tests/testthat/test-fedregs.R

In fedregs: Text Analysis of the US Code of Federal Regulations

context("test-fedregs.R")

# test_that("We get the best CFR URLs.", {
#
#   good_year <- 2000
#   good_title_number <- 15
#
#   bad_year <- 1993
#   bad_title_number <- 55
#
#   na_year <- 1996
#   na_title_number <- 1
#
#   testthat::expect_error(cfr_urls(year = bad_year,
#                           title_number = good_title_number,
#                             check_url = TRUE), "Year must be between 1996 and 2018.\n")
#
#   testthat::expect_error(cfr_urls(year = good_year,
#                           title_number = bad_title_number,
#                           check_url = TRUE), "Title must be a numeric value between 1 and 50.\n")
#
#   testthat::expect_message(cfr_urls(year = na_year,
#                             title_number = na_title_number,
#                             verbose = TRUE,
#                             check_url = FALSE), sprintf("There aren't any regulations for title %s in %s.",
#                                                         na_title_number,
#                                                         na_year))
#
#   testthat::expect_true(is.na(cfr_urls(year = na_year,
#                          title_number = na_title_number,
#                          verbose = FALSE,
#                          check_url = FALSE)))
#
#     cfr <- cfr_urls(year = good_year,
#                     title_number = good_title_number,
#                     check_url = FALSE)
#
#     testthat::expect_true(all(grepl("*..xml$", cfr)))
#     testthat::expect_true(is.character(cfr))
#
# })
#
# test_that("We can parse some parts.", {
#
#   good_year <- 2000
#   good_title_number <- 15
#
#   bad_year <- 1993
#   bad_title_number <- 55
#
#   na_year <- 1996
#   na_title_number <- 1
#
#   na_url <- cfr_urls(year = na_year, title_number = na_title_number)
#   good_url <- cfr_urls(year = good_year, title_number = good_title_number)[1]
#   bad_url <- sprintf("https://www.gpo.gov/fdsys/bulkdata/CFR/%s/title-%s/CFR-%s-title%s-vol1.xml",
#                      bad_year,
#                      bad_title_number,
#                      bad_year,
#                      bad_title_number)
#
#   testthat::expect_error(cfr_part(bad_url), "The URL is not valid.")
#
#   testthat::expect_error(cfr_part(na_url), "NA is not a valid url.")
#
#   testthat::expect_message(cfr_part(good_url,
#                           verbose = TRUE),
#                  sprintf("Pulling the chapter, part, and volume information from:\n%s.\n",
#                          good_url))
#
#   good_part <- cfr_part(good_url)
#
#   testthat::expect_true(is.data.frame(good_part))
#   testthat::expect_true(good_url == unique(good_part$url))
#   testthat::expect_true(good_year == unique(good_part$year))
#   testthat::expect_true(good_title_number == unique(good_part$title))
#   testthat::expect_true(all(grepl("*..xml$", good_part$url)))
#
# })

test_that("We can extract some numbers.", {

  bad_part <- "800 to end"
  good_part <- "Part 800 to end"
  silly_part <- "Part no digits"

  testthat::expect_error(numextract(bad_part), "Make sure you are providing a valid 'part'.")
  testthat::expect_error(numextract(silly_part), "Make sure string is a numeric value.")

  testthat::expect_true(numextract("Part 100 to 200", return = "max") == 200)
  testthat::expect_true(numextract("Part 100 to 200", return = "min") == 100)
  testthat::expect_true(numextract("Part 100 to end", return = "max") == Inf)

  testthat::expect_true(is.numeric(numextract("Part 100 to 200", return = "max")))
  testthat::expect_true(is.numeric(numextract("Part 100 to 200", return = "min")))
  testthat::expect_true(is.numeric(numextract("Part 100 to end", return = "max")))

})


test_that("We can go all the way", {

  good_year <- 2012
  bad_year <- 1995
  na_year <- 1996
  good_title_number <- 50
  bad_title_number <- 1000
  na_title_number <- 1
  good_chapter <- 6
  bad_chapter <- "BB"
  good_part <- 648
  bad_part <- "DD"

  testthat::expect_error(cfr_text(bad_year,
                                  good_title_number,
                                  good_chapter,
                                  good_part),
                         "Year must be between 1996 and 2018.\n")

  testthat::expect_error(cfr_text(good_year,
                                  bad_title_number,
                                  good_chapter,
                                  good_part),
                         "Title must be a numeric value between 1 and 50.\n")

  testthat::expect_error(cfr_text(good_year,
                                  good_title_number,
                                  bad_chapter,
                                  good_part),
                         "Chapter must be a numeric value, not a Roman Numeral.\n")

  testthat::expect_error(cfr_text(good_year,
                                  good_title_number,
                                  good_chapter,
                                  bad_part),
                         "Part must be a numeric value.\n")

  testthat::expect_error(cfr_text(na_year,
                                  na_title_number,
                                  good_chapter,
                                  good_part),
                         sprintf("There aren't any regulations for title %s in %s.\n", na_title_number, na_year))


  good_text <- cfr_text(good_year, good_title_number, good_chapter, good_part)


  testthat::expect_true(all(class(good_text) %in% c("grouped_df", "tbl", "tbl_df", "data.frame") == TRUE))

})