Installing arenalytics offline

This guide explains how to install arenalytics on a machine that has no internet connection. You will need the file arenalytics_offline_pkgs.zip provided by your administrator or a colleague (see the section Preparing the bundle at the bottom if you are the one creating it). The bundle can be downloaded [here].

---

Prerequisites

Make sure the following software is already installed on your machine before you start.

Software	Notes
R ≥ 4.1	https://cran.r-project.org
RStudio	https://posit.co/download/rstudio-desktop/
Rtools (Windows, source bundle only)	Not needed if you use the Windows binary bundle. Only required to compile the source bundle; download the version matching your R from https://cran.r-project.org/bin/windows/Rtools/
Xcode Command Line Tools (macOS only)	Only for the source bundle. Open the Terminal app and run xcode-select --install
Quarto CLI	Required for report generation. https://quarto.org/docs/get-started/

---

Step 1 — Create a new R project

1. Open RStudio.
2. Go to File › New Project › New Directory › New Project.
3. Name the project (e.g. arenalytics) and choose a folder on your machine.
4. Click Create Project.

You should now be inside a fresh RStudio project with an empty workspace.

---

Step 2 — Choose and extract the bundle

Two bundles are available. Pick one:

Bundle	Use when	Speed
arenalytics_offline_pkgs_WIN.zip	You are on Windows with R 4.6.x	Fast — pre-compiled, no Rtools needed
arenalytics_offline_pkgs.zip	Any other platform (macOS, Linux), or a different R version	Slow — compiled from source, needs a compiler

Windows users should use the _WIN bundle. It contains pre-compiled binaries, so installation takes seconds instead of an hour and does not require Rtools. It only works on R 4.6.x — check yours with R.version.string.

1. Copy your chosen zip into the project folder you just created (the same folder that contains the .Rproj file).
2. Extract it there:
   • Windows: right-click the zip → Extract All… → keep the default destination.
   • macOS / Linux: double-click the zip, or run in the Terminal:

unzip arenalytics_offline_pkgs.zip -d arenalytics_offline_pkgs

You should now have a sub-folder (arenalytics_offline_pkgs/ or arenalytics_offline_pkgs_WIN/) laid out as a standard CRAN repository that R can install from.

---

Step 3 — Install the packages

Open the RStudio Console pane and run the commands for your bundle.

Option A — Windows binary bundle (recommended on Windows)

Install the dependencies first (as pre-compiled binaries), then arenalytics itself (pure R, from its source tarball). Two explicit steps are used because a single type = "both" call does not reliably resolve a local file:// repo.

local_repo <- paste0("file:///", normalizePath("arenalytics_offline_pkgs_WIN", winslash = "/"))

deps <- c(
  "bsicons", "bslib", "DT", "dplyr", "ggplot2", "htmltools", "jsonlite",
  "purrr", "quarto", "rlang", "scales", "shiny", "shiny.i18n", "shinyjs",
  "shinyWidgets", "srvyr", "stringr", "tibble", "tidyr", "zip"
)

# 1. Dependencies as pre-compiled Windows binaries (no compiler, seconds)
install.packages(deps, repos = local_repo, type = "win.binary")

# 2. arenalytics itself (pure R — installs instantly, no Rtools)
install.packages("arenalytics", repos = local_repo, type = "source", dependencies = FALSE)

dependencies = FALSE is used because every required package is already listed explicitly, so R does not need to resolve (and fail to find) optional Suggests such as knitr and rmarkdown.

Option B — Source bundle (macOS, Linux, other R versions)

local_repo <- paste0("file:///", normalizePath("arenalytics_offline_pkgs", winslash = "/"))

install.packages(
  pkgs = c(
    "arenalytics",
    "bsicons", "bslib", "DT", "dplyr", "ggplot2", "htmltools",
    "jsonlite", "purrr", "quarto", "rlang", "scales", "shiny",
    "shiny.i18n", "shinyjs", "shinyWidgets", "srvyr",
    "stringr", "tibble", "tidyr", "zip"
  ),
  repos        = local_repo,
  type         = "source",
  dependencies = FALSE
)

This compiles every package from source and may take 10–60 minutes. A compiler is required (Rtools on Windows, Xcode CLT on macOS). A lot of console output is normal.

---

Step 4 — Launch

library(arenalytics)
shiny_run_arenalytics()

The arenalytics dashboard should open in your browser.

---

Troubleshooting

package 'XYZ' is not available
The package was not included in the bundle. Ask your administrator to regenerate the bundle using inst/extdata/make_offline_bundle.R.

Installation is very slow / stuck compiling (Windows)
You are using the source bundle. Switch to the Windows binary bundle (arenalytics_offline_pkgs_WIN.zip, Option A) — it is pre-compiled and installs in seconds.

package 'XYZ' ... is not available with the Windows bundle
Your R version does not match the bundle. The _WIN bundle targets R 4.6.x; check R.version.string and either install R 4.6.x or use the source bundle.

Compilation errors on Windows (source bundle)
Rtools is likely missing or the wrong version. Make sure the Rtools version matches your R version (e.g. Rtools46 for R 4.6.x).

Compilation errors on macOS
Run xcode-select --install in the Terminal and retry.

could not find function "shiny_run_arenalytics"
The arenalytics package itself was not installed. Repeat Step 3.

---

Preparing the bundle (for administrators)

If you are the one providing the bundles, run the following script on a machine with internet access. It downloads all required packages and their dependencies and produces both bundles:

source("data-raw/make_offline_bundle.R")

This creates, in data-raw/:

• arenalytics_offline_pkgs.zip — source packages (any platform)
• arenalytics_offline_pkgs_WIN.zip — Windows binaries for R 4.6.x

To target a different R version for the Windows bundle, change the win_ver variable near the top of the script. Share the appropriate bundle along with this vignette with your users.