Merge branch 'main' into DOC46504

Author: eshirvana

ci/code_checks.sh

@@ -78,8 +78,8 @@ fi
 ### DOCSTRINGS ###
 if [[ -z "$CHECK" || "$CHECK" == "docstrings" ]]; then
 
-    MSG='Validate docstrings (GL01, GL02, GL03, GL04, GL05, GL06, GL07, GL09, GL10, SS01, SS02, SS03, SS04, SS05, PR03, PR04, PR05, PR06, PR08, PR09, PR10, EX04, RT01, RT04, RT05, SA02, SA03)' ; echo $MSG
-    $BASE_DIR/scripts/validate_docstrings.py --format=actions --errors=GL01,GL02,GL03,GL04,GL05,GL06,GL07,GL09,GL10,SS02,SS03,SS04,SS05,PR03,PR04,PR05,PR06,PR08,PR09,PR10,EX04,RT01,RT04,RT05,SA02,SA03
+    MSG='Validate docstrings (EX04, GL01, GL02, GL03, GL04, GL05, GL06, GL07, GL09, GL10, PR03, PR04, PR05, PR06, PR08, PR09, PR10, RT01, RT04, RT05, SA02, SA03, SS01, SS02, SS03, SS04, SS05)' ; echo $MSG
+    $BASE_DIR/scripts/validate_docstrings.py --format=actions --errors=EX04,GL01,GL02,GL03,GL04,GL05,GL06,GL07,GL09,GL10,PR03,PR04,PR05,PR06,PR08,PR09,PR10,RT01,RT04,RT05,SA02,SA03,SS01,SS02,SS03,SS04,SS05
     RET=$(($RET + $?)) ; echo $MSG "DONE"
 
 fi

doc/source/getting_started/tutorials.rst

@@ -75,6 +75,16 @@ Excel charts with pandas, vincent and xlsxwriter
 
 *  `Using Pandas and XlsxWriter to create Excel charts <https://pandas-xlsxwriter-charts.readthedocs.io/>`_
 
+Joyful pandas
+-------------
+
+A tutorial written in Chinese by Yuanhao Geng. It covers the basic operations
+for NumPy and pandas, 4 main data manipulation methods (including indexing, groupby, reshaping
+and concatenation) and 4 main data types (including missing data, string data, categorical
+data and time series data). At the end of each chapter, corresponding exercises are posted.
+All the datasets and related materials can be found in the GitHub repository
+`datawhalechina/joyful-pandas <https://github.com/datawhalechina/joyful-pandas>`_.
+
 Video tutorials
 ---------------
 

doc/source/user_guide/duplicates.rst

@@ -172,7 +172,7 @@ going forward, to ensure that your data pipeline doesn't introduce duplicates.
    >>> deduplicated = raw.groupby(level=0).first()  # remove duplicates
    >>> deduplicated.flags.allows_duplicate_labels = False  # disallow going forward
 
-Setting ``allows_duplicate_labels=True`` on a ``Series`` or ``DataFrame`` with duplicate
+Setting ``allows_duplicate_labels=False`` on a ``Series`` or ``DataFrame`` with duplicate
 labels or performing an operation that introduces duplicate labels on a ``Series`` or
 ``DataFrame`` that disallows duplicates will raise an
 :class:`errors.DuplicateLabelError`.

doc/source/whatsnew/v1.5.0.rst

@@ -527,6 +527,7 @@ Groupby/resample/rolling
 - Bug in :meth:`GroupBy.cummin` and :meth:`GroupBy.cummax` with nullable dtypes incorrectly altering the original data in place (:issue:`46220`)
 - Bug in :meth:`GroupBy.cummax` with ``int64`` dtype with leading value being the smallest possible int64 (:issue:`46382`)
 - Bug in :meth:`GroupBy.max` with empty groups and ``uint64`` dtype incorrectly raising ``RuntimeError`` (:issue:`46408`)
+- Bug in :meth:`.GroupBy.apply` would fail when ``func`` was a string and args or kwargs were supplied (:issue:`46479`)
 -
 
