Question

I am puzzled by this behavior in R. I just want to do a simple string compare of a list of strings produced by strsplit. So do not understand why the following first two code pieces do what I expected, while the third is not.

> for (i in strsplit("A text I want to display with spaces", " ")) { print(i) }
[1] "A"       "text"    "I"       "want"    "to"      "display" "with"    "spaces" 

Ok, this makes sense …

> for (i in strsplit("A text I want to display with spaces", " ")) { print(i=="want") }
[1] FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE

Ok, this too. But, what is wrong with the following construction?

> for (i in strsplit("A text I want to display with spaces", " ")) { if (i=="want")     print("yes") }
Warning message:
In if (i == "want") print("yes") :
  the condition has length > 1 and only the first element will be used

Why doesn’t this just print “yes” when the fourth word is encountered? What should I change to have this desired behavior?

Answer 1

The problem is that strsplit produces a list of split strings (in this case with length 1, because you only gave it a single string to split).

ss <- strsplit("A text I want to display with spaces", " ")
for (i in ss[[1]]) {
  if (i=="want")     print("yes")
}

You can see what’s going on if you just print the elements:

for (i in ss) {
  print(i)
}

the first element is a character vector.

Depending on what you’re doing you might also consider vectorized comparisons such as ifelse(ss=="want","yes","no")