[Bug]: Transformer enters infinite recursion when lapply used to transform data.

[chlebowa]

What happened?

Following the transformer vignette, I am trying to build a transformer module that will modify my data by substituting some columns with their tranformed version.
I want to use this pattern:

data[index] <- lapply(data[index], function)

In particular, I am using scale, which is itself vectorized, so data[index] <- scale[data] is sufficient.

While this works in the console, the falls into infinite recursion.

Using lapply with any funciton has the same effect.

A for loop does works as expected.

code

library(teal)

data <- within(teal_data(), {
  iris <- iris
  mtcars <- mtcars
})

transformator_iris_scale <- teal_transform_module(
  label = "Scaling transformator for iris",
  ui = function(id) {
    ns <- NS(id)
    uiOutput(ns("scaled_columns_container"))
  },
  server = function(id, data) {
    moduleServer(id, function(input, output, session) {
      ns <- session$ns
      
      scalable_columns <- reactive({
        names(Filter(is.numeric, data()[["iris"]]))
      })

      output$scaled_columns_container <- renderUI({
        selectInput(
          inputId = ns("scaled_columns"),
          label = "Columns to scale",
          choices = scalable_columns(),
          selected = input$scaled_columns,
          multiple = TRUE
        )
      })

      # BUG HERE
      reactive({
        within(
          data(),
          {
            iris[scaled_columns] <- scale(iris[scaled_columns])
            # # these also fail:
            # iris[scaled_columns] <- lapply(iris[scaled_columns], scale)
            # iris[scaled_columns] <- lapply(iris[scaled_columns], function(x) x^2)
          },
          scaled_columns = input$scaled_columns
        )
      })
      # END BUG
      # LOOP WORKS
      # reactive({
      #   within(
      #     data(),
      #     {
      #       for (i in scaled_columns) {
      #         iris[[i]] <- scale(iris[[i]])
      #       }
      #       rm(i)
      #     },
      #     scaled_columns = input$scaled_columns
      #   )
      # })
      # END LOOP
    })
  }
)

app <- init(
  data = data,
  modules = teal::example_module(transformators = list(transformator_iris_scale))
)

shinyApp(app$ui, app$server)

sessionInfo()

> packageVersion("teal")
[1] '0.15.2.9117'

Relevant log output

Warning: Error in : evaluation nested too deeply: infinite recursion / options(expressions=)?
  97: <Anonymous>
  96: stop
  95: value[[3L]]
  94: tryCatchOne
  93: tryCatchList
  92: tryCatch
  91: do
  90: hybrid_chain
  89: renderFunc
  88: output$teal-teal_modules-example_teal_module-validate_datanames-message
   3: runApp
   2: print.shiny.appobj
   1: <Anonymous>
Warning: Error in renderUI: Error occurred during data processing. See details in the main panel.
  104: stop
  103: renderUI
  102: func
   89: renderFunc
   88: output$teal-teal_modules-example_teal_module-data_summary-table
    3: runApp
    2: print.shiny.appobj
    1: <Anonymous>
Warning: Error in : evaluation nested too deeply: infinite recursion / options(expressions=)?   
  3: runApp
  2: print.shiny.appobj
  1: <Anonymous>
Warning: Error in : evaluation nested too deeply: infinite recursion / options(expressions=)?   
  137: <Anonymous>
  136: stop
  135: data
  133: <reactive>
  113: table_data
  106: renderPrint
  105: func
   89: renderFunc
   88: output$teal-teal_modules-example_teal_module-module-text
    3: runApp
    2: print.shiny.appobj
    1: <Anonymous>

Code of Conduct

• I agree to follow this project's Code of Conduct.

Contribution Guidelines

• I agree to follow this project's Contribution Guidelines.

Security Policy

• I agree to follow this project's Security Policy.