tests/testthat/test_autoscore.R

In autoscore/autoscore: Automatically Score Word Identification in Speech Perception

df <- autoscore::example_data
acceptable_df <- tibble::tibble(
  target = c("model",
             "treason",
             "duck"),
  acceptable= c("modal, moddel",
                          "treeson",
                          "dock")
)

testthat::expect_s3_class(autoscore::autoscore(df), "data.frame")
testthat::expect_s3_class(autoscore::autoscore(df,
                                               acceptable_df = acceptable_df), "data.frame")
testthat::expect_s3_class(autoscore::autoscore(df,
                                               acceptable_df = acceptable_df,
                                               root_word_rule = TRUE), "data.frame")
testthat::expect_s3_class(autoscore::autoscore(df,
                                               root_word_rule = FALSE,
                                               double_letter_rule = FALSE), "data.frame")
testthat::expect_s3_class(autoscore::autoscore(df,
                                               suffix_rule = TRUE,
                                               double_letter_rule = FALSE), "data.frame")
testthat::expect_s3_class(autoscore::autoscore(df,
                                               suffix_rule = TRUE,
                                               tense_rule = TRUE,
                                               double_letter_rule = FALSE), "data.frame")
testthat::expect_s3_class(autoscore::autoscore(df,
                                               suffix_rule = TRUE,
                                               plural_rule = TRUE,
                                               double_letter_rule = FALSE), "data.frame")
testthat::expect_s3_class(autoscore::autoscore(df,
                                               suffix_rule = TRUE,
                                               tense_rule = FALSE,
                                               tense_add_rule = TRUE,
                                               double_letter_rule = FALSE), "data.frame")
testthat::expect_s3_class(autoscore::autoscore(df,
                                               suffix_rule = TRUE,
                                               plural_rule = FALSE,
                                               plural_add_rule = TRUE,
                                               double_letter_rule = FALSE), "data.frame")

alternate_df <- tibble::tibble(
  target = c("beat",
             "treeson"),
  acceptable = c("beet, baet",
                 "treason"))

d <- tibble::tribble(
  ~id, ~target, ~response, ~human,
  1, "the coin ate it", "a coins for it", 3,
  2, "beat the clock", "beets the clock", 3,
  3, "beated it", "beet it", 2,
  4, "beets the clock", "beat the clock", 3,
  5, "beeted the clock", "beet the clock", 3,
  6, "junkyard", "junk yard", 0,
  7, "junk yard", "junkyard", 1,
  8, "The matches are on the shelf", "23  the matches are on the shelf", 6,
  9, "The puppy played with a ball", "1  x", 0,
  10, "One two three", "1 2 3", 3,
  11, "She will each be", "She'll be", 3
)

contractions_list = tibble::tibble(
  contraction = "She'll",
  replacement = "She will"
)

autoscored <- autoscore::autoscore(d, alternate_df,
                                   plural_rule = TRUE,
                                   tense_rule = TRUE,
                                   root_word_rule = TRUE,
                                   a_the_rule = TRUE,
                                   number_text_rule = TRUE,
                                   contractions_df = contractions_list,
                                   output = "text")

testthat::expect_s3_class(autoscored,
                          "data.frame")
testthat::expect_equal(autoscored$equal, rep(TRUE, 11))

autoscored2 <- autoscore::autoscore(d, alternate_df,
                                   plural_rule = TRUE,
                                   tense_rule = TRUE,
                                   root_word_rule = FALSE,
                                   a_the_rule = TRUE,
                                   output = "text")

testthat::expect_s3_class(autoscored2,
                          "data.frame")
testthat::expect_equal(autoscored2$equal, c(rep(TRUE, 6), FALSE, TRUE, TRUE, FALSE, FALSE))

testthat::expect_equal({
  target = c(
    "we will get to know these players as we go along",
    "according to the rules, you shouldn't end any sentence with a preposition",
    "it will only work if it will do work",
    "nope"
  )
  double_word_detect(target)
  },
  c("we", NA, "it, will, work", NA)
)

testthat::expect_equal({
  d <- tibble::tribble(
    ~id, ~target, ~response, ~human,
    1, "The score was the goal", "The score was a goal", 4,
    2, "The score was the goal", "The score was the goal", 5,
    3, "The score was a goal", "The score was a goal", 5,
    4, "The score was a goal", "The score was the goal", 4,
    5, "The score was the goal", "The the score was the goal", 5,
    6, "The they score was the goal", "The score was the goal", 5,
    7, "A patient seriously hurt his ankle in a motorcycle accident", "My patient seriously hurt his ankle in a motorcycle accident", 9,
    8, "There were seven hundred things", "There were 700 stuff", 4,
    9, "One two three", "1 2 3", 3,
    10, "Junkyard is far", "Junk yard is far", 3,
    11, "The twenties were great", "The 20s were great", 4
  )
  comp <- c("junkyard" = "junk yard")
  autoscore(d, number_text_rule = TRUE, compound_rule = comp) %>%
    dplyr::pull(equal)
},
rep(TRUE, 11)
)


testthat::expect_equal({
  d <- tibble::tribble(
    ~id, ~target, ~response, ~human,
    11, "You have to feed yourself with information every day", "You have to feed yourself with information everyday", 9,
    12, "The wait was", "the weight xxxx", 2,
    13, "I have to hide it because I have young children", "I have to hit it", 4,
    14, "He will share his life's work, his happiness, and his problems with them", "he will share his life's work: his happiness and his problems with them", 13,
    15, "Tests resulted in the removal of many foods from the diets of millions of diabetics", "test results in the removal of many foods from the diets of millions of diabetics", 13,
    16, "Finally, the dog cut across the circle and jumped back into the wagon.",	"(unintelligible)", 0,
    17, "He will share his life's work, his happiness, and his problems with them.", "He won't charge all of our hey ham in ah one of.", 1
  )
  comp <- c("junkyard" = "junk yard", "every day" = "everyday")
  autoscore(d, number_text_rule = TRUE, compound_rule = comp, acceptable_df = autoscore::acceptable_spellings) %>%
    dplyr::pull(equal)
},
rep(TRUE, 7)
)