dbt-vertica

Warnings

• breaking Beginning in dbt Core v1.8, adapter installations no longer automatically include `dbt-core`. Users must explicitly install `dbt-core` alongside `dbt-vertica` to ensure a working environment.
  Fix: Update your installation command to `pip install dbt-core dbt-vertica`.
• breaking For incremental models using 'delete+insert' or 'merge' strategies, the `merge_columns` config parameter was refactored to `unique_key` and is now a required parameter. Existing models using `merge_columns` will cause errors.
  Fix: Update your incremental model configurations to use `unique_key` instead of `merge_columns` and ensure a unique key is provided.
• gotcha The `on_schema_change` parameter for incremental models in dbt-vertica does not support the `sync_all_columns` value. Only `ignore`, `fail`, and `append_new_columns` are currently supported.
  Fix: Configure your incremental models to use one of the supported `on_schema_change` values: `ignore`, `fail`, or `append_new_columns`.
• deprecated In dbt Core v1.8, custom defaults for global config flags in `profiles.yml` have been deprecated. These flags should now be set in the `flags` dictionary within your `dbt_project.yml`.
  Fix: Move any global config flags from `profiles.yml` to a `flags` dictionary in `dbt_project.yml`.

Install

• pip install dbt-core dbt-vertica Recommended Installation (dbt Core >= 1.8)
• pip install dbt-vertica Legacy Installation (dbt Core < 1.8)

Imports

• dbt-vertica

  dbt-vertica is a dbt adapter and is not typically imported directly into user Python code. Its functionality is exposed via dbt CLI commands and YAML configurations (e.g., profiles.yml, dbt_project.yml).

  dbt adapters extend dbt Core's capabilities, rather than providing Python classes for direct import by end-users.

Quickstart

To get started with dbt-vertica, first install the adapter and dbt-core. Then, initialize a dbt project, configure your Vertica connection details in `profiles.yml` (using environment variables for security is recommended), and finally, run `dbt debug` to verify connectivity. You can then create and run your dbt models.

# 1. Install dbt-vertica and dbt-core (recommended for dbt Core >= 1.8)
pip install dbt-core dbt-vertica

# 2. Initialize a new dbt project
dbt init my_vertica_project
cd my_vertica_project

# 3. Configure your profiles.yml (usually located at ~/.dbt/profiles.yml)
#    Ensure you replace placeholders or set environment variables.
#    Example profiles.yml entry (use your actual values or env vars):
# my_vertica_profile:
#   target: dev
#   outputs:
#     dev:
#       type: vertica
#       host: "{{ env_var('DBT_VERTICA_HOST') }}"
#       port: "{{ env_var('DBT_VERTICA_PORT', 5433) | as_number }}"
#       username: "{{ env_var('DBT_VERTICA_USERNAME') }}"
#       password: "{{ env_var('DBT_VERTICA_PASSWORD') }}"
#       database: "{{ env_var('DBT_VERTICA_DATABASE') }}"
#       schema: "{{ env_var('DBT_VERTICA_SCHEMA') }}"

# 4. Set environment variables (replace with your Vertica connection details)
import os
os.environ['DBT_VERTICA_HOST'] = os.environ.get('DBT_VERTICA_HOST', 'your_vertica_host')
os.environ['DBT_VERTICA_PORT'] = os.environ.get('DBT_VERTICA_PORT', '5433')
os.environ['DBT_VERTICA_USERNAME'] = os.environ.get('DBT_VERTICA_USERNAME', 'your_username')
os.environ['DBT_VERTICA_PASSWORD'] = os.environ.get('DBT_VERTICA_PASSWORD', 'your_password')
os.environ['DBT_VERTICA_DATABASE'] = os.environ.get('DBT_VERTICA_DATABASE', 'your_database')
os.environ['DBT_VERTICA_SCHEMA'] = os.environ.get('DBT_VERTICA_SCHEMA', 'your_schema')

# 5. Test the connection
dbt debug --target dev --profile my_vertica_project

# 6. Create your first model (e.g., models/my_first_model.sql)
# SELECT 1 as id, 'Hello dbt!' as message

# 7. Run your dbt models
dbt run --profile my_vertica_project