CLN: assorted cleanups (#29314)

Author: jbrockmendel

pandas/_libs/algos.pyx

@@ -379,28 +379,34 @@ ctypedef fused algos_t:
     uint8_t
 
 
+def _validate_limit(nobs: int, limit=None) -> int:
+    if limit is None:
+        lim = nobs
+    else:
+        if not util.is_integer_object(limit):
+            raise ValueError('Limit must be an integer')
+        if limit < 1:
+            raise ValueError('Limit must be greater than 0')
+        lim = limit
+
+    return lim
+
+
 @cython.boundscheck(False)
 @cython.wraparound(False)
 def pad(ndarray[algos_t] old, ndarray[algos_t] new, limit=None):
     cdef:
         Py_ssize_t i, j, nleft, nright
         ndarray[int64_t, ndim=1] indexer
-        algos_t cur, next
+        algos_t cur, next_val
         int lim, fill_count = 0
 
     nleft = len(old)
     nright = len(new)
     indexer = np.empty(nright, dtype=np.int64)
     indexer[:] = -1
 
-    if limit is None:
-        lim = nright
-    else:
-        if not util.is_integer_object(limit):
-            raise ValueError('Limit must be an integer')
-        if limit < 1:
-            raise ValueError('Limit must be greater than 0')
-        lim = limit
+    lim = _validate_limit(nright, limit)
 
     if nleft == 0 or nright == 0 or new[nright - 1] < old[0]:
         return indexer
@@ -426,9 +432,9 @@ def pad(ndarray[algos_t] old, ndarray[algos_t] new, limit=None):
                 j += 1
             break
 
-        next = old[i + 1]
+        next_val = old[i + 1]
 
-        while j < nright and cur <= new[j] < next:
+        while j < nright and cur <= new[j] < next_val:
             if new[j] == cur:
                 indexer[j] = i
             elif fill_count < lim:
@@ -438,16 +444,14 @@ def pad(ndarray[algos_t] old, ndarray[algos_t] new, limit=None):
 
         fill_count = 0
         i += 1
-        cur = next
+        cur = next_val
 
     return indexer
 
 
 @cython.boundscheck(False)
 @cython.wraparound(False)
-def pad_inplace(algos_t[:] values,
-                const uint8_t[:] mask,
-                limit=None):
+def pad_inplace(algos_t[:] values, const uint8_t[:] mask, limit=None):
     cdef:
         Py_ssize_t i, N
         algos_t val
@@ -459,14 +463,7 @@ def pad_inplace(algos_t[:] values,
     if N == 0:
         return
 
-    if limit is None:
-        lim = N
-    else:
-        if not util.is_integer_object(limit):
-            raise ValueError('Limit must be an integer')
-        if limit < 1:
-            raise ValueError('Limit must be greater than 0')
-        lim = limit
+    lim = _validate_limit(N, limit)
 
     val = values[0]
     for i in range(N):
@@ -482,9 +479,7 @@ def pad_inplace(algos_t[:] values,
 
 @cython.boundscheck(False)
 @cython.wraparound(False)
-def pad_2d_inplace(algos_t[:, :] values,
-                   const uint8_t[:, :] mask,
-                   limit=None):
+def pad_2d_inplace(algos_t[:, :] values, const uint8_t[:, :] mask, limit=None):
     cdef:
         Py_ssize_t i, j, N, K
         algos_t val
@@ -496,14 +491,7 @@ def pad_2d_inplace(algos_t[:, :] values,
     if N == 0:
         return
 
-    if limit is None:
-        lim = N
-    else:
-        if not util.is_integer_object(limit):
-            raise ValueError('Limit must be an integer')
-        if limit < 1:
-            raise ValueError('Limit must be greater than 0')
-        lim = limit
+    lim = _validate_limit(N, limit)
 
     for j in range(K):
         fill_count = 0
@@ -559,14 +547,7 @@ def backfill(ndarray[algos_t] old, ndarray[algos_t] new, limit=None):
     indexer = np.empty(nright, dtype=np.int64)
     indexer[:] = -1
 
-    if limit is None:
-        lim = nright
-    else:
-        if not util.is_integer_object(limit):
-            raise ValueError('Limit must be an integer')
-        if limit < 1:
-            raise ValueError('Limit must be greater than 0')
-        lim = limit
+    lim = _validate_limit(nright, limit)
 
     if nleft == 0 or nright == 0 or new[0] > old[nleft - 1]:
         return indexer
@@ -612,9 +593,7 @@ def backfill(ndarray[algos_t] old, ndarray[algos_t] new, limit=None):
 
