Skip to content

Update default config/allow config #16

@mikeAdamss

Description

@mikeAdamss

Currently databaker defaults the value and marker columns to OBS and DATAMARKER respectively (which the DEs then have to change).

This is a callback to the old WDA days and is defined here:

('observation', "OBS"),
('data_marking', "DATAMARKER"),

Ideally, we should refactor this to take a proper config file (eg use Value and Marker) for new pipelines without changing it globally (otherwise all the old pipelines will break).

For mvp, it does support including a structure_csv_user.py in your databaker install which will be used in preference to structure_csv_default.py (so config, just config via py file).

Ideally we'd want a solution where this can be configured via an environment variable (so different pipelines have different config) to keep everything backwards compatible.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions