REF: Make Util functions accessible from C

pandas/_libs/algos.pyx

@@ -38,6 +38,12 @@ from numpy cimport (
 cnp.import_array()
 
 cimport pandas._libs.util as util
+
+
+cdef extern from "pandas/type.h":
+    bint is_integer_object(object obj)
+    bint is_array(object obj)
+
 from pandas._libs.dtypes cimport (
     numeric_object_t,
     numeric_t,
@@ -513,7 +519,7 @@ def validate_limit(nobs: int | None, limit=None) -> int:
     if limit is None:
         lim = nobs
     else:
-        if not util.is_integer_object(limit):
+        if not is_integer_object(limit):
             raise ValueError("Limit must be an integer")
         if limit < 1:
             raise ValueError("Limit must be greater than 0")

pandas/_libs/algos_common_helper.pxi.in

@@ -12,7 +12,7 @@ WARNING: DO NOT edit .pxi FILE directly, .pxi is generated from .pxi.in
 def ensure_platform_int(object arr):
     # GH3033, GH1392
     # platform int is the size of the int pointer, e.g. np.intp
-    if util.is_array(arr):
+    if is_array(arr):
         if (<ndarray>arr).descr.type_num == cnp.NPY_INTP:
             return arr
         else:
@@ -23,7 +23,7 @@ def ensure_platform_int(object arr):
 
 
 def ensure_object(object arr):
-    if util.is_array(arr):
+    if is_array(arr):
         if (<ndarray>arr).descr.type_num == NPY_OBJECT:
             return arr
         else:
@@ -61,7 +61,7 @@ def get_dispatch(dtypes):
 
 
 def ensure_{{name}}(object arr, copy=True):
-    if util.is_array(arr):
+    if is_array(arr):
         if (<ndarray>arr).descr.type_num == NPY_{{c_type}}:
             return arr
         else:

pandas/_libs/hashing.pyx

@@ -18,7 +18,10 @@ from numpy cimport (
 
 import_array()
 
-from pandas._libs.util cimport is_nan
+cdef extern from "pandas/type.h":
+    bint is_nan(object obj)
+    bint is_datetime64_object(object obj)
+    bint is_integer_object(object obj)
 
 
 @cython.boundscheck(False)

pandas/_libs/index.pyx

@@ -14,7 +14,13 @@ from numpy cimport (
 cnp.import_array()
 
 
-from pandas._libs cimport util
+cdef extern from "pandas/type.h":
+    bint is_integer_object(object obj)
+    bint is_float_object(object obj)
+    bint is_bool_object(object obj)
+    bint is_complex_object(object obj)
+    bint is_nan(object obj)
+
 from pandas._libs.hashtable cimport HashTable
 from pandas._libs.tslibs.nattype cimport c_NaT as NaT
 from pandas._libs.tslibs.np_datetime cimport (
@@ -74,7 +80,7 @@ cdef ndarray _get_bool_indexer(ndarray values, object val):
                 indexer[i] = is_matching_na(item, val)
 
     else:
-        if util.is_nan(val):
+        if is_nan(val):
             indexer = np.isnan(values)
         else:
             indexer = values == val
@@ -836,7 +842,7 @@ include "index_class_helper.pxi"
 
 cdef class BoolEngine(UInt8Engine):
     cdef _check_type(self, object val):
-        if not util.is_bool_object(val):
+        if not is_bool_object(val):
             raise KeyError(val)
         return <uint8_t>val
 
@@ -994,7 +1000,7 @@ cdef class SharedEngine:
             except KeyError:
                 loc = -1
             else:
-                assert util.is_integer_object(loc), (loc, val)
+                assert is_integer_object(loc), (loc, val)
             res[i] = loc
 
         return res
@@ -1032,7 +1038,7 @@ cdef class SharedEngine:
                 if isinstance(locs, slice):
                     # Only needed for get_indexer_non_unique
                     locs = np.arange(locs.start, locs.stop, locs.step, dtype=np.intp)
-                elif util.is_integer_object(locs):
+                elif is_integer_object(locs):
                     locs = np.array([locs], dtype=np.intp)
                 else:
                     assert locs.dtype.kind == "b"

pandas/_libs/index_class_helper.pxi.in

@@ -36,8 +36,8 @@ cdef class {{name}}Engine(IndexEngine):
 
