Merge remote-tracking branch 'upstream/main' into cln/assorted

Author: mroeschke

ci/code_checks.sh

@@ -79,23 +79,8 @@ fi
 ### DOCSTRINGS ###
 if [[ -z "$CHECK" || "$CHECK" == "docstrings" ]]; then
 
-    MSG='Validate docstrings (EX04, GL01, GL02, GL03, GL04, GL05, GL06, GL07, GL09, GL10, PR03, PR04, PR05, PR06, PR08, PR09, PR10, RT01, RT04, RT05, SA02, SA03, SA04, SS01, SS02, SS03, SS04, SS05, SS06)' ; echo $MSG
-    $BASE_DIR/scripts/validate_docstrings.py --format=actions --errors=EX04,GL01,GL02,GL03,GL04,GL05,GL06,GL07,GL09,GL10,PR03,PR04,PR05,PR06,PR08,PR09,PR10,RT01,RT04,RT05,SA02,SA03,SA04,SS01,SS02,SS03,SS04,SS05,SS06
-    RET=$(($RET + $?)) ; echo $MSG "DONE"
-
-    MSG='Partially validate docstrings (RT02)' ; echo $MSG
-    $BASE_DIR/scripts/validate_docstrings.py --format=actions --errors=RT02 --ignore_functions \
-	    pandas.Index.all \
-	    pandas.Index.any \
-	    pandas.MultiIndex.drop \
-	    pandas.DatetimeIndex.to_pydatetime \
-	    pandas.TimedeltaIndex.to_pytimedelta \
-	    pandas.io.formats.style.Styler.export \
-	    pandas.api.extensions.ExtensionArray.astype \
-	    pandas.api.extensions.ExtensionArray.dropna \
-	    pandas.api.extensions.ExtensionArray.isna \
-	    pandas.api.extensions.ExtensionArray.repeat \
-	    pandas.api.extensions.ExtensionArray.unique
+    MSG='Validate docstrings (EX04, GL01, GL02, GL03, GL04, GL05, GL06, GL07, GL09, GL10, PR03, PR04, PR05, PR06, PR08, PR09, PR10, RT01, RT02, RT04, RT05, SA02, SA03, SA04, SS01, SS02, SS03, SS04, SS05, SS06)' ; echo $MSG
+    $BASE_DIR/scripts/validate_docstrings.py --format=actions --errors=EX04,GL01,GL02,GL03,GL04,GL05,GL06,GL07,GL09,GL10,PR03,PR04,PR05,PR06,PR08,PR09,PR10,RT01,RT02,RT04,RT05,SA02,SA03,SA04,SS01,SS02,SS03,SS04,SS05,SS06
     RET=$(($RET + $?)) ; echo $MSG "DONE"
 
 fi

doc/source/development/contributing_gitpod.rst

@@ -145,7 +145,7 @@ docs you need to run the following command in the docs directory::
 
 Alternatively you can build a single page with::
 
-    python make.py html python make.py --single development/contributing_gitpod.rst
+    python make.py --single development/contributing_gitpod.rst
 
 You have two main options to render the documentation in Gitpod.
 

doc/source/user_guide/timeseries.rst

@@ -292,7 +292,7 @@ The default behavior, ``errors='raise'``, is to raise when unparsable:
 .. code-block:: ipython
 
     In [2]: pd.to_datetime(['2009/07/31', 'asd'], errors='raise')
-    ValueError: Unknown string format
+    ValueError: Unknown datetime string format
 
 Pass ``errors='ignore'`` to return the original input when unparsable:
 

doc/source/whatsnew/v2.0.0.rst

@@ -165,6 +165,7 @@ Other enhancements
 - Improved error message in :func:`to_datetime` for non-ISO8601 formats, informing users about the position of the first error (:issue:`50361`)
 - Improved error message when trying to align :class:`DataFrame` objects (for example, in :func:`DataFrame.compare`) to clarify that "identically labelled" refers to both index and columns (:issue:`50083`)
 - Added :meth:`DatetimeIndex.as_unit` and :meth:`TimedeltaIndex.as_unit` to convert to different resolutions; supported resolutions are "s", "ms", "us", and "ns" (:issue:`50616`)
+- Added new argument ``dtype`` to :func:`read_sql` to be consistent with :func:`read_sql_query` (:issue:`50797`)
 -
 
 .. ---------------------------------------------------------------------------
@@ -950,6 +951,7 @@ Datetimelike
 - Bug in :func:`to_datetime` was failing to parse date strings ``'today'`` and ``'now'`` if ``format`` was not ISO8601 (:issue:`50359`)
 - Bug in :func:`Timestamp.utctimetuple` raising a ``TypeError`` (:issue:`32174`)
 - Bug in :func:`to_datetime` was raising ``ValueError`` when parsing mixed-offset :class:`Timestamp` with ``errors='ignore'`` (:issue:`50585`)
+- Bug in :func:`to_datetime` was incorrectly handling floating-point inputs within 1 ``unit`` of the overflow boundaries (:issue:`50183`)
 
