From d2f0c7c7ef31b691d7fe2185d5e9481937a27be3 Mon Sep 17 00:00:00 2001
From: tonyyuyiding <tonyyuyiding@gmail.com>
Date: Mon, 3 Mar 2025 14:54:21 +0800
Subject: [PATCH 1/9] make array 1D before calling maybe_infer_to_datetimelike
---
pandas/core/construction.py | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/pandas/core/construction.py b/pandas/core/construction.py
index ada492787a179..6a6ae544c5b29 100644
--- a/pandas/core/construction.py
+++ b/pandas/core/construction.py
@@ -612,7 +612,10 @@ def sanitize_array(
if dtype is None:
subarr = data
if data.dtype == object and infer_object:
- subarr = maybe_infer_to_datetimelike(data)
+ # GH#61026
+ if data.ndim != 1:
+ subarr = subarr.ravel()
+ subarr = maybe_infer_to_datetimelike(subarr)
elif data.dtype.kind == "U" and using_string_dtype():
from pandas.core.arrays.string_ import StringDtype
From afc3c231d6c0feaf5308e07f7d1535c4ccceb5be Mon Sep 17 00:00:00 2001
From: tonyyuyiding <tonyyuyiding@gmail.com>
Date: Mon, 3 Mar 2025 15:11:25 +0800
Subject: [PATCH 2/9] add desc to whatsnew
---
doc/source/whatsnew/v3.0.0.rst | 1 +
1 file changed, 1 insertion(+)
diff --git a/doc/source/whatsnew/v3.0.0.rst b/doc/source/whatsnew/v3.0.0.rst
index c967b97cb2ef6..966bb043177d9 100644
--- a/doc/source/whatsnew/v3.0.0.rst
+++ b/doc/source/whatsnew/v3.0.0.rst
@@ -696,6 +696,7 @@ Indexing
- Bug in :meth:`DataFrame.loc` with inconsistent behavior of loc-set with 2 given indexes to Series (:issue:`59933`)
- Bug in :meth:`MultiIndex.insert` when a new value inserted to a datetime-like level gets cast to ``NaT`` and fails indexing (:issue:`60388`)
- Bug in printing :attr:`Index.names` and :attr:`MultiIndex.levels` would not escape single quotes (:issue:`60190`)
+- Bug in :meth:`DataFrame.__setitem__` throwing a ``ValueError`` when setting a column with a 2D object array (:issue:`61026`)
Missing
^^^^^^^
From c967d6ba691ae438d29d2740c6afdcaec17e6ae5 Mon Sep 17 00:00:00 2001
From: tonyyuyiding <tonyyuyiding@gmail.com>
Date: Mon, 3 Mar 2025 15:30:55 +0800
Subject: [PATCH 3/9] update the order of whatsnew entries
---
doc/source/whatsnew/v3.0.0.rst | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/source/whatsnew/v3.0.0.rst b/doc/source/whatsnew/v3.0.0.rst
index 966bb043177d9..7d8679e1ec7d8 100644
--- a/doc/source/whatsnew/v3.0.0.rst
+++ b/doc/source/whatsnew/v3.0.0.rst
@@ -692,11 +692,11 @@ Interval
Indexing
^^^^^^^^
- Bug in :meth:`DataFrame.__getitem__` returning modified columns when called with ``slice`` in Python 3.12 (:issue:`57500`)
+- Bug in :meth:`DataFrame.__setitem__` throwing a ``ValueError`` when setting a column with a 2D object array (:issue:`61026`)
- Bug in :meth:`DataFrame.from_records` throwing a ``ValueError`` when passed an empty list in ``index`` (:issue:`58594`)
- Bug in :meth:`DataFrame.loc` with inconsistent behavior of loc-set with 2 given indexes to Series (:issue:`59933`)
- Bug in :meth:`MultiIndex.insert` when a new value inserted to a datetime-like level gets cast to ``NaT`` and fails indexing (:issue:`60388`)
