Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Refactor _register_dataset_changes #42343

Open
wants to merge 9 commits into
base: main
Choose a base branch
from

Conversation

uranusjr
Copy link
Member

@uranusjr uranusjr commented Sep 19, 2024

Depends on #42245. Please review that PR first. This will be rebased when that one is merged. Rebased. Ready!

This makes DatasetAliasModel only be created inside DatasetManager, similar to how DatasetModel is handled. The function create_dataset_aliases is added for this. DatasetAlias is also made to only be associated to a DatasetEvent in the manager (in function register_dataset_change). This function now takes an additional keyword argument aliases that is a list of dataset aliases associated to the DatasetEvent to be created.

All the dataset manager functions are also changed to only accept public-facing dataset classes, instead of ORM models.

@boring-cyborg boring-cyborg bot added the area:Scheduler including HA (high availability) scheduler label Sep 19, 2024
@uranusjr uranusjr changed the title Refactor bulk_save_to_db Refactor _register_dataset_changes Sep 19, 2024
@uranusjr uranusjr force-pushed the refactor-register-dataset-changes branch from 9f2101b to 24e7777 Compare September 19, 2024 10:38
Instead of fetching DatasetModel one by one, do a bulk fetch into a dict
to save roundtrips to the database.
Prior to this commit, we already only create DatasetModel rows inside
the manager. This also changes how DatasetAliasModel to only be created
inside create_dataset_aliases, and only associated them to DatasetEvent
in register_dataset_change.

All the dataset manager functions are also changed to only accept
public-facing dataset classes, instead of ORM models.

The register_dataset_change function now takes an additional keyword
argument 'aliases' that is a list of dataset aliases associated to the
DatasetEvent to be created.
Prior to this commit, we already only create DatasetModel rows inside
the manager. This also changes how DatasetAliasModel to only be created
inside create_dataset_aliases, and only associated them to DatasetEvent
in register_dataset_change.

All the dataset manager functions are also changed to only accept
public-facing dataset classes, instead of ORM models.

The register_dataset_change function now takes an additional keyword
argument 'aliases' that is a list of dataset aliases associated to the
DatasetEvent to be created.
@uranusjr uranusjr force-pushed the refactor-register-dataset-changes branch from 24e7777 to a106a8a Compare September 20, 2024 06:46
@uranusjr uranusjr marked this pull request as ready for review September 20, 2024 07:28
self.log.info(
'Creating event for %r through aliases "%s"',
dataset_obj,
", ".join(alias_names),
)
dataset_manager.register_dataset_change(
task_instance=self,
dataset=dataset_obj,
extra=extra,
dataset=dataset_obj.to_public(),
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This was a bug—dataset here should be passed a Dataset, not DatasetModel. The to_public methods on DatasetModel was added just for this.

@uranusjr uranusjr force-pushed the refactor-register-dataset-changes branch from 972b885 to 7475d1b Compare September 20, 2024 08:36
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
area:Scheduler including HA (high availability) scheduler
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant