bugraoz93 commented on PR #55282:
URL: https://github.com/apache/airflow/pull/55282#issuecomment-3263893616
> Datetime handling sometimes make Python complex :-D
Exactly :D There are three layers in the code manipulating the datetime. The
actual problem lies in the handling of datetime by httpClient and the Pydantic
model during `model_dump`.
> I am wondering a bit that you need to parse JSON after Pydantic model
dump() - I thought that is the reason for Pydantic - but if it is needed, I
trust you checked for this.
I thought the same and trusted the Pydantic `model_dump()` method. I assumed
the main was working until I started doing regression tests on beta 1 :D The
`model_dump()` in Pydantic should be able to handle datetime manipulation
compatible with clients. What happened is that httpClient couldn't convert the
date to JSON after `model_dump()`
Implementation in main
```python
self.response = self.client.post(f"dags/{dag_id}/dagRuns",
trigger_dag_run.model_dump())
```
This threw a datetime JSON that is not serializable exception because the
`model_dump` method, when creating the model dict, could not be serialised by
the httpClient.
So I tried with `model_dump_json()` in Pydantic, which returns a JSON
string, but httpClient needs a dictionary, made the code as below.
```python
self.response = self.client.post(
f"dags/{dag_id}/dagRuns",
json=json.loads(trigger_dag_run.model_dump_json())
)
```
It ended up in a method so that other operations using datetime or will use
datetime in the future should do this, not a good manipulation of the models.
Even though I didn't like the manipulation, let's keep it until I dig this deep
and find a more elegant way of solving it :)
--
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: commits-unsubscr...@airflow.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
