fix: Use primitive constructors to create a Series of lists when dtyp…

…e is defined
pola-rs · Mar 12, 2024 · f0cf33e · f0cf33e
1 parent c5b53ab
commit f0cf33e
Show file tree

Hide file tree

Showing 5 changed files with 25 additions and 6 deletions.
diff --git a/py-polars/polars/_utils/construction.py b/py-polars/polars/_utils/construction.py
@@ -444,7 +444,6 @@ def sequence_to_pyseries(
 
     # lists defer to subsequent handling; identify nested type
     elif dtype == List:
-        getattr(dtype, "inner", None)
         python_dtype = list
 
     # infer temporal type handling
@@ -459,7 +458,7 @@ def sequence_to_pyseries(
             or is_namedtuple(value.__class__)
         ) and dtype != Object:
             return pl.DataFrame(values).to_struct(name)._s
-        elif isinstance(value, range):
+        elif isinstance(value, range) and dtype is None:
             values = [range_to_series("", v) for v in values]
         else:
             # for temporal dtypes:
@@ -602,7 +601,18 @@ def sequence_to_pyseries(
             if isinstance(dtype, Object):
                 return PySeries.new_object(name, values, strict)
             if dtype:
-                srs = sequence_from_any_value_and_dtype_or_object(name, values, dtype)
+                if (inner_dtype := getattr(dtype, "inner", None)) is not None:
+                    py_srs = [
+                        None
+                        if value is None
+                        else sequence_to_pyseries("", value, inner_dtype, strict=strict)
+                        for value in values
+                    ]
+                    srs = PySeries.new_series_list(name, py_srs, strict)
+                else:
+                    srs = sequence_from_any_value_and_dtype_or_object(
+                        name, values, dtype
+                    )
                 if dtype != srs.dtype():
                     srs = srs.cast(dtype, strict=False)
                 return srs

diff --git a/py-polars/tests/unit/datatypes/test_list.py b/py-polars/tests/unit/datatypes/test_list.py
@@ -23,22 +23,29 @@ def test_dtype() -> None:
     assert a.dtype.is_(pl.List(pl.Int64))
 
     # explicit
+    u64_max = (2**64) - 1
     df = pl.DataFrame(
         data={
             "i": [[1, 2, 3]],
+            "li": [[[1, 2, 3]]],
+            "u": [[u64_max]],
             "tm": [[time(10, 30, 45)]],
             "dt": [[date(2022, 12, 31)]],
             "dtm": [[datetime(2022, 12, 31, 1, 2, 3)]],
         },
         schema=[
             ("i", pl.List(pl.Int8)),
+            ("li", pl.List(pl.List(pl.Int8))),
+            ("u", pl.List(pl.UInt64)),
             ("tm", pl.List(pl.Time)),
             ("dt", pl.List(pl.Date)),
             ("dtm", pl.List(pl.Datetime)),
         ],
     )
     assert df.schema == {
         "i": pl.List(pl.Int8),
+        "li": pl.List(pl.List(pl.Int8)),
+        "u": pl.List(pl.UInt64),
         "tm": pl.List(pl.Time),
         "dt": pl.List(pl.Date),
         "dtm": pl.List(pl.Datetime),
@@ -48,6 +55,8 @@ def test_dtype() -> None:
     assert df.rows() == [
         (
             [1, 2, 3],
+            [[1, 2, 3]],
+            [u64_max],
             [time(10, 30, 45)],
             [date(2022, 12, 31)],
             [datetime(2022, 12, 31, 1, 2, 3)],

diff --git a/py-polars/tests/unit/io/test_json.py b/py-polars/tests/unit/io/test_json.py
@@ -297,7 +297,7 @@ def test_write_json_duration() -> None:
         ([["a", "b"], [None, None]], pl.Array(pl.Utf8, width=2)),
         ([[True, False, None], [None, None, None]], pl.Array(pl.Utf8, width=3)),
         (
-            [[[1, 2, 3], [4, None]], None, [[None, None, 2]]],
+            [[[1, 2, 3], [4, None, 5]], None, [[None, None, 2]]],
             pl.List(pl.Array(pl.Int32(), width=3)),
         ),
         (

diff --git a/py-polars/tests/unit/series/test_describe.py b/py-polars/tests/unit/series/test_describe.py
@@ -117,7 +117,7 @@ def test_series_describe_null() -> None:
 def test_series_describe_nested_list() -> None:
     s = pl.Series(
         values=[[10e10, 10e15], [10e12, 10e13], [10e10, 10e15]],
-        dtype=pl.List(pl.Int64),
+        dtype=pl.List(pl.Float64),
     )
     result = s.describe()
     stats = {

diff --git a/py-polars/tests/unit/test_serde.py b/py-polars/tests/unit/test_serde.py
@@ -188,7 +188,7 @@ def test_serde_array_dtype() -> None:
     assert_series_equal(pickle.loads(pickle.dumps(s)), s)
 
     nested_s = pl.Series(
-        [[[1, 2, 3], [4, None]], None, [[None, None, 2]]],
+        [[[1, 2, 3], [4, None, 5]], None, [[None, None, 2]]],
         dtype=pl.List(pl.Array(pl.Int32(), width=3)),
     )
     assert_series_equal(pickle.loads(pickle.dumps(nested_s)), nested_s)