 Reshaping

environment.yml

@@ -86,7 +86,7 @@ dependencies:
   - bottleneck>=1.3.1
   - ipykernel
   - ipython>=7.11.1
-  - jinja2  # pandas.Styler
+  - jinja2<=3.0.3  # pandas.Styler
   - matplotlib>=3.3.2  # pandas.plotting, Series.plot, DataFrame.plot
   - numexpr>=2.7.1
   - scipy>=1.4.1

pandas/_libs/tslibs/conversion.pxd

@@ -31,5 +31,3 @@ cdef int64_t get_datetime64_nanos(object val) except? -1
 cpdef datetime localize_pydatetime(datetime dt, tzinfo tz)
 cdef int64_t cast_from_unit(object ts, str unit) except? -1
 cpdef (int64_t, int) precision_from_unit(str unit)
-
-cdef int64_t normalize_i8_stamp(int64_t local_val) nogil

pandas/_libs/tslibs/conversion.pyx

@@ -72,6 +72,7 @@ from pandas._libs.tslibs.nattype cimport (
 )
 from pandas._libs.tslibs.tzconversion cimport (
     bisect_right_i8,
+    infer_datetuil_fold,
     localize_tzinfo_api,
     tz_localize_to_utc_single,
 )
@@ -530,7 +531,7 @@ cdef _TSObject _create_tsobject_tz_using_offset(npy_datetimestruct dts,
         if typ == 'dateutil':
             tdata = <int64_t*>cnp.PyArray_DATA(trans)
             pos = bisect_right_i8(tdata, obj.value, trans.shape[0]) - 1
-            obj.fold = _infer_tsobject_fold(obj, trans, deltas, pos)
+            obj.fold = infer_datetuil_fold(obj.value, trans, deltas, pos)
 
     # Keep the converter same as PyDateTime's
     dt = datetime(obj.dts.year, obj.dts.month, obj.dts.day,
@@ -714,7 +715,7 @@ cdef inline void _localize_tso(_TSObject obj, tzinfo tz):
             local_val = obj.value + deltas[pos]
 
             # dateutil supports fold, so we infer fold from value
-            obj.fold = _infer_tsobject_fold(obj, trans, deltas, pos)
+            obj.fold = infer_datetuil_fold(obj.value, trans, deltas, pos)
         else:
             # All other cases have len(deltas) == 1. As of 2018-07-17
             #  (and 2022-03-07), all test cases that get here have
@@ -726,49 +727,6 @@ cdef inline void _localize_tso(_TSObject obj, tzinfo tz):
     obj.tzinfo = tz
 
 
-cdef inline bint _infer_tsobject_fold(
-    _TSObject obj,
-    const int64_t[:] trans,
-    const int64_t[:] deltas,
-    intp_t pos,
-):
-    """
-    Infer _TSObject fold property from value by assuming 0 and then setting
-    to 1 if necessary.
-
-    Parameters
-    ----------
-    obj : _TSObject
-    trans : ndarray[int64_t]
-        ndarray of offset transition points in nanoseconds since epoch.
-    deltas : int64_t[:]
-        array of offsets corresponding to transition points in trans.
-    pos : intp_t
-        Position of the last transition point before taking fold into account.
-
-    Returns
-    -------
-    bint
-        Due to daylight saving time, one wall clock time can occur twice
-        when shifting from summer to winter time; fold describes whether the
-        datetime-like corresponds  to the first (0) or the second time (1)
-        the wall clock hits the ambiguous time
-
-    References
-    ----------
-    .. [1] "PEP 495 - Local Time Disambiguation"
-           https://www.python.org/dev/peps/pep-0495/#the-fold-attribute
-    """
-    cdef:
-        bint fold = 0
-
-    if pos > 0:
-        fold_delta = deltas[pos - 1] - deltas[pos]
-        if obj.value - fold_delta < trans[pos]:
-            fold = 1
-
-    return fold
-
 cdef inline datetime _localize_pydatetime(datetime dt, tzinfo tz):
     """
     Take a datetime/Timestamp in UTC and localizes to timezone tz.
@@ -802,24 +760,3 @@ cpdef inline datetime localize_pydatetime(datetime dt, tzinfo tz):
     elif isinstance(dt, ABCTimestamp):
         return dt.tz_localize(tz)
     return _localize_pydatetime(dt, tz)
-
-
-# ----------------------------------------------------------------------
-# Normalization
-
-@cython.cdivision(False)
-cdef inline int64_t normalize_i8_stamp(int64_t local_val) nogil:
-    """
-    Round the localized nanosecond timestamp down to the previous midnight.
-
-    Parameters
-    ----------
-    local_val : int64_t
-
-    Returns
-    -------
-    int64_t
-    """
-    cdef:
-        int64_t day_nanos = 24 * 3600 * 1_000_000_000
-    return local_val - (local_val % day_nanos)

