What's a good term for an R value of length 1?

What's a good term for an R value of length 1? In R, the simplest kind of value is an "atomic" vector with a primitive type like double, but it can have any length, so the term "atomic" is not what I want. Here's an example: I often want to document my non-vectorized functions to indicate that, for a given argument, we're only interested in a single "value", i.e. something of length one. For instance, a function that sends data to a server would need an argument called url, but I don't want to support sending an arbitrary number of HTTP calls to an arbitrary number of urls. I want url to be an atomic character vector of length 1. Is there a term for this already in use? Anything better than "of length 1"? P.S. Note that R functions require these kinds of arguments all the time, like anyNA(x, recursive = FALSE). It would be weird to provide a logical vector like c(TRUE, FALSE, TRUE, FALSE) as a value for recursive. But the docs don't seem to discuss this. EDIT 2025 I am persuaded by Do not come up with a new word and if I encounter this again I will probably document such arguments as "character, length 1."

Feb 21, 2025 - 01:06

In R, the simplest kind of value is an "atomic" vector with a primitive type like double, but it can have any length, so the term "atomic" is not what I want.

Here's an example: I often want to document my non-vectorized functions to indicate that, for a given argument, we're only interested in a single "value", i.e. something of length one. For instance, a function that sends data to a server would need an argument called url, but I don't want to support sending an arbitrary number of HTTP calls to an arbitrary number of urls. I want url to be an atomic character vector of length 1.

Is there a term for this already in use? Anything better than "of length 1"?

P.S. Note that R functions require these kinds of arguments all the time, like anyNA(x, recursive = FALSE). It would be weird to provide a logical vector like c(TRUE, FALSE, TRUE, FALSE) as a value for recursive. But the docs don't seem to discuss this.

EDIT 2025

I am persuaded by Do not come up with a new word and if I encounter this again I will probably document such arguments as "character, length 1."