CLN: some code cleanups in pandas/_libs/ (#31808)

* CLN: some code cleanups in pandas/_libs/

* Reverted "bint"

REF:
https://github.com/pandas-dev/pandas/pull/31808/files#r376721172

* Added trailing comma to imports

REF:
https://github.com/pandas-dev/pandas/pull/31808/files#r378523656

* Reverted bad code

* Lint issues

* Reverted wrong code

REF:
#31808 (comment)

* Removed parens

REF:
#31808 (comment)

* "in fmt" in prev line

REF:
#31808 (comment)

Author: ShaharNaveh

pandas/_libs/algos.pyx

@@ -7,13 +7,30 @@ from libc.math cimport fabs, sqrt
 
 import numpy as np
 cimport numpy as cnp
-from numpy cimport (ndarray,
-                    NPY_INT64, NPY_INT32, NPY_INT16, NPY_INT8,
-                    NPY_UINT64, NPY_UINT32, NPY_UINT16, NPY_UINT8,
-                    NPY_FLOAT32, NPY_FLOAT64,
-                    NPY_OBJECT,
-                    int8_t, int16_t, int32_t, int64_t, uint8_t, uint16_t,
-                    uint32_t, uint64_t, float32_t, float64_t)
+from numpy cimport (
+    NPY_FLOAT32,
+    NPY_FLOAT64,
+    NPY_INT8,
+    NPY_INT16,
+    NPY_INT32,
+    NPY_INT64,
+    NPY_OBJECT,
+    NPY_UINT8,
+    NPY_UINT16,
+    NPY_UINT32,
+    NPY_UINT64,
+    float32_t,
+    float64_t,
+    int8_t,
+    int16_t,
+    int32_t,
+    int64_t,
+    ndarray,
+    uint8_t,
+    uint16_t,
+    uint32_t,
+    uint64_t,
+)
 cnp.import_array()
 
 

pandas/_libs/join.pyx

@@ -3,13 +3,25 @@ from cython import Py_ssize_t
 
 import numpy as np
 cimport numpy as cnp
-from numpy cimport (ndarray,
-                    int8_t, int16_t, int32_t, int64_t, uint8_t, uint16_t,
-                    uint32_t, uint64_t, float32_t, float64_t)
+from numpy cimport (
+    float32_t,
+    float64_t,
+    int8_t,
+    int16_t,
+    int32_t,
+    int64_t,
+    ndarray,
+    uint8_t,
+    uint16_t,
+    uint32_t,
+    uint64_t,
+)
 cnp.import_array()
 
 from pandas._libs.algos import (
-    groupsort_indexer, ensure_platform_int, take_1d_int64_int64
+    ensure_platform_int,
+    groupsort_indexer,
+    take_1d_int64_int64,
 )
 
 

pandas/_libs/lib.pyx

@@ -15,18 +15,33 @@ from cpython.iterator cimport PyIter_Check
 from cpython.sequence cimport PySequence_Check
 from cpython.number cimport PyNumber_Check
 
-from cpython.datetime cimport (PyDateTime_Check, PyDate_Check,
-                               PyTime_Check, PyDelta_Check,
-                               PyDateTime_IMPORT)
+from cpython.datetime cimport (
+    PyDateTime_Check,
+    PyDate_Check,
+    PyTime_Check,
+    PyDelta_Check,
+    PyDateTime_IMPORT,
+)
 PyDateTime_IMPORT
 
 import numpy as np
 cimport numpy as cnp
-from numpy cimport (ndarray, PyArray_Check, PyArray_GETITEM,
-                    PyArray_ITER_DATA, PyArray_ITER_NEXT, PyArray_IterNew,
-                    flatiter, NPY_OBJECT,
-                    int64_t, float32_t, float64_t,
-                    uint8_t, uint64_t, complex128_t)
+from numpy cimport (
+    NPY_OBJECT,
+    PyArray_Check,
+    PyArray_GETITEM,
+    PyArray_ITER_DATA,
+    PyArray_ITER_NEXT,
+    PyArray_IterNew,
+    complex128_t,
+    flatiter,
+    float32_t,
+    float64_t,
+    int64_t,
+    ndarray,
+    uint8_t,
+    uint64_t,
+)
 cnp.import_array()
 
 cdef extern from "numpy/arrayobject.h":
@@ -60,7 +75,12 @@ from pandas._libs.tslibs.timedeltas cimport convert_to_timedelta64
 from pandas._libs.tslibs.timezones cimport get_timezone, tz_compare
 
 from pandas._libs.missing cimport (
-    checknull, isnaobj, is_null_datetime64, is_null_timedelta64, is_null_period, C_NA
+    checknull,
+    isnaobj,
+    is_null_datetime64,
+    is_null_timedelta64,
+    is_null_period,
+    C_NA,
 )
 
 
@@ -246,7 +266,7 @@ def item_from_zerodim(val: object) -> object:
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def fast_unique_multiple(list arrays, sort: bool=True):
+def fast_unique_multiple(list arrays, sort: bool = True):
     """
     Generate a list of unique values from a list of arrays.
 