     cdef _check_type(self, object val):
     {{if name not in {'Float64', 'Float32', 'Complex64', 'Complex128'} }}
-        if not util.is_integer_object(val):
-            if util.is_float_object(val):
+        if not is_integer_object(val):
+            if is_float_object(val):
                 # Make sure Int64Index.get_loc(2.0) works
                 if val.is_integer():
                     return int(val)
@@ -48,13 +48,13 @@ cdef class {{name}}Engine(IndexEngine):
             raise KeyError(val)
         {{endif}}
     {{elif name not in {'Complex64', 'Complex128'} }}
-        if not util.is_integer_object(val) and not util.is_float_object(val):
+        if not is_integer_object(val) and not is_float_object(val):
             # in particular catch bool and avoid casting True -> 1.0
             raise KeyError(val)
     {{else}}
-        if (not util.is_integer_object(val)
-            and not util.is_float_object(val)
-            and not util.is_complex_object(val)
+        if (not is_integer_object(val)
+            and not is_float_object(val)
+            and not is_complex_object(val)
         ):
             # in particular catch bool and avoid casting True -> 1.0
             raise KeyError(val)

pandas/_libs/interval.pyx

@@ -31,7 +31,6 @@ from numpy cimport (
 cnp.import_array()
 
 
-from pandas._libs cimport util
 from pandas._libs.hashtable cimport Int64Vector
 from pandas._libs.tslibs.timedeltas cimport _Timedelta
 from pandas._libs.tslibs.timestamps cimport _Timestamp
@@ -42,6 +41,12 @@ from pandas._libs.tslibs.util cimport (
     is_timedelta64_object,
 )
 
+
+cdef extern from "pandas/type.h":
+    bint is_array(object obj)
+    bint is_nan(object obj)
+
+
 VALID_CLOSED = frozenset(["left", "right", "both", "neither"])
 
 
@@ -360,7 +365,7 @@ cdef class Interval(IntervalMixin):
             self_tuple = (self.left, self.right, self.closed)
             other_tuple = (other.left, other.right, other.closed)
             return PyObject_RichCompare(self_tuple, other_tuple, op)
-        elif util.is_array(other):
+        elif is_array(other):
             return np.array(
                 [PyObject_RichCompare(self, x, op) for x in other],
                 dtype=bool,
@@ -551,7 +556,7 @@ def intervals_to_interval_bounds(ndarray intervals, bint validate_closed=True):
 
     for i in range(n):
         interval = intervals[i]
-        if interval is None or util.is_nan(interval):
+        if interval is None or is_nan(interval):
             left[i] = np.nan
             right[i] = np.nan
             continue

pandas/_libs/lib.pyx

@@ -92,6 +92,17 @@ cdef extern from "src/parse_helper.h":
     int floatify(object, float64_t *result, int *maybe_int) except -1
 
 from pandas._libs cimport util
+
+
+cdef extern from "pandas/type.h":
+    bint is_integer_object(object obj)
+    bint is_float_object(object obj)
+    bint is_bool_object(object obj)
+    bint is_datetime64_object(object obj)
+    bint is_timedelta64_object(object obj)
+    bint is_array(object obj)
+
+
 from pandas._libs.util cimport (
     INT64_MAX,
     INT64_MIN,

pandas/_libs/missing.pyx

@@ -4,6 +4,7 @@ from sys import maxsize
 
 cimport cython
 from cython cimport Py_ssize_t
+
 import numpy as np
 
 cimport numpy as cnp
@@ -17,6 +18,16 @@ from numpy cimport (
 cnp.import_array()
 
