seek any true in a column from a list of data frames in R

user_7106865 0
Viewed 61

I have a list of dataframes containing only boolean (TRUE / FALSE) columns.

I need to know if any of these columns contains any TRUE value (all are expected to be false and in fact, all are FALSE), but I'm getting just nonsense.

mylist <-
  list(df1=data.frame(
    "column11" = c(FALSE, FALSE, FALSE, FALSE),
    "column12" = c(FALSE, FALSE, FALSE, FALSE)
  ),df2=data.frame(
    "column21" = c(FALSE, FALSE, FALSE, FALSE),
    "column22" = c(FALSE, FALSE, FALSE, FALSE)
  ),df3=data.frame(
    "column31" = c(FALSE, FALSE, FALSE, FALSE),
    "column32" = c(FALSE, FALSE, FALSE, TRUE)
      ))

lapply(mylist, function(x) any(TRUE))


$df1
[1] TRUE

$df2
[1] TRUE

$df3
[1] TRUE
r boolean list
4 Answers

One reason the anonymous function doesn't work as expected is because its argument is not being used, instead any() is checking if the value TRUE is TRUE. It also needs to be passed a logical vector, not a data.frame.

If you want to check if any column in the data.frame has a TRUE value use:

sapply(mylist, function(x) any(unlist(x)))

  df1   df2   df3 
FALSE FALSE  TRUE

If you want to check if any particular column has a TRUE value, use:

sapply(mylist, sapply, any)

           df1   df2   df3
column11 FALSE FALSE FALSE
column12 FALSE FALSE  TRUE
user_28352610

An elusive rapply option:

rapply(mylist, f = any, how = 'unlist')

df1.column11 df1.column12 df2.column21 df2.column22 df3.column31 df3.column32 
       FALSE        FALSE        FALSE        FALSE        FALSE         TRUE 

do.call(rbind, rapply(mylist, f = any, how = 'list'))

    column11 column12
df1 FALSE    FALSE   
df2 FALSE    FALSE   
df3 FALSE    TRUE

And with coercion:

sapply(rapply(mylist, f = any, how = 'list'), any)

  df1   df2   df3 
FALSE FALSE  TRUE
user_113550660

We use lapply() for each dataframe in mylistthen we use apply for each column of dataframes.

abc = function(x) {
  tf = apply(x, 2, FUN = any)
}

lapply(mylist, FUN =  abc )

Output:

$df1
column11 column12 
   FALSE    FALSE 

$df2
column21 column22 
   FALSE    FALSE 

$df3
column31 column32 
   FALSE     TRUE
user_31725560

According to ?any

Given a set of logical vectors, is at least one of the values true?

A matrix is a vector with dim attributes, but a data.frame is not. One option is to convert to matrix and use any

lapply(mylist, function(x) any(as.matrix(x)))
user_37322710
Related