Skip to content

Commit

Permalink
Delete some duplicated tests in test-dataset.R
Browse files Browse the repository at this point in the history
  • Loading branch information
@nealrichardson
nealrichardson committed May 10, 2021
1 parent 1f93ca1 commit 34dc1e6
Show file tree
Hide file tree
Showing 3 changed files with 0 additions and 310 deletions.
308 changes: 0 additions & 308 deletions r/tests/testthat/test-dataset.R
View file
Original file line number Diff line number Diff line change
Expand Up @@ -613,22 +613,6 @@ test_that("Creating UnionDataset", {
expect_error(c(ds1, 42), "character")
})

test_that("InMemoryDataset", {
ds <- InMemoryDataset$create(rbind(df1, df2))
expect_r6_class(ds, "InMemoryDataset")
expect_equivalent(
ds %>%
select(chr, dbl) %>%
filter(dbl > 7 & dbl < 53L) %>%
collect() %>%
arrange(dbl),
rbind(
df1[8:10, c("chr", "dbl")],
df2[1:2, c("chr", "dbl")]
)
)
})

test_that("map_batches", {
skip_if_not_available("parquet")
ds <- open_dataset(dataset_dir, partitioning = "part")
Expand All @@ -647,18 +631,6 @@ test_that("partitioning = NULL to ignore partition information (but why?)", {
expect_identical(names(ds), names(df1)) # i.e. not c(names(df1), "group", "other")
})

test_that("filter() with is.na()", {
skip_if_not_available("parquet")
ds <- open_dataset(dataset_dir, partitioning = schema(part = uint8()))
expect_equivalent(
ds %>%
select(part, lgl) %>%
filter(!is.na(lgl), part == 1) %>%
collect(),
tibble(part = 1L, lgl = df1$lgl[!is.na(df1$lgl)])
)
})

test_that("filter() with is.nan()", {
skip_if_not_available("parquet")
ds <- open_dataset(dataset_dir, partitioning = schema(part = uint8()))
Expand Down Expand Up @@ -693,103 +665,6 @@ test_that("filter() with %in%", {
)
})

test_that("filter() with negative scalar", {
skip_if_not_available("parquet")
ds <- open_dataset(dataset_dir, partitioning = schema(part = uint8()))
expect_equivalent(
ds %>%
filter(part == 1) %>%
select(chr, int) %>%
filter(int > -2) %>%
collect(),
df1[, c("chr", "int")]
)

expect_equivalent(
ds %>%
filter(part == 1) %>%
select(chr, int) %>%
filter(int %in% -2) %>%
collect(),
df1[FALSE, c("chr", "int")]
)

expect_equivalent(
ds %>%
filter(part == 1) %>%
select(chr, int) %>%
filter(-int < -2) %>%
collect(),
df1[df1$int > 2, c("chr", "int")]
)
})

test_that("filter() with strings", {
skip_if_not_available("parquet")
ds <- open_dataset(dataset_dir, partitioning = schema(part = uint8()))
expect_equivalent(
ds %>%
select(chr, part) %>%
filter(chr == "b", part == 1) %>%
collect(),
tibble(chr = "b", part = 1)
)

skip_if_not_available("utf8proc")
expect_equivalent(
ds %>%
select(chr, part) %>%
filter(toupper(chr) == "B", part == 1) %>%
collect(),
tibble(chr = "b", part = 1)
)
})

test_that("filter() with arrow compute functions by name", {
skip_if_not_available("parquet")
ds <- open_dataset(dataset_dir, partitioning = schema(part = uint8()))
expect_equivalent(
ds %>%
select(part, lgl) %>%
filter(arrow_is_valid(lgl), arrow_equal(part, 1)) %>%
collect(),
ds %>%
select(part, lgl) %>%
filter(!is.na(lgl), part == 1L) %>%
collect()
)
})

test_that("filter() with .data", {
skip_if_not_available("parquet")
ds <- open_dataset(dataset_dir, partitioning = schema(part = uint8()))
expect_equivalent(
ds %>%
select(.data$int, .data$part) %>%
filter(.data$int == 3, .data$part == 1) %>%
collect(),
tibble(int = df1$int[3], part = 1)
)

expect_equivalent(
ds %>%
select(.data$int, .data$part) %>%
filter(.data$int %in% c(6, 4, 3, 103, 107), .data$part == 1) %>%
collect(),
tibble(int = df1$int[c(3, 4, 6)], part = 1)
)

# and the .env pronoun too!
chr <- 1
expect_equivalent(
ds %>%
select(.data$int, .data$part) %>%
filter(.data$int %in% c(6, 4, 3, 103, 107), .data$part == .env$chr) %>%
collect(),
tibble(int = df1$int[c(3, 4, 6)], part = 1)
)
})

test_that("filter() on timestamp columns", {
skip_if_not_available("parquet")
ds <- open_dataset(dataset_dir, partitioning = schema(part = uint8()))
Expand Down Expand Up @@ -849,109 +724,6 @@ test_that("filter() on date32 columns", {
)
})

test_that("filter() with expressions", {
skip_if_not_available("parquet")
ds <- open_dataset(dataset_dir, partitioning = schema(part = uint8()))
expect_r6_class(ds$format, "ParquetFileFormat")
expect_r6_class(ds$filesystem, "LocalFileSystem")
expect_r6_class(ds, "Dataset")
expect_equivalent(
ds %>%
select(chr, dbl) %>%
filter(dbl * 2 > 14 & dbl - 50 < 3L) %>%
collect() %>%
arrange(dbl),
rbind(
df1[8:10, c("chr", "dbl")],
df2[1:2, c("chr", "dbl")]
)
)

