R：在 ifelse 语句中将特定行馈送到 function

Question

I am using an ifelse statement to create a new column in my dataframe.

Inside this ifelse statement I need to call a function on results which pass the initial test.

For example:

test_function=function(value){
   print(value)
   return(value)}

testdf<-data.frame(a=c(1,1,2,2,3,3),b=c(1,2,3,4,5,6))

testdf$c<-ifelse(testdf$a==2,test_function(testdf$b),NA)

This example does give the answer I would expect (testdf$c= NA,NA,3,4,NA,NA), however the print statement shows that along the way all of column a is being passed to test_function().

In my actual data this causes a fail (apologies, I'm really struggling to provide a reproducible example).

Is there any way to alter this so that only those rows that pass the initial test are passed to the function rather than all rows with the decision seemingly made after.

I'm aware I could achieve the same with an apply function, or a loop but my actual dataframe is 1.5m rows long and I have found these to be prohibitively time consuming.

Thanks in advance for your help!

Answer 1

If you want the results of a function in ifelse , save the output and pass it to the function. Otherwise use and if{}else{} construct inside the function.

x <- test_function(testdf$b)
[1] 1 2 3 4 5 6
testdf$c <- ifelse(testdf$a==2, x, NA)
testdf$c
[1] NA NA  3  4 NA NA

Answer 2

You could get the index where a = 2 and pass it to test_function .

inds <- testdf$a == 2
testdf$c <- NA
testdf$c[inds] <- test_function(testdf$b[inds])

This prints in the console:

#[1] 3 4

and testdf is:

testdf
#  a b  c
#1 1 1 NA
#2 1 2 NA
#3 2 3  3
#4 2 4  4
#5 3 5 NA
#6 3 6 NA