@@ -277,6 +297,7 @@ def fast_unique_multiple(list arrays, sort: bool=True):
             if val not in table:
                 table[val] = stub
                 uniques.append(val)
+
     if sort is None:
         try:
             uniques.sort()
@@ -289,7 +310,7 @@ def fast_unique_multiple(list arrays, sort: bool=True):
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def fast_unique_multiple_list(lists: list, sort: bool=True) -> list:
+def fast_unique_multiple_list(lists: list, sort: bool = True) -> list:
     cdef:
         list buf
         Py_ssize_t k = len(lists)

pandas/_libs/reshape.pyx

@@ -1,8 +1,20 @@
 import cython
 from cython import Py_ssize_t
 
-from numpy cimport (int8_t, int16_t, int32_t, int64_t, uint8_t, uint16_t,
-                    uint32_t, uint64_t, float32_t, float64_t, ndarray)
+from numpy cimport (
+    float32_t,
+    float64_t,
+    int8_t,
+    int16_t,
+    int32_t,
+    int64_t,
+    ndarray,
+    uint8_t,
+    uint16_t,
+    uint32_t,
+    uint64_t,
+)
+
 cimport numpy as cnp
 import numpy as np
 from pandas._libs.lib cimport c_is_list_like

pandas/_libs/sparse.pyx

@@ -448,7 +448,7 @@ cdef class BlockIndex(SparseIndex):
         ylen = y.blengths
 
         # block may be split, but can't exceed original len / 2 + 1
-        max_len = int(min(self.length, y.length) / 2) + 1
+        max_len = min(self.length, y.length) // 2 + 1
         out_bloc = np.empty(max_len, dtype=np.int32)
         out_blen = np.empty(max_len, dtype=np.int32)
 
@@ -672,7 +672,7 @@ cdef class BlockUnion(BlockMerge):
         ystart = self.ystart
         yend = self.yend
 
-        max_len = int(min(self.x.length, self.y.length) / 2) + 1
+        max_len = min(self.x.length, self.y.length) // 2 + 1
         out_bloc = np.empty(max_len, dtype=np.int32)
         out_blen = np.empty(max_len, dtype=np.int32)
 

pandas/_libs/tslibs/period.pyx

@@ -1,8 +1,6 @@
 from datetime import datetime
 
-from cpython.object cimport (
-    PyObject_RichCompareBool,
-    Py_EQ, Py_NE)
+from cpython.object cimport PyObject_RichCompareBool, Py_EQ, Py_NE
 
 from numpy cimport int64_t, import_array, ndarray
 import numpy as np
@@ -14,15 +12,25 @@ from libc.string cimport strlen, memset
 
 import cython
 
-from cpython.datetime cimport (PyDateTime_Check, PyDelta_Check, PyDate_Check,
-                               PyDateTime_IMPORT)
+from cpython.datetime cimport (
+    PyDate_Check,
+    PyDateTime_Check,
+    PyDateTime_IMPORT,
+    PyDelta_Check,
+)
 # import datetime C API
 PyDateTime_IMPORT
 
 from pandas._libs.tslibs.np_datetime cimport (
-    npy_datetimestruct, dtstruct_to_dt64, dt64_to_dtstruct,
-    pandas_datetime_to_datetimestruct, check_dts_bounds,
-    NPY_DATETIMEUNIT, NPY_FR_D, NPY_FR_us)
+    npy_datetimestruct,
+    dtstruct_to_dt64,
+    dt64_to_dtstruct,
+    pandas_datetime_to_datetimestruct,
+    check_dts_bounds,
+    NPY_DATETIMEUNIT,
+    NPY_FR_D,
+    NPY_FR_us,
+)
 
 cdef extern from "src/datetime/np_datetime.h":
     int64_t npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT fr,
@@ -37,12 +45,15 @@ from pandas._libs.tslibs.timedeltas import Timedelta
 from pandas._libs.tslibs.timedeltas cimport delta_to_nanoseconds
 
 cimport pandas._libs.tslibs.ccalendar as ccalendar
-from pandas._libs.tslibs.ccalendar cimport (
-    dayofweek, get_day_of_year, is_leapyear)
+from pandas._libs.tslibs.ccalendar cimport dayofweek, get_day_of_year, is_leapyear
 from pandas._libs.tslibs.ccalendar import MONTH_NUMBERS
 from pandas._libs.tslibs.frequencies cimport (
-    get_freq_code, get_base_alias, get_to_timestamp_base, get_freq_str,
-    get_rule_month)
+    get_base_alias,
+    get_freq_code,
+    get_freq_str,
+    get_rule_month,
+    get_to_timestamp_base,
+)
 from pandas._libs.tslibs.parsing import parse_time_string
 from pandas._libs.tslibs.resolution import Resolution
 from pandas._libs.tslibs.nattype import nat_strings
@@ -55,7 +66,7 @@ from pandas._libs.tslibs.tzconversion cimport tz_convert_utc_to_tzlocal
 
 cdef:
     enum:
-        INT32_MIN = -2147483648
+        INT32_MIN = -2_147_483_648
 
