apache / apache/airflow

Pre-populate a suggested Partition Key when manually triggering temporal partitioned DAGs

Open
#71,284 2 comments 0 reactions 1 assignee Assigned to @Lee-W View on GitHub
area:data-aware-scheduling area:UI kind:feature
Dominant language
Python
Stars
46.9k
Forks
17.8k
Avg merge
2d 9h
Merged PRs (30d)
472

Description

### Description

When manually triggering a DAG using a temporal partitioned timetable, the trigger form leaves partition_key blank.

Submitted the form without a key created an unpartitioned DagRun. Although the run may execute successfully, its asset events cannot satisfy partition-aware downstream DAGs. This is an easy operational footgun, particularly for daily asset-driven pipelines.

### Use case/motivation

Operators frequently need to manually trigger or rerun daily partitioned DAGs. Currently, they must enter the partition key manually on every trigger.

It would be useful if the form could derive a suggested key, perhaps using some custom function.

### Related issues

_No response_

### Are you willing to submit a 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)

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.