tests/testthat/test-acc_end_digits.R

In dataquieR: Data Quality in Epidemiological Research

test_that("acc_end_digits works with 2 args", {
  skip_on_cran() # slow, errors obvious
  meta_data <- prep_get_data_frame("meta_data")
  study_data <- prep_get_data_frame("study_data")
  expect_error(
    res1 <-
      acc_end_digits(study_data = study_data, meta_data = meta_data),
    regexp = "Argument resp_vars is NULL",
    perl = TRUE
  )
  expect_message(
    res1 <-
      acc_end_digits(resp_vars = "v00014", study_data = study_data,
                     meta_data = meta_data),
    regexp =
      sprintf(
        "(%s)",
        paste("Due to missing values in v00014_x_last 301",
              "observations were deleted.")
      ),
    perl = TRUE,
    all = TRUE
  )

  expect_true(all(c("SummaryData", "SummaryPlot") %in% names(res1)))
  expect_equal(
    nrow(res1$SummaryData),
    1
  )
  expect_equal(
    ncol(res1$SummaryData),
    2
  )
  expect_equal(
    suppressWarnings(sum(as.numeric(
      as.matrix(res1$SummaryData)),
      na.rm = TRUE)), 1
  )
})

test_that("acc_end_digits works with label_col", {
  skip_on_cran() # slow, errors obvious
  meta_data <- prep_get_data_frame("meta_data")
  study_data <- prep_get_data_frame("study_data")
  expect_message(
    res1 <-
      acc_end_digits(resp_vars = "CRP_0",
                     study_data = study_data, meta_data = meta_data,
                        label_col = LABEL),
    regexp =
      sprintf(
        "(%s)",
        paste("Due to missing values in CRP_0_x_last",
              "301 observations were deleted.")
      ),
    perl = TRUE,
    all = TRUE
  )
  expect_true(all(c("SummaryData", "SummaryPlot") %in% names(res1)))
  expect_equal(
    nrow(res1$SummaryData),
    1
  )
  expect_equal(
    ncol(res1$SummaryData),
    2
  )
  expect_equal(
    suppressWarnings(sum(as.numeric(
      as.matrix(res1$SummaryData)),
      na.rm = TRUE)), 1
  )
})

test_that("acc_end_digits works image check", {
  skip_on_cran()
  skip_if_not(capabilities()["long.double"])
  skip_if_not_installed("vdiffr")
  meta_data <- prep_get_data_frame("meta_data")
  study_data <- prep_get_data_frame("study_data")
  expect_message(
    res1 <-
      acc_end_digits(resp_vars = "CRP_0",
                     study_data = study_data, meta_data = meta_data,
                     label_col = LABEL),
    regexp =
      sprintf(
        "(%s)",
        paste("Due to missing values in CRP_0_x_last",
              "301 observations were deleted.")
      ),
    perl = TRUE,
    all = TRUE
  )
  skip_on_cran()
  vdiffr::expect_doppelganger("enddigits plot for CRP_0 ok",
                              res1$SummaryPlot)
})

test_that("acc_end_digits is robust", {
  skip_on_cran() # slow, errors obvious
  meta_data <- prep_get_data_frame("meta_data")
  study_data <- prep_get_data_frame("study_data")
  sd0 <- study_data
  sd0$v00014[[2]] <- Inf
  expect_error(
    res1 <-
      acc_end_digits(resp_vars = "CRP_0",
                     study_data = sd0, meta_data = meta_data,
                     label_col = LABEL),
    regexp =
      sprintf(
        "(%s)",
        paste("Values in .+resp_vars.+ must not contain infinite data")
      ),
    perl = TRUE
  )
  sd0 <- study_data
  md0 <- meta_data
  md0[[DECIMALS]] <- NULL
  expect_error(
    res1 <-
      acc_end_digits(resp_vars = "CRP_0",
                     study_data = sd0, meta_data = md0,
                     label_col = LABEL),
    regexp =
      sprintf(
        "(%s)",
        paste("The number of digits following the",
              "decimal point must be prespecified.")
      ),
    perl = TRUE
  )
  sd0 <- study_data
  sd0$v00010[[2]] <- sd0$v00010[[2]] + 0.5
  md0 <- meta_data
  expect_message(
    expect_error(
      res1 <-
        acc_end_digits(resp_vars = "ARM_CUFF_0",
                       study_data = sd0, meta_data = md0,
                       label_col = LABEL),
      regexp =
        sprintf(
          "(%s)",
          paste(".+resp_vars.+ are categorical.*")
        ),
      perl = TRUE
    ),
    regexp =
      paste(
        "Argument .+resp_vars.+: Variable .+ARM_CUFF_0.+ .integer. does not",
        "have matching data type in the study data .float."
      ),
    perl = TRUE
  )

  sd0 <- study_data
  sd0$v00014 <- c(rep(1:9, 3000/9), 1:3)
  sd0$v00014 <- sd0$v00014 / 1000
  sd0$CRP_0_x_last <- sd0$v00014
  sd0$CRP_0_x_last1 <- sd0$v00014
  md0 <- prep_add_to_meta(VAR_NAMES = "CRP_0_x_last",
                          DATA_TYPE = DATA_TYPES$STRING,
                          LABEL = "xxx",
                          VALUE_LABELS = NA,
                          meta_data)
  md0 <- prep_add_to_meta(VAR_NAMES = "CRP_0_x_last1",
                          DATA_TYPE = DATA_TYPES$STRING,
                          LABEL = "yyy",
                          VALUE_LABELS = NA,
                          md0)
  res1 <-
    acc_end_digits(resp_vars = "CRP_0",
                   study_data = sd0, meta_data = md0,
                   label_col = LABEL)

  sd0 <- study_data
  sd0$v00014 <- c(rep(1:9, 3000/9), 1:3)
  sd0$v00014 <- sd0$v00014 / 1000
  sd0$CRP_0_x_last <- sd0$v00014
  sd0$CRP_0_x_last1 <- sd0$v00014
  md0 <- prep_add_to_meta(VAR_NAMES = "CRP_0_x_last",
                          DATA_TYPE = DATA_TYPES$STRING,
                          LABEL = "xxx",
                          VALUE_LABELS = NA,
                          meta_data)
  md0 <- prep_add_to_meta(VAR_NAMES = "CRP_0_x_last1",
                          DATA_TYPE = DATA_TYPES$STRING,
                          LABEL = "yyy",
                          VALUE_LABELS = NA,
                          md0)
  md0$DISTRIBUTION <- NULL
  res1 <-
    acc_end_digits(resp_vars = "CRP_0",
                   study_data = sd0, meta_data = md0,
                   label_col = LABEL)

})