A Shiny Template System

R-CMD-check CRAN status

The goal of shidashi is to provide framework for R-shiny templates, especially for dashboard applications.

Default template (using AdminLTE) provides two themes: light vs dark


You can install the released version of shidashi from CRAN with:


Demo & Tutorial Application

The demo app requires to install the following extra packages

install.packages(c("ggExtra", "rmarkdown"))

Once you have installed these packages, run the following command from R:

project <- file.path(tools::R_user_dir('shidashi', which = "data"), "AdminLTE3")

# `use_template` only needs to be called once

Start From Existing Templates

To start a shidashi project, open RStudio menu from the navigation bar:

File > New Project… > New Directory > Shidashi Shiny Template (*)

_*You might need to scroll down to find that template option_.

Please enter the project information accordingly. By default, the Github user is dipterix, and theme is AdminLTE3, which lead to the default template.

Bare-bone Template

If you want to start from a bare-bone template, change the theme option to be AdminLTE3-bare.

File Structure

A typical shidashi project has the following file structure:

<project root_path>
│ └─<module ID>           - Module folder; folder name is module ID
│   ├─R                   - Module functions shared across UI and server
│   ├─module-ui.html      - Module HTML template
│   └─server.R            - Module-level server function
├─R/                      - Common functions shared across modules
├─views/                  - Small snippets (see `?include_view` function)
├─www/                    - Static files: css, js, img, ...
├─index.html              - Template for homepage
├─modules.yaml            - Module label, order, icon, badge..
└─server.R                - Root server function, usually no modification is required


Create your own Github repository with name shidashi-templates. Add folders named by the themes. Then people can install your themes as templates through RStudio. An easy start is to fork this repository.