tests/testthat/test-metadata.R

# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.

test_that("Schema metadata", {
  s <- schema(b = double())
  expect_equal(s$metadata, empty_named_list())
  expect_false(s$HasMetadata)
  s$metadata <- list(test = TRUE)
  expect_identical(s$metadata, list(test = "TRUE"))
  expect_true(s$HasMetadata)
  s$metadata$foo <- 42
  expect_identical(s$metadata, list(test = "TRUE", foo = "42"))
  expect_true(s$HasMetadata)
  s$metadata$foo <- NULL
  expect_identical(s$metadata, list(test = "TRUE"))
  expect_true(s$HasMetadata)
  s$metadata <- NULL
  expect_equal(s$metadata, empty_named_list())
  expect_false(s$HasMetadata)
  expect_error(
    s$metadata <- 4,
    "Key-value metadata must be a named list or character vector"
  )
})

test_that("Table metadata", {
  tab <- Table$create(x = 1:2, y = c("a", "b"))
  expect_equal(tab$metadata, empty_named_list())
  tab$metadata <- list(test = TRUE)
  expect_identical(tab$metadata, list(test = "TRUE"))
  tab$metadata$foo <- 42
  expect_identical(tab$metadata, list(test = "TRUE", foo = "42"))
  tab$metadata$foo <- NULL
  expect_identical(tab$metadata, list(test = "TRUE"))
  tab$metadata <- NULL
  expect_equal(tab$metadata, empty_named_list())
})

test_that("Table R metadata", {
  tab <- Table$create(example_with_metadata)
  expect_output(
    print(tab$metadata),
    "$r$columns$c$columns$c1$attributes$extra_attr",
    fixed = TRUE
  )
  expect_equal_data_frame(tab, example_with_metadata)
})

test_that("R metadata is not stored for types that map to Arrow types (factor, Date, etc.)", {
  tab <- Table$create(example_data[1:6])
  expect_null(tab$metadata$r)

  expect_null(Table$create(example_with_times[1:3])$metadata$r)
})

test_that("R metadata is not stored for ExtensionType columns", {
  tab <- Table$create(
    x = vctrs::new_vctr(1:5, class = "special_integer")
  )
  expect_null(tab$metadata$r)
})

test_that("classes are not stored for arrow_binary/arrow_large_binary/arrow_fixed_size_binary (ARROW-14140)", {
  raws <- charToRaw("bonjour")

  binary <- Array$create(list(raws), binary())
  large_binary <- Array$create(list(raws), large_binary())
  fixed_size_binary <- Array$create(list(raws), fixed_size_binary(7L))

  expect_null(RecordBatch$create(b = binary)$metadata$r)
  expect_null(RecordBatch$create(b = large_binary)$metadata$r)
  expect_null(RecordBatch$create(b = fixed_size_binary)$metadata$r)

  expect_null(Table$create(b = binary)$metadata$r)
  expect_null(Table$create(b = large_binary)$metadata$r)
  expect_null(Table$create(b = fixed_size_binary)$metadata$r)
})

test_that("Garbage R metadata doesn't break things", {
  tab <- Table$create(example_data[1:6])
  tab$metadata$r <- "garbage"
  expect_warning(
    as.data.frame(tab),
    "Invalid metadata$r",
    fixed = TRUE
  )
  # serialize data like .serialize_arrow_r_metadata does, but don't call that
  # directly since it checks to ensure that the data is a list
  tab <- Table$create(example_data[1:6])
  tab$metadata$r <- rawToChar(serialize("garbage", NULL, ascii = TRUE))
  expect_warning(
    as.data.frame(tab),
    "Invalid metadata$r",
    fixed = TRUE
  )
})

test_that("Metadata serialization compression", {
  # attributes that (when serialized) are just under 100kb are not compressed,
  # and simply serialized
  strings <- as.list(rep(make_string_of_size(1), 98))
  small <- .serialize_arrow_r_metadata(strings)
  expect_equal(
    object.size(small),
    object.size(rawToChar(serialize(strings, NULL, ascii = TRUE)))
  )

  # Large strings will be compressed
  large_strings <- as.list(rep(make_string_of_size(1), 100))
  large <- .serialize_arrow_r_metadata(large_strings)
  expect_lt(
    object.size(large),
    object.size(rawToChar(serialize(large_strings, NULL, ascii = TRUE)))
  )
  # and this compression ends up being smaller than even the "small" strings
  expect_lt(object.size(large), object.size(small))

