Make it easier to create a DataFrame with a single list-column (#796)

Co-authored-by: eitsupi <[email protected]>

NEWS.md

@@ -40,6 +40,11 @@
         is changed to `$versions$r_package` (#791).
     -   `$rust_polars`, which indicates the version of the dependent Rust Polars,
         is changed to `$versions$rust_crate` (#791).
+-   New behavior when creating a `DataFrame` with a single list-variable.
+    `pl$DataFrame(x = list(1:2, 3:4))` used to create a `DataFrame` with two
+    columns named "new_column" and "new_column_1", which was unexpected. It now
+    produces a `DataFrame` with a single `list` variable. This also applies to
+    list-column created in `$with_columns()` and `$select()` (#794).
 
 ### Deprecations
 

R/dotdotdot.R

@@ -45,8 +45,9 @@ unpack_list = function(..., .context = NULL, .call = sys.call(1L), skip_classes
   l = list2(..., .context = .context, .call = .call)
   if (
     length(l) == 1L &&
-      is.list(l[[1L]]) &&
-      !(!is.null(skip_classes) && inherits(l[[1L]], skip_classes))
+    is.list(l[[1L]]) &&
+    !(!is.null(skip_classes) && inherits(l[[1L]], skip_classes)) &&
+    is.null(names(l))
   ) {
     l[[1L]]
   } else {

tests/testthat/test-dataframe.R

@@ -225,6 +225,22 @@ test_that("select with list of exprs", {
   expect_equal(x6$columns, c("mpg", "hp"))
 })
 
+test_that("select: create a list variable", {
+  test = pl$DataFrame(x = 1:2)
+
+  # create one column
+  expect_identical(
+    test$select(y = list(1:2, 3:4))$to_list(),
+    list(y = list(1:2, 3:4))
+  )
+
+  # create several column
+  expect_identical(
+    test$select(y = list(1:2, 3:4), z = list(c("a", "b"), c("c", "d")))$to_list(),
+    list(y = list(1:2, 3:4), z = list(c("a", "b"), c("c", "d")))
+  )
+})
+
 test_that("map_batches unity", {
   x = pl$
     DataFrame(iris)$
@@ -375,6 +391,37 @@ test_that("get column", {
 
 # TODO implement series cast and test Series_equal
 
+test_that("with_columns: list or unlisted input", {
+  test = pl$DataFrame(mtcars)
+
+  # one element in $with_columns()
+  expect_identical(
+    test$with_columns(list(a = mtcars$drat))$to_data_frame(),
+    test$with_columns(a = mtcars$drat)$to_data_frame()
+  )
+
+  # several elements
+  expect_identical(
+    test$with_columns(list(a = 1, b = mtcars$drat))$to_data_frame(),
+    test$with_columns(a = 1, b = mtcars$drat)$to_data_frame()
+  )
+})
+
+test_that("with_columns: create a list variable", {
+  test = pl$DataFrame(x = 1:2)
+
+  # create one column
+  expect_identical(
+    test$with_columns(y = list(1:2, 3:4))$to_list(),
+    list(x = 1:2, y = list(1:2, 3:4))
+  )
+
+  # create several column
+  expect_identical(
+    test$with_columns(y = list(1:2, 3:4), z = list(c("a", "b"), c("c", "d")))$to_list(),
+    list(x = 1:2, y = list(1:2, 3:4), z = list(c("a", "b"), c("c", "d")))
+  )
+})
 
 test_that("with_columns lazy/eager", {
   l = list(

tests/testthat/test-expr_list.R

@@ -44,7 +44,7 @@ test_that("list$sum max min mean", {
     c(-Inf)
   )
 
-  df = pl$DataFrame(list(x = ints))
+  df = pl$DataFrame(x = ints)
   p_res = df$select(
     pl$col("x")$list$sum()$alias("sum"),
     pl$col("x")$list$max()$alias("max"),
@@ -63,7 +63,7 @@ test_that("list$sum max min mean", {
     r_res
   )
 
-  df = pl$DataFrame(list(x = floats))
+  df = pl$DataFrame(x = floats)
   p_res = df$select(
     pl$col("x")$list$sum()$alias("sum"),
     pl$col("x")$list$max()$alias("max"),
@@ -404,7 +404,7 @@ test_that("concat", {
 
 test_that("to_struct", {
   l = list(integer(), 1:2, 1:3, 1:2)
-  df = pl$DataFrame(list(a = l))
+  df = pl$DataFrame(a = l)
   act_1 = df$select(pl$col("a")$list$to_struct(
     n_field_strategy = "first_non_null",
     fields = \(idx) paste0("hello_you_", idx)
@@ -438,7 +438,7 @@ test_that("to_struct", {
 
 
 test_that("eval", {
-  df = pl$DataFrame(a = list(a = c(1, 8, 3), b = c(4, 5, 2)))
+  df = pl$DataFrame(a = c(1, 8, 3), b = c(4, 5, 2))
   l_act = df$with_columns(
     pl$concat_list(list("a", "b"))$list$eval(pl$element()$rank())$alias("rank")
   )$to_list()
@@ -454,7 +454,7 @@ test_that("eval", {
 
 test_that("$list$all() works", {
   df = pl$DataFrame(
-    list(a = list(c(TRUE, TRUE), c(FALSE, TRUE), c(FALSE, FALSE), NA, c()))
+    a = list(c(TRUE, TRUE), c(FALSE, TRUE), c(FALSE, FALSE), NA, c())
   )
   expect_identical(
     df$select(all = pl$col("a")$list$all())$to_list(),
@@ -464,7 +464,7 @@ test_that("$list$all() works", {
 
 test_that("$list$any() works", {
   df = pl$DataFrame(
-    list(a = list(c(TRUE, TRUE), c(FALSE, TRUE), c(FALSE, FALSE), NA, c()))
+    a = list(c(TRUE, TRUE), c(FALSE, TRUE), c(FALSE, FALSE), NA, c())
   )
   expect_identical(
     df$select(any = pl$col("a")$list$any())$to_list(),