Package 'tblcheck'

Generate a human-readable description of an object's class

Description

Generate a human-readable description of an object's class

Usage

friendly_class(object)

## S4 method for signature 'ANY'
friendly_class(object)

## S4 method for signature 'character'
friendly_class(object)

## S4 method for signature 'numeric'
friendly_class(object)

## S4 method for signature 'integer'
friendly_class(object)

## S4 method for signature 'logical'
friendly_class(object)

## S4 method for signature 'complex'
friendly_class(object)

## S4 method for signature 'raw'
friendly_class(object)

## S4 method for signature 'factor'
friendly_class(object)

## S4 method for signature 'Date'
friendly_class(object)

## S4 method for signature 'POSIXt'
friendly_class(object)

## S4 method for signature 'Period'
friendly_class(object)

## S4 method for signature 'data.frame'
friendly_class(object)

## S4 method for signature 'tbl_df'
friendly_class(object)

## S4 method for signature 'grouped_df'
friendly_class(object)

## S4 method for signature 'rowwise_df'
friendly_class(object)

## S4 method for signature 'list'
friendly_class(object)

## S4 method for signature 'matrix'
friendly_class(object)

## S4 method for signature 'array'
friendly_class(object)

Arguments

object

An object whose class will be described

Value

A character string of length 1, based on the class and length of object.

---

Grade this table

Description

Automatically grade a table resulting from student code using gradethis::grade_this() and tbl_grade() to compare the student's result with the author's solution.

Usage

grade_this_table(
  correct = NULL,
  pre_check = NULL,
  post_check = NULL,
  pass_if_equal = FALSE,
  ...,
  max_diffs = 3,
  cols = NULL,
  check_class = TRUE,
  ignore_class = NULL,
  check_names = TRUE,
  check_column_order = FALSE,
  check_dimensions = TRUE,
  check_groups = TRUE,
  check_columns = TRUE,
  check_column_class = check_columns,
  check_column_levels = check_columns,
  check_column_values = check_columns,
  tolerance = sqrt(.Machine$double.eps),
  check_row_order = check_columns,
  hint = getOption("gradethis.fail.hint", FALSE),
  encourage = getOption("gradethis.fail.encourage", FALSE),
  pass.praise = NULL
)

Arguments

correct	⁠[character(1)]⁠ The message shown to the student when their .result matches the exercise .solution, if pass_if_equal is TRUE.
pre_check, post_check	⁠[expression]⁠ Code to run before or after the table or vector grading is performed. The pre check runs before calling gradethis::pass_if_equal() so that you can modify or adjust the student's .result or the .solution if there are parts of either that need to be ignored. These arguments can also be used in conjunction with the pass_if_equal option when the grading requirements are more involved.
pass_if_equal	⁠[logical(1)]⁠ When TRUE (default for grade_this_vector() but not grade_this_table()), the .result is compared to the .solution with gradethis::pass_if_equal() after the pre check and before calling the tblcheck grading function.
...	Additional arguments passed to graded() or additional data to be included in the feedback object.
max_diffs	⁠[numeric(1)]⁠ The maximum number of mismatched values to display in an informative failure message. Passed to tbl_check_names() to determine the number of mismatched column names to display and the n_values argument of tbl_check_column() to determine the number of mismatched column values to display. Defaults to 3.
cols	[tidy-select] A selection of columns to compare between object and expected. Differences in other columns will be ignored. If NULL, the default, all columns will be checked.
check_class	⁠[logical(1)]⁠ Whether to check that object and expected have the same classes with tbl_check_class().
ignore_class	⁠[character()]⁠ A vector of classes to ignore when finding differences between object and expected. If an element is named, differences will only be ignored between the pair of the element and its name. For example, ignore_class = c("integer" = "numeric") will ignore class differences only if object has class integer and expected has class numeric, or vice versa. If all the classes of expected are included in ignore_class, a class problem will never be returned.
check_names	⁠[logical(1)]⁠ Whether to check that object and expected have the same column names with tbl_check_names().
check_column_order	⁠[logical(1)]⁠ Whether to check that the columns of object are in the same order as expected with tbl_check_names(). Defaults to FALSE.
check_dimensions	⁠[logical(1)]⁠ Whether to check that object and expected have the same number of rows and columns with tbl_check_dimensions().
check_groups	⁠[logical(1)]⁠ Whether to check that object and expected have the same groups with dplyr::group_vars().
check_columns	⁠[logical(1)]⁠ Whether to check that all columns have the same contents with tbl_check_column().
check_column_class	⁠[logical(1)]⁠ Whether to check that each column has the same class in object and expected.
check_column_levels	⁠[logical(1)]⁠ Whether to check that each column has the same factor levels in object and expected.
check_column_values	⁠[logical(1)]⁠ Whether to check that each column has the same values in object and expected.
tolerance	⁠[numeric(1) ≥ 0]⁠ values differences smaller than tolerance are ignored. The default value is close to 1.5e-8.
check_row_order	⁠[logical(1)]⁠ Whether to check that the values in each column are in the same order in object and expected.
hint	Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option.
encourage	Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.
pass.praise	Logical TRUE or FALSE to determine whether a praising phrase should be automatically prepended to any pass() or pass_if_equal() messages. Sets the gradethis.pass.praise option.

Value

The returned feedback is equivalent to gradethis grading code using grade_this() with the following components:

1. First the pre_check code, if any, is evaluated. If this code calls pass(), fail(), or their equivalents, that feedback is provided immediately.

2. If pass_if_equal is TRUE, then pass_if_equal() is called to compare the .result to the .solution. The message in correct is used for the feedback.

