Merge branch 'enh-consistency-interval-range' of https://github.com/weikhor/pandas into enh-consistency-interval-range

Author: chean.wei.khor

doc/source/whatsnew/v1.5.0.rst

@@ -342,6 +342,7 @@ Other Deprecations
 - Deprecated allowing non-keyword arguments in :meth:`ExtensionArray.argsort` (:issue:`46134`)
 - Deprecated treating all-bool ``object``-dtype columns as bool-like in :meth:`DataFrame.any` and :meth:`DataFrame.all` with ``bool_only=True``, explicitly cast to bool instead (:issue:`46188`)
 - Deprecated behavior of method :meth:`DataFrame.quantile`, attribute ``numeric_only`` will default False. Including datetime/timedelta columns in the result (:issue:`7308`).
+- Deprecated :attr:`Timedelta.freq` and :attr:`Timedelta.is_populated` (:issue:`46430`)
 - Deprecated the ``closed`` argument in :meth:`interval_range` in favor of ``inclusive`` argument; In a future version passing ``closed`` will raise (:issue:`40245`)
 -
 

pandas/_libs/tslibs/timedeltas.pxd

@@ -11,8 +11,7 @@ cdef bint is_any_td_scalar(object obj)
 cdef class _Timedelta(timedelta):
     cdef readonly:
         int64_t value      # nanoseconds
-        object freq        # frequency reference
-        bint is_populated  # are my components populated
+        bint _is_populated  # are my components populated
         int64_t _d, _h, _m, _s, _ms, _us, _ns
 
     cpdef timedelta to_pytimedelta(_Timedelta self)

pandas/_libs/tslibs/timedeltas.pyi

@@ -146,3 +146,7 @@ class Timedelta(timedelta):
     def __hash__(self) -> int: ...
     def isoformat(self) -> str: ...
     def to_numpy(self) -> np.timedelta64: ...
+    @property
+    def freq(self) -> None: ...
+    @property
+    def is_populated(self) -> bool: ...

pandas/_libs/tslibs/timedeltas.pyx

@@ -826,13 +826,32 @@ cdef _to_py_int_float(v):
 cdef class _Timedelta(timedelta):
     # cdef readonly:
     #    int64_t value      # nanoseconds
-    #    object freq        # frequency reference
-    #    bint is_populated  # are my components populated
+    #    bint _is_populated  # are my components populated
     #    int64_t _d, _h, _m, _s, _ms, _us, _ns
 
     # higher than np.ndarray and np.matrix
     __array_priority__ = 100
 
+    @property
+    def freq(self) -> None:
+        # GH#46430
+        warnings.warn(
+            "Timedelta.freq is deprecated and will be removed in a future version",
+            FutureWarning,
+            stacklevel=1,
+        )
+        return None
+
+    @property
+    def is_populated(self) -> bool:
+        # GH#46430
+        warnings.warn(
+            "Timedelta.is_populated is deprecated and will be removed in a future version",
+            FutureWarning,
+            stacklevel=1,
+        )
+        return self._is_populated
+
     def __hash__(_Timedelta self):
         if self._has_ns():
             return hash(self.value)
@@ -881,7 +900,7 @@ cdef class _Timedelta(timedelta):
         """
         compute the components
         """
-        if self.is_populated:
+        if self._is_populated:
             return
 
         cdef:
@@ -898,7 +917,7 @@ cdef class _Timedelta(timedelta):
         self._seconds = tds.seconds
         self._microseconds = tds.microseconds
 
-        self.is_populated = 1
+        self._is_populated = 1
 
     cpdef timedelta to_pytimedelta(_Timedelta self):
         """
@@ -1389,7 +1408,7 @@ class Timedelta(_Timedelta):
         # make timedelta happy
         td_base = _Timedelta.__new__(cls, microseconds=int(value) // 1000)
         td_base.value = value
-        td_base.is_populated = 0
+        td_base._is_populated = 0
         return td_base
 
     def __setstate__(self, state):

pandas/_libs/tslibs/tzconversion.pyx

@@ -15,7 +15,6 @@ from cpython.datetime cimport (
 
 import_datetime()
 
