injector: Create, Set, and Get Injectors (Modulr Internal States).

In openscienceunil/modulr: A Dependency Injection Framework for R

Description

Create an new injector, set and get the current injector, and get the default injector.

Usage

new_injector()

set_injector(injector = new_injector())

set_default_injector()

get_injector()

get_default_injector()

Arguments

injector

An injector returned by new_injector.

Details

An injector essentially carries an internal modulr state. Technically, it is is an R environment containing every piece of information needed by modulr to reflect the module definitions, the dependencies between them, the configurations, and all the associated metadata. As a PORO (Plain Old R Object), an injector can be stored to disk with the session data, or shared between Alice and Bob, for instance.

When the modulr package is loaded, a default injector is created. This injector is returned by the get_default_injector function.

Value

Every function returns an injector (R environment).

Warning

Setting an injector from within a module is not allowed and results in an error.

See Also

.SharedEnv, define, list_modules, make, and reset.

Examples

reset()
define("foo", NULL, function() NULL)
injector <- new_injector()
previous_injector <- set_injector(injector)
define("bar", NULL, function() NULL)
lsmod()
set_injector(previous_injector)
lsmod()

## Not run: 
.Last <- function() {
 # Bind the current injector (internal modulr state) to the environment.
 injector <- get_injector()
}
quit(save = "yes")
## End(Not run)

reset()
define("foo", NULL, function() "Hi Bob!")
## Alice saves its injector and sends it to ...
saveRDS(get_injector(), file = "injector.R")
## ... Bob who restores it.
injector <- readRDS("injector.R")
set_injector(injector)
make("foo")

openscienceunil/modulr documentation built on May 3, 2019, 5:49 p.m.