Grade result of exercise code (Legacy)

Please use grade_this() mixed with pass(), pass_if_equal(), fail(), and/or fail_if_equal().

grade_result() and grade_result_strict() both take a set of pass_if()/fail_if() conditions, evaluate them, and return a final graded() object. For grade_result_strict() to return a correct grade, every pass_if() condition must be met, and every fail_if() condition must not be met. On the other hand, grade_result()'s final grade reflects the first satisfied condition (if no conditions are met, the final grade can be controlled by default_correct and default_message).

Usage

grade_result(
  ...,
  correct = NULL,
  incorrect = NULL,
  glue_correct = getOption("gradethis.glue_correct"),
  glue_incorrect = getOption("gradethis.glue_incorrect"),
  default_correct = "auto",
  default_message = NULL,
  grader_args = deprecated(),
  learnr_args = deprecated()
)

grade_result_strict(
  ...,
  correct = NULL,
  incorrect = NULL,
  glue_correct = getOption("gradethis.glue_correct_test"),
  glue_incorrect = getOption("gradethis.glue_incorrect_test"),
  grader_args = deprecated(),
  learnr_args = deprecated()
)

condition(x, message, correct)

Arguments

...

pass_if()/fail_if() condition()s to check.

correct

logical whether the condition is the correct answer.

incorrect

A character string to display if the student answer matches a known incorrect answer.

glue_correct

A glue string that returns the final correct message displayed. Defaults to getOption("gradethis_glue_correct").

glue_incorrect

A glue string that returns the final incorrect message displayed. Defaults to getOption("gradethis_glue_incorrect").

default_correct

In the event that no condition()s are met, should the end result be correct? When "auto", this will be TRUE when all the conditions() are fail_if() (and FALSE otherwise).

default_message

In the event that no condition()s are met, what message should be included with the returned graded() object?

grader_args

A list of parameters passed to grader functions (provided by gradethis_exercise_checker()). This contains:

• user_quo: Quoted R code submitted by the user. For example rlang::quo(1)

• solution_quo: (Optional) Quoted solution R code provided by the *-solution chunk for an exercise.

learnr_args

A list of all parameters passed to gradethis_exercise_checker() by learnr. See https://rstudio.github.io/learnr/exercises.html#exercise_checking for more details.

x

A formula, function, or value, that returns TRUE or FALSE. When comparing objects that are greater than length 1 (e.g., vectors, dataframes, matrices, etc) A logical vector will be returned if the user uses ==, not a single logical value. gradethis will run the vector through all(..., na.rm = TRUE) to check for the logical value. It is advised that the user use identical() instead of == in this case.

message

character string for message returned (usually passed in from pass_if() or fail_if().

Value

a function whose first parameter should be an environment that contains all necessary information to compare the user's result. The result of the returned function will be a graded() object containing a formatted correct or incorrect message.

See also

grade_code()

Examples

# The student submits code that returns `5`
submitted_5 <- mock_this_exercise(.user_code = 2 + 3)

# grade_result() returns a function that is called on the student's
# submission. The *-check chunk of the exercise might contain the following
# grade_result() code. When you're writing your tutorial, you can ignore the
# `(submitted_5)`, we're using that notation here to preview what this
# `grade_result()` will do with the mock submission.
grade_result(
  fail_if(~ identical(.result, 4), "Try adding 1"),
  pass_if(~ identical(.result, 5), "You got 5, great!"),
  fail_if(~ TRUE, "Some generic failing message.")
)(submitted_5)
#> <gradethis_graded: [Correct] You rock! You got 5, great! >

grade_result_strict(
  pass_if(~ identical(.result, 5), "You got 5, great!"),
  fail_if(~ !is.integer(.result), "I expected an integer")
)(submitted_5)
#> <gradethis_graded: [Incorrect]
#>   1/2 correct! But no need to fret, try it again.
#> >

# Suppose our exercise asks the student to write a function that adds 1 to
# the value of its only argument. The student submits:
submitted_fn <- mock_this_exercise(
  .user_code = function(x) { x + 2 }
)

# We can check submissions for this exercise using `grade_result()`. Note that
# because all checks in this `grade_result()` are `fail_if()` statements, if
# the student's submission passes all of the checks, then `grade_result()`
# will return a correct grade.
grade_result(
  fail_if(~ !is.function(.result), "I expected a function."),
  fail_if(~ .result(1) != 2, "Your function should add one.")
)(submitted_fn)
#> <gradethis_graded: [Incorrect]
#>   Your function should add one.  That's okay: you learn more from
#>   mistakes than successes. Let's do it one more time.
#> >

# To learn more about using grade_result() and grade_code() with learnr, see:
if (FALSE) {
gradethis_demo()
}