Background
…Is there a low-code data-transformation solution in Python or R? Is there a visual way to transform tables in R? And a way to edit data directly in R? …
These recipes consider low-code interface solutions for simple editing and transformation problems outside of more formalised extract-transform-load pipelines.
There are a number of fantastic low code/ graphical interface solutions for bigger data load and transformation problems including Alteryx (closed source) and Apache NiFi (open source).
The solutions considered here are bolt-on options for existing R or Python processes with the benefit of not requiring much work to set up.
Shiny and R
Two very useful packages allow direct visual editing and transformation of r tables through shiny below. These methods are extendable to passing through edits to SQL tables
- Datamods is a package of shiny modules to import and manipulate data.
- rhandsontable is an htmlwidget with an Excel-like appearance. It comes with powerful features like data validation, sorting, grouping, data binding, formula support or column ordering.
A bare-bones app using the above is here:
Python
There are a few existing packages that deal with most of the standard transformation problems an end user might usually encounter, e.g.
- Pandas_ui discusses the usage of this package as a user interface tool for fast data manipulation in pandas.
- PandasGUI similarly discusses data analysis using the PandasGUI interface.
- Bamboolib discusses using the Bamboolib Python package for low-code data transformation work. Unlike the others, Bamboolib is closed source and owned by Databricks.
The first two options are considered in a demo, available on github:
