ceph/src/arrow/r/tests/testthat/test-dplyr-funcs-datetime.R

   1 # Licensed to the Apache Software Foundation (ASF) under one
   2 # or more contributor license agreements.  See the NOTICE file
   3 # distributed with this work for additional information
   4 # regarding copyright ownership.  The ASF licenses this file
   5 # to you under the Apache License, Version 2.0 (the
   6 # "License"); you may not use this file except in compliance
   7 # with the License.  You may obtain a copy of the License at
   8 #
   9 #   http://www.apache.org/licenses/LICENSE-2.0
  10 #
  11 # Unless required by applicable law or agreed to in writing,
  12 # software distributed under the License is distributed on an
  13 # "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
  14 # KIND, either express or implied.  See the License for the
  15 # specific language governing permissions and limitations
  16 # under the License.
  17
  18 skip_if_not_available("dataset")
  19
  20 library(lubridate, warn.conflicts = FALSE)
  21 library(dplyr, warn.conflicts = FALSE)
  22
  23 # base::strptime() defaults to local timezone
  24 # but arrow's strptime defaults to UTC.
  25 # So that tests are consistent, set the local timezone to UTC
  26 # TODO: consider reevaluating this workaround after ARROW-12980
  27 withr::local_timezone("UTC")
  28
  29 # TODO: We should test on windows once ARROW-13168 is resolved.
  30 if (tolower(Sys.info()[["sysname"]]) == "windows") {
  31   test_date <- as.POSIXct("2017-01-01 00:00:11.3456789", tz = "")
  32 } else {
  33   test_date <- as.POSIXct("2017-01-01 00:00:11.3456789", tz = "Pacific/Marquesas")
  34 }
  35
  36
  37 test_df <- tibble::tibble(
  38   # test_date + 1 turns the tzone = "" to NULL, which is functionally equivalent
  39   # so we can run some tests on Windows, but this skirts around
  40   # https://issues.apache.org/jira/browse/ARROW-13588
  41   # That issue is tough because in C++, "" is the "no timezone" value
  42   # due to static typing, so we can't distinguish a literal "" from NULL
  43   datetime = c(test_date, NA) + 1,
  44   date = c(as.Date("2021-09-09"), NA)
  45 )
  46
  47 # These tests test component extraction from timestamp objects
  48
  49 test_that("extract year from timestamp", {
  50   compare_dplyr_binding(
  51     .input %>%
  52       mutate(x = year(datetime)) %>%
  53       collect(),
  54     test_df
  55   )
  56 })
  57
  58 test_that("extract isoyear from timestamp", {
  59   compare_dplyr_binding(
  60     .input %>%
  61       mutate(x = isoyear(datetime)) %>%
  62       collect(),
  63     test_df
  64   )
  65 })
  66
  67 test_that("extract quarter from timestamp", {
  68   compare_dplyr_binding(
  69     .input %>%
  70       mutate(x = quarter(datetime)) %>%
  71       collect(),
  72     test_df
  73   )
  74 })
  75
  76 test_that("extract month from timestamp", {
  77   compare_dplyr_binding(
  78     .input %>%
  79       mutate(x = month(datetime)) %>%
  80       collect(),
  81     test_df
  82   )
  83 })
  84
  85 test_that("extract isoweek from timestamp", {
  86   compare_dplyr_binding(
  87     .input %>%
  88       mutate(x = isoweek(datetime)) %>%
  89       collect(),
  90     test_df
  91   )
  92 })
  93
  94 test_that("extract epiweek from timestamp", {
  95   compare_dplyr_binding(
  96     .input %>%
  97       mutate(x = epiweek(datetime)) %>%
  98       collect(),
  99     test_df
 100   )
 101 })
 102
 103 test_that("extract day from timestamp", {
 104   compare_dplyr_binding(
 105     .input %>%
 106       mutate(x = day(datetime)) %>%
 107       collect(),
 108     test_df
 109   )
 110 })
 111
 112 test_that("extract wday from timestamp", {
 113   compare_dplyr_binding(
 114     .input %>%
 115       mutate(x = wday(datetime)) %>%
 116       collect(),
 117     test_df
 118   )
 119
 120   compare_dplyr_binding(
 121     .input %>%
 122       mutate(x = wday(date, week_start = 3)) %>%
 123       collect(),
 124     test_df
 125   )
 126
 127   compare_dplyr_binding(
 128     .input %>%
 129       mutate(x = wday(date, week_start = 1)) %>%
 130       collect(),
 131     test_df
 132   )
 133
 134   skip_on_os("windows") # https://issues.apache.org/jira/browse/ARROW-13168
 135
 136   compare_dplyr_binding(
 137     .input %>%
 138       mutate(x = wday(date, label = TRUE)) %>%
 139       mutate(x = as.character(x)) %>%
 140       collect(),
 141     test_df
 142   )
 143
 144   compare_dplyr_binding(
 145     .input %>%
 146       mutate(x = wday(datetime, label = TRUE, abbr = TRUE)) %>%
 147       mutate(x = as.character(x)) %>%
 148       collect(),
 149     test_df
 150   )
 151 })
 152
 153 test_that("extract yday from timestamp", {
 154   compare_dplyr_binding(
 155     .input %>%
 156       mutate(x = yday(datetime)) %>%
 157       collect(),
 158     test_df
 159   )
 160 })
 161
 162 test_that("extract hour from timestamp", {
 163   compare_dplyr_binding(
 164     .input %>%
 165       mutate(x = hour(datetime)) %>%
 166       collect(),
 167     test_df
 168   )
 169 })
 170
 171 test_that("extract minute from timestamp", {
 172   compare_dplyr_binding(
 173     .input %>%
 174       mutate(x = minute(datetime)) %>%
 175       collect(),
 176     test_df
 177   )
 178 })
 179
 180 test_that("extract second from timestamp", {
 181   compare_dplyr_binding(
 182     .input %>%
 183       mutate(x = second(datetime)) %>%
 184       collect(),
 185     test_df,
 186     # arrow supports nanosecond resolution but lubridate does not
 187     tolerance = 1e-6
 188   )
 189 })
 190
 191 # These tests test extraction of components from date32 objects
 192
 193 test_that("extract year from date", {
 194   compare_dplyr_binding(
 195     .input %>%
 196       mutate(x = year(date)) %>%
 197       collect(),
 198     test_df
 199   )
 200 })
 201
 202 test_that("extract isoyear from date", {
 203   compare_dplyr_binding(
 204     .input %>%
 205       mutate(x = isoyear(date)) %>%
 206       collect(),
 207     test_df
 208   )
 209 })
 210
 211 test_that("extract quarter from date", {
 212   compare_dplyr_binding(
 213     .input %>%
 214       mutate(x = quarter(date)) %>%
 215       collect(),
 216     test_df
 217   )
 218 })
 219
 220 test_that("extract month from date", {
 221   compare_dplyr_binding(
 222     .input %>%
 223       mutate(x = month(date)) %>%
 224       collect(),
 225     test_df
 226   )
 227 })
 228
 229 test_that("extract isoweek from date", {
 230   compare_dplyr_binding(
 231     .input %>%
 232       mutate(x = isoweek(date)) %>%
 233       collect(),
 234     test_df
 235   )
 236 })
 237
 238 test_that("extract epiweek from date", {
 239   compare_dplyr_binding(
 240     .input %>%
 241       mutate(x = epiweek(date)) %>%
 242       collect(),
 243     test_df
 244   )
 245 })
 246
 247 test_that("extract day from date", {
 248   compare_dplyr_binding(
 249     .input %>%
 250       mutate(x = day(date)) %>%
 251       collect(),
 252     test_df
 253   )
 254 })
 255
 256 test_that("extract wday from date", {
 257   compare_dplyr_binding(
 258     .input %>%
 259       mutate(x = wday(date)) %>%
 260       collect(),
 261     test_df
 262   )
 263
 264   compare_dplyr_binding(
 265     .input %>%
 266       mutate(x = wday(date, week_start = 3)) %>%
 267       collect(),
 268     test_df
 269   )
 270
 271   compare_dplyr_binding(
 272     .input %>%
 273       mutate(x = wday(date, week_start = 1)) %>%
 274       collect(),
 275     test_df
 276   )
 277
 278   skip_on_os("windows") # https://issues.apache.org/jira/browse/ARROW-13168
 279
 280   compare_dplyr_binding(
 281     .input %>%
 282       mutate(x = wday(date, label = TRUE, abbr = TRUE)) %>%
 283       mutate(x = as.character(x)) %>%
 284       collect(),
 285     test_df
 286   )
 287
 288   compare_dplyr_binding(
 289     .input %>%
 290       mutate(x = wday(date, label = TRUE)) %>%
 291       mutate(x = as.character(x)) %>%
 292       collect(),
 293     test_df
 294   )
 295 })
 296
 297 test_that("extract yday from date", {
 298   compare_dplyr_binding(
 299     .input %>%
 300       mutate(x = yday(date)) %>%
 301       collect(),
 302     test_df
 303   )
 304 })