-from dateutil.tz import tzutc
 import numpy as np
 import pytz
 
@@ -40,7 +39,6 @@ from pandas._libs.tslibs.np_datetime cimport (
 )
 from pandas._libs.tslibs.timezones cimport (
     get_dst_info,
-    get_utcoffset,
     is_fixed_offset,
     is_tzlocal,
     is_utc,
@@ -194,6 +192,8 @@ timedelta-like}
     result_b[:] = NPY_NAT
 
     for i in range(n):
+        # This loops resembles the "Find the two best possibilities" block
+        #  in pytz's DstTZInfo.localize method.
         val = vals[i]
         if val == NPY_NAT:
             continue
@@ -339,27 +339,37 @@ cdef inline str _render_tstamp(int64_t val):
 
 cdef ndarray[int64_t] _get_dst_hours(
     # vals only needed here to potential render an exception message
-    ndarray[int64_t] vals,
+    const int64_t[:] vals,
     ndarray[int64_t] result_a,
     ndarray[int64_t] result_b,
 ):
     cdef:
-        Py_ssize_t n = vals.shape[0]
-        ndarray[uint8_t, cast=True] both_nat, both_eq
+        Py_ssize_t i, n = vals.shape[0]
+        ndarray[uint8_t, cast=True] mismatch
         ndarray[int64_t] delta, dst_hours
-        ndarray trans_idx, grp, a_idx, b_idx, one_diff
+        ndarray[intp_t] switch_idxs, trans_idx, grp, a_idx, b_idx, one_diff
         list trans_grp
+        intp_t switch_idx
+        int64_t left, right
 
     dst_hours = np.empty(n, dtype=np.int64)
     dst_hours[:] = NPY_NAT
 
-    # Get the ambiguous hours (given the above, these are the hours
-    # where result_a != result_b and neither of them are NAT)
-    both_nat = np.logical_and(result_a != NPY_NAT, result_b != NPY_NAT)
-    both_eq = result_a == result_b
-    trans_idx = np.squeeze(np.nonzero(np.logical_and(both_nat, ~both_eq)))
+    mismatch = np.zeros(n, dtype=bool)
+
+    for i in range(n):
+        left = result_a[i]
+        right = result_b[i]
+
+        # Get the ambiguous hours (given the above, these are the hours
+        # where result_a != result_b and neither of them are NAT)
+        if left != right and left != NPY_NAT and right != NPY_NAT:
+            mismatch[i] = 1
+
+    trans_idx = mismatch.nonzero()[0]
+
     if trans_idx.size == 1:
-        stamp = _render_tstamp(vals[trans_idx])
+        stamp = _render_tstamp(vals[trans_idx[0]])
         raise pytz.AmbiguousTimeError(
             f"Cannot infer dst time from {stamp} as there "
             "are no repeated times"
@@ -385,14 +395,14 @@ cdef ndarray[int64_t] _get_dst_hours(
 
             # Find the index for the switch and pull from a for dst and b
             # for standard
-            switch_idx = (delta <= 0).nonzero()[0]
-            if switch_idx.size > 1:
+            switch_idxs = (delta <= 0).nonzero()[0]
+            if switch_idxs.size > 1:
                 raise pytz.AmbiguousTimeError(
-                    f"There are {switch_idx.size} dst switches when "
+                    f"There are {switch_idxs.size} dst switches when "
                     "there should only be 1."
                 )
 
-            switch_idx = switch_idx[0] + 1  # TODO: declare type for switch_idx
+            switch_idx = switch_idxs[0] + 1
             # Pull the only index and adjust
             a_idx = grp[:switch_idx]
             b_idx = grp[switch_idx:]