Tee a pipeline to extract the data at a given point

Add a tee to a pipeline to get the data coming in through the pipe.

This is generally a useful function for debugging pipelines, and for caching data after expensive calls. It is also useful if you want the flexibility of multiple calls with the convenience of a fully-piped approach.

The name tee comes from the tee shell command within unix systems.

Usage

tee(x, .name = "teeout", .env = NULL)

Arguments

x

The data coming in (whatever it may be).

.name

The name to assign to the output within .env.

.env

The environment within which to save the output at this point. Defaults to the caller env (i.e the env which the pipeline is in).

Value

The value that came from the left hand side of the pipe.

Note

tee() does modify the external environment (if .env is not specified). This can lead to unpredictable behaviour if not carefully managed, so it is generally worthwhile restricting usage to interactive situations where the environment can be more carefully monitored.

Author

Francis Windram

Examples

pipeout <- 1:5 |> exp() |> tee("teeout") |> log()
print(pipeout)
#> [1] 1 2 3 4 5
print(teeout)
#> [1]   2.718282   7.389056  20.085537  54.598150 148.413159

myenv <- new.env()
pipeout <- 1:5 |> exp() |> tee("teeout", .env = myenv) |> log()
print(myenv$teeout)
#> [1]   2.718282   7.389056  20.085537  54.598150 148.413159