  # However strings where compression + serialization is not effective are no
  # worse than only serialization alone
  large_few_strings <- as.list(rep(make_random_string_of_size(50), 2))
  large_few <- .serialize_arrow_r_metadata(large_few_strings)
  expect_equal(
    object.size(large_few),
    object.size(rawToChar(serialize(large_few_strings, NULL, ascii = TRUE)))
  )

  # But we can disable compression
  op <- options(arrow.compress_metadata = FALSE)
  on.exit(options(op))

  large_strings <- as.list(rep(make_string_of_size(1), 100))
  large <- .serialize_arrow_r_metadata(large_strings)
  expect_equal(
    object.size(large),
    object.size(rawToChar(serialize(large_strings, NULL, ascii = TRUE)))
  )
})

test_that("RecordBatch metadata", {
  rb <- RecordBatch$create(x = 1:2, y = c("a", "b"))
  expect_equal(rb$metadata, empty_named_list())
  rb$metadata <- list(test = TRUE)
  expect_identical(rb$metadata, list(test = "TRUE"))
  rb$metadata$foo <- 42
  expect_identical(rb$metadata, list(test = "TRUE", foo = "42"))
  rb$metadata$foo <- NULL
  expect_identical(rb$metadata, list(test = "TRUE"))
  rb$metadata <- NULL
  expect_equal(rb$metadata, empty_named_list())
})

test_that("RecordBatch R metadata", {
  expect_equal_data_frame(record_batch(example_with_metadata), example_with_metadata)
})

test_that("R metadata roundtrip via parquet", {
  skip_if_not_available("parquet")
  tf <- tempfile()
  on.exit(unlink(tf))

  write_parquet(example_with_metadata, tf)
  expect_identical(read_parquet(tf), example_with_metadata)
})

test_that("R metadata roundtrip via feather", {
  tf <- tempfile()
  on.exit(unlink(tf))

  write_feather(example_with_metadata, tf)
  expect_identical(read_feather(tf), example_with_metadata)
})

test_that("haven types roundtrip via feather", {
  tf <- tempfile()
  on.exit(unlink(tf))

  write_feather(haven_data, tf)
  expect_identical(read_feather(tf), haven_data)
})

test_that("Date/time type roundtrip", {
  rb <- record_batch(example_with_times)
  expect_r6_class(rb$schema$posixlt$type, "VctrsExtensionType")
  expect_equal_data_frame(rb, example_with_times)
})

test_that("metadata keeps attribute of top level data frame", {
  df <- structure(data.frame(x = 1, y = 2), foo = "bar")
  tab <- Table$create(df)
  expect_identical(attr(as.data.frame(tab), "foo"), "bar")
  expect_equal_data_frame(tab, df)
})


test_that("metadata drops readr's problems attribute", {
  readr_like <- tibble::tibble(
    dbl = 1.1,
    not_here = NA_character_
  )
  attributes(readr_like) <- append(
    attributes(readr_like),
    list(problems = tibble::tibble(
      row = 1L,
      col = NA_character_,
      expected = "2 columns",
      actual = "1 columns",
      file = "'test'"
    ))
  )

  tab <- Table$create(readr_like)
  expect_null(attr(as.data.frame(tab), "problems"))
})

test_that("Row-level metadata (does not by default) roundtrip", {
  # First tracked at ARROW-10386, though it was later determined that row-level
  # metadata should be handled separately ARROW-14020, ARROW-12542
  df <- data.frame(x = I(list(structure(1, foo = "bar"), structure(2, baz = "qux"))))
  tab <- Table$create(df)
  r_metadata <- tab$r_metadata
  expect_type(r_metadata, "list")
  expect_null(r_metadata$columns$x$columns)

  # But we can re-enable this / read data that has already been written with
  # row-level metadata
  withr::with_options(
    list("arrow.preserve_row_level_metadata" = TRUE),
    {
      tab <- Table$create(df)
      expect_identical(attr(as.data.frame(tab)$x[[1]], "foo"), "bar")
      expect_identical(attr(as.data.frame(tab)$x[[2]], "baz"), "qux")
    }
  )
})


test_that("Row-level metadata (does not) roundtrip in datasets", {
  # First tracked at ARROW-10386, though it was later determined that row-level
  # metadata should be handled separately ARROW-14020, ARROW-12542
  skip_if_not_available("dataset")
  skip_if_not_available("parquet")

  df <- tibble::tibble(
    metadata = list(
      structure(1, my_value_as_attr = 1),
      structure(2, my_value_as_attr = 2),
      structure(3, my_value_as_attr = 3),
      structure(4, my_value_as_attr = 3)
    ),
    int = 1L:4L,
    part = c(1, 3, 2, 1)
  )

  dst_dir <- make_temp_dir()

  withr::local_options("arrow.preserve_row_level_metadata" = TRUE)

  expect_warning(
    write_dataset(df, dst_dir, partitioning = "part"),
    "Row-level metadata is not compatible with datasets and will be discarded"
  )

