RKWard Plugin: Pivot Reshape (rk.pivot.reshape)

An RKWard plugin package to easily reshape data frames by pivoting them from wide to long (pivot_longer) or long to wide (pivot_wider), powered by the tidyr package.

Overview

This plugin package provides two powerful data reshaping tools within the RKWard graphical user interface, making common data wrangling tasks more accessible. It acts as a friendly front-end for two core functions from the popular tidyr package:

1. Pivot Longer: Lengthens data by converting multiple columns into two new columns: a "key" column (containing the original column headers) and a "value" column. This is useful for tidying data for analysis and plotting.
2. Pivot Wider: Widens data by taking a key-value pair of columns and spreading them out into multiple new columns. This is the inverse of pivot_longer and is often used for creating summary tables or preparing data for specific modeling formats.

Features

Pivot Longer

• Select a data frame and two or more columns to pivot.
• Specify custom names for the new "key" (names_to) and "value" (values_to) columns.
• Choose a strategy for handling potentially non-unique column names (names_repair).
• Optionally remove rows where the new value column contains NA (values_drop_na).
• Save the reshaped, longer data frame to a new R object.

Pivot Wider (Updated in v0.01.10)

• Select a long-format data frame to widen.
• Multiple Value Columns: Select one or more columns whose values will fill the cells of the new columns (values_from).
• Automatic Sequence Generation: Automatically handles duplicate identifiers (where a single ID has multiple rows) by generating an index/sequence column.
  • Example: If a Teacher has 3 classes, it creates a counter (1, 2, 3) to spread the classes into columns Class_1, Class_2, Class_3.
  • Option to include this generated index variable in the final output.
• Specify the "key" column whose values will become the new column headers (names_from).
• Optionally select any ID columns (id_cols) to keep fixed.
• Save the reshaped, wider data frame to a new R object.

Internationalization

Internationalization (i18n): Fully localized interface available in multiple languages.

Supported Languages

As of version 0.01.10, this plugin is available in:

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

Installation

With devtools (Recommended)

You can install this plugin directly from its repository using the devtools package in R.

# If you don't have devtools installed:
# install.packages("devtools")
local({
## Prepare
require(devtools)
## Compute
  install_github(
    repo="AlfCano/rk.pivot.reshape",
    force = TRUE
  )
## Print result
rk.header ("Installation from git result")
})

Manual Installation

1. Download this repository as a .zip file.
2. In RKWard, go to Settings -> R Packages -> Install package(s) from local zip file(s) and select the downloaded file.
3. Restart RKWard (This is required for the new plugin and translations to load).
4. The plugin will be available in the Data menu.

Usage

Once installed, the plugins can be found in the Data -> Pivot reshape menu in RKWard.

Using "Pivot Longer"

1. Navigate to Data -> Pivot reshape -> Pivot Longer.
2. Select the input data.frame you want to make longer.
3. In the "Columns to pivot (cols)" slot, select at least two columns that you want to gather.
4. Optionally, change the default names for the new key and value columns.
5. Set other options as needed and specify a name for the output object.
6. Click Submit.

Using "Pivot Wider"

1. Navigate to Data -> Pivot reshape -> Pivot Wider.
2. Select the input data.frame you want to make wider.
3. Tab: Data & Values:
   • Select the ID columns (optional) and the Column(s) for cell values.
4. Tab: Column Naming:
   • Standard Method: Select the column containing the new headers in "Column for new column names".
   • Sequence Method: If you have duplicates, check "Generate Sequence/Index Column". This will disable the standard name selector and automatically create headers based on a sequence (e.g., variable_1, variable_2).
5. Tab: Output:
   • Specify a name for the output object.
   • Click Preview to check the result.
6. Click Submit.

Output

Both plugins are designed for data transformation. Their primary output is a new data frame object saved to your R workspace with the name you specified. A confirmation message will appear in the RKWard Output window upon successful completion.

Dependencies

This plugin relies on the following R packages:

• tidyr (Core logic)
• rkwarddev (Plugin generation)

Troubleshooting: Errors installing devtools or missing binary dependencies (Windows)

If you encounter errors mentioning "non-zero exit status", "namespace is already loaded", or requirements for compilation (compiling from source) when installing packages, it is likely because the R version bundled with RKWard is older than the current CRAN standard.

Workaround: Until a new, more recent version of R (current bundled version is 4.3.3) is packaged into the RKWard executable, these issues will persist. To fix this:

1. Download and install the latest version of R (e.g., 4.5.2 or newer) from CRAN.
2. Open RKWard and go to the Settings (or Preferences) menu.
3. Run the "Installation Checker".
4. Point RKWard to the newly installed R version.

This "two-step" setup (similar to how RStudio operates) ensures you have access to the latest pre-compiled binaries, avoiding the need for RTools and manual compilation.

License

This plugin is licensed under the GPL (>= 3).

Author

• Alfonso Cano (alfonso.cano@correo.buap.mx)
• Assisted by Gemini, a large language model from Google.