meta

Models

models/schema.yml

version: 2

models:
  - name: model_name
    config:
      meta: {<dictionary>}

    columns:
      - name: column_name
        meta: {<dictionary>}

The meta config can also be defined:

• under the models config block in dbt_project.yml
• in a config() Jinja macro within a model's SQL file

See configs and properties for details.

Sources

models/schema.yml

version: 2

sources:
  - name: model_name
    config:
      meta: {<dictionary>}

    tables:
      - name: table_name
        config:
          meta: {<dictionary>}

        columns:
          - name: column_name
            meta: {<dictionary>}

Seeds

seeds/schema.yml

version: 2

seeds:
  - name: seed_name
    config:
      meta: {<dictionary>}

    columns:
      - name: column_name
        meta: {<dictionary>}

The meta config can also be defined under the seeds config block in dbt_project.yml. See configs and properties for details.

Snapshots

snapshots/schema.yml

version: 2

snapshots:
  - name: snapshot_name
    config:
      meta: {<dictionary>}

    columns:
      - name: column_name
        meta: {<dictionary>}

The meta config can also be defined:

• under the snapshots config block in dbt_project.yml
• in a config() Jinja macro within a snapshot's SQL block

See configs and properties for details.

Tests

You can't add YAML meta configs for generic tests. However, you can add meta properties to singular tests using config() at the top of the test file.

Analyses

The meta config is not currently supported for analyses.

Macros

macros/schema.yml

version: 2

macros:
  - name: macro_name
    meta: {<dictionary>}

    arguments:
      - name: argument_name

Exposures

models/exposures.yml

version: 2

exposures:
  - name: exposure_name
    meta: {<dictionary>}

Semantic models

Configure meta in the your semantic models YAML file or under the semantic-models config block in the dbt_project.yml file.

Dimensions, entities, and measures can also have their own meta configurations.

models/semantic_models.yml

semantic_models:
  - name: semantic_model_name
    config:
      meta: {<dictionary>}

    dimensions:
      - name: dimension_name
        config:
          meta: {<dictionary>}

    entities:
      - name: entity_name
        config:
          meta: {<dictionary>}

    measures:
      - name: measure_name
        config:
          meta: {<dictionary>}

The meta config can also be defined under the semantic-models config block in dbt_project.yml. See configs and properties for details.

Metrics

models/metrics.yml

metrics:
  - name: number_of_people
    label: "Number of people"
    description: Total count of people
    type: simple
    type_params:
      measure: people
    config:
      meta:
        my_meta_config: 'config_value'

Saved queries

models/semantic_models.yml

saved_queries:
  - name: saved_query_name
    config:
      meta: {<dictionary>}

Definition

The meta field can be used to set metadata for a resource and accepts any key-value pairs. This metadata is compiled into the manifest.json file generated by dbt, and is viewable in the auto-generated documentation.

Depending on the resource you're configuring, meta may be available within the config property, and/or as a top-level key. (For backwards compatibility, meta is often (but not always) supported as a top-level key, though without the capabilities of config inheritance.)

Examples

Designate a model owner

Additionally, indicate the maturity of a model using a model_maturity: key.

models/schema.yml

version: 2

models:
  - name: users
    meta:
      owner: "@alice"
      model_maturity: in dev

Designate a source column as containing PII

models/schema.yml

version: 2

sources:
  - name: salesforce

    tables:
      - name: account
        meta:
          contains_pii: true
        columns:
          - name: email
            meta:
              contains_pii: true

Configure one meta attribute for all seeds

dbt_project.yml

seeds:
  +meta:
    favorite_color: red

Override one meta attribute for a single model

models/my_model.sql

{{ config(meta = {
    'single_key': 'override'
}) }}

select 1 as id

Assign owner and favorite_color in the dbt_project.yml as a config property

dbt_project.yml

models:
  jaffle_shop:
    +meta:
      owner: "@alice"
      favorite_color: red

Assign meta to semantic model

The following example shows how to assign a meta value to a semantic model in the semantic_model.yml file and dbt_project.yml file:

Semantic model

semantic_models:
  - name: transaction 
    model: ref('fact_transactions')
    description: "Transaction fact table at the transaction level. This table contains one row per transaction and includes the transaction timestamp."
    defaults:
      agg_time_dimension: transaction_date
    config:
      meta:
        data_owner: "Finance team"
        used_in_reporting: true

dbt_project.yml

semantic-models:
  jaffle_shop:
    +meta:
      used_in_reporting: true

Assign meta to dimensions, measures, entities

Semantic model

The following example shows how to assign a meta value to a dimension, entity, and measure in a semantic model:

semantic_model.yml

semantic_models:
  - name: semantic_model
    ...
    dimensions:
      - name: order_date
        type: time
        config:
          meta:
            data_owner: "Finance team"
            used_in_reporting: true
    entities:
      - name: customer_id
        type: primary
        config:
          meta:
            description: "Unique identifier for customers"
            data_owner: "Sales team"
            used_in_reporting: false
    measures:
      - name: count_of_users
        expr: user_id
        config:
          meta:
            used_in_reporting: true

dbt_project.yml

This second example shows how to assign a data_owner and additional metadata value to a dimension in the dbt_project.yml file using the +meta syntax. The similar syntax can be used for entities and measures.

dbt_project.yml

semantic-models:
  jaffle_shop:
    ...
    dimensions:
      - name: order_date
        config:
          meta:
            data_owner: "Finance team"
            used_in_reporting: true