Introduction

In Mantle, data types specify the metadata and filetypes present on a dataset. You can think of them similar to email filters; a data type matches a dataset if the field names and types specified in the data type configuration are present on the dataset as properties. You can have multiple data types that match a dataset, and you can use data types to more efficiently query for relevant datasets based on their metadata.

Create a data type

Navigate to the Datasets section using the sidebar. On the Datasets page, the Data types section will appear on the left-hand side. To create a new data type, click on the + button next to the Data types label.

A form will appear with the following fields to fill out:

Unique ID
required

The unique identifier for your data type. This is how Mantle will keep track of the data type. This should be unique and should be one word with no no spaces.

Label
required

The name that will appear in the Mantle UI

Description

The description of the data type and what it will be used for. This field is optional.

Properties
  • Define the properties associated with this data type. Every dataset of this type will have these properties.
  • Add as many properties as you like.
  • The supported types for properties are: string, integer, float, double, Boolean, and file.

Once you have filled out the form, click OK at the bottom to create the data type.

Auto-tagging

Mantle automatically tags datasets with data types if the field names and types specified in the data type configuration are present on the dataset as properties. Within a few minutes of creating a dataset or data type, you should see your datasets be auto-tagged with the appropriate data type.