tests/testthat/test_mgsub_censor.R

In mgsub: Safe, Multiple, Simultaneous String Substitution

#mgsub_censor--------
context("Censor Call")
suppressWarnings(RNGversion("3.5.0"))

test_that("basic functionality", {
  expect_equal(mgsub_censor("hey, ho", pattern = c("hey", "ho"), censor = "*"), "***, **")
  expect_equal(mgsub_censor("Production Workers, All Other",
                            c("s$", "s([[:punct:] ])", ", ? All Other$"), "*"),
               "Production Workers***********")
})

test_that("non-named mgsub_censor and named sub works", {
  expect_equal(mgsub_censor("Dopazamine and dopaloramide are fake chemicals."
                            , c("dopa", "fake"), "?", ignore.case = TRUE)
               , "????zamine and ????loramide are ???? chemicals.")
  expect_equal(mgsub_censor("Hey, ho", ignore.case = TRUE, c("hey"), "&"), "&&&, ho")
})

test_that("partially named mgsub_censor inputs works", {
  expect_equal(mgsub_censor("Dopazamine and dopaloramide are fake chemicals."
                            , pattern = c("dopa", "fake"), "*", ignore.case = TRUE)
               , "****zamine and ****loramide are **** chemicals.")
  expect_equal(mgsub_censor("Dopazamine and dopaloramide are fake chemicals."
                            , "*", pattern = c("dopa", "fake"), ignore.case = TRUE)
               , "****zamine and ****loramide are **** chemicals.")
})

test_that("multiple input strings are processed", {
  expect_equal(mgsub_censor(c("string", "test"), c("t"), "*")
               , c("s*ring", "*es*"))
  expect_equal(mgsub_censor(c("Dopazamine is not the same as Dopachloride and is still fake."
                              , "dopazamine is undergoing a review by the fake news arm of the Dopazamine government"),
                            c("[Dd]opa(.*?mine)", "fake"), "*"),
               c("********** is not the same as Dopachloride and is still ****.",
                 "********** is undergoing a review by the **** news arm of the ********** government"))
})

test_that("NAs are correctly handled", {
  expect_equal(mgsub_censor(c("string", NA, "test"), c("t"), "*")
               , c("s*ring", NA, "*es*"))
})

test_that("works even with start/end symbols", {
  expect_equal(mgsub_censor(c("hi there", "who said hi to me?", "who said hi"), c("^hi"), "*")
               , c("** there", "who said hi to me?", "who said hi"))
  expect_equal(mgsub_censor(c("hi there", "who said hi to me?", "who said hi"), c("hi$"), "*")
               , c("hi there", "who said hi to me?", "who said **"))
})

test_that("all NA fails quickly", {
  expect_equal(mgsub_censor(rep(NA, 4), "A", "b"), rep(NA, 4))
  expect_equal(mgsub_censor(NA, "A", "b"), NA)
})

test_that("multicharacter censor works", {
  expect_equal(mgsub_censor("What the flip?", "flip", "?#!*", seed = 10), "What the !##!?")
  expect_equal(mgsub_censor("What the flip?", "flip", c("?", "#", "!", "*"), seed = 10), "What the !##!?")
  expect_equal(mgsub_censor("What the flip?", "flip", c("?#", "!*"), seed = 10), "What the !##!?")
})

test_that("multicharacter censor can be used badly", {
  expect_equal(mgsub_censor("What the flip?", "flip", "?#!*", seed = 10, split = FALSE)
               , "What the ?#!*?#!*?#!*?#!*?")
  expect_equal(mgsub_censor("What the flip?", "flip", c("?#", "!*"), seed = 10, split = FALSE)
               , "What the !*?#?#!*?")
})

#censor_worker --------
context("censor_worker")

test_that("Letter substitution works", {
  expect_equal(unlist(censor_worker("ho ho hoot", c("h", "o"), "*")), "** ** ***t")
  expect_equal(unlist(censor_worker("developer", c("e", "p"), "*")), "d*v*lo**r")
})

test_that("Substring matches are superceded by longer matches", {
  expect_equal(unlist(censor_worker("they don't understand the value of what they seek."
                                    , c("the", "they"), "*"))
               , "**** don't understand *** value of what **** seek.")
})

test_that("Regular expression matches work", {
  expect_equal(unlist(censor_worker(
    "Dopazamine (of dopazamines family) is not the same as Dopachloride and is still fake."
                                    , c("[Dd]opa", "fake"), "*"))
               , "****zamine (of ****zamines family) is not the same as ****chloride and is still ****.")
})

test_that("Options passed to sub family work", {
  expect_equal(unlist(censor_worker("Dopazamine and dopaloramide are fake chemicals."
                                    , c("dopa", "fake"), "*", ignore.case = TRUE))
               , "****zamine and ****loramide are **** chemicals.")
  expect_equal(unlist(censor_worker("Where are the \\bspaces\\b - not the spaces I want?"
                                    , c("\\bspaces\\b", "[Ww]here"), "*", fixed = TRUE))
               , "Where are the ********** - not the spaces I want?")
})

test_that("Priority is based on matched length", {
  expect_equal(censor_worker("Dopazamine is a fake chemical", c("do.*ne", "dopazamin"), "*", ignore.case = TRUE)
               , "********** is a fake chemical")
})

test_that("all missing patterns works", {
  expect_equal(censor_worker("hi there", c("why", "not", "go"), "*"), "hi there")
})

test_that("some missing patterns work", {
  expect_equal(censor_worker("hi there", c("hi", "bye"), "*"), "** there")
})