3. The appropriate tblcheck grading function is called, returning any feedback:

   1. grade_this_table() returns the results from tbl_grade()

   2. grade_this_vector() returns the results from vec_grade()

4. The post_check code, if any, is evaluated and any feedback from a call to pass(), fail(), or their equivalents is returned.

5. Finally, if no other feedback is returned, the feedback from gradethis::fail() is provided to the student, using the options fail.message, fail.hint and fail.encourage.

See Also

tbl_grade()

Other graders: grade_this_vector()

Examples

ex <- gradethis::mock_this_exercise(
	.solution_code = tibble::tibble(x = 1:3, y = letters[x]),
	.user_code = tibble::tibble(x = 1:3, y = c("A", "b", "c"))
)

## Grading Tables ----
grade_this_table()(ex)

# Roughly equivalent to...
gradethis::grade_this({
	gradethis::pass_if_equal()
	tbl_grade()
	gradethis::fail()
})(ex)

---

Grade this vector

Description

Automatically grade a vector resulting from student code using gradethis::grade_this() and vec_grade() to compare the student's result with the author's solution.

Usage

grade_this_vector(
  correct = NULL,
  pre_check = NULL,
  post_check = NULL,
  pass_if_equal = TRUE,
  ...,
  max_diffs = 3,
  check_class = TRUE,
  ignore_class = NULL,
  check_length = TRUE,
  check_levels = TRUE,
  check_values = TRUE,
  tolerance = sqrt(.Machine$double.eps),
  check_names = TRUE,
  hint = getOption("gradethis.fail.hint", FALSE),
  encourage = getOption("gradethis.fail.encourage", FALSE),
  pass.praise = NULL
)

Arguments

correct	⁠[character(1)]⁠ The message shown to the student when their .result matches the exercise .solution, if pass_if_equal is TRUE.
pre_check, post_check	⁠[expression]⁠ Code to run before or after the table or vector grading is performed. The pre check runs before calling gradethis::pass_if_equal() so that you can modify or adjust the student's .result or the .solution if there are parts of either that need to be ignored. These arguments can also be used in conjunction with the pass_if_equal option when the grading requirements are more involved.
pass_if_equal	⁠[logical(1)]⁠ When TRUE (default for grade_this_vector() but not grade_this_table()), the .result is compared to the .solution with gradethis::pass_if_equal() after the pre check and before calling the tblcheck grading function.
...	Additional arguments passed to graded() or additional data to be included in the feedback object.
max_diffs	⁠[numeric(1)]⁠ The maximum number of mismatched values to print. Defaults to 3.
check_class	⁠[logical(1)]⁠ Whether to check that object and expected have the same classes.
ignore_class	⁠[character()]⁠ A vector of classes to ignore when finding differences between object and expected. If an element is named, differences will only be ignored between the pair of the element and its name. For example, ignore_class = c("integer" = "numeric") will ignore class differences only if object has class integer and expected has class numeric, or vice versa. If all the classes of expected are included in ignore_class, a class problem will never be returned.
check_length	⁠[logical(1)]⁠ Whether to check that object and expected have the same length.
check_levels	⁠[logical(1)]⁠ Whether to check that object and expected have the same factor levels.
check_values	⁠[logical(1)]⁠ Whether to check that object and expected contain the same values.
tolerance	⁠[numeric(1) ≥ 0]⁠ values differences smaller than tolerance are ignored. The default value is close to 1.5e-8.
check_names	⁠[logical(1)]⁠ Whether to check that object and expected have the same names.
hint	Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option.
encourage	Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.
pass.praise	Logical TRUE or FALSE to determine whether a praising phrase should be automatically prepended to any pass() or pass_if_equal() messages. Sets the gradethis.pass.praise option.

Value

The returned feedback is equivalent to gradethis grading code using grade_this() with the following components:

1. First the pre_check code, if any, is evaluated. If this code calls pass(), fail(), or their equivalents, that feedback is provided immediately.

2. If pass_if_equal is TRUE, then pass_if_equal() is called to compare the .result to the .solution. The message in correct is used for the feedback.

3. The appropriate tblcheck grading function is called, returning any feedback:

   1. grade_this_table() returns the results from tbl_grade()

   2. grade_this_vector() returns the results from vec_grade()

4. The post_check code, if any, is evaluated and any feedback from a call to pass(), fail(), or their equivalents is returned.

5. Finally, if no other feedback is returned, the feedback from gradethis::fail() is provided to the student, using the options fail.message, fail.hint and fail.encourage.

See Also

vec_grade()

Other graders: grade_this_table()

Examples

ex <- gradethis::mock_this_exercise(
	.solution_code = tibble::tibble(x = 1:3, y = letters[x]),
	.user_code = tibble::tibble(x = 1:3, y = c("A", "b", "c"))
)

#' ## Grading Vectors ----
# Here we use `pre_check` to modify `.result` and
grade_this_vector(
	pre_check = {
		.result <- .result$y
		.solution <- .solution$y
	}
)(ex)

# Roughly equivalent to...
gradethis::grade_this({
	.result <- .result$y
	.solution <- .solution$y
	gradethis::pass_if_equal()
	vec_grade()
	gradethis::fail()
})(ex)

---

Generate a hint for how to convert one object type to another

Description

Generate a hint for how to convert one object type to another

Usage

hinted_class_message(object, expected)

## S4 method for signature 'ANY,ANY'
hinted_class_message(object, expected)

## S4 method for signature 'rowwise_df,grouped_df'
hinted_class_message(object, expected)

## S4 method for signature 'data.frame,grouped_df'
hinted_class_message(object, expected)

## S4 method for signature 'grouped_df,data.frame'
hinted_class_message(object, expected)

## S4 method for signature 'data.frame,rowwise_df'
hinted_class_message(object, expected)

## S4 method for signature 'rowwise_df,data.frame'
hinted_class_message(object, expected)

Arguments

object	An object to be compared to expected
expected	An object of the expected class

Value

A character string of length 1

---

Declare a problem

Description

Useful for constructing a small list to communicate the problem that was discovered during checking.

