Shiny Data Config Example

This Shiny app shows researchers how to keep code and data separated by putting paths in a config file instead of in the code. The same pattern makes eventual Azure deployment easier: point the config at wherever the data is mounted, no code edits required.

How it works

• The app reads config/config.yml via the config package; you pick the environment with R_CONFIG_ACTIVE (default/development/production).
• Each environment defines data_dir and data_file; the app builds a CSV path from those values and loads it.
• If the config package or file is missing, it falls back to data/default_data.csv.

Current sample config (config/config.yml):

default:
  data_dir: "data"
  data_file: "default_data.csv"

development:
  data_dir: "data"
  data_file: "alternate_data.csv"

production:
  data_dir: "/opt/data"
  data_file: "prod_data.csv"

Quick start (local)

1. Install config if needed: install.packages("config").

2. Run from the project root:

   R -e "shiny::runApp('app.R')"

3. By default it uses the default config (data/default_data.csv).

Switch environments with R_CONFIG_ACTIVE, e.g.:

R_CONFIG_ACTIVE=development R -e "shiny::runApp('app.R')"
R_CONFIG_ACTIVE=production  R -e "shiny::runApp('app.R')"

Edit config/config.yml to point to your own directories/files as needed.

What the UI shows

• The resolved data source (path + which config/env was active).
• A data preview and simple summaries/plots to confirm you loaded the expected file.
• A Reload button to re-read config/config.yml after edits.

Deploying to Azure (conceptual)

• Keep code (the app) and data (CSV files) separate; mount or map your data location (Azure Files, Blob via fuse, or disk).
• Set R_CONFIG_ACTIVE=production in the Azure service and adjust the production block in config/config.yml to the mounted path/filename before deploying.
• No code changes between local and cloud—only the config differs.

Files of interest

• app.R – Shiny app; reads config, resolves the data path, and renders the UI.
• config/config.yml – environment-specific data locations.
• data/default_data.csv and data/alternate_data.csv – sample datasets for local runs.

Key takeaway

Keep data locations in configuration, not code. Swap configs (or environments) to move from local to Azure without touching the Shiny logic.