pandas/_libs/tslibs/timestamps.pxd

@@ -28,3 +28,5 @@ cdef class _Timestamp(ABCTimestamp):
                                          int op) except -1
     cpdef void _set_freq(self, freq)
     cdef _warn_on_field_deprecation(_Timestamp self, freq, str field)
+
+cdef int64_t normalize_i8_stamp(int64_t local_val) nogil

pandas/_libs/tslibs/timestamps.pyx

@@ -51,7 +51,6 @@ from pandas._libs.tslibs.conversion cimport (
     _TSObject,
     convert_datetime_to_tsobject,
     convert_to_tsobject,
-    normalize_i8_stamp,
 )
 from pandas._libs.tslibs.util cimport (
     is_array,
@@ -2116,3 +2115,23 @@ cdef int64_t _NS_LOWER_BOUND = NPY_NAT + 1
 Timestamp.min = Timestamp(_NS_LOWER_BOUND)
 Timestamp.max = Timestamp(_NS_UPPER_BOUND)
 Timestamp.resolution = Timedelta(nanoseconds=1)  # GH#21336, GH#21365
+
+
+# ----------------------------------------------------------------------
+# Scalar analogues to functions in vectorized.pyx
+
+
+@cython.cdivision(False)
+cdef inline int64_t normalize_i8_stamp(int64_t local_val) nogil:
+    """
+    Round the localized nanosecond timestamp down to the previous midnight.
+
+    Parameters
+    ----------
+    local_val : int64_t
+
+    Returns
+    -------
+    int64_t
+    """
+    return local_val - (local_val % ccalendar.DAY_NANOS)

pandas/_libs/tslibs/tzconversion.pxd

@@ -1,5 +1,8 @@
 from cpython.datetime cimport tzinfo
-from numpy cimport int64_t
+from numpy cimport (
+    int64_t,
+    intp_t,
+)
 
 
 cdef int64_t localize_tzinfo_api(
@@ -11,3 +14,10 @@ cdef int64_t tz_localize_to_utc_single(
 ) except? -1
 
 cdef Py_ssize_t bisect_right_i8(int64_t *data, int64_t val, Py_ssize_t n)
+
+cdef bint infer_datetuil_fold(
+    int64_t value,
+    const int64_t[::1] trans,
+    const int64_t[::1] deltas,
+    intp_t pos,
+)