 @cython.boundscheck(False)
 @cython.wraparound(False)
-def backfill_inplace(algos_t[:] values,
-                     const uint8_t[:] mask,
-                     limit=None):
+def backfill_inplace(algos_t[:] values, const uint8_t[:] mask, limit=None):
     cdef:
         Py_ssize_t i, N
         algos_t val
@@ -626,14 +605,7 @@ def backfill_inplace(algos_t[:] values,
     if N == 0:
         return
 
-    if limit is None:
-        lim = N
-    else:
-        if not util.is_integer_object(limit):
-            raise ValueError('Limit must be an integer')
-        if limit < 1:
-            raise ValueError('Limit must be greater than 0')
-        lim = limit
+    lim = _validate_limit(N, limit)
 
     val = values[N - 1]
     for i in range(N - 1, -1, -1):
@@ -663,14 +635,7 @@ def backfill_2d_inplace(algos_t[:, :] values,
     if N == 0:
         return
 
-    if limit is None:
-        lim = N
-    else:
-        if not util.is_integer_object(limit):
-            raise ValueError('Limit must be an integer')
-        if limit < 1:
-            raise ValueError('Limit must be greater than 0')
-        lim = limit
+    lim = _validate_limit(N, limit)
 
     for j in range(K):
         fill_count = 0

pandas/_libs/algos_common_helper.pxi.in

@@ -23,7 +23,7 @@ def diff_2d(ndarray[diff_t, ndim=2] arr,
             ndarray[out_t, ndim=2] out,
             Py_ssize_t periods, int axis):
     cdef:
-        Py_ssize_t i, j, sx, sy
+        Py_ssize_t i, j, sx, sy, start, stop
 
     # Disable for unsupported dtype combinations,
     #  see https://github.com/cython/cython/issues/2646

pandas/_libs/groupby.pxd

@@ -27,6 +27,13 @@ _int64_max = np.iinfo(np.int64).max
 
 cdef float64_t NaN = <float64_t>np.NaN
 
+cdef enum InterpolationEnumType:
+    INTERPOLATION_LINEAR,
+    INTERPOLATION_LOWER,
+    INTERPOLATION_HIGHER,
+    INTERPOLATION_NEAREST,
+    INTERPOLATION_MIDPOINT
+
 
 cdef inline float64_t median_linear(float64_t* a, int n) nogil:
     cdef:

pandas/_libs/groupby.pyx

@@ -698,8 +698,7 @@ def generate_bins_dt64(ndarray[int64_t] values, const int64_t[:] binner,
 
 @cython.boundscheck(False)
 @cython.wraparound(False)
-def get_level_sorter(const int64_t[:] label,
-                     const int64_t[:] starts):
+def get_level_sorter(const int64_t[:] label, const int64_t[:] starts):
     """
     argsort for a single level of a multi-index, keeping the order of higher
     levels unchanged. `starts` points to starts of same-key indices w.r.t
@@ -1677,6 +1676,7 @@ cpdef bint is_datetime64_array(ndarray values):
     return validator.validate(values)
 
 
+# TODO: only non-here use is in test
 def is_datetime_with_singletz_array(values: ndarray) -> bool:
     """
     Check values have the same tzinfo attribute.
@@ -1720,6 +1720,7 @@ cdef class AnyTimedeltaValidator(TimedeltaValidator):
         return is_timedelta(value)
 
 
+# TODO: only non-here use is in test
 cpdef bint is_timedelta_or_timedelta64_array(ndarray values):
     """ infer with timedeltas and/or nat/none """
     cdef:

pandas/_libs/lib.pyx

@@ -278,7 +278,7 @@ cdef class TextReader:
         object true_values, false_values
         object handle
         bint na_filter, keep_default_na, verbose, has_usecols, has_mi_columns
-        int64_t parser_start
+        uint64_t parser_start
         list clocks
         char *c_encoding
         kh_str_starts_t *false_set
@@ -710,11 +710,11 @@ cdef class TextReader:
         # header is now a list of lists, so field_count should use header[0]
 
         cdef:
-            Py_ssize_t i, start, field_count, passed_count, unnamed_count  # noqa
+            Py_ssize_t i, start, field_count, passed_count, unnamed_count
             char *word
             object name, old_name
             int status
-            int64_t hr, data_line
+            uint64_t hr, data_line
             char *errors = "strict"
             StringPath path = _string_path(self.c_encoding)
 
@@ -1015,12 +1015,14 @@ cdef class TextReader:
         else:
             end = min(start + rows, self.parser.lines)
 
+        # FIXME: dont leave commented-out
         # # skip footer
         # if footer > 0:
         #     end -= footer
 
         num_cols = -1
-        for i in range(self.parser.lines):
+        # Py_ssize_t cast prevents build warning
+        for i in range(<Py_ssize_t>self.parser.lines):
             num_cols = (num_cols < self.parser.line_fields[i]) * \
                 self.parser.line_fields[i] + \
                 (num_cols >= self.parser.line_fields[i]) * num_cols