- Bug in printing :attr:`Index.names` and :attr:`MultiIndex.levels` would not escape single quotes (:issue:`60190`)
-- Bug in :meth:`DataFrame.__setitem__` throwing a ``ValueError`` when setting a column with a 2D object array (:issue:`61026`)
Missing
^^^^^^^
From fe1a1d1414afdcbd75e9af95636dbe13bbb81b27 Mon Sep 17 00:00:00 2001
From: tonyyuyiding <tonyyuyiding@gmail.com>
Date: Tue, 4 Mar 2025 09:19:43 +0800
Subject: [PATCH 4/9] reshape the array back after maybe_infer_to_datetimelike
---
pandas/core/construction.py | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/pandas/core/construction.py b/pandas/core/construction.py
index 6a6ae544c5b29..e60c545e83b77 100644
--- a/pandas/core/construction.py
+++ b/pandas/core/construction.py
@@ -612,10 +612,8 @@ def sanitize_array(
if dtype is None:
subarr = data
if data.dtype == object and infer_object:
- # GH#61026
- if data.ndim != 1:
- subarr = subarr.ravel()
- subarr = maybe_infer_to_datetimelike(subarr)
+ fltarr = data.ravel() # GH#61026
+ subarr = maybe_infer_to_datetimelike(fltarr).reshape(data.shape)
elif data.dtype.kind == "U" and using_string_dtype():
from pandas.core.arrays.string_ import StringDtype
From e6cdec8c5cccccb22cb1c45f8d6dfddf11aa5684 Mon Sep 17 00:00:00 2001
From: tonyyuyiding <tonyyuyiding@gmail.com>
Date: Tue, 4 Mar 2025 09:20:33 +0800
Subject: [PATCH 5/9] add a test
---
pandas/tests/frame/indexing/test_setitem.py | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/pandas/tests/frame/indexing/test_setitem.py b/pandas/tests/frame/indexing/test_setitem.py
index 20dd7b0c4d3e7..db2c24e97893b 100644
--- a/pandas/tests/frame/indexing/test_setitem.py
+++ b/pandas/tests/frame/indexing/test_setitem.py
@@ -815,6 +815,19 @@ def test_setitem_index_object_dtype_not_inferring(self):
}
)
tm.assert_frame_equal(df, expected)
+
+ def test_setitem_2d_object_array(self):
+ # GH#61026
+ df = DataFrame({
+ "c1": [1, 2, 3, 4, 5],
+ })
+ arr = np.array([["A"], ["B"], ["C"], ["D"], ["E"]], dtype=object)
+ df["c1"] = arr
+
+ expected = DataFrame({
+ "c1": ["A", "B", "C", "D", "E"],
+ })
+ tm.assert_frame_equal(df, expected)
class TestSetitemTZAwareValues:
From 68dfd0ecc4c57e065f0a64a8a85f7113c3e4235f Mon Sep 17 00:00:00 2001
From: tonyyuyiding <tonyyuyiding@gmail.com>
Date: Tue, 4 Mar 2025 09:59:12 +0800
Subject: [PATCH 6/9] formatting by pre-commit
---
pandas/core/construction.py | 2 +-
pandas/tests/frame/indexing/test_setitem.py | 20 ++++++++++++--------
2 files changed, 13 insertions(+), 9 deletions(-)
diff --git a/pandas/core/construction.py b/pandas/core/construction.py
index e60c545e83b77..3da78a2972e77 100644
--- a/pandas/core/construction.py
+++ b/pandas/core/construction.py
@@ -612,7 +612,7 @@ def sanitize_array(
if dtype is None:
subarr = data
if data.dtype == object and infer_object:
- fltarr = data.ravel() # GH#61026
+ fltarr = data.ravel() # GH#61026
subarr = maybe_infer_to_datetimelike(fltarr).reshape(data.shape)
elif data.dtype.kind == "U" and using_string_dtype():
from pandas.core.arrays.string_ import StringDtype
diff --git a/pandas/tests/frame/indexing/test_setitem.py b/pandas/tests/frame/indexing/test_setitem.py