pandas/_libs/tslibs/tzconversion.pyx

@@ -632,3 +632,48 @@ cdef int64_t _tz_localize_using_tzinfo_api(
     td = tz.utcoffset(dt)
     delta = int(td.total_seconds() * 1_000_000_000)
     return delta
+
+
+# NB: relies on dateutil internals, subject to change.
+cdef bint infer_datetuil_fold(
+    int64_t value,
+    const int64_t[::1] trans,
+    const int64_t[::1] deltas,
+    intp_t pos,
+):
+    """
+    Infer _TSObject fold property from value by assuming 0 and then setting
+    to 1 if necessary.
+
+    Parameters
+    ----------
+    value : int64_t
+    trans : ndarray[int64_t]
+        ndarray of offset transition points in nanoseconds since epoch.
+    deltas : int64_t[:]
+        array of offsets corresponding to transition points in trans.
+    pos : intp_t
+        Position of the last transition point before taking fold into account.
+
+    Returns
+    -------
+    bint
+        Due to daylight saving time, one wall clock time can occur twice
+        when shifting from summer to winter time; fold describes whether the
+        datetime-like corresponds  to the first (0) or the second time (1)
+        the wall clock hits the ambiguous time
+
+    References
+    ----------
+    .. [1] "PEP 495 - Local Time Disambiguation"
+           https://www.python.org/dev/peps/pep-0495/#the-fold-attribute
+    """
+    cdef:
+        bint fold = 0
+
+    if pos > 0:
+        fold_delta = deltas[pos - 1] - deltas[pos]
+        if value - fold_delta < trans[pos]:
+            fold = 1
+
+    return fold

pandas/_libs/tslibs/vectorized.pyx

@@ -18,8 +18,6 @@ from numpy cimport (
 
 cnp.import_array()
 
-from .conversion cimport normalize_i8_stamp
-
 from .dtypes import Resolution
 
 from .ccalendar cimport DAY_NANOS
@@ -34,7 +32,10 @@ from .np_datetime cimport (
 )
 from .offsets cimport BaseOffset
 from .period cimport get_period_ordinal
-from .timestamps cimport create_timestamp_from_ts
+from .timestamps cimport (
+    create_timestamp_from_ts,
+    normalize_i8_stamp,
+)
 from .timezones cimport (
     get_dst_info,
     is_tzlocal,

pandas/core/groupby/groupby.py

@@ -1363,10 +1363,19 @@ def apply(self, func, *args, **kwargs) -> NDFrameT:
 
         func = com.is_builtin_func(func)
 
-        # this is needed so we don't try and wrap strings. If we could
-        # resolve functions to their callable functions prior, this
-        # wouldn't be needed
-        if args or kwargs:
+        if isinstance(func, str):
+            if hasattr(self, func):
+                res = getattr(self, func)
+                if callable(res):
+                    return res(*args, **kwargs)
+                elif args or kwargs:
+                    raise ValueError(f"Cannot pass arguments to property {func}")
+                return res
+
+            else:
+                raise TypeError(f"apply func should be callable, not '{func}'")
+
+        elif args or kwargs:
             if callable(func):
 
                 @wraps(func)
@@ -1382,15 +1391,6 @@ def f(g):
                 raise ValueError(
                     "func must be a callable if args or kwargs are supplied"
                 )
-        elif isinstance(func, str):
-            if hasattr(self, func):
-                res = getattr(self, func)
-                if callable(res):
-                    return res()
-                return res
-
-            else:
-                raise TypeError(f"apply func should be callable, not '{func}'")
         else:
 
             f = func

pandas/io/formats/info.py

@@ -258,8 +258,6 @@
 
     Parameters
     ----------
-    data : {klass}
-        {klass} to print information about.
     verbose : bool, optional
         Whether to print the full summary. By default, the setting in
         ``pandas.options.display.max_info_columns`` is followed.

pandas/tests/groupby/test_apply.py

@@ -1247,3 +1247,12 @@ def test_apply_nonmonotonic_float_index(arg, idx):
     expected = DataFrame({"col": arg}, index=idx)
     result = expected.groupby("col").apply(lambda x: x)
     tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("args, kwargs", [([True], {}), ([], {"numeric_only": True})])
+def test_apply_str_with_args(df, args, kwargs):
+    # GH#46479
+    gb = df.groupby("A")
+    result = gb.apply("sum", *args, **kwargs)
+    expected = gb.sum(numeric_only=True)
+    tm.assert_frame_equal(result, expected)

requirements-dev.txt

@@ -58,7 +58,7 @@ blosc
 bottleneck>=1.3.1
 ipykernel
 ipython>=7.11.1
-jinja2
+jinja2<=3.0.3
 matplotlib>=3.3.2
 numexpr>=2.7.1
 scipy>=1.4.1