dbt Cloud

Woven pulls the catalog.json and manifest.json from the dbt Cloud CI job to accurately detect schema changes.

Did you know? If you use Woven on your Software engineer repositories, we can also sync your dbt Source layer as changes are made to your application source tables. Talk to us to turn this feature on.

You must be on the Team or Enterprise plan with dbt Cloud. API access is not available with the Developer Tier

Woven currently only supports GitHub repositories that are connected with dbt Cloud. Additionally, Woven requires the following setup for each repository

• A documentation job must be enabled on your production env

1. Provision a service token

Follow this guide from dbt to create a service token for Woven. This token needs JOB_ADMIN access to allow accessing the following APIs - accounts (read), jobs (read and trigger), runs (read), and artifacts (read).

2. Share the token with Woven

1. ACCESS_URL (use this guide)

2. API Key (The Service token generated in step 1)

Go to Woven's Portal and enter the following:

3. Create a new CI job for Woven to trigger on-demand

As part of your PRs, Woven will trigger this job with dbt build --state:modified+ --empty and then pull down the catalog and manifest files. This will not affect your existing CI job. Here are some pre-requisites

1. The job must be created in the CI environment

2. The job must have opendapi_ci_fast_generate_docs in the name

3. Disable Triggered by pull requests

4. Set the compare changes against environment to Yes; Defer to prod

4. [Optional] Set up post-CI jobs

If you'd like for Woven to trigger your normal dbt Cloud CI jobs upon successful Woven metadata updates, update your CI job with the following

1. Include opendapi_post_ci_trigger in your job's names

2. Disable Triggered by pull requests

FAQs

I have many dbt Cloud projects. Does Woven work?

Yes! just repeat steps (3) and (4) for all the projects. Make sure your service token works for all projects.

Does Woven write back into dbt semantic layer?

Yes, Woven syncs the metadata back into dbt's schema.yml files on every Pull Request so that your dbt explorer always stays in sync with Woven.

Is dbt-core supported?

Woven supports but a customized integration is required based on how your dbt PRs are tested in an integration environment. Reach out to us and we can figure it out for you.