 from pandas._libs cimport util
+
+
+cdef extern from "pandas/type.h":
+    bint is_timedelta64_object(object obj)
+    bint is_float_object(object obj)
+    bint is_complex_object(object obj)
+    bint is_datetime64_object(object obj)
+    bint is_array(object obj)
+    bint is_nan(object obj)
+
 from pandas._libs.tslibs.nattype cimport (
     c_NaT as NaT,
     checknull_with_nat,
@@ -89,38 +100,38 @@ cpdef bint is_matching_na(object left, object right, bint nan_matches_none=False
     bool
     """
     if left is None:
-        if nan_matches_none and util.is_nan(right):
+        if nan_matches_none and is_nan(right):
             return True
         return right is None
     elif left is C_NA:
         return right is C_NA
     elif left is NaT:
         return right is NaT
-    elif util.is_float_object(left):
-        if nan_matches_none and right is None and util.is_nan(left):
+    elif is_float_object(left):
+        if nan_matches_none and right is None and is_nan(left):
             return True
         return (
-            util.is_nan(left)
-            and util.is_float_object(right)
-            and util.is_nan(right)
+            is_nan(left)
+            and is_float_object(right)
+            and is_nan(right)
         )
-    elif util.is_complex_object(left):
+    elif is_complex_object(left):
         return (
-            util.is_nan(left)
-            and util.is_complex_object(right)
-            and util.is_nan(right)
+            is_nan(left)
+            and is_complex_object(right)
+            and is_nan(right)
         )
-    elif util.is_datetime64_object(left):
+    elif is_datetime64_object(left):
         return (
             get_datetime64_value(left) == NPY_NAT
-            and util.is_datetime64_object(right)
+            and is_datetime64_object(right)
             and get_datetime64_value(right) == NPY_NAT
             and get_datetime64_unit(left) == get_datetime64_unit(right)
         )
-    elif util.is_timedelta64_object(left):
+    elif is_timedelta64_object(left):
         return (
             get_timedelta64_value(left) == NPY_NAT
-            and util.is_timedelta64_object(right)
+            and is_timedelta64_object(right)
             and get_timedelta64_value(right) == NPY_NAT
             and get_datetime64_unit(left) == get_datetime64_unit(right)
         )
@@ -153,15 +164,15 @@ cpdef bint checknull(object val, bint inf_as_na=False):
     """
     if val is None or val is NaT or val is C_NA:
         return True
-    elif util.is_float_object(val) or util.is_complex_object(val):
+    elif is_float_object(val) or is_complex_object(val):
         if val != val:
             return True
         elif inf_as_na:
             return val == INF or val == NEGINF
         return False
-    elif util.is_timedelta64_object(val):
+    elif is_timedelta64_object(val):
         return get_timedelta64_value(val) == NPY_NAT
-    elif util.is_datetime64_object(val):
+    elif is_datetime64_object(val):
         return get_datetime64_value(val) == NPY_NAT
     else:
         return is_decimal_na(val)
@@ -251,11 +262,11 @@ def isnaobj2d(arr: ndarray, inf_as_na: bool = False) -> ndarray:
 
 
 def isposinf_scalar(val: object) -> bool:
-    return util.is_float_object(val) and val == INF
+    return is_float_object(val) and val == INF
 
 
 def isneginf_scalar(val: object) -> bool:
-    return util.is_float_object(val) and val == NEGINF
+    return is_float_object(val) and val == NEGINF
 
 
 cdef bint is_null_datetime64(v):
@@ -299,7 +310,7 @@ def is_float_nan(values: ndarray) -> ndarray:
 
     for i in range(N):
         val = values[i]
-        if util.is_nan(val):
+        if is_nan(val):
             result[i] = True
     return result.view(bool)
 
@@ -327,7 +338,7 @@ def is_numeric_na(values: ndarray) -> ndarray:
     for i in range(N):
         val = values[i]
         if checknull(val):
-            if val is None or val is C_NA or util.is_nan(val) or is_decimal_na(val):
+            if val is None or val is C_NA or is_nan(val) or is_decimal_na(val):
                 result[i] = True
             else:
                 raise TypeError(f"'values' contains non-numeric NA {val}")
@@ -343,7 +354,7 @@ def _create_binary_propagating_op(name, is_divmod=False):
     def method(self, other):
         if (other is C_NA or isinstance(other, (str, bytes))
                 or isinstance(other, (numbers.Number, np.bool_))
-                or util.is_array(other) and not other.shape):
+                or is_array(other) and not other.shape):
             # Need the other.shape clause to handle NumPy scalars,
             # since we do a setitem on `out` below, which
             # won't work for NumPy scalars.
@@ -352,7 +363,7 @@ def _create_binary_propagating_op(name, is_divmod=False):
             else:
                 return NA
 
-        elif util.is_array(other):
+        elif is_array(other):
             out = np.empty(other.shape, dtype=object)
             out[:] = NA
 
@@ -464,7 +475,7 @@ class NAType(C_NAType):
                 return type(other)(1)
             else:
                 return NA
-        elif util.is_array(other):
+        elif is_array(other):
             return np.where(other == 0, other.dtype.type(1), NA)
 
         return NotImplemented
@@ -477,7 +488,7 @@ class NAType(C_NAType):
                 return other
             else:
                 return NA
-        elif util.is_array(other):
+        elif is_array(other):
             return np.where(other == 1, other, NA)
         return NotImplemented