rk.data.wrangling: Tidy Data Manipulation for RKWard

rk.data.wrangling brings modern, "tidy" data manipulation tools to the RKWard GUI. It provides a user-friendly interface for the powerful dplyr package, allowing users to perform complex batch operations—transformations, recoding, and scoring—on multiple variables simultaneously without writing complex code.

---

✨ Features

1. Batch Transform Variables

• Multi-Column Operations: Apply functions like Log, Exponential, Z-score (scale), or custom formulas to $N$ variables instantly using dplyr::across().
• Grouped Calculations: Automatically calculate statistics per group (e.g., group means) and append them to the dataset.
• Smart Naming: Customize output variable names using glue syntax (e.g., {.col}_{.fn} -> variable_mean).
• NA Handling: Toggle na.rm with a single click.

2. Batch Recode Variables

• Spreadsheet Interface: Define recoding rules ("Old Value" -> "New Value") in a clean grid.
• Type Safety: Strict handling of Numeric vs. Character/Factor types to prevent R errors.
• Default Behavior: Choose how to handle unmatched values: Copy original, set to NA, or use a specific fallback value.
• Output Control: Convert results to Factors automatically and add suffixes to prevent overwriting raw data.

3. Create Composite Score

• Row-wise Aggregation: Calculate scores for each subject/row based on a selection of items (e.g., survey scales).
• Methods: Supports Mean (rowMeans), Sum (rowSums), Median, SD, Variance, Min/Max, and Valid Count.
• Safety: Explicit control over Missing Value removal (na.rm).

---

🚀 What's New in Version 0.1.6

• Batch Transform Bug Fix: Fixed a critical JavaScript ReferenceError (func_call is not defined) that prevented R code generation in the Batch Transform component.
• Grouped Operations Restored: Resolved an issue where grouped calculations (dplyr::group_by and ungroup) were accidentally dropped from the generated script. Grouped batch transformations now work perfectly.
• Internal Cleanup: Removed duplicated code blocks in the Batch Recode JS generator to improve plugin stability and execution speed.

🚀 What's New in Version 0.1.5

• Attribute & Label Preservation: Fixed an issue where recoding or coercing variables to factors destroyed their metadata. The generated R code now uses a safe wrapper that automatically preserves label and .rk.meta attributes, ensuring your dataset remains properly documented within the RKWard environment.

🚀 What's New in Version 0.1.4

• Extreme Variable Name Safety: Fixed a critical bug where variables containing reserved R symbols (like $, e.g., filter_$) would break the generated script. All variable names are now safely wrapped in backticks (`) during code generation.
• Smart NA Handling in Recode: The plugin now automatically detects NA text inputs and intelligently translates them to NA_character_ when the output type is Character/Factor, preventing strict type-matching errors in dplyr::case_match.
• UI Enhancements: Added explicit on-screen instructions in the Batch Recode dialog for setting missing values (fully translatable across all supported languages).

🚀 What's New in Version 0.1.3

• Recode Fix for Factors: Resolved a strict type error in dplyr::case_match when recoding Factor variables. The plugin now automatically handles the necessary type conversions (to character) when defining text-based rules, preventing "loss of generality" errors during batch recoding.

🚀 What's New in Version 0.1.2

This maintenance release focuses on stability and code generation accuracy:

• Composite Score Fix: Resolved a syntax error in the "Create Composite Score" component that occurred when saving results to complex object paths (e.g., survey_list[["score"]]). The generated R code now correctly escapes these names.
• Internal Stability: Enforced strict separation between internal calculation variables and user-defined output names across all components, preventing potential naming collisions.

🚀 What's New in Version 0.1.1

• Robust Variable Handling: Fixed issues where variables with spaces or special characters (e.g., Mucha confianza) caused syntax errors. Variable names are now correctly quoted in generated R code.
• Complex Object Output: Fixed bugs when saving results to complex object paths. Quotes in object names are now properly escaped.
• Recode Matrix Improvements: The spreadsheet interface for recoding now correctly accepts mixed data types (text/numbers) without blocking the "Submit" button or showing validation errors.

---

🛡️ Universal Features

• Live Preview: All plugins include a data preview window that shows the result of the operation on the first 50 rows, allowing you to verify logic before processing the full dataset.
• Non-Destructive: By default, operations create new variables or new objects, preserving your original data unless you explicitly choose to overwrite.
• Internationalization: Fully localized interface available in:
  • 🇺🇸 English (Default)
  • 🇪🇸 Spanish (es)
  • 🇫🇷 French (fr)
  • 🇩🇪 German (de)
  • 🇧🇷 Portuguese (Brazil) (pt_BR)

📦 Installation

This plugin is not yet on CRAN. To install it, use the remotes or devtools package in RKWard.

1. Open RKWard.

2. Run the following command in the R Console:

   # If you don't have remotes installed:
   # install.packages("remotes")
   
   local({
     require(remotes)
     install_github("AlfCano/rk.data.wrangling", force = TRUE)
   })
   

3. Restart RKWard to load the new menu entries.

💻 Usage

Once installed, the tools are organized under:

Data -> Data Wrangling

1. Batch Transform
2. Batch Recode
3. Create Composite Score

🛠️ Dependencies

This plugin relies on the following R packages:

• dplyr (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.

✍️ Author & License

• Author: Alfonso Cano (alfonso.cano@correo.buap.mx)
• Assisted by: Gemini, a large language model from Google.
• License: GPL (>= 3)