Merge branch 'master' into bugfix/replace_recursion

* master:
  BENCH: asv csv reading benchmarks no longer read StringIO objects off the end (pandas-dev#21807)
  BUG: df.agg, df.transform and df.apply use different methods when axis=1 than when axis=0 (pandas-dev#21224)
  BUG: bug in GroupBy.count where arg minlength passed to np.bincount must be None for np<1.13 (pandas-dev#21957)
  CLN: Vbench to asv conversion script (pandas-dev#22089)
  consistent docstring (pandas-dev#22066)
  TST: skip pytables test with not-updated pytables conda package (pandas-dev#22099)
  CLN: Remove Legacy MultiIndex Index Compatibility (pandas-dev#21740)
  DOC: Reword doc for filepath_or_buffer in read_csv (pandas-dev#22058)
  BUG: rolling with MSVC 2017 build (pandas-dev#21813)

Author: minggli

appveyor.yml

@@ -20,12 +20,14 @@ environment:
   matrix:
 
     - CONDA_ROOT: "C:\\Miniconda3_64"
+      APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2017
       PYTHON_VERSION: "3.6"
       PYTHON_ARCH: "64"
       CONDA_PY: "36"
       CONDA_NPY: "113"
 
     - CONDA_ROOT: "C:\\Miniconda3_64"
+      APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2015
       PYTHON_VERSION: "2.7"
       PYTHON_ARCH: "64"
       CONDA_PY: "27"

asv_bench/benchmarks/io/csv.py

@@ -54,7 +54,14 @@ def time_frame_date_formatting(self):
         self.data.to_csv(self.fname, date_format='%Y%m%d')
 
 
-class ReadCSVDInferDatetimeFormat(object):
+class StringIORewind(object):
+
+    def data(self, stringio_object):
+        stringio_object.seek(0)
+        return stringio_object
+
+
+class ReadCSVDInferDatetimeFormat(StringIORewind):
 
     goal_time = 0.2
     params = ([True, False], ['custom', 'iso8601', 'ymd'])
@@ -66,10 +73,12 @@ def setup(self, infer_datetime_format, format):
                    'iso8601': '%Y-%m-%d %H:%M:%S',
                    'ymd': '%Y%m%d'}
         dt_format = formats[format]
-        self.data = StringIO('\n'.join(rng.strftime(dt_format).tolist()))
+        self.StringIO_input = StringIO('\n'.join(
+                                       rng.strftime(dt_format).tolist()))
 
     def time_read_csv(self, infer_datetime_format, format):
-        read_csv(self.data, header=None, names=['foo'], parse_dates=['foo'],
+        read_csv(self.data(self.StringIO_input),
+                 header=None, names=['foo'], parse_dates=['foo'],
                  infer_datetime_format=infer_datetime_format)
 
 
@@ -95,7 +104,7 @@ def time_skipprows(self, skiprows):
         read_csv(self.fname, skiprows=skiprows)
 
 
-class ReadUint64Integers(object):
+class ReadUint64Integers(StringIORewind):
 
     goal_time = 0.2
 
@@ -108,13 +117,13 @@ def setup(self):
         self.data2 = StringIO('\n'.join(arr.astype(str).tolist()))
 
     def time_read_uint64(self):
-        read_csv(self.data1, header=None, names=['foo'])
+        read_csv(self.data(self.data1), header=None, names=['foo'])
 
     def time_read_uint64_neg_values(self):
-        read_csv(self.data2, header=None, names=['foo'])
+        read_csv(self.data(self.data2), header=None, names=['foo'])
 
     def time_read_uint64_na_values(self):
-        read_csv(self.data1, header=None, names=['foo'],
+        read_csv(self.data(self.data1), header=None, names=['foo'],
                  na_values=self.na_values)
 
 
@@ -140,19 +149,20 @@ def time_thousands(self, sep, thousands):
         read_csv(self.fname, sep=sep, thousands=thousands)
 
 
-class ReadCSVComment(object):
+class ReadCSVComment(StringIORewind):
 
     goal_time = 0.2
 
     def setup(self):
         data = ['A,B,C'] + (['1,2,3 # comment'] * 100000)
-        self.s_data = StringIO('\n'.join(data))
+        self.StringIO_input = StringIO('\n'.join(data))
 
     def time_comment(self):
-        read_csv(self.s_data, comment='#', header=None, names=list('abc'))
+        read_csv(self.data(self.StringIO_input), comment='#',
+                 header=None, names=list('abc'))
 
 
-class ReadCSVFloatPrecision(object):
+class ReadCSVFloatPrecision(StringIORewind):
 
     goal_time = 0.2
     params = ([',', ';'], ['.', '_'], [None, 'high', 'round_trip'])
@@ -164,14 +174,14 @@ def setup(self, sep, decimal, float_precision):
         rows = sep.join(['0{}'.format(decimal) + '{}'] * 3) + '\n'
         data = rows * 5
         data = data.format(*floats) * 200  # 1000 x 3 strings csv
-        self.s_data = StringIO(data)
+        self.StringIO_input = StringIO(data)
 
     def time_read_csv(self, sep, decimal, float_precision):
-        read_csv(self.s_data, sep=sep, header=None, names=list('abc'),
-                 float_precision=float_precision)
+        read_csv(self.data(self.StringIO_input), sep=sep, header=None,
+                 names=list('abc'), float_precision=float_precision)
 
     def time_read_csv_python_engine(self, sep, decimal, float_precision):
-        read_csv(self.s_data, sep=sep, header=None, engine='python',
+        read_csv(self.data(self.StringIO_input), sep=sep, header=None, engine='python',
                  float_precision=None, names=list('abc'))
 
