Skip to content

feat: Add translation logic from UUID to Source ID and back in Indicator Selection Chain #172#173

Open
veryniceuser wants to merge 5 commits intodevelopmentfrom
feature/translate_dataset_uuids_for_indicator_selection_prompt
Open

feat: Add translation logic from UUID to Source ID and back in Indicator Selection Chain #172#173
veryniceuser wants to merge 5 commits intodevelopmentfrom
feature/translate_dataset_uuids_for_indicator_selection_prompt

Conversation

@veryniceuser
Copy link
Contributor

@veryniceuser veryniceuser commented Feb 27, 2026

Applicable issues

Description of changes

Checklist

By submitting this pull request, I confirm that my contribution is made under the terms of the MIT license.

@veryniceuser veryniceuser requested a review from ypldan as a code owner February 27, 2026 09:31
@veryniceuser veryniceuser removed the request for review from ypldan February 27, 2026 09:38
@veryniceuser veryniceuser changed the title Add translation logic from UUID to Source ID and back in Indicator Selection Chain feat: Add translation logic from UUID to Source ID and back in Indicator Selection Chain #172 Feb 27, 2026
dataset_alias_to_name: dict[str, str],
dataset_id_to_source_id: dict[str, str] | None = None,
):
self.dataset_alias_to_name = dataset_alias_to_name
Copy link
Contributor

Choose a reason for hiding this comment

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

let's avoid using "alias" term - to avoid confusion with dimension_alias dataset config field

pass

@abstractmethod
def translate_dataset_ids(self, source_id_to_dataset_id: dict[str, str]) -> None:
Copy link
Contributor

Choose a reason for hiding this comment

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

let's use translate_forward and translate_backward for clarity

@navalnica
Copy link
Contributor

let's get back to this PR after we move to gpt-5* models - to simplify transition to gpt-5* (avoid unnecessary evals). we'll need to run following evals and compare:

  • data query, before translation changes
  • data query, after translation changes

to think: on full set of tests / subset, for one/multiple clients, with/without repro

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Avoid using Dataset UUIDs in prompt for indicator selection chain

3 participants