[BUG] Series.__setitem__ that expands is not displayable

[mroeschke]

Describe the bug
Series.setitem that expands is not displayable

Steps/Code to reproduce bug

In [12]: import cudf

In [13]: se = cudf.Series([1, 2, 3])

In [14]: se[5] = 5.

In [15]: repr(se)
File ~/mambaforge/envs/xdf_dev/lib/python3.10/site-packages/cudf/core/series.py:1442, in Series.__repr__(self)
   1433     output = pd_series.to_string(
   1434         name=self.name,
   1435         dtype=self.dtype,
   (...)
   1439         na_rep=cudf._NA_REP,
   1440     )
   1441 else:
-> 1442     output = repr(preprocess.to_pandas())
   1444 lines = output.split("\n")
   1445 if isinstance(preprocess._column, cudf.core.column.CategoricalColumn):

File ~/mambaforge/envs/xdf_dev/lib/python3.10/site-packages/nvtx/nvtx.py:101, in annotate.__call__.<locals>.inner(*args, **kwargs)
     98 @wraps(func)
     99 def inner(*args, **kwargs):
    100     libnvtx_push_range(self.attributes, self.domain.handle)
--> 101     result = func(*args, **kwargs)
    102     libnvtx_pop_range(self.domain.handle)
    103     return result

File ~/mambaforge/envs/xdf_dev/lib/python3.10/site-packages/cudf/core/series.py:2004, in Series.to_pandas(self, index, nullable, **kwargs)
   2002 if index is True:
   2003     index = self.index.to_pandas()
-> 2004 s = self._column.to_pandas(index=index, nullable=nullable)
   2005 s.name = self.name
   2006 return s

File ~/mambaforge/envs/xdf_dev/lib/python3.10/site-packages/cudf/core/column/numerical.py:690, in NumericalColumn.to_pandas(self, index, nullable, **kwargs)
    687     pd_series = self.to_arrow().to_pandas(**kwargs)
    689 if index is not None:
--> 690     pd_series.index = index
    691 return pd_series

File ~/mambaforge/envs/xdf_dev/lib/python3.10/site-packages/pandas/core/generic.py:5915, in NDFrame.__setattr__(self, name, value)
   5913 try:
   5914     object.__getattribute__(self, name)
-> 5915     return object.__setattr__(self, name, value)
   5916 except AttributeError:
   5917     pass

File ~/mambaforge/envs/xdf_dev/lib/python3.10/site-packages/pandas/_libs/properties.pyx:69, in pandas._libs.properties.AxisProperty.__set__()

File ~/mambaforge/envs/xdf_dev/lib/python3.10/site-packages/pandas/core/series.py:593, in Series._set_axis(self, axis, labels)
    590             pass
    592 # The ensure_index call above ensures we have an Index object
--> 593 self._mgr.set_axis(axis, labels)

File ~/mambaforge/envs/xdf_dev/lib/python3.10/site-packages/pandas/core/internals/managers.py:230, in BaseBlockManager.set_axis(self, axis, new_labels)
    228 def set_axis(self, axis: int, new_labels: Index) -> None:
    229     # Caller is responsible for ensuring we have an Index object.
--> 230     self._validate_set_axis(axis, new_labels)
    231     self.axes[axis] = new_labels

File ~/mambaforge/envs/xdf_dev/lib/python3.10/site-packages/pandas/core/internals/base.py:70, in DataManager._validate_set_axis(self, axis, new_labels)
     67     pass
     69 elif new_len != old_len:
---> 70     raise ValueError(
     71         f"Length mismatch: Expected axis has {old_len} elements, new "
     72         f"values have {new_len} elements"
     73     )

ValueError: Length mismatch: Expected axis has 4 elements, new values have 3 elements

EDIT: I notice something similar happens with loc.__setitem__ and iloc.__seitem__

Expected behavior
This object repr should not be corrupted

In [17]: import pandas

In [18]: se = pandas.Series([1, 2, 3])

In [19]: se[5] = 5.

In [20]: se
Out[20]: 
0    1.0
1    2.0
2    3.0
5    5.0
dtype: float64

Environment overview (please complete the following information)

• Environment location: Bare-metal
• Method of cuDF install: conda
  • If method of install is [Docker], provide docker pull & docker run commands used

Environment details
Please run and paste the output of the cudf/print_env.sh script here, to gather any other relevant environment details

Additional context
Add any other context about the problem here.

[wence-]

EDIT: I notice something similar happens with loc.setitem and iloc.seitem

iloc.__setitem__ should fail right? With positional indexing one can't expand unknown indices AFAIU.

