Reputation: 5403
In R
, I'm defining the function lengths
depending on the value of a parameter previously set:
if(condition == 1){
lengths <- function(vector) {
n <- ceiling(length(vector)/2)
}
}
else if(condition == 2){
lengths <- function(vector) {
n <- length(vector)
}
}
else if(condition == 3){
lengths <- function(vector) {
n <- length(vector)*2
}
}
else{
lengths <- function(vector) {
n <- length(vector)+10
}
}
Defining a function conditionally in this way seems just a bit... messy. Is there a better way?
Upvotes: 7
Views: 2029
Reputation: 21502
A quick hack based on CSGillespie's suggestion:
length.lut <- cbind(c(0.5,1,2,1,),c(0,0,0,10))
lengths <- function(vector, condition) vector*length.lut[condition,1] + length.lut[condition,2]
Upvotes: 0
Reputation: 176648
You could use switch
:
lengths <- function(vector, condition) {
switch(condition,
ceiling(length(vector)/2),
length(vector),
length(vector)*2,
length(vector)*+10)
}
This function provides behavior more like your example code:
lengths <- function(vector, condition) {
switch(as.character(condition),
`1`=ceiling(length(vector)/2),
`2`=length(vector),
`3`=length(vector)*2,
length(vector)*+10)
}
...and this function will be defined by the value of condition
:
condition <- 1
lengths <- switch(as.character(condition),
`1`=function(vector) ceiling(length(vector)/2),
`2`=function(vector) length(vector),
`3`=function(vector) length(vector)*2,
function(vector) length(vector)*+10)
lengths
# function(vector) ceiling(length(vector)/2)
Upvotes: 9