Skip to content

CLAUDE.md

Latest commit

 

History

History
74 lines (60 loc) · 2.61 KB

CLAUDE.md

File metadata and controls

74 lines (60 loc) · 2.61 KB

Claude Code Assistant Instructions

Project Overview

This is the forestly R package, which creates interactive forest plots for clinical trial data analysis. The package is built on top of metalite and metalite.ae packages and uses reactable for interactive tables with Plotly.js for interactive visualizations.

Development Guidelines

Testing

  • load project using devtools::load_all()
  • Run tests using: devtools::test()
  • Run specific test files using: devtools::test(filter = "filename")
  • Before running tests, ensure required packages are installed (metalite, metalite.ae, reactable, reactR)

Code Style

  • Follow tidyverse style guide
  • Use |> pipe operator (R 4.1+)
  • Functions should handle both character vectors and factors robustly

Key Functions

  • ae_forestly(): Main function to create interactive forest plots
  • format_ae_forestly(): Formats AE data and configures Plotly visualizations
  • format_ae_listing(): Formats AE listing data
  • sparkline_point_js(): Generates JavaScript/Plotly code for interactive sparkline plots
  • propercase(): Converts strings to proper case (handles factors)
  • titlecase(): Converts strings to title case using tools::toTitleCase (handles factors)

Before Committing

  • Run linting: Check for any linting issues in the IDE
  • Run tests: devtools::test() to ensure all tests pass
  • Check documentation: devtools::document() if roxygen comments are updated

Branch Strategy

  • Main branch: main
  • Feature branches: Use descriptive names like fix-factor-handling or add-new-feature
  • Always create pull requests for merging into main

Common Commands

# Load all functions for development
devtools::load_all()

# Run all tests
devtools::test()

# Check package
devtools::check()

# Build documentation
devtools::document()

Package Dependencies

  • metalite
  • metalite.ae
  • reactable
  • reactR
  • plotly (via JavaScript integration)
  • brew (for template processing)
  • tools (base R)

Testing Data

The package includes test data in data/:

  • forestly_adae.rda
  • forestly_adae_3grp.rda
  • forestly_adsl.rda
  • forestly_adsl_3grp.rda

Notes for Future Development

  • The ae_listing.R file contains functions that handle factor inputs, which was a recent fix
  • Test files should use devtools::load_all() or source the R files directly for testing
  • The package uses testthat for unit testing framework
  • Interactive plots use Plotly.js via JavaScript templates in inst/js/
  • The y-axis ordering in format_ae_forestly.R affects the visual display in interactive plots
  • When debugging interactive plot issues, check both R code and JavaScript template files