Start with a Template

Overview

A template is a pre-defined set of guidelines for creating the objects needed for a flow that solves a specific use case. Templates enable you to quickly create a flow by guiding you through the process of importing a dataset, transforming the data, and publishing outputs to the specified destination. A template consists of placeholders for the following flow objects:

• Imported dataset
• Recipe
• Output

To create a working flow, into which you can import data, transform it, and then run a job to generate the desired output, you must specify one of each of the above objects in the template. Details are below.

How to use a template

Follow these steps to use a template.

Steps :

1. From the Home page, select the required template. For more information, see Home Page.

   NOTE: If you want to use a blank flow, you can select the Create a new flow option. For more information, see Create Flow Page.

2. A new flow is created, with the Template Name  - X, where:
   1. Template Name is the name of the template you have selected

   2. X is a number. 

      Tip: Click the Template Name  - X,  to enter a flow name and description.

3. Review the guidelines to populate each placeholder in the template. A template has been specified when you have configured one instance of each of the following:
   1. Dataset: Click the Dataset placeholder to select a dataset to import. For more information, see Import Data Page.
   2. Recipe: After you import a dataset, an empty recipe is created for you. To build your recipe, click the recipe placeholder. For more information, see Transformer Page.

   3. Output: Click the output placeholder. Specify the file or table output to which results are written. For more information, see Create Outputs.

      The Create Output dialog box is displayed.
      

      Figure: Create Output dialog

      1. Select the required loading option. For more information, see Loading Options below.
      2. Then, specify the table that you are loading. For example, for BigQuery, after you select the Project, you can specify the Dataset and Table to load. 
      3. To add the output, click Save.
      4. The output is saved in the selected destination.
4. When finished, the specified objects are displayed in the flow. For more information, see Flow View Page.
5. To run a job, select the output object, and click Run. For more information, see Run Job Page.

Loading Options

The following options are available for loading a table:

• Create a new table: For each job run, a new table is created with the same base name in the selected publishing destination.
• Replace data only (Truncate): For each job run, all data in the table is truncated and replaced with any new results.
• Append to table: For each job run,  new results are added to the end of the table.
• Drop the table: With each run, the table is dropped (deleted). A new table with the same name is created, and any new results are added to it.