 
 ctypedef struct asfreq_info:
@@ -179,8 +190,7 @@ cdef freq_conv_func get_asfreq_func(int from_freq, int to_freq) nogil:
             return <freq_conv_func>asfreq_MtoB
         elif from_group == FR_WK:
             return <freq_conv_func>asfreq_WtoB
-        elif from_group in [FR_DAY, FR_HR, FR_MIN, FR_SEC,
-                            FR_MS, FR_US, FR_NS]:
+        elif from_group in [FR_DAY, FR_HR, FR_MIN, FR_SEC, FR_MS, FR_US, FR_NS]:
             return <freq_conv_func>asfreq_DTtoB
         else:
             return <freq_conv_func>nofunc
@@ -289,17 +299,15 @@ cdef int64_t DtoB(npy_datetimestruct *dts, int roll_back,
     return DtoB_weekday(unix_date)
 
 
-cdef inline int64_t upsample_daytime(int64_t ordinal,
-                                     asfreq_info *af_info) nogil:
-    if (af_info.is_end):
+cdef inline int64_t upsample_daytime(int64_t ordinal, asfreq_info *af_info) nogil:
+    if af_info.is_end:
         return (ordinal + 1) * af_info.intraday_conversion_factor - 1
     else:
         return ordinal * af_info.intraday_conversion_factor
 
 
-cdef inline int64_t downsample_daytime(int64_t ordinal,
-                                       asfreq_info *af_info) nogil:
-    return ordinal // (af_info.intraday_conversion_factor)
+cdef inline int64_t downsample_daytime(int64_t ordinal, asfreq_info *af_info) nogil:
+    return ordinal // af_info.intraday_conversion_factor
 
 
 cdef inline int64_t transform_via_day(int64_t ordinal,
@@ -1464,24 +1472,24 @@ def extract_freq(ndarray[object] values):
 
     cdef:
         Py_ssize_t i, n = len(values)
-        object p
+        object value
 
     for i in range(n):
-        p = values[i]
+        value = values[i]
 
         try:
             # now Timestamp / NaT has freq attr
-            if is_period_object(p):
-                return p.freq
+            if is_period_object(value):
+                return value.freq
         except AttributeError:
             pass
 
     raise ValueError('freq not specified and cannot be inferred')
 
-
 # -----------------------------------------------------------------------
 # period helpers
 
+
 @cython.wraparound(False)
 @cython.boundscheck(False)
 cdef int64_t[:] localize_dt64arr_to_period(const int64_t[:] stamps,

pandas/_libs/tslibs/strptime.pyx

@@ -45,8 +45,7 @@ cdef dict _parse_code_table = {'y': 0,
                                'u': 22}
 
 
-def array_strptime(object[:] values, object fmt,
-                   bint exact=True, errors='raise'):
+def array_strptime(object[:] values, object fmt, bint exact=True, errors='raise'):
     """
     Calculates the datetime structs represented by the passed array of strings
 
@@ -78,12 +77,9 @@ def array_strptime(object[:] values, object fmt,
     if fmt is not None:
         if '%W' in fmt or '%U' in fmt:
             if '%Y' not in fmt and '%y' not in fmt:
-                raise ValueError("Cannot use '%W' or '%U' without "
-                                 "day and year")
-            if ('%A' not in fmt and '%a' not in fmt and '%w' not
-                    in fmt):
-                raise ValueError("Cannot use '%W' or '%U' without "
-                                 "day and year")
+                raise ValueError("Cannot use '%W' or '%U' without day and year")
+            if '%A' not in fmt and '%a' not in fmt and '%w' not in fmt:
+                raise ValueError("Cannot use '%W' or '%U' without day and year")
         elif '%Z' in fmt and '%z' in fmt:
             raise ValueError("Cannot parse both %Z and %z")
 
@@ -749,6 +745,6 @@ cdef parse_timezone_directive(str z):
     microseconds = int(gmtoff_remainder + gmtoff_remainder_padding)
 
     total_minutes = ((hours * 60) + minutes + (seconds // 60) +
-                     (microseconds // 60000000))
+                     (microseconds // 60_000_000))
     total_minutes = -total_minutes if z.startswith("-") else total_minutes
     return pytz.FixedOffset(total_minutes)

pandas/_libs/tslibs/timezones.pyx

@@ -196,7 +196,7 @@ cdef int64_t[:] unbox_utcoffsets(object transinfo):
     arr = np.empty(sz, dtype='i8')
 
     for i in range(sz):
-        arr[i] = int(transinfo[i][0].total_seconds()) * 1000000000
+        arr[i] = int(transinfo[i][0].total_seconds()) * 1_000_000_000
 
     return arr
 
@@ -217,7 +217,7 @@ cdef object get_dst_info(object tz):
     if cache_key is None:
         # e.g. pytz.FixedOffset, matplotlib.dates._UTC,
         # psycopg2.tz.FixedOffsetTimezone
-        num = int(get_utcoffset(tz, None).total_seconds()) * 1000000000
+        num = int(get_utcoffset(tz, None).total_seconds()) * 1_000_000_000
         return (np.array([NPY_NAT + 1], dtype=np.int64),
                 np.array([num], dtype=np.int64),
                 None)