REF: handle 2D in array_to_datetime (#50549)

* REF: handle 2D in array_to_datetime

* troubleshoot build failures

* avoid build error

Author: jbrockmendel

pandas/_libs/tslib.pyx

@@ -439,7 +439,7 @@ def first_non_null(values: ndarray) -> int:
 @cython.wraparound(False)
 @cython.boundscheck(False)
 cpdef array_to_datetime(
-    ndarray[object] values,
+    ndarray values,  # object dtype, arbitrary ndim
     str errors="raise",
     bint dayfirst=False,
     bint yearfirst=False,
@@ -478,7 +478,7 @@ cpdef array_to_datetime(
     tzinfo or None
     """
     cdef:
-        Py_ssize_t i, n = len(values)
+        Py_ssize_t i, n = values.size
         object val, tz
         ndarray[int64_t] iresult
         npy_datetimestruct dts
@@ -498,15 +498,18 @@ cpdef array_to_datetime(
         datetime py_dt
         tzinfo tz_out = None
         bint found_tz = False, found_naive = False
+        cnp.broadcast mi
 
     # specify error conditions
     assert is_raise or is_ignore or is_coerce
 
-    result = np.empty(n, dtype="M8[ns]")
-    iresult = result.view("i8")
+    result = np.empty((<object>values).shape, dtype="M8[ns]")
+    mi = cnp.PyArray_MultiIterNew2(result, values)
+    iresult = result.view("i8").ravel()
 
     for i in range(n):
-        val = values[i]
+        # Analogous to `val = values[i]`
+        val = <object>(<PyObject**>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
 
         try:
             if checknull_with_nat_and_na(val):
@@ -524,7 +527,7 @@ cpdef array_to_datetime(
                     found_tz,
                     utc_convert,
                 )
-                result[i] = parse_pydatetime(val, &dts, utc_convert)
+                iresult[i] = parse_pydatetime(val, &dts, utc_convert)
 
             elif PyDate_Check(val):
                 iresult[i] = pydate_to_dt64(val, &dts)
@@ -559,6 +562,7 @@ cpdef array_to_datetime(
 
                 if len(val) == 0 or val in nat_strings:
                     iresult[i] = NPY_NAT
+                    cnp.PyArray_MultiIter_NEXT(mi)
                     continue
 
                 string_to_dts_failed = string_to_dts(
@@ -569,6 +573,7 @@ cpdef array_to_datetime(
                     # An error at this point is a _parsing_ error
                     # specifically _not_ OutOfBoundsDatetime
                     if parse_today_now(val, &iresult[i], utc):
+                        cnp.PyArray_MultiIter_NEXT(mi)
                         continue
 
                     py_dt = parse_datetime_string(val,
@@ -614,10 +619,13 @@ cpdef array_to_datetime(
             else:
                 raise TypeError(f"{type(val)} is not convertible to datetime")
 
+            cnp.PyArray_MultiIter_NEXT(mi)
+
         except OutOfBoundsDatetime as ex:
             ex.args = (f"{ex}, at position {i}",)
             if is_coerce:
                 iresult[i] = NPY_NAT
+                cnp.PyArray_MultiIter_NEXT(mi)
                 continue
             elif is_raise:
                 raise
@@ -627,6 +635,7 @@ cpdef array_to_datetime(
             ex.args = (f"{ex}, at position {i}",)
             if is_coerce:
                 iresult[i] = NPY_NAT
+                cnp.PyArray_MultiIter_NEXT(mi)
                 continue
             elif is_raise:
                 raise
@@ -650,7 +659,7 @@ cpdef array_to_datetime(
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-cdef ndarray[object] ignore_errors_out_of_bounds_fallback(ndarray[object] values):
+cdef ndarray ignore_errors_out_of_bounds_fallback(ndarray values):
     """
     Fallback for array_to_datetime if an OutOfBoundsDatetime is raised
     and errors == "ignore"
@@ -664,27 +673,36 @@ cdef ndarray[object] ignore_errors_out_of_bounds_fallback(ndarray[object] values
     ndarray[object]
     """
     cdef:
-        Py_ssize_t i, n = len(values)
+        Py_ssize_t i, n = values.size
         object val
+        cnp.broadcast mi
+        ndarray[object] oresult
+        ndarray oresult_nd
 
