Multiple, Unpacking, and Destructuring Assignment
Provides a %<-% operator to perform multiple,
unpacking, and destructuring assignment in R. The
operator unpacks the right-hand side of an assignment
into multiple values and assigns these values to
variables on the left-hand side of the assignment.
Variable assignment with zeal!
What's there to be excited about?
zeallot allows multiple, unpacking, or destructuring assignment in R by
%<-% operator. With zeallot you can tighten code with explicit
variable names, unpack pieces of a lengthy list or the entirety of a small list,
destructure and assign object elements, or do it all at once.
Unpack a vector of values.
c(x, y) %<-% c(0, 1) #> x # 0 #> y # 1
Unpack a list of values.
c(r, d) %<-% list(2, 2) #> r # 2 #> d # 2
Destructure a data frame and assign its columns.
c(duration, wait) %<-% head(faithful) #> duration # 3.600 1.800 3.333 2.283 4.533 2.883 #> wait # 79 54 74 62 85 55
Unpack a nested list into nested left-hand side variables.
c(c(a, b), c(c, d)) %<-% list(list(1, 2), list(3, 4)) #> a # 1 #> b # 2 #> c # 3 #> d # 4
Destructure and partially unpack a list. "a" is assigned to
"b", "c", "d", and "e" are grouped and assigned to one variable.
c(first, ...rest) %<-% list("a", "b", "c", "d", "e") first # "a" rest #[] # "b" # #[] # "c" # #[] # "d" # #[] # "e"
You can install zeallot from CRAN.
Use devtools to install the latest, development version of zeallot from GitHub.
Below is a simple example using the purrr package and the safely function.
purrr::safely function returns a "safe" version of a function. The
following example is borrowed from the safely documentation. In this example a
safe version of the log function is created.
safe_log <- purrr::safely(log) safe_log(10) #$result # 2.302585 # #$error #NULL safe_log("a") #$result #NULL # #$error #<simpleError in .f(...): non-numeric argument to mathematical function>
A safe function always returns a list of two elements and will not throw an error. Instead of throwing an error, the error element of the return list is set and the value element is NULL. When called successfully the result element is set and the error element is NULL.
Safe functions are a great way to write self-documenting code. However, dealing with a return value that is always a list could prove tedious and may undo efforts to write concise, readable code. Enter zeallot.
With zeallot's unpacking operator
%<-% we can unpack a safe function's return
value into two explicit variables and avoid dealing with the list return value
c(res, err) %<-% safe_log(10) res # 2.302585 err #NULL
The name structure of the operator is a flat or nested set of bare variable
names built with calls to
c(). . These variables do not need to be previously
defined. On the right-hand side is a vector, list, or other R object to unpack.
%<-% unpacks the right-hand side, checks the number of variable names against
the number of unpacked values, and then assigns each unpacked value to a
variable. The result, instead of dealing with a list of values there are two
Further Reading and Examples
For more on the above example, other examples, and a thorough introduction to zeallot check out the vignette on unpacking assignment.
Below are links to discussions about multiple, unpacking, and destructuring assignment in R,
The vadr package includes a
with much of the same functionality as
%<-%. As the author states, "[they]
strongly prefer there to be a
<- anywhere that there is a modification to the
environment." If you feel similarly I suggest looking at vadr. Unfortunately the
vadr package is not on CRAN and will need to be installed using
Thank you to Paul Teetor for inspiring me to build zeallot.
Without his encouragement nothing would have gotten off the ground.
Functions in zeallot
|operator||Multiple assignment operators|
|zeallot||Multiple, unpacking, and destructuring assignment in R|
|destructure-methods||Included Implementations of destructure|
|destructure||Destructure an object|
Vignettes of zeallot
|License||MIT + file LICENSE|
|Roxygen||list(markdown = TRUE)|
Include our badge in your README