Usage

problem(
  type,
  expected,
  actual,
  ...,
  .class = c(paste0(type, "_problem"), "tblcheck_problem")
)

Arguments

type	A character string, e.g. column_values or table_rows, that describes the problem that was discovered.
expected, actual	The expected and actual values. These should be included when the value is a summary, e.g. nrow(expected) or length(actual). Be careful not to include large amounts of data.
...	Additional elements to be included in the problem object.
.class	The class of the problem. Typically, we expect the problem class to be ⁠<type>_problem⁠, but if you are building custom classes you may set these classes as desired.

Value

Returns a problem with class ⁠<type>_problem⁠ and the base classes tblcheck_problem and gradethis_problem.

See Also

Other Problem functions: problem_grade(), problem_message(), problem_type()

Examples

problem(
	type = "class",
	expected = "character",
	actual = "numeric",
	expected_length = 1,
	actual_length = 2
)

---

Apply automatic grading to a problem object

Description

Automatically converts a problem() object into a gradethis grade. problem_grade() is an S4 generic and tblcheck provides an internal method for problems with class "tblcheck_problem". In tblcheck, or for problems with this class, any problems are automatically turned into failing grades with gradethis::fail() and using the message provided by problem_message().

Usage

problem_grade(problem, max_diffs = 3, env = parent.frame(), ...)

## Default S3 method:
problem_grade(problem, max_diffs = 3, env = parent.frame(), ...)

## S3 method for class 'list'
problem_grade(problem, max_diffs = 3, env = parent.frame(), ...)

## S3 method for class 'gradethis_problem'
problem_grade(problem, max_diffs = 3, env = parent.frame(), ...)

## S3 method for class 'tblcheck_problem'
problem_grade(problem, max_diffs = 3, env = parent.frame(), ...)

Arguments

problem	A problem generated by tbl_check(), vec_check() or their related helper functions.
max_diffs	⁠[numeric(1)]⁠ The maximum number of mismatched values to display in an informative failure message. Passed to tbl_check_names() to determine the number of mismatched column names to display and the n_values argument of tbl_check_column() to determine the number of mismatched column values to display. Defaults to 3.
env	The environment used for grading.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

A gradethis::fail() message or NULL invisibly.

See Also

Other Problem functions: problem_message(), problem_type(), problem()

Examples

.result <- 1:10
.solution <- letters[1:10]
problem <- vec_check()
problem_grade(problem)

---

Create a message from a problem object

Description

problem_message() is an S3 generic that powers the conversion of problems detected by tbl_check(), vec_check(), and their related helper functions into a human-readable message.

Usage

problem_message(problem, ...)

Arguments

problem	An object with base class gradethis_problem. Problems identified by tblcheck also include tblcheck_problem, plus additional classes that more specifically identify the problem type.
...	Additional arguments passed to the underlying methods.

Value

A length-1 character string with a message describing the problem.

See Also

Other Problem functions: problem_grade(), problem_type(), problem()

Examples

problem <- problem(
	type = "class",
	expected = "character",
	actual = "numeric",
	expected_length = 1,
	actual_length = 2
)

problem_message(problem)

---

Problem helper functions

Description

• problem_type() returns a problem's type, or NULL if the input is not a problem.

• is_problem() tests whether an object is a gradethis problem.

• is_tblcheck_problem() tests whether an object is a problem created by tblcheck.

• as_problem() converts a list to a tblcheck_problem.

Usage

problem_type(x)

is_problem(x, type = NULL)

is_tblcheck_problem(x, type = NULL)

as_problem(x)

Arguments

x	An object
type	⁠[character(1)]⁠ A problem type

Details

If type is specified, is_problem() and is_tblcheck_problem() test whether an object is a problem of the specified type.

Value

is_problem() and is_tblcheck_problem() return a logical of length 1. problem_type() returns a character of length 1. as_problem() returns a tblcheck_problem.

See Also

Other Problem functions: problem_grade(), problem_message(), problem()

Examples

problem_type(vec_check(1, "1"))
is_problem(vec_check(1, "1"), "vector_class")
is_tblcheck_problem(vec_check(1, "1"), "class")

---

Check that the rows and columns of two tables are the same

Description

Checks for differences between object and expected in the following order:

1. Check table class with tbl_check_class()

2. Check column names with tbl_check_names()

3. Check number of rows and columns with tbl_check_dimensions()

4. Check groups with tbl_check_groups()

5. Check that each column is the same with tbl_check_column()

If the tables differ

• tbl_check() returns a list describing the problem

• tbl_grade() returns a failing grade and informative message with gradethis::fail()

