Skip to main content

How to update Data Docs after validating a Checkpoint

This guide will explain how to use a Validation Action to update Data Docs sites with new Validation Results from running a Checkpoint.

Prerequisites: This how-to guide assumes you have:

Steps#

  1. Update your Checkpoint

    A Checkpoint's action_list contains a list of Validation Actions. After the checkpoint is validated, these actions are called in order. Add a Validation Action to the end of the action_list and name it update_data_docs. Validation Actions are required to have a single field, action. Inside of the action field, a class_name field must be defined, which determines which class will be instantiated to execute this action. Add class_name: UpdateDataDocsAction to the action.

    Note:

    The StoreValidationResultAction action must appear before UpdateDataDocsAction action, since Data Docs are rendered from Validation Results from the store.

     action_list:   - name: store_validation_result     action:       class_name: StoreValidationResultAction   - name: store_evaluation_params     action:       class_name: StoreEvaluationParametersAction   - name: update_data_docs     action:       class_name: UpdateDataDocsAction

  2. Specify Data Docs sites (optional):

    • By default, the UpdateDataDocsAction updates all Data Docs sites found within your project. To specify which Data Docs sites to update, provide a site_names key to the action config inside your UpdateDataDocsAction. This field accepts a list of Data Docs site names, and when provided, will only update the specified sites.
     action_list:   - name: store_validation_result     action:       class_name: StoreValidationResultAction   - name: store_evaluation_params     action:       class_name: StoreEvaluationParametersAction   - name: update_data_docs     action:       class_name: UpdateDataDocsAction       site_names:         - team_site

  3. Test your configuration.

    Test that your new Validation Action is configured correctly:

    1. Run the Checkpoint from your code or the CLI and verify that no errors are thrown.
    import great_expectations as gecontext = ge.get_context()checkpoint_name = "your checkpoint name here"context.run_checkpoint(checkpoint_name=checkpoint_name)
    $ great_expectations checkpoint run <your checkpoint name>
    1. Check your Data Docs sites to confirm that a new Validation Result has been added.

Additional notes#

The UpdateDataDocsAction generates an HTML file for the latest Validation Result and updates the index page to link to the new file, and re-renders pages for the suite used for that validation. It does not perform a full rebuild of Data Docs sites. This means that if you wish to render older Validation Results, you should run full Data Docs rebuild (via CLI's great_expectations docs build command or by calling context.build_data_docs()).

Additional resources#