FIX: Many warning fixes for ACT. (#822)

* FIX: Many warning fixes for ACT.
These fixes include datetime utc changes, missing images, xarray.dims to sizes changes and more.

* FIX: Forgot to remove 0 index.

* STY: PEP8 fixes.

* FIX: Use timezone import.

* FIX: Need to check length of coord arrays, when casting to float.

* STY: PEP8 fixes.

act/discovery/cropscape.py

@@ -89,10 +89,9 @@ def get_crop_type(lat=None, lon=None, year=None):
     # Add year, lat, and lon as parameters
     params = {'year': str(year), 'x': str(x), 'y': str(y)}
 
-    # Perform the request.  Note, verify set to False until
-    # server SSL errors can be worked out
+    # Perform the request.
     try:
-        req = requests.get(url, params=params, verify=False, timeout=1)
+        req = requests.get(url, params=params, timeout=1)
     except Exception:
         return
 

act/io/neon.py

@@ -82,9 +82,18 @@ def read_neon_csv(files, variable_files=None, position_files=None):
             dloc = loc_df.loc[loc_df['HOR.VER'] == hor_loc + '.' + ver_loc]
             idx = dloc.index.values
             if len(idx) > 0:
-                ds['lat'] = xr.DataArray(data=float(loc_df['referenceLatitude'].values[idx]))
-                ds['lon'] = xr.DataArray(data=float(loc_df['referenceLongitude'].values[idx]))
-                ds['alt'] = xr.DataArray(data=float(loc_df['referenceElevation'].values[idx]))
+                if len(loc_df['referenceLatitude'].values) > 1:
+                    ds['lat'] = xr.DataArray(data=float(loc_df['referenceLatitude'].values[idx][0]))
+                    ds['lon'] = xr.DataArray(
+                        data=float(loc_df['referenceLongitude'].values[idx][0])
+                    )
+                    ds['alt'] = xr.DataArray(
+                        data=float(loc_df['referenceElevation'].values[idx][0])
+                    )
+                else:
+                    ds['lat'] = xr.DataArray(data=float(loc_df['referenceLatitude'].values[idx]))
+                    ds['lon'] = xr.DataArray(data=float(loc_df['referenceLongitude'].values[idx]))
+                    ds['alt'] = xr.DataArray(data=float(loc_df['referenceElevation'].values[idx]))
                 variables = [
                     'xOffset',
                     'yOffset',
@@ -98,7 +107,10 @@ def read_neon_csv(files, variable_files=None, position_files=None):
                     'yAzimuth',
                 ]
                 for v in variables:
-                    ds[v] = xr.DataArray(data=float(loc_df[v].values[idx]))
+                    if len(loc_df[v].values) > 1:
+                        ds[v] = xr.DataArray(data=float(loc_df[v].values[idx][0]))
+                    else:
+                        ds[v] = xr.DataArray(data=float(loc_df[v].values[idx]))
         multi_ds.append(ds)
 
     ds = xr.merge(multi_ds)

act/io/noaapsl.py

@@ -188,7 +188,7 @@ def _parse_psl_wind_lines(filepath, lines, line_offset=0):
     beam_elevation = np.array([beam_elevation1, beam_elevation2, beam_elevation3], dtype='float32')
 
     # Read in the data table section using pandas
-    df = pd.read_csv(filepath, skiprows=line_offset + 10, delim_whitespace=True)
+    df = pd.read_csv(filepath, skiprows=line_offset + 10, sep=r'\s+')
 
     # Only read in the number of rows for a given set of gates
     df = df.iloc[: int(number_of_range_gates)]
@@ -326,7 +326,7 @@ def _parse_psl_temperature_lines(filepath, lines, line_offset=0):
     beam_azimuth, beam_elevation = filter_list(lines[8].split(' ')).astype(float)
 
     # Read in the data table section using pandas
-    df = pd.read_csv(filepath, skiprows=line_offset + 10, delim_whitespace=True)
+    df = pd.read_csv(filepath, skiprows=line_offset + 10, sep=r'\s+')
 
     # Only read in the number of rows for a given set of gates
     df = df.iloc[: int(number_of_gates)]
@@ -1115,7 +1115,7 @@ def _parse_psl_radar_moments(files):
                         f,
                         skiprows=[0, 1, 2],
                         nrows=int(data['n_gates']['data'][-1]) - 1,
-                        delim_whitespace=True,
+                        sep=r'\s+',
                         names=list(names.keys()),
                     )
                     index2 = 0
@@ -1143,7 +1143,7 @@ def _parse_psl_radar_moments(files):
                         f,
                         skiprows=list(range(index2 + 1)),
                         nrows=int(data['n_gates']['data'][-1]) - 1,
-                        delim_whitespace=True,
+                        sep=r'\s+',
                         names=list(names.keys()),
                     )
 

act/utils/data_utils.py

@@ -383,7 +383,7 @@ def assign_coordinates(ds, coord_list):
         if coord not in ds.variables.keys():
             raise KeyError(coord + ' is not a variable in the Dataset.')
 
