PEP8

xarray/backends/zarr.py

@@ -8,7 +8,7 @@
 from ..core import indexing
 from ..core.pycompat import OrderedDict, integer_types, iteritems
 from ..core.utils import FrozenOrderedDict, HiddenKeyDict
-from .common import AbstractWritableDataStore, ArrayWriter, BackendArray
+from .common import AbstractWritableDataStore, BackendArray
 
 # need some special secret attributes to tell us the dimensions
 _DIMENSION_KEY = '_ARRAY_DIMENSIONS'

xarray/core/dataarray.py

@@ -16,7 +16,7 @@
     assert_coordinate_consistent, remap_label_indexers)
 from .dataset import Dataset, merge_indexes, split_indexes
 from .formatting import format_item
-from .options import OPTIONS, _get_keep_attrs
+from .options import OPTIONS
 from .pycompat import OrderedDict, basestring, iteritems, range, zip
 from .utils import (
     decode_numpy_dict_values, either_dict_or_kwargs, ensure_us_time_resolution)

xarray/core/options.py

@@ -63,7 +63,8 @@ def _get_keep_attrs(default):
     elif global_choice in [True, False]:
         return global_choice
     else:
-        raise ValueError("The global option keep_attrs must be one of True, False or 'default'.")
+        raise ValueError("The global option keep_attrs must be one of"
+                         " True, False or 'default'.")
 
 
 class set_options(object):

xarray/core/resample.py

@@ -194,7 +194,7 @@ def _interpolate(self, kind='linear'):
                 .format(self._obj.data.name)
             )
 
-        from ..coding.cftimeindex import CFTimeIndex
+        # from ..coding.cftimeindex import CFTimeIndex
         import cftime as cf
         import numpy as np
         if isinstance(self._obj[self._dim].values[0], cf.datetime):

xarray/plot/plot.py

@@ -14,7 +14,7 @@
 import numpy as np
 import pandas as pd
 
-from xarray.core.alignment import align
+# from xarray.core.alignment import align
 from xarray.core.common import contains_cftime_datetimes
 from xarray.core.pycompat import basestring
 
@@ -255,7 +255,6 @@ def _infer_line_data(darray, x, y, hue):
         huelabel = label_from_attrs(darray[huename])
         hueplt = darray[huename]
 
-
     xlabel = label_from_attrs(xplt)
     ylabel = label_from_attrs(yplt)
 

xarray/tests/temp/cftime_resample_pandas_comparison.py

@@ -1,11 +1,11 @@
-import xarray as xr
-import pandas as pd
 import numpy as np
+import pandas as pd
 
+import xarray as xr
 
 # Equal sampling comparisons:
 ti = pd.date_range('2000-01-01', periods=9, freq='T', tz='UTC')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)
 print(ps.resample('T', label='left', closed='left').mean())
 print(ps.resample('T', label='right', closed='left').mean())
 print(ps.resample('T', label='left', closed='right').mean())
@@ -15,7 +15,7 @@
 print(ps.resample('60S', label='left', closed='right').mean())
 print(ps.resample('60S', label='right', closed='right').mean())
 ti = pd.date_range('2000', periods=30, freq='MS', tz='UTC')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)
 print(ps.resample('M', label='left', closed='left').max())
 print(ps.resample('M', label='right', closed='left').max())
 print(ps.resample('M', label='left', closed='right').max())
@@ -25,10 +25,9 @@
 print(ps.resample('MS', label='left', closed='right').max())
 print(ps.resample('MS', label='right', closed='right').max())
 
