The generate_model_yaml function is only writing column descriptions for tables with table materializations

[daviibf]

Describe the bug

For non-table materialized models, the generate_model_yaml method does not generate columns description.

Steps to reproduce

1. Write an intermediary model of type materialized_view:

-- models/int_orders.sql

{{ config(
    materialized='view'  -- Specifies that this model should be materialized as a view
) }}

WITH orders AS (
    SELECT
        customer_id,
        order_id,
        order_date,
        order_status,
        total_amount
    FROM {{ ref('raw_orders') }}  -- Reference a raw source table
)

SELECT *
FROM orders

2. Then, execute generate_model_yaml for this model.

version: 2

models:
  - name: int_orders
    description: ""
    columns:
      - name: customer_id
        data_type: uuid
        description: ""

      - name: order_id
        data_type: uuid
        description: ""

      - name: order_date
        data_type: timestamp
        description: ""

      - name: order_status
        data_type: character varying
        description: ""

      - name: total_amount
        data_type: numeric
        description: ""

Actual results

version: 2

models:
  - name: int_orders
    description: ""
    columns:

System information

The contents of your packages.yml file:

packages:
  - package: calogica/dbt_expectations
    version: [">=0.1.0", "<1.0.0"]
  - package: dbt-labs/audit_helper
    version: [">=0.3.0", "<1.0.0"]
  # - package: jpmmcneill/dbt_graph_theory
  #   version: [">=0.2.0", "<1.0.0"]
  - package: calogica/dbt_date
    version: [">=0.1.0", "<1.0.0"]
  - package: dbt-labs/logging
    version: [">=0.1.0", "<1.0.0"]
  - package: dbt-labs/dbt_utils
    version: [">=1.1.0", "<2.0.0"]
  - package: elementary-data/elementary
    version: 0.16.1
  - package: dbt-labs/codegen
    version: 0.13.1

Which database are you using dbt with?

• postgres
• redshift
• bigquery
• snowflake
• other (specify: ____________)

The output of dbt --version:

Core:
  - installed: 1.9.1
  - latest:    1.9.1 - Up to date!

Plugins:
  - postgres: 1.9.0 - Up to date!

The operating system you're using:
Apple M3 Pro

The output of python --version:
Python 3.11.9

[henry-yu-gce]

Wanted to chime in and say that I'm also experiencing this issue using BigQuery and materialized as anything but table (in this schema or any dependent schema)

[gabeschenz]

+1 on this defect. I am seeing the same with Snowflake, using the dbt cloud cli.

packages:
  - package: dbt-labs/codegen
    version: 0.13.1
  - package: dbt-labs/dbt_utils
    version: 1.3.0

❯ dbt --version
dbt Cloud CLI - 0.38.32 (c75d8c4750469a2ce02e61939c29dc753dec1b94 2025-03-12T21:03:52Z)

[enzoaccelerator]

I have the same issue with snowflake

  - package: dbt-labs/codegen
    version: 0.13.1
  - package: dbt-labs/dbt_utils
    version: 1.3.0```

[M0h3eN]

Same here with bigquery adapter. In my case it's actually a materialized table with microbatch incremental_strategy. I'm using

  - package: dbt-labs/codegen
    version: 0.14.0

Running:

dbt run-operation generate_model_yaml --args '{"model_names"["a_materialized_table_with_microbatch_incremental_strategy"]}'

I get the following yaml:

version: 2

models:
  - name: a_materialized_table_with_microbatch_incremental_strategy
    description: ""
    columns: