pro: Standard Evaluated Versions of Some Common...

In tinycodet: Functions to Help in your Coding Etiquette

Standard Evaluated Versions of Some Common Expression-Evaluation Functions

Description

The with_pro() and aes_pro() functions are standard-evaluated versions of the expression-evaluation functions with and ggplot2::aes, respectively.

These alternative functions are more programmatically friendly:
They use proper standard evaluation, through the usage of one-sided formulas, instead of non-standard evaluation, tidy evaluation, or similar programmatically unfriendly evaluations.

Usage

with_pro(data, form)

aes_pro(...)

Arguments

data	a list, environment, or data.frame.
form	a one-sided formula giving the expression to evaluate in with_pro. If the formula has an environment, that environment is used to find any variables or objects not present in data.
...	arguments to be passed to ggplot2::aes, but given as one-sided formulas.

Details

The aes_pro() function is the standard evaluated alternative to ggplot2::aes.
Due to the way aes_pro() is programmed, it should work even if the tidy evaluation technique changes in 'ggplot2'.
To support functions in combinations with references of the variables, the input used here are formula inputs, rather than string inputs.
See the Examples section below.

Value

For with_pro(): see with.
For aes_pro(): see ggplot2::aes.

Non-Standard Evaluation

Non-Standard Evaluation (sometimes abbreviated as "NSE"), is somewhat controversial.
Consider the following example:

aplot <- "ggplot2"
library(aplot)

What package will be attached? It will not be 'ggplot2', nor will an error occur. Instead, the package 'aplot' will be attached.
This is due to evaluating the expression 'aplot' as a quoted expression, instead of evaluating the contents (i.e. string or formula) of the variable. In other words: Non-Standard Evaluation.

Regular Standard Evaluation does not have the above problem.

Note

The with_pro() function, like the original with function, is made for primarily for convenience.
When using modelling or graphics functions with an explicit data argument (and typically using formulas), it is typically preferred to use the data argument of that function, rather than to use either
with(data, ...) or with_pro(data, ...).

See Also

tinycodet_safer

Examples

requireNamespace("ggplot2")


d <- import_data("ggplot2", "mpg")

# mutate data:
myform <- ~ displ + cyl + cty + hwy
d$mysum <- with_pro(d, myform)
summary(d)

# plotting data:
x <- ~ cty
y <- ~ sqrt(hwy)
color <- ~ drv

ggplot2::ggplot(d, aes_pro(x, y, color = color)) +
  ggplot2::geom_point()

tinycodet documentation built on Sept. 12, 2024, 7:03 a.m.