As we have mentioned earlier, not all functions are pipe-friendly. While %>>% accepts {} and () on the right side, Pipe object provides .() to perform exactly the same set of features with x %>>% (...) which is powerful enough to meet a wide variety of demands in piping.
If you supply an ordinary function call in .() it will evaluate it with . being the piped value.
Pipe(mtcars)$
.(lm(mpg ~ cyl + wt, data = .))
# <Pipe: lm>
#
# Call:
# lm(formula = mpg ~ cyl + wt, data = .)
#
# Coefficients:
# (Intercept) cyl wt
# 39.686 -1.508 -3.191
To avoid ambiguity, one can write a formula in .() to define the symbol representing mtcars.
Pipe(mtcars)$
.(df ~ lm(mpg ~ cyl + wt, data = df))
# <Pipe: lm>
#
# Call:
# lm(formula = mpg ~ cyl + wt, data = df)
#
# Coefficients:
# (Intercept) cyl wt
# 39.686 -1.508 -3.191
The syntax of one-sided formula is also interpreted as side effect in .().
Pipe(mtcars)$
subset(mpg >= quantile(mpg, 0.05) & mpg <= quantile(mpg, 0.95),
c(mpg, cyl, wt))$
.(~ cat("rows:",nrow(.),"\n"))$
lm(formula = mpg ~ .)$
summary()$
coef()
# rows: 28
# <Pipe: matrix>
# Estimate Std. Error t value Pr(>|t|)
# (Intercept) 36.630834 1.6127431 22.713372 3.299463e-18
# cyl -1.418216 0.3533452 -4.013684 4.783302e-04
# wt -2.528175 0.7657771 -3.301450 2.894825e-03
The question mark works too.
Pipe(mtcars)$
subset(vs == 1, c(mpg, cyl, wt))$
.(? nrow(.))$
.(? data ~ ncol(data))$
summary()
# ? nrow(.)
# [1] 14
# ? data ~ ncol(data)
# [1] 3
# <Pipe: table>
# mpg cyl wt
# Min. :17.80 Min. :4.000 Min. :1.513
# 1st Qu.:21.40 1st Qu.:4.000 1st Qu.:2.001
# Median :22.80 Median :4.000 Median :2.623
# Mean :24.56 Mean :4.571 Mean :2.611
# 3rd Qu.:29.62 3rd Qu.:5.500 3rd Qu.:3.209
# Max. :33.90 Max. :6.000 Max. :3.460
The debugging function browser() works exactly the same.
mtcars %>>%
subset(vs == 1, c(mpg, cyl, wt)) %>>%
(~ browser()) %>>%
lm(formula = mpg ~ cyl + wt) %>>%
summary
As we have previously mentioned that assignment may be the most important side effect so it deserves a syntax. One-sided formula with a symbol on the right indicates the input value should be assigned to that symbol. The syntax works in .() too.
pmodel <- Pipe(mtcars)$
subset(select = c(mpg, wt, cyl))$
.(~ sub_mtcars)$ # assign subsetted mtcars to sub_mtcars
lm(formula = mpg ~ wt + cyl)
summary(sub_mtcars)
# mpg wt cyl
# Min. :10.40 Min. :1.513 Min. :4.000
# 1st Qu.:15.43 1st Qu.:2.581 1st Qu.:4.000
# Median :19.20 Median :3.325 Median :6.000
# Mean :20.09 Mean :3.217 Mean :6.188
# 3rd Qu.:22.80 3rd Qu.:3.610 3rd Qu.:8.000
# Max. :33.90 Max. :5.424 Max. :8.000
Note that = in function argument will be interpreted as argument selection, therefore = cannot be used to assign value in .() function.
The way to perform side-effect assignment is to use <- or -> introduced by pipeR v0.5 with ~, like the following:
Pipe(mtcars)$
lm(formula = mpg ~ wt + cyl)$
.(~ summary(.) -> summ)$
coef()
# <Pipe: numeric>
# (Intercept) wt cyl
# 39.686261 -3.190972 -1.507795
summ$r.squared
# [1] 0.8302274
If ~ in is removed, the assignment will be done and the value of summary() will go on piping.
Pipe(mtcars)$
lm(formula = mpg ~ wt + cyl)$
.(summary(.) -> summ)$
coef()
# <Pipe: matrix>
# Estimate Std. Error t value Pr(>|t|)
# (Intercept) 39.686261 1.7149840 23.140893 3.043182e-20
# wt -3.190972 0.7569065 -4.215808 2.220200e-04
# cyl -1.507795 0.4146883 -3.635972 1.064282e-03
summ$r.squared
# [1] 0.8302274
Note that the output of two Pipes are different. The Pipe with ~ summary(.) results in a numeric vector of the coeffecients of the linear model because the summary is only a side effect while the Pipe without ~ results in a comprehensive matrix showing the summary of the linear coefficients because coef() works on the summary.
Element extraction also works in .() if a symbol name is supplied.
Pipe(mtcars)$
lm(formula = mpg ~ wt + cyl)$
summary()$
.(r.squared)
# <Pipe: numeric>
# [1] 0.8302274
Note that .() is only defined within a Pipe object and is not an exported function in pipeR, you cannot directly call .() outside a Pipe().
In fact, if you are familiar with functional programming, it is called a closure, which is basically defined as a function returned by a function. However, you can save .() in a Pipe object to a symbol as a shortcut to call it elsewhere.
model <- Pipe(mtcars)$
lm(formula = mpg ~ wt + cyl)$
.
model()
# <Pipe: lm>
#
# Call:
# lm(formula = mpg ~ wt + cyl, data = .)
#
# Coefficients:
# (Intercept) wt cyl
# 39.686 -3.191 -1.508
model(coefficients)
# <Pipe: numeric>
# (Intercept) wt cyl
# 39.686261 -3.190972 -1.507795
model()$summary()$coef()
# <Pipe: matrix>
# Estimate Std. Error t value Pr(>|t|)
# (Intercept) 39.686261 1.7149840 23.140893 3.043182e-20
# wt -3.190972 0.7569065 -4.215808 2.220200e-04
# cyl -1.507795 0.4146883 -3.635972 1.064282e-03