atul-astronomer opened a new issue, #46665:
URL: https://github.com/apache/airflow/issues/46665
### Apache Airflow version
3.0.0a1
### If "Other Airflow 2 version" selected, which one?
_No response_
### What happened?
While pulling XCom, if user passes multiple map_indexes in argument then
task is failing.
```javascript
{"timestamp":"2025-02-12T06:55:44.726594","level":"error","event":"Task
failed with
exception","logger":"task","error_detail":[{"exc_type":"NotImplementedError","exc_value":"Multiple
map_indexes are not supported
yet","syntax_error":null,"is_cause":false,"frames":[{"filename":"/opt/airflow/task_sdk/src/airflow/sdk/execution_time/task_runner.py","lineno":567,"name":"run"},
{"filename":"/opt/airflow/task_sdk/src/airflow/sdk/execution_time/task_runner.py","lineno":671,"name":"_execute_task"},{"filename":"/opt/airflow/airflow/models/baseoperator.py","lineno":173,"name":"wrapper"},{"filename":"/opt/airflow/airflow/decorators/base.py","lineno":252,"name":"execute"},
{"filename":"/opt/airflow/airflow/models/baseoperator.py","lineno":173,"name":"wrapper"},{"filename":"/opt/airflow/providers/standard/src/airflow/providers/standard/operators/python.py","lineno":196,"name":"execute"},
{"filename":"/opt/airflow/providers/standard/src/airflow/providers/standard/operators/python.py","lineno":222,"name":"execute_callable"},{"filename":"/opt/airflow/airflow/utils/operator_helpers.py","lineno":261,"name":"run"},{"filename":"/files/dags/taskmap/xcom_pull.py","lineno":49,"name":"xcom_pull"},
{"filename":"/opt/airflow/task_sdk/src/airflow/sdk/execution_time/task_runner.py","lineno":277,"name":"xcom_pull"}]}]}
{"timestamp":"2025-02-12T06:55:44.726849","level":"debug","event":"Sending
request","json":"{\"state\":\"failed\",\"end_date\":\"2025-02-12T06:55:44.726813Z\",\"type\":\"TaskState\"}\n","logger":"task"}
```
### What you think should happen instead?
DAG should pass as its passing in AF2.
### How to reproduce
Run the below DAG code on AF3 tag 3.0.0a2 and notice error on task
`xcom_pull`
```python
from airflow import DAG
from airflow.models.taskinstance import TaskInstance
from datetime import datetime, timedelta
with DAG(
dag_id="taskmap_xcom_pull",
start_date=datetime(1970, 1, 1),
schedule=None,
) as dag:
@dag.task
def foo():
return "foo"
@dag.task
def identity(thing):
return thing
@dag.task
def uppercase(thing):
return thing.upper()
@dag.task(multiple_outputs=True)
def two_parts(thing):
return {"first": thing[0:3], "rest": thing[3:]}
@dag.task
def xcom_pull(**context):
ti: TaskInstance = context["ti"]
for x in [
ti.xcom_pull(task_ids="identity"),
# <airflow.models.taskinstance._LazyXComAccess object at
0x7f5d39f92430>
ti.xcom_pull(task_ids="foo"),
# foo
ti.xcom_pull(task_ids="bar"),
# []
ti.xcom_pull(task_ids=["identity", "foo", "bar", "uppercase",
"two_parts"]),
# [2, 3, 1, 'foo', [], 'A', 'B', 'C', {'first': 'ips', 'rest':
'um'}, {'first': 'lor', 'rest': 'em'}, {'first': 'dol', 'rest': 'or'}]
ti.xcom_pull(task_ids="identity", map_indexes=2),
# 3
ti.xcom_pull(task_ids="foo", map_indexes=0),
# None
ti.xcom_pull(task_ids="foo", map_indexes=2),
# None
ti.xcom_pull(task_ids=["identity", "plusfive"], map_indexes=2),
# [3]
ti.xcom_pull(task_ids=["identity", "foo", "plusfive"],
map_indexes=[0, 2]),
# [1, 3]
ti.xcom_pull(task_ids=["identity", "plusfive", "two_parts"]),
]:
print(x)
# works
(
foo()
>> identity.expand(thing=[1, 2, 3])
>> uppercase.expand(thing=["a", "b", "c"])
>> two_parts.expand(thing=["lorem", "ipsum", "dolor"])
>> xcom_pull()
)
```
### Operating System
Linux
### Versions of Apache Airflow Providers
_No response_
### Deployment
Other
### Deployment details
_No response_
### Anything else?
_No response_
### Are you willing to submit PR?
- [ ] Yes I am willing to submit a PR!
### Code of Conduct
- [x] I agree to follow this project's [Code of
Conduct](https://github.com/apache/airflow/blob/main/CODE_OF_CONDUCT.md)
--
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]
