CODEWITHSUNDEEP
rshinydt
colef55
colef55

Reputation: 45

Integrating DTedit with "new row" ActionButton

I am trying to append rows to a dtedit object using a seperate actionbutton. The button creates a row and I want it appended to the dtedit table. I have it working so that this all happens, but the dtedit function doesn't seem to recognize the data, because when I try using the "Edit" button that's built in, the fields don't pre-populate in the modal as they should.

I have included a reproducible example below. You will see that the "Add Row" functionality works, but when you click on a row and press edit, the modal does not recognize the data.

*Edited with DTedit package load and the removal of view.label.cols

devtools::install_github('jbryer/DTedit')
library(DTedit)

ui <- fluidPage(
      actionButton('addrow', "New Row"),
      uiOutput("table")
)

server <- function(input, output) {

  df <- data.frame('Order' = integer(), 'Category' = character(), 'Name' = character(), stringsAsFactors = F)
  table.reactive <- reactiveVal(df)
  dtedit(input, output,
         name = 'table',
         thedata = df,
         edit.cols = c('Order', 'Category', 'Name'),
         edit.label.cols = c('Order', 'Category', 'Name'),
         input.types = c( Order = 'numericInput',
                          Category='textInput',
                          Name='textInput'
         ),
         view.cols = c('Order', 'Category', 'Name'),
         callback.update = plandrill.update.callback,
         callback.insert = plandrill.insert.callback,
         callback.delete = plandrill.delete.callback,
         show.copy = F, show.insert = F)

  dt.proxy <- DT::dataTableProxy("tabledt")

  ## append row
  observeEvent(input$addrow, {
    row <- data.frame(nrow(table.reactive()) + 1, "ABC", "DEF", stringsAsFactors = F)
    appended_table = rbind(table.reactive(), row)
    table.reactive(appended_table)
    replaceData(dt.proxy,
               appended_table,
               rownames = FALSE)
  })

}

shinyApp(ui = ui, server = server)

Upvotes: 0

Views: 288

Answers (1)

David Fong
David Fong

Reputation: 546

jbryer/DTedit makes a local copy of the dataframe passed through the thedata argument, and stores it internally in results$thedata. results$thedata is used for subsequent datatable editing and is not changed by replaceData.

The effect you require can be achieved through a modified version of DTedit, which I wrote (with the help of others) and is available on Github at DavidPatShuiFong/DTedit.

library(DTedit)

ui <- fluidPage(
  actionButton('addrow', "New Row"),
  uiOutput("table")
)

server <- function(input, output) {
  
  df <- data.frame('Order' = integer(),
                   'Category' = character(),
                   'Name' = character(),
                   stringsAsFactors = F)
  table.reactive <- reactiveVal(df)
  dtedit(input, output,
         name = 'table',
         thedata = table.reactive,
         edit.cols = c('Order', 'Category', 'Name'),
         edit.label.cols = c('Order', 'Category', 'Name'),
         input.types = c( Order = 'numericInput',
                          Category='textInput',
                          Name='textInput'
         ),
         view.cols = c('Order', 'Category', 'Name'),
         # callback.update = plandrill.update.callback,
         # callback.insert = plandrill.insert.callback,
         # callback.delete = plandrill.delete.callback,
         show.copy = F, show.insert = F)
  
  dt.proxy <- DT::dataTableProxy("tabledt")
  
  ## append row
  observeEvent(input$addrow, {
    row <- data.frame(Order = nrow(table.reactive()) + 1,
                      Category = "ABC",
                      Name = "DEF",
                      stringsAsFactors = F)
    appended_table = rbind(table.reactive(), row)
    table.reactive(appended_table)
  })
}

shinyApp(ui = ui, server = server)

In this version of DTedit, the dataframe is passed as a reactiveVal (table.reactive), and DTedit's copy changes when table.reactive changes.

I needed to disable defining the various callbacks in your example, as the callbacks were not defined. I also need to be explicit about labeling the columns in the row dataframe.

Further examples can be seen in a vignette, and linked through README.md on Github.

Upvotes: 1

Related Questions