Copy
Ask AI
version: 2
models:
- name: dim_data_dictionary
description: >
Metadata table used for table availability, data freshness, and column coverage hints. It is queried by SourceMedium systems to determine which tables are present and whether they have recent data. Grain varies by row: table-level availability fields can appear alongside column-level coverage fields.
columns:
- name: dataset_name
description: >
Dataset containing the table (e.g., 'sm_transformed_v2').
- name: table_name
description: >
Table name (e.g., 'obt_orders').
- name: table_has_data
description: >
Boolean indicating whether the table has data for at least one store.
- name: table_has_fresh_data_14d
description: >
Boolean indicating whether the table has had data in the last 14 days for at least one store.
- name: table_last_data_date
description: >
Most recent date with data for the table (best-effort).
- name: table_description
description: >
Human-readable table description, when available.
- name: column_name
description: >
Column name within the table, when the row contains column-level metadata.
- name: column_null_percentage
description: >
Percent of rows where the column is null (0–100), when available.
- name: column_distinct_count
description: >
Number of distinct values observed for the column (best-effort), when available.
- name: categorical_value_distribution
description: >
Array of structs describing top categorical values and their distribution, when available.
- name: categorical_value_distribution.value
description: >
Categorical value.
- name: categorical_value_distribution.pct
description: >
Percent share for the categorical value.