# check division's special casing.
expect_equivalent(
ds %>%
select(chr, dbl) %>%
filter(dbl / 2 > 3.5 & dbl < 53) %>%
collect() %>%
arrange(dbl),
rbind(
df1[8:10, c("chr", "dbl")],
df2[1:2, c("chr", "dbl")]
)
)

expect_equivalent(
ds %>%
select(chr, dbl, int) %>%
filter(int %/% 2L > 3 & dbl < 53) %>%
collect() %>%
arrange(dbl),
rbind(
df1[8:10, c("chr", "dbl", "int")],
df2[1:2, c("chr", "dbl", "int")]
)
)

expect_equivalent(
ds %>%
select(chr, dbl, int) %>%
filter(int %/% 2 > 3 & dbl < 53) %>%
collect() %>%
arrange(dbl),
rbind(
df1[8:10, c("chr", "dbl", "int")],
df2[1:2, c("chr", "dbl", "int")]
)
)

expect_equivalent(
ds %>%
select(chr, dbl, int) %>%
filter(int %% 2L > 0 & dbl < 53) %>%
collect() %>%
arrange(dbl),
rbind(
df1[c(1, 3, 5, 7, 9), c("chr", "dbl", "int")],
df2[1, c("chr", "dbl", "int")]
)
)

expect_equivalent(
ds %>%
select(chr, dbl, int) %>%
filter(int %% 2L > 0 & dbl < 53) %>%
collect() %>%
arrange(dbl),
rbind(
df1[c(1, 3, 5, 7, 9), c("chr", "dbl", "int")],
df2[1, c("chr", "dbl", "int")]
)
)

expect_equivalent(
ds %>%
select(chr, dbl, int) %>%
filter(int %% 2 > 0 & dbl < 53) %>%
collect() %>%
arrange(dbl),
rbind(
df1[c(1, 3, 5, 7, 9), c("chr", "dbl", "int")],
df2[1, c("chr", "dbl", "int")]
)
)

expect_equivalent(
ds %>%
select(chr, dbl, int) %>%
filter(dbl + int > 15 & dbl < 53L) %>%
collect() %>%
arrange(dbl),
rbind(
df1[8:10, c("chr", "dbl", "int")],
df2[1:2, c("chr", "dbl", "int")]
)
)
})

test_that("mutate()", {
ds <- open_dataset(dataset_dir, partitioning = schema(part = uint8()))
Expand Down Expand Up @@ -985,26 +757,6 @@ See $.data for the source Arrow object",
)
})

test_that("transmute()", {
ds <- open_dataset(dataset_dir, partitioning = schema(part = uint8()))
mutated <-
expect_equivalent(
ds %>%
select(chr, dbl, int) %>%
filter(dbl * 2 > 14 & dbl - 50 < 3L) %>%
transmute(twice = int * 2) %>%
collect() %>%
arrange(twice),
rbind(
df1[8:10, "int", drop = FALSE],
df2[1:2, "int", drop = FALSE]
) %>%
transmute(
twice = int * 2
)
)
})

test_that("mutate() features not yet implemented", {
expect_error(
ds %>%
Expand All @@ -1015,66 +767,6 @@ test_that("mutate() features not yet implemented", {
)
})


test_that("mutate() with scalar (length 1) literal inputs", {
expect_equal(
ds %>%
mutate(the_answer = 42) %>%
collect() %>%
pull(the_answer),
rep(42, nrow(ds))
)

expect_error(
ds %>% mutate(the_answer = c(42, 42)),
"In the_answer = c(42, 42), only values of size one are recycled\nCall collect() first to pull data into R.",
fixed = TRUE
)
})

test_that("mutate() with NULL inputs", {
expect_equal(
ds %>%
mutate(int = NULL) %>%
collect(),
ds %>%
select(-int) %>%
collect()
)
})

test_that("empty mutate()", {
expect_equal(
ds %>%
mutate() %>%
collect(),
ds %>%
collect()
)
})

test_that("transmute() with NULL inputs", {
expect_equal(
ds %>%
transmute(int = NULL) %>%
collect(),
ds %>%
select() %>%
collect()
)
})

test_that("empty transmute()", {
expect_equal(
ds %>%
transmute() %>%
collect(),
ds %>%
select() %>%
collect()
)
})

test_that("filter scalar validation doesn't crash (ARROW-7772)", {
expect_error(
ds %>%
Expand Down
1 change: 0 additions & 1 deletion r/tests/testthat/test-dplyr-filter.R
View file
Original file line number Diff line number Diff line change
Expand Up @@ -193,7 +193,6 @@ test_that("Negative scalar values", {
)
})


test_that("filter() with between()", {
expect_dplyr_equal(
input %>%
Expand Down
1 change: 0 additions & 1 deletion r/tests/testthat/test-dplyr-mutate.R
View file
Original file line number Diff line number Diff line change
Expand Up @@ -173,7 +173,6 @@ test_that("mutate with reassigning same name", {
})

test_that("mutate with single value for recycling", {
skip("Not implemented (ARROW-11705")
expect_dplyr_equal(
input %>%
select(int, padded_strings) %>%
Expand Down

0 comments on commit 34dc1e6

Please sign in to comment.