dplyr：if_else很贪心

Question

I am trying to mutate a new variable conditionally on the value of another variable using if_else , but if_else insists on computing all the values, and then conditionally replacing the values in the final vector.

This behaviour is problematic when the function that results in the final values for some cases cannot be computed for other cases:

df_foo = data_frame(
  rate = sample(c(0, 0.1), size = 100, replace = TRUE),
  pmt = 1000,
  nper = 10
)

df_foo %>% 
  mutate(
    if_else(
      rate > 0,
      optiRum::PV(rate = rate, pmt = -pmt, nper = nper),
      0
    )
  )

Is there any way to only compute the TRUE values when the condition is TRUE ?

Answer 1

This is not a problem specific to if_else , the base version ifelse acts the same. From the help file of ifelse we can read:

If yes or no are too short, their elements are recycled. yes will be evaluated if and only if any element of test is true, and analogously for no .

There is a number of cases where ifelse use is inappropriate, and the raising of warnings of errors in one of the conditions is one.

A solution to do what you want (without dplyr ) is:

df_foo$x <- 0
df_foo$x[df_foo$rate > 0] <- with(df_foo[df_foo$rate > 0, ], optiRum::PV(rate, nper, -pmt))