Formatting DT::datatables based on logical vector

I currently want to use colorformatting for my dataFrames using shiny and the DT package. My table looks roughly like this.

| val1     | val2  | 
| ---------|------ |
| a        | b     |
| a        | a     |
| b        | b     |

I want to change the textcolor of val1 as red whenever val1[i] == val2[i] is satisfied. So the resulting table should look like this.

| val1     | val2  | 
| ---------|------ |
| a        | b     |
| a*       | a     |    * = red
| b*       | b     |

From the documentation, I see that conditional formatting can be applied with DT::formatstyle. However, the conditions specified there must be dependent on one single column of the table. Of course, I could create a new column with

dat$condition = ( val1 == val2 )

but then this column is also displayed in the widget which I don't want to happen. (Unless there is a way of removing columns of datatables objects). I am looking for a function addcolor_conditional that does something like this

condition_vector = ( val1 == val2 )
datatable( mtcars ) %>% addcolor_conditional( 
  condition_vector, color = "red" )

Any help would be appreciated

You can use the rowCallback argument and a little js code to do that. Have a look at the following minimal code:

library(shiny)
 library(DT)
 
 
 data_ <- data.frame(A = 1:5, B=c(1,3,6,4,5))
 
 ui <- fluidPage(
   
   
   dataTableOutput("table")
 )

 server <- function(input, output){
   
   output$table <- renderDataTable({
    datatable(data_, rownames = FALSE, options = list(rowCallback = JS(
                     "function(nRow, aData) {
                     if (parseFloat(aData[0]) >= parseFloat(aData[1])){
                     $('td:eq(0)', nRow).css('color', 'red');
                     }
                     }")
                   ))
     
     
   }) 
   
 }
 
 shinyApp(ui = ui, server = server)

The output that you get using the code is as follows:

Hope it helps!