Building functions with `compose()` and `negate()`

Intermediate Functional Programming with purrr

Colin Fay

Data Scientist & R Hacker at ThinkR

compose() at will

Limitless compositions:

library(broom)
library(purrr)
clean_aov <- compose(tidy, anova, lm)

clean_aov(Sepal.Length ~ Sepal.Width, data = iris)
# A tibble: 2 x 6
  term           df  sumsq meansq statistic p.value
  <chr>       <int>  <dbl>  <dbl>     <dbl>   <dbl>
1 Sepal.Width     1   1.41  1.41       2.07   0.152
2 Residuals     148 101.    0.681     NA     NA

negate()

Flip the logical:

is_not_na <- negate(is.na)
x <- c(1,2,3,4, NA)
is.na(x)

FALSE FALSE FALSE FALSE  TRUE

is_not_na(x)

TRUE  TRUE  TRUE  TRUE FALSE

With mappers

negates() & mappers:

under_hundred <- as_mapper(~ mean(.x) < 100)
not_under_hundred <- negate(under_hundred)
map_lgl(98:102, under_hundred)

TRUE  TRUE FALSE FALSE FALSE

map_lgl(98:102, not_under_hundred)

FALSE FALSE  TRUE  TRUE  TRUE

Digression on http codes

From Hypertext Transfer Protocol (HTTP) Status Code Registry , fixerrorcode.com

Test if in

Is x %in% y?

status <- 201

good_status <- c(200, 201, 202, 203)

status %in% good_status

TRUE

Let's practice!

Intermediate Functional Programming with purrr