Difference between inherits(x,"data.frame") and "class" %in% class("data.fr-CodePudding

Are these any different from each other? Assume x is defined:

inherits(x,"data.frame")

"data.frame" %in% class(x)

Also, it need not be "data.frame". It can be something else like "workflow". I just want to know as these seem to do the same thing (i.e. check if x is a data frame), but I feel like they might be different.

CodePudding user response：

The main difference is in vectorized class checking, for example if you want to check whether an object belongs to one of a number of different classes. Then inherits will return a length-one logical vector, whereas %in% returns multiple results (one for each test class)

df  <- data.frame(a = 1:3)

inherits(df, c("data.frame", "foo"))
#> [1] TRUE

c("data.frame", "foo") %in% class(df)
#> [1]  TRUE FALSE

This could be relevant when using a logical test of class type. For example, suppose I want to check that df is either a data frame or of class foo. I can safely put inherits inside an if statement:

if(inherits(df, c("data.frame", "foo"))) {
  cat("Object df is either a data frame or a foo")
}
#> Object df is either a data frame or a foo

But I will get an error if I do the same thing with %in%

if(c("data.frame", "foo") %in% class(df)) {
  cat("Object df is either a data frame or a foo")
}
#> Error in if (c("data.frame", "foo") %in% class(df)) {
#> : the condition has length > 1

If you only want to check for a single class, the two are essentially equivalent, or in the case of tests for multiple classes, wrapping the %in% with any will achieve the same effect.

^{Created on 2022-06-15 by the reprex package (v2.0.1)}