Page tree

 

Contents:


An exported flow can be imported into  Trifacta Wrangler Enterprise.

  • Dev instance: If you are using an instance of the platform for developing and testing your flows, you can import a new flow through the Flows page. 

    NOTE: Unless your instance of the platform has been specifically configured to support deployment management, you are using a Dev instance of the platform.

    NOTE: If you are attempting to share a flow with other users on the same instance of the platform, you should use the sharing functions. See Overview of Sharing.

  • Prod instance: If you are importing a flow into a Production instance of the platform, you import it as a package through the Deployment Manager.

    NOTE: Deployment Manager is a feature that enables segmentation of platform usage between Dev instances and Prod instances. This feature must be enabled and configured. For more information, see Overview of Deployment Manager.



Limitations

You cannot import flows that were exported from a different edition, release, or build of the product.

NOTE: You cannot import flows into a Dev instance that is not the same instance from which it was exported.

Imported flows do not contain the following objects:

  • Reference datasets
  • Samples

  • Connections

    NOTE: Exported flows do not contain connections. If your flow relies on a connection to the source, you must create the connection in the Prod environment and create an import mapping rule to assign the local connection ID to the import package. Flows that do not require connections may not require remapping before import. See Define Import Mapping Rules.

    NOTE: If the flow's output object uses connections that are not used for importing datasets in the flow, the output is broken on import. Those outputs and their associated connections must be recreated in the environment into which the flow is imported.


Define import rules

Before you import a package, you may need to create import mapping rules to apply to your package. For example, if the Development data is stored in a different location than the Production data, you may need to create import rules to remap paths and connections to use to acquire the data from the Production environment. 

NOTE: Import rules are applied at the time of import. They cannot be retroactively applied to releases that have already been imported.

For more information, see Define Import Mapping Rules.



Import


NOTE: You cannot import into a Dev instance if your account for the instance contains the Deployment role.



NOTE: When you import a package, you import a ZIP file. Please re-ZIP before you import.

Steps:

  1. Export the flow from the source system. See Export Flow.
  2. Login to the import system, if needed.
  3. Click Flows.
  4. From the context menu in the Flow page, select Import Flow.
  5. Select the ZIP file containing the exported flow. Click Open.

The flow is imported and available for use in the Flows page.

NOTE: After import, the createdAt timestamp for the flow references the date of creation of the flow that was originally exported. This is a known issue.

Import into Prod instance

After creating any import rules in your Prod instance, please do the following.

Steps:

  1. Export the flow from the source system. See Export Flow.
  2. Login to the Prod instance. The Deployment Manager is displayed.
  3. Click Deployments.
  4. Select or create the deployment into which to import the package.
  5. Within the deployment, click Import Package.
  6. Select the ZIP file containing the exported flow. Click Open.
    1. Any defined import rules are applied to the package during import.
    2. The package is selected as the active one for the deployment.

Tip: After you import, you should open the flow in Flow View and run a job to verify that the import was successful and the rules were applied. See Flow View Page.


Troubleshooting

Problem - Imported flow containing imported Excel dataset fails to load

If you have loaded an imported flow that contains an imported Excel dataset, the recipe for it may fail to load in the Transformer page with the following error:

Input validation failed: (Cannot read property 'filter' of undefined)

Additionally, the preview in Flow View may not display any sampled data.

Solution:

NOTE: This is a known issue that will be addressed in a future release.

For now, the workaround is the following:

  1. Open the imported flow.
  2. If a recipe doesn't exist that depends on this dataset, create a new one for it.
  3. Click the recipe for the imported Excel dataset.
  4. Click the output object for the recipe.
  5. Run the job. The success and results of the job are immaterial.
  6. After the job completes, load the recipe in the Transformer page.
  7. If no data appears in the grid, create a new sample. See Samples Panel.

For more information, see Flow View Page.


This page has no comments.