Learn R Programming

lintr (version 3.1.2)

vector_logic_linter: Enforce usage of scalar logical operators in conditional statements

Description

Usage of & in conditional statements is error-prone and inefficient. condition in if (condition) expr must always be of length 1, in which case && is to be preferred. Ditto for | vs. ||.

Usage

vector_logic_linter()

Arguments

Tags

best_practices, default, efficiency

Details

This linter covers inputs to if() and while() conditions and to testthat::expect_true() and testthat::expect_false().

Note that because & and | are generics, it is possible that && / || are not perfect substitutes because & is doing method dispatch in an incompatible way.

Moreover, be wary of code that may have side effects, most commonly assignments. Consider if ((a <- foo(x)) | (b <- bar(y))) { ... } vs. if ((a <- foo(x)) || (b <- bar(y))) { ... }. Because || exits early, if a is TRUE, the second condition will never be evaluated and b will not be assigned. Such usage is not allowed by the Tidyverse style guide, and the code can easily be refactored by pulling the assignment outside the condition, so using || is still preferable.

See Also

Examples

Run this code
# will produce lints
lint(
  text = "if (TRUE & FALSE) 1",
  linters = vector_logic_linter()
)

lint(
  text = "if (TRUE && (TRUE | FALSE)) 4",
  linters = vector_logic_linter()
)

# okay
lint(
  text = "if (TRUE && FALSE) 1",
  linters = vector_logic_linter()
)

lint(
  text = "if (TRUE && (TRUE || FALSE)) 4",
  linters = vector_logic_linter()
)

Run the code above in your browser using DataLab