-
 # Downsampling comparisons:
 ti = pd.date_range('2000-01-01', periods=9, freq='MS', tz='UTC')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)
 print(ps.resample('3M', label='left', closed='left').max())
 print(ps.resample('3M', label='right', closed='left').max())
 print(ps.resample('3M', label='left', closed='right').max())
@@ -55,33 +54,32 @@
 print(ps.resample('2MS', label='right', closed='right').mean())
 # Checking how label and closed args affect outputs
 ti = pd.date_range('2000-01-01', periods=9, freq='T', tz='UTC')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)
 print(ps.resample('3T', label='left', closed='left').mean())
 print(ps.resample('3T', label='right', closed='left').mean())
 print(ps.resample('3T', label='left', closed='right').mean())
 print(ps.resample('3T', label='right', closed='right').mean())
 ti = pd.date_range('2000', periods=30, freq='MS')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)
 print(ps.resample('6MS', label='left', closed='left').max())
 print(ps.resample('6MS', label='right', closed='left').max())
 print(ps.resample('6MS', label='left', closed='right').max())
 print(ps.resample('6MS', label='right', closed='right').max())
 # Checking different aggregation funcs, also checking cases when label and closed == None
 ti = pd.date_range('2000', periods=30, freq='MS')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)
 print(ps.resample('MS').mean())  # essentially doing no resampling, should return original data
 print(ps.resample('6MS').mean())
 print(ps.resample('6MS').asfreq())  # results do not match since xarray makes asfreq = mean (see resample.py)
 print(ps.resample('6MS').sum())
 print(ps.resample('6MS').min())
 print(ps.resample('6MS').max())
 
-
 # Upsampling comparisons:
 # At seconds-resolution, xr.cftime_range is 1 second off from pd.date_range
 ti = pd.date_range('2011-01-01T13:02:03', '2012-01-01T00:00:00', freq='D', tz='UTC')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)
-print(ps.resample('12T', base=0).interpolate().index) # testing T vs min
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)
+print(ps.resample('12T', base=0).interpolate().index)  # testing T vs min
 print(ps.resample('12min', base=0).interpolate().index)
 print(ps.resample('12min', base=1).interpolate().index)
 print(ps.resample('12min', base=5).interpolate().index)
@@ -91,7 +89,7 @@
 print(ps.resample('1D', base=0).mean().values)  # essentially doing no resampling, should return original data
 # Pandas' upsampling behave aberrantly if start times for dates are not neat, should we replicate?
 ti = pd.date_range('2000-01-01T13:02:03', '2000-02-01T00:00:00', freq='D', tz='UTC')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)  # results unchanged if array of floats used
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)  # results unchanged if array of floats used
 print(ps.resample('8H', base=0, closed='left').interpolate().values)
 print(ps.resample('8H', base=0, closed='left').sum().values)
 print(ps.resample('8H', base=0, closed='left').mean().values)
@@ -100,19 +98,18 @@
 print(ps.resample('8H', base=0, closed='right').mean().values)
 # Neat start times (00:00:00) produces expected behavior when upsampling with pandas
 ti = pd.date_range('2000-01-01T00:00:00', '2000-02-01T00:00:00', freq='D', tz='UTC')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)
 print(ps.resample('12T', base=0).interpolate())
 print(ps.resample('12T', base=24, closed='left').interpolate())
 print(ps.resample('12T', base=24, closed='right', label='left').interpolate())
 ti = pd.date_range('2000', periods=30, freq='MS')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)
 print(ps.resample('D').interpolate())
 
-
 # Shows how resample-apply produces different results with Series and DataArray
 ti = pd.date_range('2000', periods=30, freq='MS')
-da = xr.DataArray(np.arange(100, 100+ti.size), [('time', ti)])
+da = xr.DataArray(np.arange(100, 100 + ti.size), [('time', ti)])
 print(da.resample(time='6MS').sum())
 ti = pd.date_range('2000', periods=30, freq='MS')
-ps = pd.Series(np.arange(100, 100+ti.size), index=ti)
+ps = pd.Series(np.arange(100, 100 + ti.size), index=ti)
 print(ps.resample('6MS').sum())