-    oresult = cnp.PyArray_EMPTY(values.ndim, values.shape, cnp.NPY_OBJECT, 0)
+    oresult_nd = cnp.PyArray_EMPTY(values.ndim, values.shape, cnp.NPY_OBJECT, 0)
+    mi = cnp.PyArray_MultiIterNew2(oresult_nd, values)
+    oresult = oresult_nd.ravel()
 
     for i in range(n):
-        val = values[i]
+        # Analogous to `val = values[i]`
+        val = <object>(<PyObject**>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
 
         # set as nan except if its a NaT
         if checknull_with_nat_and_na(val):
             if isinstance(val, float):
                 oresult[i] = np.nan
             else:
-                oresult[i] = NaT
+                oresult[i] = <object>NaT
         elif is_datetime64_object(val):
             if get_datetime64_value(val) == NPY_NAT:
-                oresult[i] = NaT
+                oresult[i] = <object>NaT
             else:
                 oresult[i] = val.item()
         else:
             oresult[i] = val
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
     return oresult
 
 
@@ -719,24 +737,30 @@ cdef _array_to_datetime_object(
     Literal[None]
     """
     cdef:
-        Py_ssize_t i, n = len(values)
+        Py_ssize_t i, n = values.size
         object val
         bint is_ignore = errors == "ignore"
         bint is_coerce = errors == "coerce"
         bint is_raise = errors == "raise"
+        ndarray oresult_nd
         ndarray[object] oresult
         npy_datetimestruct dts
+        cnp.broadcast mi
 
     assert is_raise or is_ignore or is_coerce
 
-    oresult = cnp.PyArray_EMPTY(values.ndim, values.shape, cnp.NPY_OBJECT, 0)
+    oresult_nd = cnp.PyArray_EMPTY(values.ndim, values.shape, cnp.NPY_OBJECT, 0)
+    mi = cnp.PyArray_MultiIterNew2(oresult_nd, values)
+    oresult = oresult_nd.ravel()
 
     # We return an object array and only attempt to parse:
     # 1) NaT or NaT-like values
     # 2) datetime strings, which we return as datetime.datetime
     # 3) special strings - "now" & "today"
     for i in range(n):
-        val = values[i]
+        # Analogous to: val = values[i]
+        val = <object>(<PyObject**>cnp.PyArray_MultiIter_DATA(mi, 1))[0]
+
         if checknull_with_nat_and_na(val) or PyDateTime_Check(val):
             # GH 25978. No need to parse NaT-like or datetime-like vals
             oresult[i] = val
@@ -747,6 +771,7 @@ cdef _array_to_datetime_object(
 
             if len(val) == 0 or val in nat_strings:
                 oresult[i] = "NaT"
+                cnp.PyArray_MultiIter_NEXT(mi)
                 continue
             try:
                 oresult[i] = parse_datetime_string(val, dayfirst=dayfirst,
@@ -757,6 +782,7 @@ cdef _array_to_datetime_object(
                 ex.args = (f"{ex}, at position {i}", )
                 if is_coerce:
                     oresult[i] = <object>NaT
+                    cnp.PyArray_MultiIter_NEXT(mi)
                     continue
                 if is_raise:
                     raise
@@ -765,7 +791,10 @@ cdef _array_to_datetime_object(
             if is_raise:
                 raise
             return values, None
-    return oresult, None
+
+        cnp.PyArray_MultiIter_NEXT(mi)
+
+    return oresult_nd, None
 
 
 def array_to_datetime_with_tz(ndarray values, tzinfo tz):

pandas/core/arrays/datetimes.py

@@ -9,7 +9,6 @@
 from typing import (
     TYPE_CHECKING,
     Iterator,
-    Literal,
     cast,
 )
 import warnings
@@ -2149,16 +2148,13 @@ def objects_to_datetime64ns(
     # if str-dtype, convert
     data = np.array(data, copy=False, dtype=np.object_)
 
-    flags = data.flags
-    order: Literal["F", "C"] = "F" if flags.f_contiguous else "C"
     result, tz_parsed = tslib.array_to_datetime(
-        data.ravel("K"),
+        data,
         errors=errors,
         utc=utc,
         dayfirst=dayfirst,
         yearfirst=yearfirst,
     )
-    result = result.reshape(data.shape, order=order)
 
     if tz_parsed is not None:
         # We can take a shortcut since the datetime64 numpy array