  # Reset directory as previous write will have created some files and the default
  # behavior is to error on existing
  dst_dir <- make_temp_dir()
  # but we need to write a dataset with row-level metadata to make sure when
  # reading ones that have been written with them we warn appropriately
  fake_func_name <- write_dataset
  fake_func_name(df, dst_dir, partitioning = "part")

  ds <- open_dataset(dst_dir)
  expect_warning(
    df_from_ds <- collect(ds),
    "Row-level metadata is not compatible with this operation and has been ignored"
  )
  expect_equal(
    dplyr::arrange(df_from_ds, int),
    dplyr::arrange(df, int),
    ignore_attr = TRUE
  )

  # however there is *no* warning if we don't select the metadata column
  expect_warning(
    df_from_ds <- ds %>% dplyr::select(int) %>% dplyr::collect(),
    NA
  )
})

test_that("Dataset writing does handle other metadata", {
  skip_if_not_available("dataset")
  skip_if_not_available("parquet")

  dst_dir <- make_temp_dir()
  tab <- Table$create(example_with_metadata)
  # Tack on extra non-R metadata: sfarrow 0.4.1 relies on this
  tab$metadata[["other_stuff"]] <- "hello"

  write_dataset(tab, dst_dir, partitioning = "b")

  ds <- open_dataset(dst_dir)
  expect_equal(
    ds %>%
      # partitioning on b puts it last, so move it back
      select(a, b, c, d) %>%
      collect(),
    example_with_metadata
  )

  # Check for that extra metadata in the schema:
  expect_equal(ds$metadata$other_stuff, "hello")
})

test_that("dplyr with metadata", {
  skip_if_not_available("dataset")

  compare_dplyr_binding(
    .input %>%
      collect(),
    example_with_metadata
  )
  compare_dplyr_binding(
    .input %>%
      select(a) %>%
      collect(),
    example_with_metadata
  )
  compare_dplyr_binding(
    .input %>%
      mutate(z = b * 4) %>%
      select(z, a) %>%
      collect(),
    example_with_metadata
  )
  compare_dplyr_binding(
    .input %>%
      mutate(z = nchar(d)) %>%
      select(z, a) %>%
      collect(),
    example_with_metadata
  )
  # dplyr drops top-level attributes if you do summarize, though attributes
  # of grouping columns appear to come through
  compare_dplyr_binding(
    .input %>%
      group_by(d) %>%
      summarize(n()) %>%
      collect(),
    example_with_metadata
  )
  # Same name in output but different data, so the column metadata shouldn't
  # carry through
  compare_dplyr_binding(
    .input %>%
      mutate(a = b) %>%
      select(a) %>%
      collect(),
    example_with_metadata
  )
})

test_that("grouped_df metadata is recorded (efficiently)", {
  grouped <- group_by(tibble(a = 1:2, b = 3:4), a)
  expect_s3_class(grouped, "grouped_df")
  grouped_tab <- Table$create(grouped)
  expect_r6_class(grouped_tab, "Table")
  expect_equal(grouped_tab$metadata$r$attributes$.group_vars, "a")
})

test_that("grouped_df non-arrow metadata is preserved", {
  simple_tbl <- tibble(a = 1:2, b = 3:4)
  attr(simple_tbl, "other_metadata") <- "look I'm still here!"
  grouped <- group_by(simple_tbl, a)
  grouped_tab <- arrow_table(grouped)

  expect_equal(
    attributes(collect.ArrowTabular(grouped_tab))$other_metadata,
    "look I'm still here!"
  )
})

test_that("data.frame class attribute is not saved", {
  df <- data.frame(x = 1:5)
  df_arrow <- arrow_table(df)
  expect_null(df_arrow$r_metadata$attributes)

  df <- data.frame(x = 1:5)
  attributes(df)$foo <- "bar"
  df_arrow <- arrow_table(df)
  expect_identical(df_arrow$r_metadata, list(attributes = list(foo = "bar"), columns = list(x = NULL)))
})

Try the arrow package in your browser

Any scripts or data that you put into this service are public.

arrow documentation built on Nov. 25, 2023, 1:09 a.m.