atul-astronomer opened a new issue, #69173:
URL: https://github.com/apache/airflow/issues/69173
### Under which category would you file this issue?
Airflow Core
### Apache Airflow version
3.3.0rc1
### What happened and how to reproduce it?
Use the dag below:
```python
from airflow.sdk import DAG
from airflow.decorators import task
from airflow.providers.standard.operators.empty import EmptyOperator
from airflow.utils.task_group import TaskGroup
from datetime import datetime
from dataclasses import dataclass
from dataclasses_json import dataclass_json
from random import seed, randint
# I've seen, but been unable to recreate, a bug were data in a dataclass was
getting truncated part way through a string
# so lets make some string data and see if any of it gets lost
@dataclass_json
@dataclass
class Payload:
seed: int
a: str
b: str
c: str
d: str
def make(i):
return Payload(
seed=i, a=str(i) * 5, b=str(i) * 10, c=str(i) * 15, d=str(i) * 20
)
def transform(self):
self.a = str(int(int(self.a) / 5))
self.b = str(int(int(self.a) / 10))
self.c = str(int(int(self.a) / 15))
self.d = str(int(int(self.a) / 20))
@task
def mk_payloads(i):
return [Payload.make(j).to_dict() for j in range(1, i)]
@task
def transform(_payload):
payload = Payload.from_dict(_payload)
payload.transform()
return payload.to_dict()
@task
def check_transformed(_payload):
payload: Payload = Payload.from_dict(_payload)
orig = Payload.make(payload.seed)
orig.transform()
# same transformation made to same data
# results should be same, otherwise something got lost along the way
assert orig == payload
minmax_map_indices_per_lane = (5, 60)
lanes = 10
with DAG(
dag_id="many_expand", schedule=None, start_date=datetime(1970, 1, 1),
catchup=False,tags=["taskmap"]
) as dag:
seed(42) # be deterministic
for i in range(lanes):
with TaskGroup(group_id=f"lane{i+1}"):
min_map, max_map = minmax_map_indices_per_lane
mapped = randint(min_map, max_map)
check_transformed.expand(
_payload=transform.expand(_payload=mk_payloads(mapped))
)
```
https://github.com/user-attachments/assets/f18ccc65-1d5b-4a99-a5d2-736727c10a76
### What you think should happen instead?
_No response_
### Operating System
Linux
### Deployment
None
### Apache Airflow Provider(s)
_No response_
### Versions of Apache Airflow Providers
_No response_
### Official Helm Chart version
Not Applicable
### Kubernetes Version
_No response_
### Helm Chart configuration
_No response_
### Docker Image customizations
_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]