xarray/tests/temp/cftime_resample_tests.py

@@ -1,10 +1,10 @@
-import xarray as xr
 import numpy as np
 
+import xarray as xr
 
 # Equal sampling comparisons:
 times = xr.cftime_range('2000-01-01', periods=9, freq='T', tz='UTC')
-da = xr.DataArray(np.arange(100, 100+times.size), [('time', times)])
+da = xr.DataArray(np.arange(100, 100 + times.size), [('time', times)])
 print(da.resample(time='T', label='left', closed='left').mean())
 print(da.resample(time='T', label='right', closed='left').mean())
 print(da.resample(time='T', label='left', closed='right').mean())
@@ -14,7 +14,7 @@
 print(da.resample(time='60S', label='left', closed='right').mean())
 print(da.resample(time='60S', label='right', closed='right').mean())
 times = xr.cftime_range('2000', periods=30, freq='MS')
-da = xr.DataArray(np.arange(100, 100+times.size), [('time', times)])
+da = xr.DataArray(np.arange(100, 100 + times.size), [('time', times)])
 print(da.resample(time='M', label='left', closed='left').max())
 print(da.resample(time='M', label='right', closed='left').max())
 print(da.resample(time='M', label='left', closed='right').max())
@@ -24,10 +24,9 @@
 print(da.resample(time='MS', label='left', closed='right').max())
 print(da.resample(time='MS', label='right', closed='right').max())
 
-
 # Downsampling comparisons:
 times = xr.cftime_range('2000-01-01', periods=9, freq='MS', tz='UTC')
-da = xr.DataArray(np.arange(100, 100+times.size), [('time', times)])
+da = xr.DataArray(np.arange(100, 100 + times.size), [('time', times)])
 print(da.resample(time='3M', label='left', closed='left').max())
 print(da.resample(time='3M', label='right', closed='left').max())
 print(da.resample(time='3M', label='left', closed='right').max())
@@ -54,43 +53,43 @@
 print(da.resample(time='2MS', label='right', closed='right').mean())
 # Checking how label and closed args affect outputs
 times = xr.cftime_range('2000-01-01', periods=9, freq='T', tz='UTC')
-da = xr.DataArray(np.arange(100, 100+times.size), [('time', times)])
+da = xr.DataArray(np.arange(100, 100 + times.size), [('time', times)])
 print(da.resample(time='3T', label='left', closed='left').mean())
 print(da.resample(time='3T', label='right', closed='left').mean())
 print(da.resample(time='3T', label='left', closed='right').mean())
 print(da.resample(time='3T', label='right', closed='right').mean())
 times = xr.cftime_range('2000', periods=30, freq='MS')
-da = xr.DataArray(np.arange(100, 100+times.size), [('time', times)])
+da = xr.DataArray(np.arange(100, 100 + times.size), [('time', times)])
 print(da.resample(time='6MS', label='left', closed='left').max())
 print(da.resample(time='6MS', label='right', closed='left').max())
 print(da.resample(time='6MS', label='left', closed='right').max())
 print(da.resample(time='6MS', label='right', closed='right').max())
 # Checking different aggregation funcs, also checking cases when label and closed == None
 times = xr.cftime_range('2000', periods=30, freq='MS')
-da = xr.DataArray(np.arange(100, 100+times.size), [('time', times)])
+da = xr.DataArray(np.arange(100, 100 + times.size), [('time', times)])
 print(da.resample(time='MS').mean())  # essentially doing no resampling, should return original data
 print(da.resample(time='6MS').mean())
 print(da.resample(time='6MS').asfreq())  # results do not match since xarray makes asfreq = mean (see resample.py)
 print(da.resample(time='6MS').sum())
 print(da.resample(time='6MS').min())
 print(da.resample(time='6MS').max())
 