Usage

tbl_check(
  object = .result,
  expected = .solution,
  cols = NULL,
  check_class = TRUE,
  ignore_class = NULL,
  check_names = TRUE,
  check_column_order = FALSE,
  check_dimensions = TRUE,
  check_groups = TRUE,
  check_columns = TRUE,
  check_column_class = check_columns,
  check_column_levels = check_columns,
  check_column_values = check_columns,
  tolerance = sqrt(.Machine$double.eps),
  check_row_order = check_columns,
  env = parent.frame()
)

tbl_grade(
  object = .result,
  expected = .solution,
  cols = NULL,
  max_diffs = 3,
  check_class = TRUE,
  ignore_class = NULL,
  check_names = TRUE,
  check_column_order = FALSE,
  check_dimensions = TRUE,
  check_groups = TRUE,
  check_columns = TRUE,
  check_column_class = check_columns,
  check_column_levels = check_columns,
  check_column_values = check_columns,
  tolerance = sqrt(.Machine$double.eps),
  check_row_order = check_columns,
  env = parent.frame(),
  ...
)

Arguments

object	A data frame to be compared to expected.
expected	A data frame containing the expected result.
cols	[tidy-select] A selection of columns to compare between object and expected. Differences in other columns will be ignored. If NULL, the default, all columns will be checked.
check_class	⁠[logical(1)]⁠ Whether to check that object and expected have the same classes with tbl_check_class().
ignore_class	⁠[character()]⁠ A vector of classes to ignore when finding differences between object and expected. If an element is named, differences will only be ignored between the pair of the element and its name. For example, ignore_class = c("integer" = "numeric") will ignore class differences only if object has class integer and expected has class numeric, or vice versa. If all the classes of expected are included in ignore_class, a class problem will never be returned.
check_names	⁠[logical(1)]⁠ Whether to check that object and expected have the same column names with tbl_check_names().
check_column_order	⁠[logical(1)]⁠ Whether to check that the columns of object are in the same order as expected with tbl_check_names(). Defaults to FALSE.
check_dimensions	⁠[logical(1)]⁠ Whether to check that object and expected have the same number of rows and columns with tbl_check_dimensions().
check_groups	⁠[logical(1)]⁠ Whether to check that object and expected have the same groups with dplyr::group_vars().
check_columns	⁠[logical(1)]⁠ Whether to check that all columns have the same contents with tbl_check_column().
check_column_class	⁠[logical(1)]⁠ Whether to check that each column has the same class in object and expected.
check_column_levels	⁠[logical(1)]⁠ Whether to check that each column has the same factor levels in object and expected.
check_column_values	⁠[logical(1)]⁠ Whether to check that each column has the same values in object and expected.
tolerance	⁠[numeric(1) ≥ 0]⁠ values differences smaller than tolerance are ignored. The default value is close to 1.5e-8.
check_row_order	⁠[logical(1)]⁠ Whether to check that the values in each column are in the same order in object and expected.
env	The environment in which to find .result and .solution.
max_diffs	⁠[numeric(1)]⁠ The maximum number of mismatched values to display in an informative failure message. Passed to tbl_check_names() to determine the number of mismatched column names to display and the n_values argument of tbl_check_column() to determine the number of mismatched column values to display. Defaults to 3.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

If there are any issues, a list from tbl_check() or a gradethis::fail() message from tbl_grade(). Otherwise, invisibly returns NULL.

Problems

1. class: The table does not have the expected classes.

2. not_table: object does not inherit the data.frame class.

3. names: The table has column names that are not expected, or is missing names that are expected.

4. names_order: The table has the same column names as expected, but in a different order.

5. ncol: The table doesn't have the expected number of columns.

6. nrow: The table doesn't have the expected number of rows.

7. groups: The table has groups that are not expected, or is missing groups that are expected.

Additional problems may be produced by tbl_check_column().

Examples

.result <- data.frame(a = 1:10, b = 11:20)
.solution <- tibble::tibble(a = 1:10, b = 11:20)
tbl_check()
tbl_grade()

.result <- tibble::tibble(a = 1:10, b = a, c = a, d = a, e = a, f = a)
.solution <- tibble::tibble(z = 1:10, y = z, x = z, w = z, v = z, u = z)
tbl_check()
tbl_grade()
tbl_grade(max_diffs = 5)
tbl_grade(max_diffs = Inf)

.result <- tibble::tibble(a = 1:10, b = 11:20)
.solution <- tibble::tibble(a = 1:11, b = 12:22)
tbl_check()
tbl_grade()

.result <- tibble::tibble(a = 1:10, b = 11:20)
.solution <- tibble::tibble(a = letters[1:10], b = letters[11:20])
tbl_check()
tbl_grade()

.result <- tibble::tibble(a = 1:10, intermediate = 6:15, b = 11:20)
.solution <- tibble::tibble(a = 1:10, b = 11:20)
tbl_check(cols = any_of(names(.solution)))
tbl_grade(cols = any_of(names(.solution)))

.result <- tibble::tibble(a = 1:10, b = 11:20)
.solution <- tibble::tibble(a = 11:20, b = 1:10)
tbl_check()
tbl_grade()
tbl_grade(max_diffs = 5)
tbl_grade(max_diffs = Inf)

.result <- tibble::tibble(a = 1:10, b = rep(1:2, 5))
.solution <- dplyr::group_by(tibble::tibble(a = 1:10, b = rep(1:2, 5)), b)
tbl_check()
tbl_grade()
tbl_grade(check_groups = FALSE)

---

Checks that two objects have the same classes

Description

Checks if object and expected have the same class. If the classes differ

• tbl_check_class() and vec_check_class() return a list describing the problem

• tbl_grade_class() and vec_grade_class() return a failing grade and informative message with gradethis::fail()

