Applying Transformers on Charts and Dashboards in DataHub - Customization and Future Possibilities

Original Slack Thread

<@U06TM7M40RK> is it possible to apply transformers on charts or dashboards?

It seems likely that applying transformers to charts or dashboards in DataHub is possible, but it might require some customization. According to a <Slack conversation>, most of the existing transformers are tailored for datasets, and there isn’t a specific transformer for Tableau charts. However, you can develop a custom transformer to achieve this.

Based on the <https://datahubproject.io/docs/0.14.0/actions/guides/developing-a-transformer|DataHub documentation>, you can create a custom transformer by extending the Transformer base class in Python. This involves defining the create() and transform() methods to handle the specific logic you need for charts or dashboards.

Additionally, a <Slack message> suggests that you can include a custom transformer in a UI-based metadata ingestion by editing the YAML configuration in the UI.

Would you like more detailed guidance on how to create and apply a custom transformer for charts or dashboards?

Sources:

as runLLM states the default transformers are currently only working on datasets but its easy to extend them by inheriting and just overwriting the entity_types function

Hi :slight_smile: ! Did this interaction resolve your issue or do you still require assistance ?

Please tag @ RunLLM if you still need further assistance !

Or tag @Datahub Community Support if you require a person to take a look :slight_smile:

Hope this helps ! ><

I found some commits yesterday that were showing a change related to that. it looks like transformers will apply to dashboards and charts in the next release. unfortunately I can’t find them anymore :sweat_smile:

sorry, here it is https://github.com/datahub-project/datahub/pull/11700/files