-
 # Upsampling comparisons:
 # At seconds-resolution, xr.cftime_range is 1 second off from pd.date_range
 times = xr.cftime_range('2011-01-01T13:02:03', '2012-01-01T00:00:00', freq='D')
-da = xr.DataArray(np.arange(100, 100+times.size), [('time', times)])
+da = xr.DataArray(np.arange(100, 100 + times.size), [('time', times)])
 print(da.resample(time='12T', base=0).interpolate().indexes)  # testing 'T' vs 'min'
 print(da.resample(time='12min', base=0).interpolate().indexes)
 print(da.resample(time='12min', base=1).interpolate().indexes)
 print(da.resample(time='12min', base=5).mean().indexes)
 print(da.resample(time='12min', base=17).mean().indexes)
 print(da.resample(time='12S', base=17).interpolate().indexes)
-print(da.resample(time='1D', base=0).interpolate().values)  # essentially doing no resampling, should return original data
+print(
+    da.resample(time='1D', base=0).interpolate().values)  # essentially doing no resampling, should return original data
 print(da.resample(time='1D', base=0).mean().values)  # essentially doing no resampling, should return original data
 # Upsampling with non 00:00:00 dates. Sum and mean matches pandas behavior but interpolate doesn't.
 times = xr.cftime_range('2000-01-01T13:02:03', '2000-02-01T00:00:00', freq='D')
-da = xr.DataArray(np.arange(100, 100+times.size), [('time', times)])
+da = xr.DataArray(np.arange(100, 100 + times.size), [('time', times)])
 print(da.resample(time='8H', base=0, closed='left').interpolate().values)
 print(da.resample(time='8H', base=0, closed='left').sum().values)
 print(da.resample(time='8H', base=0, closed='left').mean().values)
@@ -99,41 +98,42 @@
 print(da.resample(time='8H', base=0, closed='right').mean().values)
 # Neat start times (00:00:00) produces behavior matching pandas'
 times = xr.cftime_range('2000-01-01T00:00:00', '2000-02-01T00:00:00', freq='D')
-da = xr.DataArray(np.arange(100, 100+times.size), [('time', times)])
+da = xr.DataArray(np.arange(100, 100 + times.size), [('time', times)])
 print(da.resample(time='12T', base=0).interpolate())
 print(da.resample(time='12T', base=24, closed='left').interpolate())
 print(da.resample(time='12T', base=24, closed='right', label='left').interpolate())
 times = xr.cftime_range('2000', periods=30, freq='MS')
-da = xr.DataArray(np.arange(100, 100+times.size), [('time', times)])
+da = xr.DataArray(np.arange(100, 100 + times.size), [('time', times)])
 print(da.resample(time='D').interpolate())
 
-
 # Check that Dataset and DataArray returns the same resampling results
 times = xr.cftime_range('2000-01-01', periods=9, freq='T', tz='UTC')
-ds = xr.Dataset(data_vars={'data1': ('time', np.arange(100, 100+times.size)),
-                           'data2': ('time', np.arange(500, 500+times.size))},
+ds = xr.Dataset(data_vars={'data1': ('time', np.arange(100, 100 + times.size)),
+                           'data2': ('time', np.arange(500, 500 + times.size))},
                 coords={'time': times})
 print(ds.resample(time='3T', label='left', closed='left').mean())
 print(ds.resample(time='3T', label='right', closed='left').mean())
 print(ds.resample(time='3T', label='left', closed='right').mean())
 print(ds.resample(time='3T', label='right', closed='right').mean())
 times = xr.cftime_range('2000', periods=30, freq='MS')
-ds = xr.Dataset(data_vars={'data1': ('time', np.arange(100, 100+times.size)),
-                           'data2': ('time', np.arange(500, 500+times.size))},
+ds = xr.Dataset(data_vars={'data1': ('time', np.arange(100, 100 + times.size)),
+                           'data2': ('time', np.arange(500, 500 + times.size))},
                 coords={'time': times})
 print(ds.resample(time='6MS', label='left', closed='left').max())
 print(ds.resample(time='6MS', label='right', closed='left').max())
 print(ds.resample(time='6MS', label='left', closed='right').max())
 print(ds.resample(time='6MS', label='right', closed='right').max())
 
-
 # Check that nc files read as dask arrays can be resampled
 #
 import os
-testfilepath = os.path.join(os.path.expanduser('~'), 'Dropbox', 'code', 'Ouranos', 'testdata', 'NRCANdaily', 'nrcan_canada_daily_tasmax_1990.nc')
+
+testfilepath = os.path.join(os.path.expanduser('~'), 'Dropbox', 'code', 'Ouranos', 'testdata', 'NRCANdaily',
+                            'nrcan_canada_daily_tasmax_1990.nc')
 xr.set_options(enable_cftimeindex=True)
 test_ds = xr.open_dataset(testfilepath, chunks={'time': 10})
-test_ds['time'] = xr.cftime_range('1999-01-01', '1999-12-31', freq='D')  # regular calendars are still read as pandas date_range even though enable_cftimeindex=True
+test_ds['time'] = xr.cftime_range('1999-01-01', '1999-12-31',
+                                  freq='D')  # regular calendars are still read as pandas date_range even though enable_cftimeindex=True
 # test_ds.fillna(0).resample(time='3MS')  # NaN in results still present
 print(test_ds.resample(time='MS').mean())
 monthly_avg = test_ds.resample(time='MS').mean()