 Timedelta
 ^^^^^^^^^

pandas/_libs/lib.pyi

@@ -240,6 +240,6 @@ def get_reverse_indexer(
 ) -> npt.NDArray[np.intp]: ...
 def is_bool_list(obj: list) -> bool: ...
 def dtypes_all_equal(types: list[DtypeObj]) -> bool: ...
-def array_equal_fast(
-    left: np.ndarray, right: np.ndarray  # np.ndarray[np.int64, ndim=1]
+def is_range_indexer(
+    left: np.ndarray, n: int  # np.ndarray[np.int64, ndim=1]
 ) -> bool: ...

pandas/_libs/lib.pyx

@@ -650,22 +650,20 @@ ctypedef fused int6432_t:
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def array_equal_fast(
-    ndarray[int6432_t, ndim=1] left, ndarray[int6432_t, ndim=1] right,
-) -> bool:
+def is_range_indexer(ndarray[int6432_t, ndim=1] left, int n) -> bool:
     """
     Perform an element by element comparison on 1-d integer arrays, meant for indexer
     comparisons
     """
     cdef:
-        Py_ssize_t i, n = left.size
+        Py_ssize_t i
 
-    if left.size != right.size:
+    if left.size != n:
         return False
 
     for i in range(n):
 
-        if left[i] != right[i]:
+        if left[i] != i:
             return False
 
     return True

pandas/_libs/tslib.pyx

@@ -273,7 +273,7 @@ def array_with_unit_to_datetime(
         bint is_coerce = errors=="coerce"
         bint is_raise = errors=="raise"
         ndarray[int64_t] iresult
-        object tz = None
+        tzinfo tz = None
         float fval
 
     assert is_ignore or is_coerce or is_raise
@@ -346,7 +346,7 @@ cdef _array_with_unit_to_datetime_object_fallback(ndarray[object] values, str un
     cdef:
         Py_ssize_t i, n = len(values)
         ndarray[object] oresult
-        object tz = None
+        tzinfo tz = None
 
     # TODO: fix subtle differences between this and no-unit code
     oresult = cnp.PyArray_EMPTY(values.ndim, values.shape, cnp.NPY_OBJECT, 0)
@@ -497,15 +497,9 @@ cpdef array_to_datetime(
 
                 if val != val or val == NPY_NAT:
                     iresult[i] = NPY_NAT
-                elif is_raise or is_ignore:
-                    iresult[i] = val
                 else:
-                    # coerce
                     # we now need to parse this as if unit='ns'
-                    try:
-                        iresult[i] = cast_from_unit(val, "ns")
-                    except OverflowError:
-                        iresult[i] = NPY_NAT
+                    iresult[i] = cast_from_unit(val, "ns")
 
             elif isinstance(val, str):
                 # string

pandas/_libs/tslibs/conversion.pyx

@@ -544,16 +544,9 @@ cdef _TSObject convert_str_to_tsobject(str ts, tzinfo tz, str unit,
                 maybe_localize_tso(obj, tz, obj.creso)
                 return obj
 
-        try:
-            dt = parse_datetime_string(
-                ts, dayfirst=dayfirst, yearfirst=yearfirst
-            )
-        except ValueError as err:
-            if "out of range for month" in str(err):
-                # dateutil raised when constructing a datetime object,
-                #  let's give a nicer exception message
-                raise ValueError("could not convert string to Timestamp") from err
-            raise
+        dt = parse_datetime_string(
+            ts, dayfirst=dayfirst, yearfirst=yearfirst
+        )
 
     return convert_datetime_to_tsobject(dt, tz)
 

pandas/_libs/tslibs/dtypes.pxd

@@ -11,6 +11,7 @@ cpdef int64_t periods_per_second(NPY_DATETIMEUNIT reso) except? -1
 cpdef NPY_DATETIMEUNIT get_supported_reso(NPY_DATETIMEUNIT reso)
 
 cdef dict attrname_to_abbrevs
+cdef dict npy_unit_to_attrname
 
 cdef enum c_FreqGroup:
     # Mirrors FreqGroup in the .pyx file

pandas/_libs/tslibs/dtypes.pyx

@@ -423,3 +423,15 @@ cdef dict _reso_str_map = {
 }
 
 cdef dict _str_reso_map = {v: k for k, v in _reso_str_map.items()}
+
+cdef dict npy_unit_to_attrname = {
+    NPY_DATETIMEUNIT.NPY_FR_Y: "year",
+    NPY_DATETIMEUNIT.NPY_FR_M: "month",
+    NPY_DATETIMEUNIT.NPY_FR_D: "day",
+    NPY_DATETIMEUNIT.NPY_FR_h: "hour",
+    NPY_DATETIMEUNIT.NPY_FR_m: "minute",
+    NPY_DATETIMEUNIT.NPY_FR_s: "second",
+    NPY_DATETIMEUNIT.NPY_FR_ms: "millisecond",
+    NPY_DATETIMEUNIT.NPY_FR_us: "microsecond",
+    NPY_DATETIMEUNIT.NPY_FR_ns: "nanosecond",
+}

pandas/_libs/tslibs/offsets.pyx

@@ -2792,7 +2792,7 @@ cdef class Week(SingleConstructorOffset):
      2nd week of each month.
 
     Examples
-    ---------
+    --------
 
     >>> date_object = pd.Timestamp("2023-01-13")
     >>> date_object