tests/testthat/test-func_repairHtPatches.r

In pratikunterwegs/watlastools: Tools for High Resolution Tracking Data from WATLAS

context("repairing high tide patches")
testthat::test_that("high tide repair works", {

  # read in data
  files_list <- list.files("../testdata/", pattern = "413_", full.names = T)
  data_list <- lapply(files_list[1:10], data.table::fread)

  # assume all patches are real
  data_list <- lapply(data_list, function(df) {
    df <- watlastools::wat_infer_residence(df)
    df <- watlastools::wat_classify_points(
      data = df,
      lim_res_time = 2,
      min_fix_warning = 100
    )
    df <- watlastools::wat_make_res_patch(
      data = df,
      buffer_radius = 10
    )
  })

  repaired_data <- watlastools::wat_repair_ht_patches(
    patch_data_list = data_list,
    lim_spat_indep = 100,
    lim_time_indep = 30,
    buffer_radius = 10
  )

  # do tests
  # test that the sf output class is at least sf
  testthat::expect_s3_class(
    object = repaired_data,
    class = c("data.frame", "data.table")
  )

  # test that names are present in output cols
  expnames <- c(
    "id", "tide_number", "type", "patch", "time_mean",
    "tidaltime_mean", "x_mean", "y_mean", "duration", "distInPatch",
    "distBwPatch", "dispInPatch",
    "patchdata"
  )

  # check that the data columns are not list
  testthat::expect_true(!"list" %in%
    purrr::map_chr(
      repaired_data[, .(setdiff(
        colnames(repaired_data),
        "patchdata"
      ))],
      class
    ))

  # check that expected column names are present
  watlastools:::wat_check_data(
    repaired_data,
    expnames
  )

  # check that accessing polygons works
  testthat::expect_silent(wat_get_patch_summary(repaired_data,
    which_data = "spatial"
  ))


  # check that data are ordered in time
  testthat::expect_gt(min(as.numeric(diff(repaired_data$time_mean)),
    na.rm = TRUE
  ), 0)

  # check that patches across tidal cycles are indpendent
  time_end <- repaired_data$time_end
  time_end <- time_end[2:length(time_end)]
  time_start <- repaired_data$time_start
  time_start <- time_start[seq_len(length(time_start) - 1)]

  temp_indep <- c(NA, as.numeric((time_end - time_start) / 60)) >= 30
  spat_indep <- wat_bw_patch_dist(repaired_data) >= 100
  rest_indep <- TRUE # for patches separated by residence time

  testthat::expect_true(all(purrr::pmap_int(
    list(
      temp_indep, spat_indep,
      rest_indep
    ),
    any
  )[-1]))
})