blockr

Appearance

Extend blockr ​

blockr's plugin system lets you customize how the application behaves: add new UI panels, modify block interactions, or change how the board works.

Plugin architecture ​

A plugin is an R object with hooks that fire at specific points in the blockr lifecycle. Plugins can:

  • Add custom UI to the board (sidebars, toolbars, modals)
  • Intercept block creation and deletion
  • Modify how blocks connect
  • Add new controls to individual blocks
r
new_my_plugin <- function() {
  new_plugin(
    # UI added to the board
    ui = function(id) {
      ns <- shiny::NS(id)
      shiny::actionButton(ns("my_btn"), "Custom action")
    },
    # Server logic
    server = function(id, board) {
      shiny::moduleServer(id, function(input, output, session) {
        shiny::observeEvent(input$my_btn, {
          # React to custom UI
        })
      })
    },
    class = "my_plugin"
  )
}

Custom block output ​

How a block displays its result is an S3 method pair on the block class: block_ui(id, x, ...) supplies the output area, block_output(x, result, session) renders into it. Override both to change how a block class shows its result. The block's input controls are not affected; those come from the ui function passed to the constructor.

r
#' @export
block_ui.my_block <- function(id, x, ...) {
  shiny::tagList(
    shiny::plotOutput(shiny::NS(id, "result"))
  )
}

#' @export
block_output.my_block <- function(x, result, session) {
  shiny::renderPlot(print(result))
}

This is the same mechanism blockr.core uses itself: plot_block pairs plotOutput() with renderPlot(), transform_block renders a table (see blockr.core/R/plot-block.R).

Block registry ​

The registry is the "supermarket" for blocks. It tracks all available blocks with metadata (name, description, category, package). When you load a blockr extension package, its blocks are registered via .onLoad() and appear in the block menu, and in the AI/MCP discovery surface, which reads the same registry.

r
# Query available blocks
list_blocks()

# Register a new block (in R/zzz.R or at runtime)
register_block(
  ctor = "new_my_block",
  name = "My block",
  description = "Does something useful",
  category = "transform"
)

# Register many at once (vectorised)
register_blocks(
  ctor = c("new_filter_block", "new_select_block"),
  name = c("Filter rows", "Select columns"),
  description = c("Filter by predicate", "Pick a subset of columns"),
  category = c("transform", "transform")
)

# Unregister
unregister_blocks("my_block")

category must come from blockr.core::suggested_categories() (input, transform, structured, plot, table, model, output, utility, uncategorized). Anything else warns. Data-fetching blocks register as input, not data.

This makes collaboration easy: one team builds a package of domain-specific blocks, registers them, and they appear in every user's block menu. They also become available to the AI assistant for configuration.

Further reading ​