 
@@ -193,7 +203,7 @@ def time_convert_direct(self):
         read_csv(self.fname, dtype='category')
 
 
-class ReadCSVParseDates(object):
+class ReadCSVParseDates(StringIORewind):
 
     goal_time = 0.2
 
@@ -206,12 +216,14 @@ def setup(self):
                """
         two_cols = ['KORD,19990127'] * 5
         data = data.format(*two_cols)
-        self.s_data = StringIO(data)
+        self.StringIO_input = StringIO(data)
 
     def time_multiple_date(self):
-        read_csv(self.s_data, sep=',', header=None,
-                 names=list(string.digits[:9]), parse_dates=[[1, 2], [1, 3]])
+        read_csv(self.data(self.StringIO_input), sep=',', header=None,
+                 names=list(string.digits[:9]),
+                 parse_dates=[[1, 2], [1, 3]])
 
     def time_baseline(self):
-        read_csv(self.s_data, sep=',', header=None, parse_dates=[1],
+        read_csv(self.data(self.StringIO_input), sep=',', header=None,
+                 parse_dates=[1],
                  names=list(string.digits[:9]))

asv_bench/vbench_to_asv.py

@@ -20,7 +20,7 @@ and bug fixes. We recommend that all users upgrade to this version.
 Fixed Regressions
 ~~~~~~~~~~~~~~~~~
 
-- 
+- Python 3.7 with Windows gave all missing values for rolling variance calculations (:issue:`21813`)
 -
 
 .. _whatsnew_0234.bug_fixes:

doc/source/whatsnew/v0.23.4.txt

@@ -475,7 +475,9 @@ Numeric
 - Bug in :class:`Series` ``__rmatmul__`` doesn't support matrix vector multiplication (:issue:`21530`)
 - Bug in :func:`factorize` fails with read-only array (:issue:`12813`)
 - Fixed bug in :func:`unique` handled signed zeros inconsistently: for some inputs 0.0 and -0.0 were treated as equal and for some inputs as different. Now they are treated as equal for all inputs (:issue:`21866`)
--
+- Bug in :meth:`DataFrame.agg`, :meth:`DataFrame.transform` and :meth:`DataFrame.apply` where,
+  when supplied with a list of functions and ``axis=1`` (e.g. ``df.apply(['sum', 'mean'], axis=1)``),
+  a ``TypeError`` was wrongly raised. For all three methods such calculation are now done correctly. (:issue:`16679`).
 -
 
 Strings
@@ -514,7 +516,7 @@ Missing
 MultiIndex
 ^^^^^^^^^^
 
--
+- Removed compatibility for MultiIndex pickles prior to version 0.8.0; compatibility with MultiIndex pickles from version 0.13 forward is maintained (:issue:`21654`)
 -
 -
 
@@ -536,11 +538,11 @@ Groupby/Resample/Rolling
 
 - Bug in :func:`pandas.core.groupby.GroupBy.first` and :func:`pandas.core.groupby.GroupBy.last` with ``as_index=False`` leading to the loss of timezone information (:issue:`15884`)
 - Bug in :meth:`DatetimeIndex.resample` when downsampling across a DST boundary (:issue:`8531`)
--
--
-
+- Bug where ``ValueError`` is wrongly raised when calling :func:`~pandas.core.groupby.SeriesGroupBy.count` method of a
+  ``SeriesGroupBy`` when the grouping variable only contains NaNs and numpy version < 1.13 (:issue:`21956`).
 - Multiple bugs in :func:`pandas.core.Rolling.min` with ``closed='left'` and a
   datetime-like index leading to incorrect results and also segfault. (:issue:`21704`)
+-
 
 Sparse
 ^^^^^^

doc/source/whatsnew/v0.24.0.txt

@@ -6,6 +6,7 @@
 #if defined(_MSC_VER) && (_MSC_VER < 1800)
 #include <cmath>
 namespace std {
+  __inline int isnan(double x) { return _isnan(x); }
   __inline int signbit(double num) { return _copysign(1.0, num) < 0; }
 }
 #else

pandas/_libs/src/headers/cmath

@@ -14,6 +14,7 @@ cnp.import_array()
 
 
 cdef extern from "src/headers/cmath" namespace "std":
+    bint isnan(double) nogil
     int signbit(double) nogil
     double sqrt(double x) nogil
 
@@ -653,16 +654,16 @@ cdef inline void add_var(double val, double *nobs, double *mean_x,
                          double *ssqdm_x) nogil:
     """ add a value from the var calc """
     cdef double delta
-
-    # Not NaN
-    if val == val:
-        nobs[0] = nobs[0] + 1
-
-        # a part of Welford's method for the online variance-calculation
-        # https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance
-        delta = val - mean_x[0]
-        mean_x[0] = mean_x[0] + delta / nobs[0]
-        ssqdm_x[0] = ssqdm_x[0] + ((nobs[0] - 1) * delta ** 2) / nobs[0]
+    # `isnan` instead of equality as fix for GH-21813, msvc 2017 bug
+    if isnan(val):
+        return
+
+    nobs[0] = nobs[0] + 1
+    # a part of Welford's method for the online variance-calculation
+    # https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance
+    delta = val - mean_x[0]
+    mean_x[0] = mean_x[0] + delta / nobs[0]
+    ssqdm_x[0] = ssqdm_x[0] + ((nobs[0] - 1) * delta ** 2) / nobs[0]
 
 
 cdef inline void remove_var(double val, double *nobs, double *mean_x,