Usage

tbl_check_class(
  object = .result,
  expected = .solution,
  ignore_class = NULL,
  env = parent.frame()
)

vec_check_class(
  object = .result,
  expected = .solution,
  ignore_class = NULL,
  env = parent.frame()
)

tbl_grade_class(
  object = .result,
  expected = .solution,
  ignore_class = NULL,
  env = parent.frame(),
  ...
)

vec_grade_class(
  object = .result,
  expected = .solution,
  ignore_class = NULL,
  env = parent.frame(),
  ...
)

Arguments

object	An object to be compared to expected.
expected	An object containing the expected result.
ignore_class	⁠[character()]⁠ A vector of classes to ignore when finding differences between object and expected. If an element is named, differences will only be ignored between the pair of the element and its name. For example, ignore_class = c("integer" = "numeric") will ignore class differences only if object has class integer and expected has class numeric, or vice versa. If all the classes of expected are included in ignore_class, a class problem will never be returned.
env	The environment in which to find .result and .solution.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

If there are any issues, a list from tbl_check_class() and vec_check_class() or a gradethis::fail() message from tbl_grade_class() and vec_grade_class(). Otherwise, invisibly returns NULL.

Problems

1. class: The object does not have the expected classes

Examples

.result <- 1:10
.solution <- as.character(1:10)
vec_check_class()
vec_grade_class()

.result <- data.frame(a = 1:10)
.solution <- tibble::tibble(a = 1:10)
tbl_check_class()
tbl_grade_class()

.result <- tibble::tibble(a = 1:10, b = a %% 2 == 0)
.solution <- dplyr::group_by(tibble::tibble(a = 1:10, b = a %% 2 == 0), b)
tbl_check_class()
tbl_grade_class()

# Ignore the difference between tibble and data frame
.result <- data.frame(a = 1:10)
.solution <- tibble::tibble(a = 1:10)
tbl_check_class(ignore_class = c("tbl_df", "tbl"))
tbl_grade_class(ignore_class = c("tbl_df", "tbl"))

# Ignore the difference between integer and double
.result <- 1L
.solution <- 1
vec_check_class(ignore_class = c("integer" = "numeric"))
vec_grade_class(ignore_class = c("integer" = "numeric"))

---

Checks that a column is identical across two tables

Description

Checks for differences between the name column in object and in expected in the following order:

1. Check that the name column exists in object

2. Check class with vec_check_class()

3. Check length with vec_check_dimensions()

4. If the column is a factor, check factor levels with vec_check_levels()

5. Check column values with vec_check_values()

If the columns differ

• tbl_check_column() returns a list describing the problem

• tbl_grade_column() returns a failing grade and informative message with gradethis::fail()

Usage

tbl_check_column(
  column,
  object = .result,
  expected = .solution,
  check_class = TRUE,
  ignore_class = NULL,
  check_length = TRUE,
  check_levels = TRUE,
  check_values = TRUE,
  tolerance = sqrt(.Machine$double.eps),
  check_names = FALSE,
  env = parent.frame()
)

tbl_grade_column(
  column,
  object = .result,
  expected = .solution,
  max_diffs = 3,
  check_class = TRUE,
  ignore_class = NULL,
  check_length = TRUE,
  check_levels = TRUE,
  check_values = TRUE,
  tolerance = sqrt(.Machine$double.eps),
  check_names = FALSE,
  env = parent.frame(),
  ...
)

Arguments

column	⁠[character(1)]⁠ The name of the column to check.
object	A data frame to be compared to expected.
expected	A data frame containing the expected result.
check_class	⁠[logical(1)]⁠ Whether to check that column has the same class in object and expected.
ignore_class	⁠[character()]⁠ A vector of classes to ignore when finding differences between object and expected. If an element is named, differences will only be ignored between the pair of the element and its name. For example, ignore_class = c("integer" = "numeric") will ignore class differences only if object has class integer and expected has class numeric, or vice versa. If all the classes of expected are included in ignore_class, a class problem will never be returned.
check_length	⁠[logical(1)]⁠ Whether to check that column has the same length in object and expected.
check_levels	⁠[logical(1)]⁠ Whether to check that column and has the same factor levels in object and expected.
check_values	⁠[logical(1)]⁠ Whether to check that column has the same values in object and expected.
tolerance	⁠[numeric(1) ≥ 0]⁠ values differences smaller than tolerance are ignored. The default value is close to 1.5e-8.
check_names	⁠[logical(1)]⁠ Whether to check that column has the same names in object and expected. Defaults to FALSE.
env	The environment in which to find .result and .solution.
max_diffs	⁠[numeric(1)]⁠ The maximum number of mismatched values to print. Defaults to 3.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

If there are any issues, a list from tbl_check_column() or a gradethis::fail() message from tbl_grade_column(). Otherwise, invisibly returns NULL.

Problems

1. names (table_problem): object doesn't contain a column named column.

2. class: Any mismatch in the classes of the column.

3. length: The column doesn't have the expected length.

4. levels_n, levels, levels_reversed, levels_order: See vec_check_levels().

5. values: The column doesn't have the expected values.

6. names (column_problem): The column has different names than expected.

7. names_order: The column has the same names as expected, but in a different order.

