Merge branch 'main' into issue-50977

Author: kostyafarber

doc/source/whatsnew/v2.0.0.rst

@@ -620,6 +620,7 @@ Other API changes
   new DataFrame (shallow copy) instead of the original DataFrame, consistent with other
   methods to get a full slice (for example ``df.loc[:]`` or ``df[:]``) (:issue:`49469`)
 - Disallow computing ``cumprod`` for :class:`Timedelta` object; previously this returned incorrect values (:issue:`50246`)
+- :class:`DataFrame` objects read from a :class:`HDFStore` file without an index now have a :class:`RangeIndex` instead of an ``int64`` index (:issue:`51076`)
 - Instantiating an :class:`Index` with an numeric numpy dtype with data containing :class:`NA` and/or :class:`NaT` now raises a ``ValueError``. Previously a ``TypeError`` was raised (:issue:`51050`)
 - Loading a JSON file with duplicate columns using ``read_json(orient='split')`` renames columns to avoid duplicates, as :func:`read_csv` and the other readers do (:issue:`50370`)
 - The levels of the index of the :class:`Series` returned from ``Series.sparse.from_coo`` now always have dtype ``int32``. Previously they had dtype ``int64`` (:issue:`50926`)

pandas/_libs/index.pyx

@@ -238,8 +238,6 @@ cdef class IndexEngine:
         return self.unique == 1
 
     cdef _do_unique_check(self):
-
-        # this de-facto the same
         self._ensure_mapping_populated()
 
     @property

pandas/_libs/indexing.pyi

@@ -9,7 +9,7 @@ _IndexingMixinT = TypeVar("_IndexingMixinT", bound=IndexingMixin)
 
 class NDFrameIndexerBase(Generic[_IndexingMixinT]):
     name: str
-    # in practise obj is either a DataFrame or a Series
+    # in practice obj is either a DataFrame or a Series
     obj: _IndexingMixinT
 
     def __init__(self, name: str, obj: _IndexingMixinT) -> None: ...

pandas/_libs/internals.pyx

@@ -33,7 +33,6 @@ from pandas._libs.util cimport (
 @cython.final
 @cython.freelist(32)
 cdef class BlockPlacement:
-    # __slots__ = '_as_slice', '_as_array', '_len'
     cdef:
         slice _as_slice
         ndarray _as_array  # Note: this still allows `None`; will be intp_t
@@ -621,7 +620,7 @@ cdef class NumpyBlock(SharedBlock):
         public ndarray values
 
     def __cinit__(self, ndarray values, BlockPlacement placement, int ndim):
-        # set values here the (implicit) call to SharedBlock.__cinit__ will
+        # set values here; the (implicit) call to SharedBlock.__cinit__ will
         #  set placement and ndim
         self.values = values
 
@@ -643,7 +642,7 @@ cdef class NDArrayBackedBlock(SharedBlock):
         NDArrayBacked values
 
     def __cinit__(self, NDArrayBacked values, BlockPlacement placement, int ndim):
-        # set values here the (implicit) call to SharedBlock.__cinit__ will
+        # set values here; the (implicit) call to SharedBlock.__cinit__ will
         #  set placement and ndim
         self.values = values
 
@@ -662,7 +661,7 @@ cdef class Block(SharedBlock):
         public object values
 
     def __cinit__(self, object values, BlockPlacement placement, int ndim):
-        # set values here the (implicit) call to SharedBlock.__cinit__ will
+        # set values here; the (implicit) call to SharedBlock.__cinit__ will
         #  set placement and ndim
         self.values = values
 

pandas/_libs/lib.pyx

@@ -492,7 +492,7 @@ def get_reverse_indexer(const intp_t[:] indexer, Py_ssize_t length) -> ndarray:
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-# Can add const once https://github.com/cython/cython/issues/1772 resolved
+# TODO(cython3): Can add const once cython#1772 is resolved
 def has_infs(floating[:] arr) -> bool:
     cdef:
         Py_ssize_t i, n = len(arr)

pandas/_libs/parsers.pyx

@@ -46,6 +46,7 @@ from libc.string cimport (
 
 
 cdef extern from "Python.h":
+    # TODO(cython3): get this from cpython.unicode
     object PyUnicode_FromString(char *v)
 
 
@@ -453,14 +454,12 @@ cdef class TextReader:
 
         self.skipfooter = skipfooter
 
-        # suboptimal
         if usecols is not None:
             self.has_usecols = 1
             # GH-20558, validate usecols at higher level and only pass clean
             # usecols into TextReader.
             self.usecols = usecols
 
