[GitHub] [airflow] uranusjr commented on pull request #18622: Use seconds for schedule_delay metric

2021-10-08 Thread GitBox
uranusjr commented on pull request #18622: URL: https://github.com/apache/airflow/pull/18622#issuecomment-939181879 You need to first fix the static checks and tests. Looks like your Union import is breaking things. -- This is an automated message from the Apache Git Service. To respond

[GitHub] [airflow] uranusjr commented on pull request #18622: Use seconds for schedule_delay metric

2021-10-07 Thread GitBox
uranusjr commented on pull request #18622: URL: https://github.com/apache/airflow/pull/18622#issuecomment-937318611 -- 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

[GitHub] [airflow] uranusjr commented on pull request #18622: Use seconds for schedule_delay metric

2021-10-06 Thread GitBox
uranusjr commented on pull request #18622: URL: https://github.com/apache/airflow/pull/18622#issuecomment-937380123 [Function parameter annotations](https://www.python.org/dev/peps/pep-3107/#parameters). The functions are already partially annotated, but we should annotate `dt` with

[GitHub] [airflow] uranusjr commented on pull request #18622: Use seconds for schedule_delay metric

2021-10-06 Thread GitBox
uranusjr commented on pull request #18622: URL: https://github.com/apache/airflow/pull/18622#issuecomment-937318611 Could you add a test for this in `tests/core/test_stats.py`? And while we're at it, we should also add annotations to `dt` in the `stats` module to make it explicit it can

[GitHub] [airflow] uranusjr commented on pull request #18622: Use seconds for schedule_delay metric

2021-10-03 Thread GitBox
uranusjr commented on pull request #18622: URL: https://github.com/apache/airflow/pull/18622#issuecomment-932884363 Agreed, the stat backend should be able to handle timedelta instead. -- This is an automated message from the Apache Git Service. To respond to the message, please log on