Examples

.result <- tibble::tibble(a = 1:10, b = 11:20)
.solution <- tibble::tibble(a = letters[1:10], b = letters[11:20])
tbl_check_column("a")
tbl_grade_column("a")

.result <- tibble::tibble(a = 1:10, b = 11:20)
.solution <- tibble::tibble(a = 1:11, b = 12:22)
tbl_check_column("a")
tbl_grade_column("a")

.result <- tibble::tibble(a = 1:10, b = 11:20)
.solution <- tibble::tibble(a = 11:20, b = 1:10)
tbl_check_column("a")
tbl_grade_column("a")
tbl_grade_column("a", max_diffs = 5)
tbl_grade_column("a", max_diffs = Inf)

---

Check that the dimensions of two object are the same

Description

Checks if object and expected have the same dimenisons. If the dimensions differ

• tbl_check_dimensions() returns a list describing the problem

• tbl_grade_dimensions() returns a failing grade and informative message with gradethis::fail()

Usage

tbl_check_dimensions(
  object = .result,
  expected = .solution,
  check_ncol = TRUE,
  env = parent.frame()
)

vec_check_dimensions(
  object = .result,
  expected = .solution,
  check_ncol = TRUE,
  env = parent.frame()
)

vec_check_length(
  object = .result,
  expected = .solution,
  check_ncol = TRUE,
  env = parent.frame()
)

tbl_grade_dimensions(
  object = .result,
  expected = .solution,
  check_ncol = TRUE,
  env = parent.frame(),
  ...
)

vec_grade_dimensions(
  object = .result,
  expected = .solution,
  check_ncol = TRUE,
  env = parent.frame(),
  ...
)

vec_grade_length(
  object = .result,
  expected = .solution,
  check_ncol = TRUE,
  env = parent.frame(),
  ...
)

Arguments

object	An object to be compared to expected.
expected	An object containing the expected result.
check_ncol	⁠[logical(1)]⁠ Whether to check that object and expected have the same number of columns.
env	The environment in which to find .result and .solution.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

If there are any issues, a list from tbl_check_dimensions() or a gradethis::fail() message from tbl_grade_dimensions(). Otherwise, invisibly returns NULL.

Problems

1. dimensions_n: object and expected have a different number of dimensions

2. length: object and expected are one-dimensional vectors of different lengths

3. ncol: object and expected are two-dimensional objects with a different number of columns

4. nrow: object and expected are two-dimensional objects with a different number of rows

5. dimensions: object and expected are multi-dimensional arrays with different dimensions

Examples

.result <- 1:10
.solution <- 1:5
tbl_check_dimensions()
tbl_grade_dimensions()

.result <- tibble::tibble(a = 1:10, b = 1:10, c = 1:10)
.solution <- tibble::tibble(a = 1:10, b = 1:10)
tbl_check_dimensions()
tbl_grade_dimensions()

.result <- tibble::tibble(a = 1:10, b = 1:10)
.solution <- tibble::tibble(a = 1:5, b = 1:5)
tbl_check_dimensions()
tbl_grade_dimensions()

.result <- 1:12
.solution <- matrix(1:12, 3)
tbl_check_dimensions()
tbl_grade_dimensions()

---

Check that the groups of two object are the same

Description

Checks if object and expected have the same groups. If the groups differ

• tbl_check_groups() returns a list describing the problem

• tbl_grade_groups() returns a failing grade and informative message with gradethis::fail()

Usage

tbl_check_groups(object = .result, expected = .solution, env = parent.frame())

tbl_grade_groups(
  object = .result,
  expected = .solution,
  max_diffs = 3,
  env = parent.frame(),
  ...
)

Arguments

object	An object to be compared to expected.
expected	An object containing the expected result.
env	The environment in which to find .result and .solution.
max_diffs	⁠[numeric(1)]⁠ The maximum number of missing and/or unexpected names to include in an informative failure message. Defaults to 3.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

If there are any issues, a list from tbl_check_groups() or a gradethis::fail() message from tbl_grade_groups(). Otherwise, invisibly returns NULL.

Problems

1. groups: The object has groups that are not expected, or is missing groups that are expected.

Examples

.result <- dplyr::group_by(tibble::tibble(a = 1:10, b = 11:20), a)
.solution <- dplyr::group_by(tibble::tibble(a = 1:10, b = 11:20), b)
tbl_check_groups()
tbl_grade_groups()

---

Checks that an object is a table

Description

Checks if object inherits the data.frame class. If the not

• tbl_check_is_table() returns a list describing the problem

• tbl_grade_is_table() returns a failing grade and informative message with gradethis::fail()

Usage

tbl_check_is_table(object = .result, env = parent.frame())

tbl_grade_is_table(object = .result, env = parent.frame(), ...)

Arguments

object	An object to be compared to expected.
env	The environment in which to find .result and .solution.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

If there are any issues, a list from tbl_check_is_table() or a gradethis::fail() message from tbl_grade_is_table(). Otherwise, invisibly returns NULL.

Problems

1. not_table: The object is not a table

Examples

.result <- data.frame(a = 1:10)
tbl_check_is_table()
tbl_grade_is_table()

.result <- tibble::tibble(a = 1:10)
tbl_check_is_table()
tbl_grade_is_table()

.result <- list(a = 1:10)
tbl_check_is_table()
tbl_grade_is_table()

---

Check that the names of two object are the same

Description

Checks if object and expected have the same names. If the names differ

• tbl_check_names() and vec_check_names() returns a list describing the problem

• tbl_grade_names() and vec_grade_names() returns a failing grade and informative message with gradethis::fail()

Usage

tbl_check_names(
  object = .result,
  expected = .solution,
  check_order = TRUE,
  env = parent.frame()
)

vec_check_names(
  object = .result,
  expected = .solution,
  check_order = TRUE,
  env = parent.frame()
)

tbl_grade_names(
  object = .result,
  expected = .solution,
  max_diffs = 3,
  check_order = TRUE,
  env = parent.frame(),
  ...
)

vec_grade_names(
  object = .result,
  expected = .solution,
  max_diffs = 3,
  check_order = TRUE,
  env = parent.frame(),
  ...
)

Arguments

object	An object to be compared to expected.
expected	An object containing the expected result.
check_order	⁠[logical(1)]⁠ Whether to check that the names of object and expected are in the same order.
env	The environment in which to find .result and .solution.
max_diffs	⁠[numeric(1)]⁠ The maximum number of missing and/or unexpected names to include in an informative failure message. Defaults to 3.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

If there are any issues, a list from tbl_check_names() and vec_check_names() or a gradethis::fail() message from tbl_grade_names() and vec_grade_names(). Otherwise, invisibly returns NULL.

