rk.janitor: Data Cleaning Tools for RKWard

This package provides a suite of RKWard plugins that create a graphical user interface for the janitor R package. It implements a "Inspect -> Clean -> Validate" workflow, allowing users to sanitize dataframes, manage duplicates, and create publication-ready frequency tables without writing code.

Features / Included Plugins

This package installs a new submenu in RKWard: Data > Data Cleaning (janitor), which contains the following plugins:

• Sanitize Data: Rapidly clean "dirty" dataframes imported from Excel/CSV.

  • remove_empty(): Remove rows and columns that are entirely missing (NA).
  • remove_constant(): Remove columns containing a single constant value (zero variance).

• Manage Duplicates: A detective tool for row-level validation.

  • get_dupes(): Find and highlight duplicate rows based on specific ID columns.
  • distinct(): Remove duplicates to enforce uniqueness.

• Frequency Tables (tabyl): A modern, pipe-friendly alternative to table().

  • tabyl(): Generates frequency tables and crosstabs.
  • adorn_*() functions: Add totals, percentages, and formatting (e.g., "50.0%") with simple checkboxes.

• Inspect Cases: Quick diagnostics for large datasets.

  • head() / tail(): View the first or last N rows.
  • slice_sample(): View a random set of N rows to sanity-check data quality.

🌍 Internationalization

As of version 0.0.2, the interface is fully localized to match your RKWard language settings:

• 🇺🇸 English (Default)
• 🇪🇸 Spanish (es)
• 🇫🇷 French (fr)
• 🇩🇪 German (de)
• 🇧🇷 Portuguese (Brazil) (pt_BR)

Requirements

1. A working installation of RKWard.
2. The R packages janitor and dplyr.

   install.packages(c("janitor", "dplyr"))
   

3. The R package devtools (for installation from source).

Installation

1. Open R in RKWard.
2. Run the following commands in the R console:

local({
## Preparar
require(devtools)
## Computar
  install_github(
    repo="AlfCano/rk.janitor"
  )
## Imprimir el resultado
rk.header ("Resultados de Instalar desde git")
})

3. Restart RKWard to ensure the new menu items appear correctly.

Usage & Examples

To try the examples below, you will need the penguins dataset (included in R 4.5.2+ or the palmerpenguins package) and a custom "messy" dataset.

Step 0: Prepare Data

Run this code in the RKWard console to create a dataset with errors:

# A dataset with duplicates, empty rows, and constant columns
messy_sales <- data.frame(
  id = c(101, 102, 103, 101, NA),
  product = c("Apple", "Banana", "Orange", "Apple", NA),
  amount = c(10, 20, 15, 10, NA),
  status = c("Active", "Active", "Active", "Active", NA) # Constant column
)

# Load penguins if not already loaded
data("penguins") # or library(palmerpenguins)

Example 1: Sanitize Data

1. Navigate to Data > Data Cleaning (janitor) > Sanitize Data.
2. Select messy_sales.
3. Check Remove empty rows, Remove empty columns, and Remove constant columns.
4. Click Submit.
   • Result: The row of NAs is gone, and the status column (which contained only "Active" and NA) is removed.

Example 2: Manage Duplicates

1. Navigate to Data > Data Cleaning (janitor) > Manage Duplicates.
2. Select messy_sales.
3. Action: Select "Find Duplicates".
4. Variables to check: Select id and product.
5. Click Submit.
   • Result: It identifies that ID 101 (Apple) appears twice.

Example 3: Frequency Tables (tabyl)

1. Navigate to Data > Data Cleaning (janitor) > Frequency Tables (tabyl).
2. Row Variable: Select species (from penguins).
3. Column Variable: Select island.
4. Adornments: Check Show Totals, Show Percentages, and Format Percentages.
5. Click Submit.
   • Result: A neat crosstab showing the percentage breakdown of penguin species across islands, formatted as percentages (e.g., "30.5%") with totals.

Author

Alfonso Cano Robles (alfonso.cano@correo.buap.mx)

Assisted by Gemini, a large language model from Google.