-        # TODO: XXX?
         if skipfooter > 0:
             self.parser.on_bad_lines = SKIP
 
@@ -501,7 +500,6 @@ cdef class TextReader:
         self.dtype = dtype
         self.use_nullable_dtypes = use_nullable_dtypes
 
-        # XXX
         self.noconvert = set()
 
         self.index_col = index_col
@@ -761,7 +759,7 @@ cdef class TextReader:
         # Corner case, not enough lines in the file
         if self.parser.lines < data_line + 1:
             field_count = len(header[0])
-        else:  # not self.has_usecols:
+        else:
 
             field_count = self.parser.line_fields[data_line]
 
@@ -1409,6 +1407,8 @@ def _maybe_upcast(arr, use_nullable_dtypes: bool = False):
     The casted array.
     """
     if is_extension_array_dtype(arr.dtype):
+        # TODO: the docstring says arr is an ndarray, in which case this cannot
+        #  be reached. Is that incorrect?
         return arr
 
     na_value = na_values[arr.dtype]

pandas/_libs/reduction.pyi

@@ -1,8 +1,6 @@
 from typing import Any
 
-import numpy as np
+from pandas._typing import DtypeObj
 
-from pandas._typing import ExtensionDtype
-
-def check_result_array(obj: object, dtype: np.dtype | ExtensionDtype) -> None: ...
+def check_result_array(obj: object, dtype: DtypeObj) -> None: ...
 def extract_result(res: object) -> Any: ...

pandas/_libs/sparse.pyx

@@ -301,9 +301,6 @@ cdef class BlockIndex(SparseIndex):
         self.nblocks = np.int32(len(self.blocs))
         self.npoints = self.blengths.sum()
 
-        # self.block_start = blocs
-        # self.block_end = blocs + blengths
-
         self.check_integrity()
 
     def __reduce__(self):

pandas/_libs/sparse_op_helper.pxi.in

@@ -137,16 +137,16 @@ cdef tuple block_op_{{opname}}_{{dtype}}({{dtype}}_t[:] x_,
                                                 {{dtype}}_t[:] y_,
                                                 BlockIndex yindex,
                                                 {{dtype}}_t yfill):
-    '''
+    """
     Binary operator on BlockIndex objects with fill values
-    '''
+    """
 
     cdef:
         BlockIndex out_index
-        Py_ssize_t xi = 0, yi = 0, out_i = 0 # fp buf indices
-        int32_t xbp = 0, ybp = 0 # block positions
+        Py_ssize_t xi = 0, yi = 0, out_i = 0  # fp buf indices
+        int32_t xbp = 0, ybp = 0  # block positions
         int32_t xloc, yloc
-        Py_ssize_t xblock = 0, yblock = 0 # block numbers
+        Py_ssize_t xblock = 0, yblock = 0  # block numbers
 
         {{dtype}}_t[:] x, y
         ndarray[{{rdtype}}_t, ndim=1] out

pandas/_libs/tslib.pyx

@@ -115,7 +115,7 @@ def format_array_from_datetime(
 
     Parameters
     ----------
-    values : a 1-d i8 array
+    values : ndarray[int64_t], arbitrary ndim
     tz : tzinfo or None, default None
     format : str or None, default None
           a strftime capable string
@@ -260,9 +260,9 @@ def array_with_unit_to_datetime(
     cdef:
         Py_ssize_t i, n=len(values)
         int64_t mult
-        bint is_ignore = errors=="ignore"
-        bint is_coerce = errors=="coerce"
-        bint is_raise = errors=="raise"
+        bint is_ignore = errors == "ignore"
+        bint is_coerce = errors == "coerce"
+        bint is_raise = errors == "raise"
         ndarray[int64_t] iresult
         tzinfo tz = None
         float fval
@@ -446,9 +446,9 @@ cpdef array_to_datetime(
         npy_datetimestruct dts
         bint utc_convert = bool(utc)
         bint seen_datetime_offset = False
-        bint is_raise = errors=="raise"
-        bint is_ignore = errors=="ignore"
-        bint is_coerce = errors=="coerce"
+        bint is_raise = errors == "raise"
+        bint is_ignore = errors == "ignore"
+        bint is_coerce = errors == "coerce"
         bint is_same_offsets
         _TSObject _ts
         float tz_offset