Skip to contents

Check if an argument is a matrix

Usage

checkMatrix(
  x,
  mode = NULL,
  any.missing = TRUE,
  all.missing = TRUE,
  min.rows = NULL,
  max.rows = NULL,
  min.cols = NULL,
  max.cols = NULL,
  nrows = NULL,
  ncols = NULL,
  row.names = NULL,
  col.names = NULL,
  null.ok = FALSE
)

check_matrix(
  x,
  mode = NULL,
  any.missing = TRUE,
  all.missing = TRUE,
  min.rows = NULL,
  max.rows = NULL,
  min.cols = NULL,
  max.cols = NULL,
  nrows = NULL,
  ncols = NULL,
  row.names = NULL,
  col.names = NULL,
  null.ok = FALSE
)

assertMatrix(
  x,
  mode = NULL,
  any.missing = TRUE,
  all.missing = TRUE,
  min.rows = NULL,
  max.rows = NULL,
  min.cols = NULL,
  max.cols = NULL,
  nrows = NULL,
  ncols = NULL,
  row.names = NULL,
  col.names = NULL,
  null.ok = FALSE,
  .var.name = vname(x),
  add = NULL
)

assert_matrix(
  x,
  mode = NULL,
  any.missing = TRUE,
  all.missing = TRUE,
  min.rows = NULL,
  max.rows = NULL,
  min.cols = NULL,
  max.cols = NULL,
  nrows = NULL,
  ncols = NULL,
  row.names = NULL,
  col.names = NULL,
  null.ok = FALSE,
  .var.name = vname(x),
  add = NULL
)

testMatrix(
  x,
  mode = NULL,
  any.missing = TRUE,
  all.missing = TRUE,
  min.rows = NULL,
  max.rows = NULL,
  min.cols = NULL,
  max.cols = NULL,
  nrows = NULL,
  ncols = NULL,
  row.names = NULL,
  col.names = NULL,
  null.ok = FALSE
)

test_matrix(
  x,
  mode = NULL,
  any.missing = TRUE,
  all.missing = TRUE,
  min.rows = NULL,
  max.rows = NULL,
  min.cols = NULL,
  max.cols = NULL,
  nrows = NULL,
  ncols = NULL,
  row.names = NULL,
  col.names = NULL,
  null.ok = FALSE
)

expect_matrix(
  x,
  mode = NULL,
  any.missing = TRUE,
  all.missing = TRUE,
  min.rows = NULL,
  max.rows = NULL,
  min.cols = NULL,
  max.cols = NULL,
  nrows = NULL,
  ncols = NULL,
  row.names = NULL,
  col.names = NULL,
  null.ok = FALSE,
  info = NULL,
  label = vname(x)
)

Arguments

x

[any]
Object to check.

mode

[character(1)]
Storage mode of the array. Arrays can hold vectors, i.e. “logical”, “integer”, “integerish”, “double”, “numeric”, “complex”, “character” and “list”. You can also specify “atomic” here to explicitly prohibit lists. Default is NULL (no check). If all values of x are missing, this check is skipped.

any.missing

[logical(1)]
Are missing values allowed? Default is TRUE.

all.missing

[logical(1)]
Are matrices with only missing values allowed? Default is TRUE.

min.rows

[integer(1)]
Minimum number of rows.

max.rows

[integer(1)]
Maximum number of rows.

min.cols

[integer(1)]
Minimum number of columns.

max.cols

[integer(1)]
Maximum number of columns.

nrows

[integer(1)]
Exact number of rows.

ncols

[integer(1)]
Exact number of columns.

row.names

[character(1)]
Check for row names. Default is “NULL” (no check). See checkNamed for possible values. Note that you can use checkSubset to check for a specific set of names.

col.names

[character(1)]
Check for column names. Default is “NULL” (no check). See checkNamed for possible values. Note that you can use checkSubset to test for a specific set of names.

null.ok

[logical(1)]
If set to TRUE, x may also be NULL. In this case only a type check of x is performed, all additional checks are disabled.

.var.name

[character(1)]
Name of the checked object to print in assertions. Defaults to the heuristic implemented in vname.

add

[AssertCollection]
Collection to store assertion messages. See AssertCollection.

info

[character(1)]
Extra information to be included in the message for the testthat reporter. See expect_that.

label

[character(1)]
Name of the checked object to print in messages. Defaults to the heuristic implemented in vname.

Value

Depending on the function prefix: If the check is successful, the functions assertMatrix/assert_matrix return x invisibly, whereas checkMatrix/check_matrix and testMatrix/test_matrix return TRUE. If the check is not successful, assertMatrix/assert_matrix throws an error message, testMatrix/test_matrix returns FALSE, and checkMatrix/check_matrix return a string with the error message. The function expect_matrix always returns an expectation.

Examples

x = matrix(1:9, 3)
colnames(x) = letters[1:3]
testMatrix(x, nrows = 3, min.cols = 1, col.names = "named")
#> [1] TRUE