1040,21 @@ def _get_hook(self, caller: str) -> DatabricksHook:
caller=caller,
)
- def _get_databricks_task_id(self, task_id: str) -> str:
- """Get the databricks task ID using dag_id and task_id. Removes
illegal characters."""
- task_id = f"{self.dag_id}__{task_id.replace('.', '__')}"
- if len(task_id) > 100:
- self.log.warning(
- "The generated task_key '%s' exceeds 100 characters and will
be truncated by the Databricks API. "
- "This will cause failure when trying to monitor the task.
task_key is generated by ",
- "concatenating dag_id and task_id.",
- task_id,
+ @cached_property
+ def databricks_task_key(self) -> str:
+ return self._generate_databricks_task_key()
+
+ def _generate_databricks_task_key(self, task_id: str | None = None) -> str:
+ """Create a databricks task key using the hash of dag_id and
task_id."""
+ if not self._databricks_task_key or len(self._databricks_task_key) >
100:
+ self.log.info(
+ "databricks_task_key has not be provided or the provided one
exceeds 100 characters and will be truncated by the Databricks API. This will
cause failure when trying to monitor the task. A task_key will be generated
using the hash value of dag_id+task_id"
)
- return task_id
+ task_id = task_id or self.task_id
+ task_key = f"{self.dag_id}__{task_id}".encode()
+ self._databricks_task_key = hashlib.md5(task_key).hexdigest()
+ self.log.info("Generated databricks task_key: %s",
self._databricks_task_key)
+ return self._databricks_task_key
Review Comment:
Do we need `_generate_databricks_task_key` separately? `databricks_task_key`
can have the implementation and can be reused in the rest of the code
--
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]
