Why is reactive dataset, in RShiny, with NA value considered as "empty" dataset in validation?

Question

Can somebody help me please? In my application, the user can upload a file from the device. The application has a number of features that do not matter now. The point is that when performing validation is reactive dataset, in RShiny, with NA value considered as "empty" dataset and "warning message" does not disappear despite the uploaded dataset. Is there some way how to resolve it ? (In my example: After upload dataset with NA value there is still message Please, upload dataset at first! )

SERVER:

##########################################
#dataset upload

df1 <- reactiveVal()
  observeEvent(input$file1,{
    df <- read.csv(input$file1$datapath,
                   header = input$header,
                   sep = input$sep,
                   quote = input$quote)
    df1(df)
  })
##########################################


##########################################
#example of function where I use validation  


#Delete column
  output$to_delete<- renderUI({
     validate(
       need(df1() != "", "Please, upload dataset at first!")
     )
    
    choice <- names(df1())
    selectInput('to_delete', label = 'Choose column to delete: ', choices = choice)
  })
  
  observeEvent(input$del_col, {
    df1(df1()[ , !names(df1()) %in% c(input$to_delete)])
  })
  ##########################################