The #TidyTuesday data viewer

The #TidyTuesday data viewer (ttdviewer) is an Shiny app-package¹ for viewing the data from the #TidyTuesday project. This package started out as a small application to demonstrate various examples of downloadable reports (R Markdown, Quarto, etc.), but it’s grown to include additional features:

1. The data are loaded directly from GitHub (i.e., with help from the tidytuesdayR and ttmeta packages).
2. The returned object is visible as a list on the first nav_panel() via the listviewerlite package.
   
3. The data are visible as a table compliments of the reactable package in the second nav_panel().
   
4. Data visualizations from the inspectdf package ‘summarise missingness, categorical levels, numeric distribution, correlation, column types and memory usage’ are available on the third nav_panel().

Architecture

The app is packaged as a Shiny application with a single entry point that wires UI and server logic together. The server loads the selected dataset, routes it through module servers, and then exposes the reactive outputs to the UI panels.

%%{init: {'theme': 'neutral', 'look': 'handDrawn', 'themeVariables': { 'fontFamily': 'monospace', 'fontSize':'11px'}}}%%
flowchart LR
  Launch["launch()"] --> UI["app_ui()"]
  Launch --> Server["app_server()"]
  Server --> Input["mod_input_server()"]
  Input --> Data["load_tt_data()"]
  Server --> ReportInput["mod_report_input_server()"]
  ReportInput --> ReportDesc["mod_report_desc_server()"]
  ReportInput --> ReportDownload["mod_report_download_server()"]
  ReportDownload --> RenderReport["render_report()"]
  Server --> List["mod_list_server()"]
  Server --> Table["mod_table_server()"]
  Server --> Plot["mod_plot_server()"]
  Plot --> InspectPlot["inspect_plot()"]
  Data --> List
  Data --> Table
  Data --> Plot

The UI arranges module outputs into panels while the sidebar hosts the dataset selector and report generation options.

%%{init: {'theme': 'neutral', 'look': 'handDrawn', 'themeVariables': { 'fontFamily': 'monospace', 'fontSize':'13px'}}}%%
flowchart LR
  Sidebar["Sidebar inputs"] --> InputUI["mod_input_ui()"]
  Sidebar --> ReportUI["mod_report_input_ui()"]
  Sidebar --> ReportDescUI["mod_report_desc_ui()"]
  Sidebar --> ReportDownloadUI["mod_report_download_ui()"]
  Tabs["Tab panels"] --> ListUI["mod_list_ui()"]
  Tabs --> TableUI["mod_table_ui()"]
  Tabs --> PlotUI["mod_plot_ui()"]

Supporting utilities and tests wrap the core modules. Logging, data helpers, and report builders are kept in standalone R files, while tests/testthat validates module behavior and data helpers.

%%{init: {'theme': 'neutral', 'look': 'handDrawn', 'themeVariables': { 'fontFamily': 'monospace', 'fontSize':'14px'}}}%%
flowchart TB
  subgraph Utilities
    Log["logr_msg()"]
    Ctr["ctr()"]
    Load["load_tt_data()"]
    Inspect["inspect_plot()"]
    Render["render_report()"]
  end
  subgraph Modules
    ModInput["mod_input_*"]
    ModList["mod_list_*"]
    ModTable["mod_table_*"]
    ModPlot["mod_plot_*"]
    ModReport["mod_report_*"]
  end
  subgraph Tests
    TestApp["test-app_*"]
    TestModules["test-mod_*"]
    TestUtils["test-load_tt_data / test-inspect_plot / test-render_report / test-logr_msg"]
  end
  Utilities --> Modules
  Modules --> Tests
  Utilities --> Tests

Installation

You can install the development version of ttdviewer from GitHub with:

# install.packages("pak")
pak::pak("mjfrigaard/ttdviewer")

Launch app

Launch the application using the standalone app function below:

library(ttdviewer)
launch()