Merge branch 'master' into validation-test-for-sorting

Author: Matias Heikkilä

asv_bench/benchmarks/frame_methods.py

@@ -96,6 +96,8 @@ def time_dict_rename_both_axes(self):
 
 
 class Iteration:
+    # mem_itertuples_* benchmarks are slow
+    timeout = 120
 
     def setup(self):
         N = 1000

asv_bench/benchmarks/groupby.py

@@ -1,12 +1,11 @@
 from functools import partial
 from itertools import product
 from string import ascii_letters
-import warnings
 
 import numpy as np
 
 from pandas import (
-    Categorical, DataFrame, MultiIndex, Series, TimeGrouper, Timestamp,
+    Categorical, DataFrame, MultiIndex, Series, Timestamp,
     date_range, period_range)
 import pandas.util.testing as tm
 
@@ -301,10 +300,6 @@ def setup(self):
     def time_multi_size(self):
         self.df.groupby(['key1', 'key2']).size()
 
-    def time_dt_timegrouper_size(self):
-        with warnings.catch_warnings(record=True):
-            self.df.groupby(TimeGrouper(key='dates', freq='M')).size()
-
     def time_category_size(self):
         self.draws.groupby(self.cats).size()
 

asv_bench/benchmarks/index_object.py

@@ -52,7 +52,6 @@ def time_is_dates_only(self):
 
 class Ops:
 
-    sample_time = 0.2
     params = ['float', 'int']
     param_names = ['dtype']
 
@@ -95,6 +94,12 @@ def time_min(self):
     def time_min_trivial(self):
         self.idx_inc.min()
 
+    def time_get_loc_inc(self):
+        self.idx_inc.get_loc(900000)
+
+    def time_get_loc_dec(self):
+        self.idx_dec.get_loc(100000)
+
 
 class IndexAppend:
 
@@ -183,7 +188,7 @@ def time_get_loc(self):
 
 class IntervalIndexMethod:
     # GH 24813
-    params = [10**3, 10**5]
+    params = [10**3, 10**5, 10**7]
 
     def setup(self, N):
         left = np.append(np.arange(N), np.array(0))
@@ -194,5 +199,8 @@ def setup(self, N):
     def time_monotonic_inc(self, N):
         self.intv.is_monotonic_increasing
 
+    def time_is_unique(self, N):
+        self.intv.is_unique
+
 
 from .pandas_vb_common import setup  # noqa: F401

asv_bench/benchmarks/io/parsers.py

@@ -1,7 +1,11 @@
 import numpy as np
 
-from pandas._libs.tslibs.parsing import (
-    _concat_date_cols, _does_string_look_like_datetime)
+try:
+    from pandas._libs.tslibs.parsing import (
+        _concat_date_cols, _does_string_look_like_datetime)
+except ImportError:
+    # Avoid whole benchmark suite import failure on asv (currently 0.4)
+    pass
 
 
 class DoesStringLookLikeDatetime(object):

asv_bench/benchmarks/rolling.py

@@ -4,7 +4,6 @@
 
 class Methods:
 
-    sample_time = 0.2
     params = (['DataFrame', 'Series'],
               [10, 1000],
               ['int', 'float'],
@@ -23,7 +22,6 @@ def time_rolling(self, constructor, window, dtype, method):
 
 class ExpandingMethods:
 
-    sample_time = 0.2
     params = (['DataFrame', 'Series'],
               ['int', 'float'],
               ['median', 'mean', 'max', 'min', 'std', 'count', 'skew', 'kurt',
@@ -41,7 +39,6 @@ def time_expanding(self, constructor, dtype, method):
 
 class EWMMethods:
 
-    sample_time = 0.2
     params = (['DataFrame', 'Series'],
               [10, 1000],
               ['int', 'float'],
@@ -58,7 +55,6 @@ def time_ewm(self, constructor, window, dtype, method):
 
 
 class VariableWindowMethods(Methods):
-    sample_time = 0.2
     params = (['DataFrame', 'Series'],
               ['50s', '1h', '1d'],
               ['int', 'float'],
@@ -75,7 +71,6 @@ def setup(self, constructor, window, dtype, method):
 
 class Pairwise:
 
-    sample_time = 0.2
     params = ([10, 1000, None],
               ['corr', 'cov'],
               [True, False])
@@ -95,7 +90,6 @@ def time_pairwise(self, window, method, pairwise):
 
 
 class Quantile:
-    sample_time = 0.2
     params = (['DataFrame', 'Series'],
               [10, 1000],
               ['int', 'float'],

azure-pipelines.yml

@@ -15,7 +15,7 @@ jobs:
     name: Windows
     vmImage: vs2017-win2016
 
-- job: 'Checks_and_doc'
+- job: 'Checks'
   pool:
     vmImage: ubuntu-16.04
   timeoutInMinutes: 90
@@ -97,10 +97,11 @@ jobs:
   - script: |
       export PATH=$HOME/miniconda3/bin:$PATH
       source activate pandas-dev
+      cd asv_bench
+      asv check -E existing
       git remote add upstream https://github.com/pandas-dev/pandas.git
       git fetch upstream
       if git diff upstream/master --name-only | grep -q "^asv_bench/"; then
-          cd asv_bench
           asv machine --yes
           ASV_OUTPUT="$(asv dev)"
           if [[ $(echo "$ASV_OUTPUT" | grep "failed") ]]; then

ci/deps/azure-35-compat.yaml

@@ -26,5 +26,5 @@ dependencies:
   - pip
   - pip:
     # for python 3.5, pytest>=4.0.2 is not available in conda
-    - pytest>=4.0.2
+    - pytest==4.5.0
     - html5lib==1.0b2

ci/deps/azure-macos-35.yaml

@@ -25,7 +25,7 @@ dependencies:
   - pip:
     - python-dateutil==2.5.3
     # universal
-    - pytest>=4.0.2
+    - pytest==4.5.0
     - pytest-xdist
     - pytest-mock
     - hypothesis>=3.58.0

ci/setup_env.sh

@@ -118,12 +118,12 @@ echo "conda list"
 conda list
 
 # Install DB for Linux
-if [ ${TRAVIS_OS_NAME} == "linux" ]; then
+if [ "${TRAVIS_OS_NAME}" == "linux" ]; then
   echo "installing dbs"
   mysql -e 'create database pandas_nosetest;'
   psql -c 'create database pandas_nosetest;' -U postgres
 else
-   echo "not using dbs on non-linux"
+   echo "not using dbs on non-linux Travis builds or Azure Pipelines"
 fi
 
 echo "done"

doc/source/conf.py

@@ -319,7 +319,7 @@
    pd.options.display.max_rows = 15
 
    import os
-   os.chdir('{}')
+   os.chdir(r'{}')
 """.format(os.path.dirname(os.path.dirname(__file__)))