Problems

1. names: The object has names that are not expected, or is missing names that are expected.

2. names_order: The object has the same names as expected, but in a different order.

Examples

.result <- c(1, 2, 3, 4, 5, 6, 7)
.solution <- c(a = 1, b = 2, c = 3, d = 4, e = 5, f = 6, g = 7)
vec_check_names()
vec_grade_names()
vec_grade_names(max_diffs = 5)
vec_grade_names(max_diffs = Inf)

.result <- tibble::tibble(a = 1:5, b = 6:10, c = 11:15)
.solution <- tibble::tibble(a = 1:5, x = 6:10, y = 11:15)
tbl_check_names()
tbl_grade_names()

---

Check that the rows and columns of two tables are the same

Description

Test if two tables are equivalent using the same process as tbl_check(). Unlike tbl_check(), which returns either a problem object or NULL, tbl_equal() returns either TRUE or FALSE.

Usage

tbl_equal(
  object = .result,
  expected = .solution,
  cols = NULL,
  check_class = TRUE,
  ignore_class = NULL,
  check_names = TRUE,
  check_column_order = FALSE,
  check_dimensions = TRUE,
  check_groups = TRUE,
  check_columns = TRUE,
  check_column_class = check_columns,
  check_column_levels = check_columns,
  check_column_values = check_columns,
  tolerance = sqrt(.Machine$double.eps),
  check_row_order = check_columns,
  env = parent.frame()
)

Arguments

object	A data frame to be compared to expected.
expected	A data frame containing the expected result.
cols	[tidy-select] A selection of columns to compare between object and expected. Differences in other columns will be ignored. If NULL, the default, all columns will be checked.
check_class	⁠[logical(1)]⁠ Whether to check that object and expected have the same classes with tbl_check_class().
ignore_class	⁠[character()]⁠ A vector of classes to ignore when finding differences between object and expected. If an element is named, differences will only be ignored between the pair of the element and its name. For example, ignore_class = c("integer" = "numeric") will ignore class differences only if object has class integer and expected has class numeric, or vice versa. If all the classes of expected are included in ignore_class, a class problem will never be returned.
check_names	⁠[logical(1)]⁠ Whether to check that object and expected have the same column names with tbl_check_names().
check_column_order	⁠[logical(1)]⁠ Whether to check that the columns of object are in the same order as expected with tbl_check_names(). Defaults to FALSE.
check_dimensions	⁠[logical(1)]⁠ Whether to check that object and expected have the same number of rows and columns with tbl_check_dimensions().
check_groups	⁠[logical(1)]⁠ Whether to check that object and expected have the same groups with dplyr::group_vars().
check_columns	⁠[logical(1)]⁠ Whether to check that all columns have the same contents with tbl_check_column().
check_column_class	⁠[logical(1)]⁠ Whether to check that each column has the same class in object and expected.
check_column_levels	⁠[logical(1)]⁠ Whether to check that each column has the same factor levels in object and expected.
check_column_values	⁠[logical(1)]⁠ Whether to check that each column has the same values in object and expected.
tolerance	⁠[numeric(1) ≥ 0]⁠ values differences smaller than tolerance are ignored. The default value is close to 1.5e-8.
check_row_order	⁠[logical(1)]⁠ Whether to check that the values in each column are in the same order in object and expected.
env	The environment in which to find .result and .solution.

Value

A TRUE or FALSE value.

Examples

tbl_equal(
	data.frame(a = 1:10, b = 11:20),
	data.frame(b = 11:20, a = 1:10)
)

---

Checks that two vectors are the same

Description

Checks for differences between object and expected in the following order:

1. Check class with vec_check_class()

2. Check length with vec_check_dimensions()

3. If the vector is a factor, check factor levels are the same with vec_check_levels()

4. Check vector values are the same with vec_check_values()

5. Check names with vec_check_names()

If the vectors differ

• vec_check() returns a list describing the problem

• vec_grade() returns a failing grade and informative message with gradethis::fail()

Usage

vec_check(
  object = .result,
  expected = .solution,
  check_class = TRUE,
  ignore_class = NULL,
  check_length = TRUE,
  check_levels = TRUE,
  check_values = TRUE,
  tolerance = sqrt(.Machine$double.eps),
  check_names = TRUE,
  env = parent.frame()
)

vec_grade(
  object = .result,
  expected = .solution,
  max_diffs = 3,
  check_class = TRUE,
  ignore_class = NULL,
  check_length = TRUE,
  check_levels = TRUE,
  check_values = TRUE,
  tolerance = sqrt(.Machine$double.eps),
  check_names = TRUE,
  env = parent.frame(),
  ...
)

