Use proper initialization of offsets after pickling

Author: fjetter

pandas/_libs/properties.pyx

@@ -23,30 +23,30 @@ cdef class CachedProperty:
             return self
 
         # Get the cache or set a default one if needed
-        cache = getattr(obj, '_cache', None)
+        cache = instance_cache = getattr(obj, '_cache', None)
 
-        if cache is not None:
+        if cache is None:
+            try:
+                cache = obj._cache = {}
+            except (AttributeError):
+                raise TypeError(
+                    f"Cython extension type {type(obj)} must declare attribute "
+                    "`_cache` to use @cache_readonly."
+                )
+
+        if instance_cache is not None:
             # When accessing cython extension types, the attribute is already
             # registered and known to the class, unlike for python object. To
             # ensure we're not accidentally using a global scope / class level
             # cache we'll need to check whether the instance and class
             # attribute is identical
             cache_class = getattr(typ, "_cache", None)
-            if cache_class is not None and cache_class is cache:
+            if cache_class is not None and cache_class is instance_cache:
                 raise TypeError(
                     f"Class {typ} defines a `_cache` attribute on class level "
                     "which is forbidden in combination with @cache_readonly."
                 )
 
-        if cache is None:
-            try:
-                cache = obj._cache = {}
-            except (AttributeError):
-                raise TypeError(
-                    f"Cython extension type {type(obj)} must declare attribute "
-                    "`_cache` to use @cache_readonly."
-                )
-
         if PyDict_Contains(cache, self.name):
             # not necessary to Py_INCREF
             val = <object>PyDict_GetItem(cache, self.name)

pandas/_libs/tslibs/offsets.pyx

@@ -895,13 +895,6 @@ cdef class Tick(SingleConstructorOffset):
 
         raise ApplyTypeError(f"Unhandled type: {type(other).__name__}")
 
-    # --------------------------------------------------------------------
-    # Pickle Methods
-
-    def __setstate__(self, state):
-        self.n = state["n"]
-        self.normalize = False
-
 
 cdef class Day(Tick):
     _nanos_inc = 24 * 3600 * 1_000_000_000
@@ -2005,11 +1998,6 @@ cdef class QuarterOffset(SingleConstructorOffset):
             startingMonth = self._default_starting_month
         self.startingMonth = startingMonth
 
-    cpdef __setstate__(self, state):
-        self.startingMonth = state.pop("startingMonth")
-        self.n = state.pop("n")
-        self.normalize = state.pop("normalize")
-
     @classmethod
     def _from_name(cls, suffix=None):
         kwargs = {}
@@ -2265,11 +2253,6 @@ cdef class SemiMonthOffset(SingleConstructorOffset):
                 f"got {self.day_of_month}"
             )
 
-    cpdef __setstate__(self, state):
-        self.n = state.pop("n")
-        self.normalize = state.pop("normalize")
-        self.day_of_month = state.pop("day_of_month")
-
     @classmethod
     def _from_name(cls, suffix=None):
         return cls(day_of_month=suffix)
@@ -2587,12 +2570,6 @@ cdef class WeekOfMonth(WeekOfMonthMixin):
         if self.week < 0 or self.week > 3:
             raise ValueError(f"Week must be 0<=week<=3, got {self.week}")
 
-    cpdef __setstate__(self, state):
-        self.n = state.pop("n")
-        self.normalize = state.pop("normalize")
-        self.weekday = state.pop("weekday")
-        self.week = state.pop("week")
-
     def _get_offset_day(self, other: datetime) -> int:
         """
         Find the day in the same month as other that has the same
@@ -2652,12 +2629,6 @@ cdef class LastWeekOfMonth(WeekOfMonthMixin):
         if self.n == 0:
             raise ValueError("N cannot be 0")
 
-    cpdef __setstate__(self, state):
-        self.n = state.pop("n")
-        self.normalize = state.pop("normalize")
-        self.weekday = state.pop("weekday")
-        self.week = -1
-
     def _get_offset_day(self, other: datetime) -> int:
         """
         Find the day in the same month as other that has the same
@@ -2709,12 +2680,6 @@ cdef class FY5253Mixin(SingleConstructorOffset):
         if self.variation not in ["nearest", "last"]:
             raise ValueError(f"{self.variation} is not a valid variation")
 
-    cpdef __setstate__(self, state):
-        self.n = state.pop("n")
-        self.normalize = state.pop("normalize")
-        self.weekday = state.pop("weekday")
-        self.variation = state.pop("variation")
-
     def is_anchored(self) -> bool:
         return (
             self.n == 1 and self.startingMonth is not None and self.weekday is not None
@@ -2995,10 +2960,6 @@ cdef class FY5253Quarter(FY5253Mixin):
         )
         self.qtr_with_extra_week = qtr_with_extra_week
 
-    cpdef __setstate__(self, state):
-        FY5253Mixin.__setstate__(self, state)
-        self.qtr_with_extra_week = state.pop("qtr_with_extra_week")
-
     @cache_readonly
     def _offset(self):
         return FY5253(
@@ -3141,10 +3102,6 @@ cdef class Easter(SingleConstructorOffset):
     Right now uses the revised method which is valid in years 1583-4099.
     """
 
-    cpdef __setstate__(self, state):
-        self.n = state.pop("n")
-        self.normalize = state.pop("normalize")
-
     @apply_wraps
     def apply(self, other: datetime) -> datetime:
         current_easter = easter(other.year)

pandas/tests/libs/test_lib.py

@@ -217,25 +217,24 @@ def test_no_default_pickle():
 
 def test_cache_readonly_no_global_cache():
     """
-    Ensure that the definition of a global attribute `_cache` does not interfere with
+    Ensure that the definition of a global attribute `_cache` does
+    not interfere with the property cache
     """
 
     from typing import Dict
 
-    # Just declaring the type is sufficient for the attribute to be considered global
     class MyDtypeGlobalAttr:
-        _cache: Dict
+        _cache: Dict = {}
 
         def __init__(self, value):
             self._value = value
-            self._cache = {}
 
         @properties.cache_readonly
         def cached_attr(self):
             return self._value
 
     a = MyDtypeGlobalAttr("A")
-    with pytest.raises(TypeError, match=r"Class \w* defines a `_cache` " r"attribute"):
+    with pytest.raises(TypeError, match=r"defines a `_cache` attribute"):
         a.cached_attr
 
     class MyDtypeInstanceAttr:
@@ -251,7 +250,8 @@ def cached_attr(self):
     b = MyDtypeInstanceAttr("B")
     b._cache = {}
 
-    # Assert our assumption that the property cache actually uses an attribute called _cache
+    # Assert our assumption that the property cache
+    # actually uses an attribute called _cache
     assert len(a._cache) == 0
 
     assert a.cached_attr == "A"