Order of the newly assigned classes

[ddsjoberg]

Quoting from the Advanced R book:

Objects are not limited to a single class, and can have many classes:
class(x) <- c("A", "B")
class(x) <- LETTERS
As discussed in the next section, R looks for methods in the order in which they appear in the class vector. So in this example, it would be like class A inherits from class B - if a method isn’t defined for A, it will fall back to B. However, if you switched the order of the classes, the opposite would be true!
http://adv-r.had.co.nz/S3.html

In visR, I noticed that some classes are defined in the opposite order: the most generic class is defined first and the specific visR classes are last. In the example below, we see that the ggplot classes are listed first, meaning that if ggplot2 were to ever implement a visr() method for class "ggplot" it would take precedent over the visR-defined method functions.

Generally, you want the most broad class listed last, and the most narrow and specific class should be listed first to ensure the S3 dispatch system selects the most appropriate method. So the classes below should be defined as c("ggsurvfit", "gg", "ggplot")

library(visR)

adtte %>%
  estimate_KM(strata = "TRTP", conf.int = 0.90) %>%
  visr(legend_position = "right", x_unit = "Days") %>%
  class()
#> [1] "gg"        "ggplot"    "ggsurvfit"

adtte %>% 
  ggplot2::ggplot() %>% 
  class()
#> [1] "gg"     "ggplot"

^{Created on 2022-03-24 by the reprex package (v2.0.1)}

As an example, the Hmisc package adds the class "labelled" to labelled objects. The new labelled class is added to the beginning.

Hmisc::label(mtcars$mpg) <- "Miles per Gallon"
class(mtcars$mpg)
#> [1] "labelled" "numeric"

^{Created on 2022-03-26 by the reprex package (v2.0.1)}