# ==================================================================== # # TITLE # # cleaner: Fast and Easy Data Cleaning # # # # SOURCE # # https://github.com/msberends/cleaner # # # # LICENCE # # (c) 2022 Berends MS (m.s.berends@umcg.nl) # # # # This R package is free software; you can freely use and distribute # # it for both personal and commercial purposes under the terms of the # # GNU General Public License version 2.0 (GNU GPL-2), as published by # # the Free Software Foundation. # # # # This R package was publicly released in the hope that it will be # # useful, but it comes WITHOUT ANY WARRANTY OR LIABILITY. # # ==================================================================== # context("freq.R") test_that("frequency table works", { expect_equal(nrow(freq(c(1, 1, 2, 2, 3, 3, 4, 4, 5, 5))), 5) nrs <- as.integer(runif(2000, min = 0, max = 20)) expect_output(print(freq(nrs))) expect_output(print(freq(nrs, nmax = 5))) expect_output(print(freq(nrs, nmax = Inf, markdown = FALSE))) expect_output(print(freq(nrs, nmax = Inf))) expect_output(print(freq(nrs, nmax = NA))) expect_output(print(freq(nrs, nmax = NULL))) expect_output(print(freq(nrs, sort.count = FALSE))) expect_output(print(freq(nrs, markdown = TRUE))) expect_output(print(freq(nrs, markdown = TRUE), markdown = FALSE)) expect_output(print(freq(nrs, markdown = TRUE), markdown = TRUE)) expect_output(print(freq(nrs, quote = TRUE))) # character expect_output(print(freq(unclean$gender))) # with weights df <- data.frame(x = sample(letters, size = 100, replace = TRUE), y = round(runif(100) * 100)) expect_gt(sum(freq(df, wt = y)$count), sum(freq(df)$count)) })