Skip to content

holstbone/dss-plugin-sort-column-names

Repository files navigation

Description

This plugin contains different tools for preparing your data

How to Set up

When installing the plugin in DSS, you will be prompted to build its code environment. Note that this plugin requires one of these Python versions: Python 3.6, Python 3.7, Python 3.8, Python 3.9, Python 3.10, Python 3.11.

How to Use

  • In your DSS project flow, click on a dataset to select it, then click on the DataPrep-Add-ons plugin on the right panel in the plugin recipe section.
  • select the component you need for your data preparation and proceed until you create and run the recipe.

Plugin Components

This plugin currently has two components:

Sort Column Names for any Dataset

This recipe component can use any DSS dataset as input. The recipe runs just like a Python recipe. Hence the recipe component can run using a local execution or containerized execution.

Sort Column Names - SQL Datasets Only

This recipe component can use only an SQL dataset as input. The recipe is executed with the SQL execution engine. The SQL execution engine could be more efficient for larger datasets than the DSS local execution engine.

License

This plugin is distributed under the Apache License version 2.0.

About

No description, website, or topics provided.

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published