A decorator is a function that receives a function, extends its behaviour, and returned the altered function. Any caller that uses the decorated function uses the same interface as it were the original, undecorated function.
# Running the original Sys.sleep() produces no output Sys.sleep(0.1) # Decorating the Sys.sleep() with time_it() measures Sys.sleep() execution time # and outputs the measurement <- decorators::time_it(base::Sys.sleep) Sys.sleep Sys.sleep(0.1) #> Time difference of 0.11 secs
## Define a functions that averages two numbers <- function(a = NA_real_, b = NA_real_) mean(c(a, b)) average_two_numbers ## Before: Averaging two nun-numeric values returns NA and prompts a warning average_two_numbers(a = "1", b = "2") #>  NA
## Decorating average_two_numbers() with validate_arguments() checks input types <- decorators::validate_arguments(average_two_numbers) average_two_numbers ## After: Averaging two nun-numeric prompts an informative error average_two_numbers(a = "1", b = "2")
#> Error in average_two_numbers(a = "1", b = "2") : #> a is of type `character` rather than `numeric`!
You can install the released version of
And the development version from GitHub with:
# install.packages("devtools") ::install_github("tidylab/decorators")devtools