-        if ds.dims[coord_list[coord]] != len(ds.variables[coord]):
+        if ds.sizes[coord_list[coord]] != len(ds.variables[coord]):
             raise IndexError(
                 coord + ' must have the same ' + 'value as length of ' + coord_list[coord]
             )

tests/io/test_hysplit.py

@@ -10,7 +10,7 @@ def test_read_hysplit():
     assert 'lon' in ds.variables.keys()
     assert 'alt' in ds.variables.keys()
     assert 'PRESSURE' in ds.variables.keys()
-    assert ds.dims["num_grids"] == 8
-    assert ds.dims["num_trajectories"] == 1
-    assert ds.dims['time'] == 121
+    assert ds.sizes["num_grids"] == 8
+    assert ds.sizes["num_trajectories"] == 1
+    assert ds.sizes['time'] == 121
     assert ds['age'].min() == -120

tests/io/test_noaapsl.py

@@ -11,12 +11,12 @@ def test_read_psl_wind_profiler():
         act.tests.EXAMPLE_NOAA_PSL, transpose=False
     )
     # test dimensions
-    assert 'time' and 'HT' in test_ds_low.dims.keys()
-    assert 'time' and 'HT' in test_ds_hi.dims.keys()
-    assert test_ds_low.dims['time'] == 4
-    assert test_ds_hi.dims['time'] == 4
-    assert test_ds_low.dims['HT'] == 49
-    assert test_ds_hi.dims['HT'] == 50
+    assert 'time' and 'HT' in test_ds_low.sizes.keys()
+    assert 'time' and 'HT' in test_ds_hi.sizes.keys()
+    assert test_ds_low.sizes['time'] == 4
+    assert test_ds_hi.sizes['time'] == 4
+    assert test_ds_low.sizes['HT'] == 49
+    assert test_ds_hi.sizes['HT'] == 50
 
     # test coordinates
     assert (test_ds_low.coords['HT'][0:5] == np.array([0.151, 0.254, 0.356, 0.458, 0.561])).all()

tests/plotting/test_timeseriesdisplay.py

@@ -423,6 +423,7 @@ def test_plot_barbs_from_u_v2():
         matplotlib.pyplot.close(BarbDisplay.fig)
 
 
+@pytest.mark.mpl_image_compare(tolerance=10)
 def test_plot_barbs_from_u_v3():
     bins = list(np.linspace(0, 1, 10))
     xbins = list(pd.date_range(pd.to_datetime('2020-01-01'), pd.to_datetime('2020-01-02'), 12))
@@ -446,6 +447,7 @@ def test_plot_barbs_from_u_v3():
         matplotlib.pyplot.close(BarbDisplay.fig)
 
 
+@pytest.mark.mpl_image_compare(tolerance=10)
 def test_plot_barbs_from_u_v4():
     bins = list(np.linspace(0, 1, 10))
     xbins = [pd.to_datetime('2020-01-01')]
@@ -471,6 +473,7 @@ def test_plot_barbs_from_u_v4():
         matplotlib.pyplot.close(BarbDisplay.fig)
 
 
+@pytest.mark.mpl_image_compare(tolerance=10)
 def test_plot_barbs_from_u_v5():
     bins = list(np.linspace(0, 1, 10))
     xbins = [pd.to_datetime('2020-01-01')]

tests/qc/test_qcfilter.py

@@ -1,5 +1,5 @@
 import copy
-from datetime import datetime
+from datetime import datetime, timezone
 
 import dask.array as da
 import numpy as np
@@ -390,12 +390,12 @@ def test_qc_speed():
         coords={'time': time},
     )
 
-    start = datetime.utcnow()
+    start = datetime.now(timezone.utc)
     for name, var in noisy_data_mapping.items():
         failed_qc = var > 0.75  # Consider data above 0.75 as bad. Negligible time here.
         ds.qcfilter.add_test(name, index=failed_qc, test_meaning='Value above threshold')
 
-    time_diff = datetime.utcnow() - start
+    time_diff = datetime.now(timezone.utc) - start
     assert time_diff.seconds <= 4
 
 

tests/utils/test_datetime_utils.py

@@ -1,4 +1,4 @@
-from datetime import datetime
+from datetime import datetime, timezone
 
 import numpy as np
 import pandas as pd
@@ -15,7 +15,7 @@ def test_dates_between():
     answer = np.arange(start_string, end_string, dtype='datetime64[D]')
     answer = np.append(answer, answer[-1] + 1)
     answer = answer.astype('datetime64[s]').astype(int)
-    answer = [datetime.utcfromtimestamp(ii) for ii in answer]
+    answer = [datetime.fromtimestamp(ii, tz=timezone.utc).replace(tzinfo=None) for ii in answer]
 
     assert date_list == answer