index db2c24e97893b..e7e119b8bf32c 100644
--- a/pandas/tests/frame/indexing/test_setitem.py
+++ b/pandas/tests/frame/indexing/test_setitem.py
@@ -815,18 +815,22 @@ def test_setitem_index_object_dtype_not_inferring(self):
}
)
tm.assert_frame_equal(df, expected)
-
+
def test_setitem_2d_object_array(self):
# GH#61026
- df = DataFrame({
- "c1": [1, 2, 3, 4, 5],
- })
+ df = DataFrame(
+ {
+ "c1": [1, 2, 3, 4, 5],
+ }
+ )
arr = np.array([["A"], ["B"], ["C"], ["D"], ["E"]], dtype=object)
df["c1"] = arr
-
- expected = DataFrame({
- "c1": ["A", "B", "C", "D", "E"],
- })
+
+ expected = DataFrame(
+ {
+ "c1": ["A", "B", "C", "D", "E"],
+ }
+ )
tm.assert_frame_equal(df, expected)
From 537671a7a063c28cc1bb23a273d86c764fd0e628 Mon Sep 17 00:00:00 2001
From: Tony Ding <tonyyuyiding@gmail.com>
Date: Tue, 4 Mar 2025 10:31:20 +0800
Subject: [PATCH 7/9] remove reshape
---
pandas/core/construction.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/pandas/core/construction.py b/pandas/core/construction.py
index 3da78a2972e77..3e08d6d59b56e 100644
--- a/pandas/core/construction.py
+++ b/pandas/core/construction.py
@@ -613,7 +613,7 @@ def sanitize_array(
subarr = data
if data.dtype == object and infer_object:
fltarr = data.ravel() # GH#61026
- subarr = maybe_infer_to_datetimelike(fltarr).reshape(data.shape)
+ subarr = maybe_infer_to_datetimelike(fltarr)
elif data.dtype.kind == "U" and using_string_dtype():
from pandas.core.arrays.string_ import StringDtype
From b7ab1559c4bea6a3474571c6410b608b869c38ed Mon Sep 17 00:00:00 2001
From: Tony Ding <tonyyuyiding@gmail.com>
Date: Tue, 4 Mar 2025 11:26:16 +0800
Subject: [PATCH 8/9] fix testcase issue
---
pandas/core/construction.py | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/pandas/core/construction.py b/pandas/core/construction.py
index 3e08d6d59b56e..826692424c1cf 100644
--- a/pandas/core/construction.py
+++ b/pandas/core/construction.py
@@ -612,8 +612,10 @@ def sanitize_array(
if dtype is None:
subarr = data
if data.dtype == object and infer_object:
- fltarr = data.ravel() # GH#61026
- subarr = maybe_infer_to_datetimelike(fltarr)
+ if data.ndim != 1:
+ # GH#61026
+ subarr = data.ravel()
+ subarr = maybe_infer_to_datetimelike(subarr)
elif data.dtype.kind == "U" and using_string_dtype():
from pandas.core.arrays.string_ import StringDtype
From fc8d55d00de58a45173524af1fe9ad446ca1083a Mon Sep 17 00:00:00 2001
From: Tony Ding <tonyyuyiding@gmail.com>
Date: Tue, 4 Mar 2025 11:47:13 +0800
Subject: [PATCH 9/9] fix testcase issue (copy)
---
pandas/core/construction.py | 2 ++
1 file changed, 2 insertions(+)
diff --git a/pandas/core/construction.py b/pandas/core/construction.py
index 826692424c1cf..23b292c34eca4 100644
--- a/pandas/core/construction.py
+++ b/pandas/core/construction.py
@@ -615,6 +615,8 @@ def sanitize_array(
if data.ndim != 1:
# GH#61026
subarr = data.ravel()
+ if copy:
+ subarr = subarr.copy()
subarr = maybe_infer_to_datetimelike(subarr)
elif data.dtype.kind == "U" and using_string_dtype():
from pandas.core.arrays.string_ import StringDtype