[wence-]

@galipremsagar not sure where you got to, but maybe:

diff --git a/python/cudf/cudf/core/series.py b/python/cudf/cudf/core/series.py
index 6ff0584538..931c67f63d 100644
--- a/python/cudf/cudf/core/series.py
+++ b/python/cudf/cudf/core/series.py
@@ -166,7 +166,7 @@ def _describe_categorical(obj, percentiles):
     return data
 
 
-def _append_new_row_inplace(col: ColumnLike, value: ScalarLike):
+def _append_new_row(col: ColumnLike, value: ScalarLike, *, inplace):
     """Append a scalar `value` to the end of `col` inplace.
     Cast to common type if possible
     """
@@ -174,7 +174,9 @@ def _append_new_row_inplace(col: ColumnLike, value: ScalarLike):
     val_col = as_column(value, dtype=to_type)
     old_col = col.astype(to_type)
 
-    col._mimic_inplace(concat_columns([old_col, val_col]), inplace=True)
+    return col._mimic_inplace(
+        concat_columns([old_col, val_col]), inplace=inplace
+    )
 
 
 class _SeriesIlocIndexer(_FrameIndexer):
@@ -274,8 +276,13 @@ class _SeriesLocIndexer(_FrameIndexer):
                 and not isinstance(self._frame.index, cudf.MultiIndex)
                 and is_scalar(value)
             ):
-                _append_new_row_inplace(self._frame.index._values, key)
-                _append_new_row_inplace(self._frame._column, value)
+                index = self._frame._index._values
+                col = self._frame._column
+                col = _append_new_row(col, value, inplace=False)
+                index = _append_new_row(index, key, inplace=False)
+                self._frame._mimic_inplace(
+                    Series(col, index=index), inplace=True
+                )
                 return
             else:
                 raise e

[mroeschke]

EDIT: I notice something similar happens with loc.setitem and iloc.seitem

iloc.__setitem__ should fail right? With positional indexing one can't expand unknown indices AFAIU.

Correct

In [4]: se.iloc[3] = 4
---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
Cell In[4], line 1
----> 1 se.iloc[3] = 4

File ~/pandas-mroeschke/pandas/core/indexing.py:882, in _LocationIndexer.__setitem__(self, key, value)
    880     key = com.apply_if_callable(key, self.obj)
    881 indexer = self._get_setitem_indexer(key)
--> 882 self._has_valid_setitem_indexer(key)
    884 iloc = self if self.name == "iloc" else self.obj.iloc
    885 iloc._setitem_with_indexer(indexer, value, self.name)

File ~/pandas-mroeschke/pandas/core/indexing.py:1607, in _iLocIndexer._has_valid_setitem_indexer(self, indexer)
   1605 elif is_integer(i):
   1606     if i >= len(ax):
-> 1607         raise IndexError("iloc cannot enlarge its target object")
   1608 elif isinstance(i, dict):
   1609     raise IndexError("iloc cannot enlarge its target object")

IndexError: iloc cannot enlarge its target object

[galipremsagar]

@galipremsagar not sure where you got to, but maybe:

diff --git a/python/cudf/cudf/core/series.py b/python/cudf/cudf/core/series.py
index 6ff0584538..931c67f63d 100644
--- a/python/cudf/cudf/core/series.py
+++ b/python/cudf/cudf/core/series.py
@@ -166,7 +166,7 @@ def _describe_categorical(obj, percentiles):
     return data
 
 
-def _append_new_row_inplace(col: ColumnLike, value: ScalarLike):
+def _append_new_row(col: ColumnLike, value: ScalarLike, *, inplace):
     """Append a scalar `value` to the end of `col` inplace.
     Cast to common type if possible
     """
@@ -174,7 +174,9 @@ def _append_new_row_inplace(col: ColumnLike, value: ScalarLike):
     val_col = as_column(value, dtype=to_type)
     old_col = col.astype(to_type)
 
-    col._mimic_inplace(concat_columns([old_col, val_col]), inplace=True)
+    return col._mimic_inplace(
+        concat_columns([old_col, val_col]), inplace=inplace
+    )
 
 
 class _SeriesIlocIndexer(_FrameIndexer):
@@ -274,8 +276,13 @@ class _SeriesLocIndexer(_FrameIndexer):
                 and not isinstance(self._frame.index, cudf.MultiIndex)
                 and is_scalar(value)
             ):
-                _append_new_row_inplace(self._frame.index._values, key)
-                _append_new_row_inplace(self._frame._column, value)
+                index = self._frame._index._values
+                col = self._frame._column
+                col = _append_new_row(col, value, inplace=False)
+                index = _append_new_row(index, key, inplace=False)
+                self._frame._mimic_inplace(
+                    Series(col, index=index), inplace=True
+                )
                 return
             else:
                 raise e

Yup, that + we need to do some handling for RangeIndex alone. Opened a WIP PR: #13888

[vyasr]

This issue was resolved in 66fdbe7