Arguments

object	A vector to be compared to expected.
expected	A vector containing the expected result.
check_class	⁠[logical(1)]⁠ Whether to check that object and expected have the same classes.
ignore_class	⁠[character()]⁠ A vector of classes to ignore when finding differences between object and expected. If an element is named, differences will only be ignored between the pair of the element and its name. For example, ignore_class = c("integer" = "numeric") will ignore class differences only if object has class integer and expected has class numeric, or vice versa. If all the classes of expected are included in ignore_class, a class problem will never be returned.
check_length	⁠[logical(1)]⁠ Whether to check that object and expected have the same length.
check_levels	⁠[logical(1)]⁠ Whether to check that object and expected have the same factor levels.
check_values	⁠[logical(1)]⁠ Whether to check that object and expected contain the same values.
tolerance	⁠[numeric(1) ≥ 0]⁠ values differences smaller than tolerance are ignored. The default value is close to 1.5e-8.
check_names	⁠[logical(1)]⁠ Whether to check that object and expected have the same names.
env	The environment in which to find .result and .solution.
max_diffs	⁠[numeric(1)]⁠ The maximum number of mismatched values to print. Defaults to 3.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

If there are any issues, a list from vec_check() or a gradethis::fail() message from vec_grade(). Otherwise, invisibly returns NULL.

Problems

1. class: object doesn't have the same classes as expected.

2. length: object doesn't have the same length as expected.

3. levels_n, levels, levels_reversed, levels_order: See vec_check_levels().

4. values: object doesn't contain the same values as expected.

5. names: object has different names than expected.

6. names_order: object has the same names as expected, but in a different order.

Examples

.result <- 1:10
.solution <- letters[1:10]
vec_check()
vec_grade()

.result <- 1:10
.solution <- 1:11
vec_check()
vec_grade()

.result <- 1:10
.solution <- rlang::set_names(1:10, letters[1:10])
vec_check()
vec_grade()
vec_grade(max_diffs = 5)
vec_grade(max_diffs = Inf)

.result <- 1:10
.solution <- 11:20
vec_check()
vec_grade()
vec_grade(max_diffs = 5)
vec_grade(max_diffs = Inf)

---

Check that the levels of two factors are the same

Description

Checks if object and expected have the same levels. If the levels differ

• vec_check_levels() returns a list describing the problem

• vec_grade_levels() returns a failing grade and informative message with gradethis::fail()

Usage

vec_check_levels(object = .result, expected = .solution, env = parent.frame())

vec_grade_levels(
  object = .result,
  expected = .solution,
  max_diffs = 3,
  env = parent.frame(),
  ...
)

Arguments

object	An object to be compared to expected.
expected	An object containing the expected result.
env	The environment in which to find .result and .solution.
max_diffs	⁠[numeric(1)]⁠ The maximum number of missing and/or unexpected names to include in an informative failure message. Defaults to 3.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

If there are any issues, a list from vec_check_levels() or a gradethis::fail() message from vec_grade_levels(). Otherwise, invisibly returns NULL.

Problems

1. levels_n: object and expected have a different number of levels.

2. levels: The object has levels that are not expected, or is missing levels that are expected.

3. levels_reversed: The levels of object are in the opposite order of expected.

4. level_order: The levels of object are not in the same order as expected.

Examples

.result <- as.factor(rep_len(letters[1:3], 6))
.solution <- as.factor(rep_len(letters[1:2], 6))
vec_check_levels()
vec_grade_levels()

.result <- as.factor(letters[1:6])
.solution <- as.factor(letters[21:26])
vec_check_levels()
vec_grade_levels()
vec_grade_levels(max_diffs = 5)
vec_grade_levels(max_diffs = Inf)

---

Checks that two vectors are contain the same values

Description

Check if two vectors contain the same values. If the values differ

• vec_check_values() returns a list describing the problem

• vec_grade_values() returns a failing grade and informative message with gradethis::fail()

Usage

vec_check_values(
  object = .result,
  expected = .solution,
  tolerance = sqrt(.Machine$double.eps),
  env = parent.frame()
)

vec_grade_values(
  object = .result,
  expected = .solution,
  tolerance = sqrt(.Machine$double.eps),
  max_diffs = 3,
  env = parent.frame(),
  ...
)

Arguments

object	A vector to be compared to expected.
expected	A vector containing the expected result.
tolerance	⁠[numeric(1) ≥ 0]⁠ values differences smaller than tolerance are ignored. The default value is close to 1.5e-8.
env	The environment in which to find .result and .solution.
max_diffs	⁠[numeric(1)]⁠ The maximum number of mismatched values to print. Defaults to 3.
...	Arguments passed on to gradethis::fail hint Include a code feedback hint with the failing message? This argument only applies to fail() and fail_if_equal() and the message is added using the default options of give_code_feedback() and maybe_code_feedback(). The default value of hint can be set using gradethis_setup() or the gradethis.fail.hint option. encourage Include a random encouraging phrase with random_encouragement()? The default value of encourage can be set using gradethis_setup() or the gradethis.fail.encourage option.

Value

If there are any issues, a list from vec_check_values() or a gradethis::fail() message from vec_grade_values(). Otherwise, invisibly returns NULL.

Problems

1. values: object doesn't contain the same values as expected

Examples

.result <- 1:10
.solution <- letters[1:10]
vec_check_values()
vec_grade_values()

.result <- 1:10
.solution <- 1:11
vec_check_values()
vec_grade_values()

.result <- 1:10
.solution <- rlang::set_names(1:10, letters[1:10])
vec_check_values()
vec_grade_values()
vec_grade_values(max_diffs = 5)
vec_grade_values(max_diffs = Inf)

.result <- 1:10
.solution <- 11:20
vec_check_values()
vec_grade_values()
vec_grade_values(max_diffs = 5)
vec_grade_values(max_diffs = Inf)

---