rlang (version 1.1.4)

defusing-advanced: Advanced defusal operators


These advanced operators defuse R expressions. expr(), enquo(), and enquos() are sufficient for most purposes but rlang provides these other operations, either for completeness or because they are useful to experts.

  • exprs() is the plural variant of expr(). It returns a list of expressions. It is like base::alist() but with injection support.

  • quo() and quos() are like expr() and exprs() but return quosures instead of naked expressions. When you are defusing your own local expressions (by opposition to function arguments where non-local expressions are supplied by your users), there is generally no need to attach the current environment in a quosure. See What are quosures and when are they needed?.

  • enexpr() and enexprs() are like enquo() and enquos() but return naked expressions instead of quosures. These operators should very rarely be used because they lose track of the environment of defused arguments.

  • ensym() and ensyms() are like enexpr() and enexprs() but they throw an error when the defused expressions are not simple symbols. They also support strings which are interpreted as symbols. These functions are modelled on the behaviour of the left-hand side of = and <- where you can supply symbols and strings interchangeably.

    "foo" <- NULL
    list("foo" = NULL)

  • enquo0 and enquos0() are like enquo() and enquos() but without injection support. The injection operators !!, !!!, and {{ are not processed, instead they are preserved in the defused expression. This makes it possible to defuse expressions that potentially contain injection operators meant for later use. The trade off is that it makes it harder for users to inject expressions in your function. They have to enable injection explicitly with inject().

    None of the features of dynamic dots are available when defusing with enquos0(). For instance, trailing empty arguments are not automatically trimmed.



exprs( ..., .named = FALSE, .ignore_empty = c("trailing", "none", "all"), .unquote_names = TRUE )

enexprs( ..., .named = FALSE, .ignore_empty = c("trailing", "none", "all"), .ignore_null = c("none", "all"), .unquote_names = TRUE, .homonyms = c("keep", "first", "last", "error"), .check_assign = FALSE )


ensyms( ..., .named = FALSE, .ignore_empty = c("trailing", "none", "all"), .ignore_null = c("none", "all"), .unquote_names = TRUE, .homonyms = c("keep", "first", "last", "error"), .check_assign = FALSE )


quos( ..., .named = FALSE, .ignore_empty = c("trailing", "none", "all"), .unquote_names = TRUE )





An unquoted argument name. The expression supplied to that argument is defused and returned.


For enexprs(), ensyms() and enquos(), names of arguments to defuse. For exprs() and quos(), expressions to defuse.


If TRUE, unnamed inputs are automatically named with as_label(). This is equivalent to applying exprs_auto_name() on the result. If FALSE, unnamed elements are left as is and, if fully unnamed, the list is given minimal names (a vector of ""). If NULL, fully unnamed results are left with NULL names.


Whether to ignore empty arguments. Can be one of "trailing", "none", "all". If "trailing", only the last argument is ignored if it is empty. Named arguments are not considered empty.


Whether to treat := as =. Unlike =, the := syntax supports names injection.


Whether to ignore unnamed null arguments. Can be "none" or "all".


How to treat arguments with the same name. The default, "keep", preserves these arguments. Set .homonyms to "first" to only keep the first occurrences, to "last" to keep the last occurrences, and to "error" to raise an informative error and indicate what arguments have duplicated names.


Whether to check for <- calls. When TRUE a warning recommends users to use = if they meant to match a function parameter or wrap the <- call in curly braces otherwise. This ensures assignments are explicit.


An expression to defuse.


Run this code
# `exprs()` is the plural variant of `expr()`
exprs(foo, bar, bar)

# `quo()` and `quos()` are the quosure variants of `expr()` and `exprs()`
quos(foo, bar)

# `enexpr()` and `enexprs()` are the naked variants of `enquo()` and `enquos()`
my_function1 <- function(arg) enexpr(arg)
my_function2 <- function(arg, ...) enexprs(arg, ...)
my_function1(1 + 1)
my_function2(1 + 1, 10 * 2)

# `ensym()` and `ensyms()` are symbol variants of `enexpr()` and `enexprs()`
my_function3 <- function(arg) ensym(arg)
my_function4 <- function(arg, ...) ensyms(arg, ...)

# The user must supply symbols
my_function4(foo, bar)

# Complex expressions are an error
try(my_function3(1 + 1))
try(my_function4(1 + 1, 10 * 2))

# `enquo0()` and `enquos0()` disable injection operators
automatic_injection <- function(x) enquo(x)
no_injection <- function(x) enquo0(x)


# Injection can still be done explicitly

Run the code above in your browser using DataLab