tests/testthat/test-get-text.R

In q-w-a/regulationsgov: Accessing Regulations.gov Data

test_that("add text works as expected", {
  skip("Only run if key is available & user has installed textreadr and tesseract")
  comment_metadata <- get_all_comments(
    docketId = "CMS-2014-0063",
    test = TRUE
  )
  text <- add_text(comment_metadata)
  expect_equal(ncol(comment_metadata) + 1, ncol(text))
})

test_that("get_text chooses desired element when all=FALSE", {
  skip("Only run if key is available & user has installed textreadr and tesseract")
  text <- extract_text("https://downloads.regulations.gov/FAA-2018-1084-0130/attachment_1.docx")
  expect_equal(text, "I believe this is a Great rule.")

  # correct function is called
  text <- extract_text("https://downloads.regulations.gov/FDA-2008-N-0115-0014/attachment_1.pdf")
  text_pdf <- textreadr::read_pdf("https://downloads.regulations.gov/FDA-2008-N-0115-0014/attachment_1.pdf",
    ocr = TRUE
  )[, "text"] %>%
    paste0(collapse = " ")

  expect_equal(text, text_pdf)
})

test_that("choose_element chooses the correct element", {
  skip("Only run if key is available & user has installed textreadr and tesseract")
  links <- c(
    "https://downloads.regulations.gov/FAA-2018-1084-0130/attachment_1.docx",
    "https://downloads.regulations.gov/FAA-2018-1084-0130/attachment_1.pdf"
  )
  link <- choose_element(links)
  expect_equal(link, "https://downloads.regulations.gov/FAA-2018-1084-0130/attachment_1.docx")
})


test_that("extract_text works as expected for invalid argument", {
  skip("Only run if key is available & user has installed textreadr and tesseract")
  # returns NA when invalid url provided
  invalid <- extract_text(NA)
  expect_true(is.na(invalid))

  invalid <- extract_text("not a link")
  expect_true(is.na(invalid))
})


test_that("add_text works as expected with all = TRUE", {
  skip("Run only if modifying add_text functions; 38 requests")

  comment_metadata <- get_all_comments(documentId = "CMS-2014-0063-0001")
  sub <- comment_metadata[25:26, ]
  text <- add_text(sub, all = TRUE)
  expect_false(is.null(text["text"]))
})

test_that("add_text works with document data", {
  skip("Only run if key is available & user has installed textreadr and tesseract")
  skip_if_no_key()
  result <- get_all_documents(docketId = "FDA-2009-N-0501")
  text <- add_text(result)
  expect_false(any(is.na(text["text"])))
})