Skip to contents

Thanks for helping make Analytics Workstation more trustworthy, reproducible, and useful.

Start Here

Read these first:

  1. README.md
  2. docs/architecture_orientation.md
  3. docs/development_principles.md
  4. docs/contributor_roadmap.md

Local Setup

Use R 4.5.x when possible.

& "C:\Program Files\R\R-4.5.2\bin\Rscript.exe" scripts/install_app_dependencies.R

Then launch either from source:

shiny::runApp(".")

or from the installed package:

install.packages("release/AnalyticsShinyApp_1.0.0.tar.gz", repos = NULL, type = "source")
library(AnalyticsShinyApp)
run_workstation()

Development Rules

  • Do not add analytical features as part of cleanup PRs.
  • Keep deterministic computation separate from GenAI synthesis.
  • Use package/resource path helpers instead of hard-coded local paths.
  • Keep mutable runtime state outside the installed package.
  • Prefer small, testable changes over broad rewrites.
  • Make unsupported capabilities degrade explicitly instead of failing silently.

Validation Before Pull Request

Run the checks that match your change:

& "C:\Program Files\R\R-4.5.2\bin\Rscript.exe" -e 'source("app.R"); q <- app_env$qa_package_distribution(); print(q); stopifnot(all(q$status %in% c("success","warning")))'
& "C:\Program Files\R\R-4.5.2\bin\R.exe" CMD build .
& "C:\Program Files\R\R-4.5.2\bin\R.exe" CMD check --no-manual AnalyticsShinyApp_1.0.0.tar.gz
git diff --check

For focused work, also run the relevant testthat file under tests/testthat/.

Pull Request Checklist

  • Explain the user or contributor problem.
  • List the files changed.
  • Include validation output.
  • Note any remaining limitation honestly.
  • Include screenshots for UI changes.
  • Do not include generated runtime files, local projects, secrets, or bulky recordings.