Package {datasetviewer}

datasetviewer: 'SAS Studio'-Style Interactive Dataset Viewer

Description

An interactive dataset viewer that renders a fast, scrollable grid with a column-selection panel, per-column property metadata, and a names-versus-labels header toggle, modelled on the 'SAS Studio' table viewer. Runs from one codebase in interactive 'Shiny' apps and in static HTML documents, with a free-text row filter, header sort, and CSV export, and handles large datasets without row sampling by querying them in the browser with 'DuckDB-WASM'.

Author(s)

Maintainer: Vignesh Thanikachalam about.vignesh@gmail.com [copyright holder]

Authors:

• Vignesh Thanikachalam about.vignesh@gmail.com [copyright holder]

See Also

Useful links:

• https://github.com/vthanik/datasetviewer

• https://vthanik.github.io/datasetviewer/

• Report bugs at https://github.com/vthanik/datasetviewer/issues

View a dataset in an interactive SAS Studio-style grid

Description

Renders a fast, scrollable data grid with a column-selection panel and per-column property metadata. The same widget renders in interactive Shiny apps and in static HTML documents.

Usage

dataset_viewer(
  x,
  ...,
  view = c("names", "labels"),
  width = NULL,
  height = NULL,
  elementId = NULL
)

Arguments

Details

Query engine. The data is sent to the browser once as 'Parquet' and queried in place with DuckDB-WASM, so filter, sort, and paging run over the whole dataset with no row sampling. The engine (~35 MB) loads from a CDN by default but is fetched into the package at install time when reachable, so a Shiny app can serve it to browsers with no internet at runtime. Set options(datasetviewer.use_local_engine = FALSE) to force the CDN (for small self-contained HTML). See vignette("datasetviewer") for offline and corporate deployment, including the ⁠DATASETVIEWER_DUCKDB_*⁠ install-time environment variables.

Value

An htmlwidget. Print it to render, or use it as a Shiny output.

See Also

Shiny bindings: datasetviewerOutput(), renderDatasetViewer().

Examples

# ---- Example 1: view a plain data frame ----
#
# Wrap any data frame to get the interactive grid. Printing the widget in
# an interactive session or a rendered document shows it; here we inspect
# the payload so the example stays headless and self-contained (printing a
# widget would launch a browser under R CMD check).
viewer <- dataset_viewer(mtcars)
viewer$x$n_rows

# ---- Example 2: CDISC labels as headers ----
#
# With the sibling artoo package, a CDISC-conformed frame supplies column
# labels, formats, and storage lengths to the property pane and the
# names-versus-labels header toggle. Start on labels with view = "labels".
if (requireNamespace("artoo", quietly = TRUE)) {
  labelled <- dataset_viewer(artoo::cdisc_adsl, view = "labels")
  labelled$x$columns[[1]]$label
}

Shiny bindings for the dataset viewer

Description

Output and render functions to embed dataset_viewer() in a Shiny app. The widget pushes its live view state back to Shiny as inputs, so the app can reuse the user's filter, sort, and column selection server-side.

Usage

datasetviewerOutput(outputId, width = "100%", height = "500px")

renderDatasetViewer(expr, env = parent.frame(), quoted = FALSE)

Arguments

Details

Inputs published. For an output with id "viewer" the widget sets, on every change:

• input$viewer_columns (⁠<character>⁠): selected column names.

• input$viewer_filter (⁠<character(1)>⁠): the filter expression.

• input$viewer_sort (⁠<list>⁠): sort keys (name, dir).

• input$viewer_view (⁠<character(1)>⁠): "names" or "labels".

Value

datasetviewerOutput() returns a Shiny output UI element; renderDatasetViewer() returns a Shiny render function.

See Also

dataset_viewer() for the widget these bindings embed.

Examples

# ---- Example 1: read the viewer's filter and selection server-side ----
#
# The app shows a dataset and echoes the filter expression and selected
# columns the user builds in the grid. Runs only in an interactive session.
if (interactive()) {
  library(shiny)
  ui <- fluidPage(
    datasetviewerOutput("viewer", height = "500px"),
    verbatimTextOutput("state")
  )
  server <- function(input, output, session) {
    output$viewer <- renderDatasetViewer(dataset_viewer(mtcars))
    output$state <- renderText({
      paste0(
        "filter: ", input$viewer_filter, "\n",
        "columns: ", paste(input$viewer_columns, collapse = ", ")
      )
    })
  }
  shinyApp(ui, server)
}