AVED: Mario strikes back (magrittr)
Štěpán Mikula
Contents
Some code examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
A concept of the pipe %>% . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Pipes and aliases of magrittr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Some code examples
Let’s see a code:
x <- round(sum(colSums(t(matrix(data = rnorm(50), ncol = 5, nrow = 10))), na.rm = TRUE), digits = 3)
. . . and x is of course equal to -3.345.
Hmm, what about writing it nicely:
x <- rnorm(50)
x <- matrix(data = x, ncol = 5, nrow = 10)
x <- t(x)
x <- colSums(x)
x <- sum(x, na.rm = TRUE)
x <- round(x, digits = 3)
We have a picture now, but it is still quite verbose.
Are you ready for a plumbing job?
rnorm(50) %>%
matrix(data = ., ncol = 5, nrow = 10) %>%
t() %>%
colSums() %>%
sum(na.rm = TRUE) %>%
round(digits = 3) -> x
A concept of the pipe %>%
(Almost) every function has inputs (arguments) and outputs (values). Pipe %>% channels output of a function
to input another function.
rnorm(50) %>%
matrix(data = ., ncol = 5, nrow = 10) %>%
t() %>%
colSums() %>%
sum(na.rm = TRUE) %>%
round(digits = 3) -> x
1
Figure 1:
2
By default an object runnig through a pipe is fed to the first argument of the right-hand side function. If
you want to fed another argument (or you want to be explicit) you need to use “placehoder” . to determine
the destination of the pipe. You can use . more then once:
multiply <- function(x = 1, y = -1) x*y
multiply()
## [1] -1
5 %>% multiply()
## [1] -5
5 %>% multiply(y = .)
## [1] 5
5 %>% multiply(x = ., y = .)
## [1] 25
Pipe is not implemented in base R. It is part of packages dplyr and magrittr. dplyr contains only basic
pipe (%>%). magrittr allows user to use more advanced options.
Where have I seen it before? Of course you know the pipe from (a) Sherlock Holmes, and
(b) your Unix/Linux shell (|). (Sherlock Holmes is a fictional detective with a funny hat.)
Pipes and aliases of magrittr
library(magrittr)
magrittr provides extra three pipes (their descriptions are from vignette). We will illustrate their function
using famous iris dataset:
# Print few lines
iris %>% head
## Sepal.Length Sepal.Width Petal.Length Petal.Width Species
## 1 5.1 3.5 1.4 0.2 setosa
## 2 4.9 3.0 1.4 0.2 setosa
## 3 4.7 3.2 1.3 0.2 setosa
## 4 4.6 3.1 1.5 0.2 setosa
## 5 5.0 3.6 1.4 0.2 setosa
## 6 5.4 3.9 1.7 0.4 setosa
The “tee” operator, %T>% works like %>%, except it returns the left-hand side value, and not the result of
the right-hand side operation. This is useful when a step in a pipeline is used for its side-effect (printing,
plotting, logging, etc.).
3
iris[,1:2] %>% plot -> v1
print(v1)
## NULL
iris[,1:2] %T>% plot -> v2
4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
2.02.53.03.54.0
Sepal.Length
Sepal.Width
print(head(v2))
## Sepal.Length Sepal.Width
## 1 5.1 3.5
## 2 4.9 3.0
## 3 4.7 3.2
## 4 4.6 3.1
## 5 5.0 3.6
## 6 5.4 3.9
The “exposition” pipe operator, %$% exposes the names within the left-hand side object to the right-hand
side expression.
Let’s sum up Sepal.Length column:
iris %>% sum(Sepal.Length)
In this case it ends up with an error. (You can try it.)
iris %$% sum(Sepal.Length)
## [1] 876.5
4
It worked this time.
Compound assignment pipe operator %<>% (aka Moebius pipe) can be used as the first pipe in a chain.
The effect will be that the result of the pipeline is assigned to the left-hand side object, rather than returning
the result as usual.
x %>% sqrt -> x
is equivalent to
x %<>% sqrt
## Warning in sqrt(.): NaNs produced
Notice that it is not necessary to write functions with (). x %<>% sqrt() is the same as x %<>%
sqrt.
Aliases
When chaining functions you might need to use some pretty basic operations. It is impossible with bare +, -,
or /. One need a series of functions similar to user-defined function multiply() defined above. magrittr
provides a lot of such functions:
Alias (magrittr) Base R
extract [
extract2 [[
inset [<inset2
[[<use_series
$
add +
subtract multiply_by
*
raise_to_power ˆ
multiply_by_matrix %*%
divide_by /
divide_by_int %/%
mod %%
is_in %in%
and &
or |
equals ==
is_greater_than >
is_weakly_greater_than >=
is_less_than <
is_weakly_less_than <=
not !
set_colnames colnames<set_rownames
rownames<set_names
names<See
a complex example:
5
iris[1:10,] %T>% print %$%
raise_to_power(Sepal.Width,2) %T>% print %>%
multiply_by(-1) %T>% print %>%
is_weakly_greater_than(0) %T>% print %>%
sum -> x
## Sepal.Length Sepal.Width Petal.Length Petal.Width Species
## 1 5.1 3.5 1.4 0.2 setosa
## 2 4.9 3.0 1.4 0.2 setosa
## 3 4.7 3.2 1.3 0.2 setosa
## 4 4.6 3.1 1.5 0.2 setosa
## 5 5.0 3.6 1.4 0.2 setosa
## 6 5.4 3.9 1.7 0.4 setosa
## 7 4.6 3.4 1.4 0.3 setosa
## 8 5.0 3.4 1.5 0.2 setosa
## 9 4.4 2.9 1.4 0.2 setosa
## 10 4.9 3.1 1.5 0.1 setosa
## [1] 12.25 9.00 10.24 9.61 12.96 15.21 11.56 11.56 8.41 9.61
## [1] -12.25 -9.00 -10.24 -9.61 -12.96 -15.21 -11.56 -11.56 -8.41 -9.61
## [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
Tee pipe returns a mirror image of the content flowing in pipes. It is printed on the screen using print. The
content itself remains intact.
In the first phase the whole data.frame is flowing in the tube but %$% allows raise_to_power to use specific
(named) part of the iris table.
And naturally x is equal to 0.
6
Figure 2:
7