Merge pull request #163 from jtkiley/euro-oecd-tests

FIX Eurostat and OECD tests that fail.

Author: femtotrader

pandas_datareader/compat/__init__.py

@@ -1 +1,2 @@
+# flake8: noqa
 from io import BytesIO

pandas_datareader/tests/test_eurostat.py

@@ -1,3 +1,4 @@
+import nose
 import sys
 
 import numpy as np
@@ -11,13 +12,14 @@ class TestEurostat(tm.TestCase):
     def setUp(self):
 
         if sys.version_info < (2, 7, 0):
-            import nose
-            raise nose.SkipTest("Doesn't support Python 2.6 because of ElementTree incompat")
+            raise nose.SkipTest("Doesn't support Python 2.6 because of"
+                                "ElementTree incompat")
 
     def test_get_cdh_e_fos(self):
         # Employed doctorate holders in non managerial and non professional
         # occupations by fields of science (%)
-        df = web.DataReader('cdh_e_fos', 'eurostat', start=pd.Timestamp('2005-01-01'),
+        df = web.DataReader('cdh_e_fos', 'eurostat',
+                            start=pd.Timestamp('2005-01-01'),
                             end=pd.Timestamp('2010-01-01'))
 
         self.assertTrue(isinstance(df, pd.DataFrame))
@@ -26,10 +28,11 @@ def test_get_cdh_e_fos(self):
         df = df['Percentage']['Total']['Natural sciences']
         df = df[['Norway', 'Poland', 'Portugal', 'Russia']]
 
-        exp_col = pd.MultiIndex.from_product([['Norway', 'Poland', 'Portugal', 'Russia'],
-                                              ['Annual']],
+        exp_col = pd.MultiIndex.from_product([['Norway', 'Poland', 'Portugal',
+                                             'Russia'], ['Annual']],
                                              names=['GEO', 'FREQ'])
-        exp_idx = pd.DatetimeIndex(['2006-01-01', '2009-01-01'], name='TIME_PERIOD')
+        exp_idx = pd.DatetimeIndex(['2006-01-01', '2009-01-01'],
+                                   name='TIME_PERIOD')
 
         values = np.array([[25.49, np.nan, 39.05, np.nan],
                            [20.38, 25.1, 27.77, 38.1]])
@@ -38,33 +41,37 @@ def test_get_cdh_e_fos(self):
 
     def test_get_sts_cobp_a(self):
         # Building permits - annual data (2010 = 100)
-        df = web.DataReader('sts_cobp_a', 'eurostat', start=pd.Timestamp('1992-01-01'),
+        df = web.DataReader('sts_cobp_a', 'eurostat',
+                            start=pd.Timestamp('1992-01-01'),
                             end=pd.Timestamp('2013-01-01'))
 
-        idx = pd.date_range('1992-01-01', '2013-01-01', freq='AS', name='TIME_PERIOD')
-        ne = pd.Series([np.nan, np.nan, np.nan, 144.53, 136.97, 180.02, 198.36,
-                        215.12, 200.05, 186.46, 127.33, 130.67, 143.26, 147.83,
-                        176.69, 227.36, 199.45, 128.49, 100.0, 113.83, 89.33, 77.57],
-                       name=('Building permits - m2 of useful floor area',
-                             'Gross data',
-                             'Non-residential buildings, except office buildings',
-                             'Netherlands', 'Annual'),
-                       index=idx)
-
-        uk = pd.Series([np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, 120.37,
-                        115.93, 112.53, 113.32, 110.18, 112.14, 119.06, 112.66,
-                        113.05, 121.8, 113.97, 105.88, 100.0, 98.56, 103.69, 81.32],
-                       name=('Building permits - m2 of useful floor area',
-                             'Gross data',
-                             'Non-residential buildings, except office buildings',
-                             'United Kingdom', 'Annual'),
-                       index=idx)
+        idx = pd.date_range('1992-01-01', '2013-01-01', freq='AS',
+                            name='TIME_PERIOD')
+        ne_name = ('Building permits - m2 of useful floor area',
+                   'Not seasonally adjusted data',
+                   'Non-residential buildings, except office buildings',
+                   'Netherlands', 'Annual')
+        ne_values = [np.nan, np.nan, np.nan, 144.53, 136.97, 180.02, 198.36,
+                     215.12, 200.05, 186.46, 127.33, 130.67, 143.26, 147.83,
+                     176.69, 227.36, 199.45, 128.49, 100.0, 113.83, 89.33,
+                     77.57]
+        ne = pd.Series(ne_values, name=ne_name, index=idx)
+
+        uk_name = ('Building permits - m2 of useful floor area',
+                   'Not seasonally adjusted data',
+                   'Non-residential buildings, except office buildings',
+                   'United Kingdom', 'Annual')
+        uk_values = [np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, 120.37,
+                     115.93, 112.53, 113.32, 110.18, 112.14, 119.06, 112.66,
+                     113.05, 121.8, 113.97, 105.88, 100.0, 98.56, 103.69,
+                     81.32]
+        uk = pd.Series(uk_values, name=uk_name, index=idx)
 
         for expected in [ne, uk]:
             result = df[expected.name]
             tm.assert_series_equal(result, expected)
 
 
 if __name__ == '__main__':
-    import nose
-    nose.runmodule(argv=[__file__, '-vvs', '-x', '--pdb', '--pdb-failure'], exit=False)
+    nose.runmodule(argv=[__file__, '-vvs', '-x', '--pdb', '--pdb-failure'],
+                   exit=False)

pandas_datareader/tests/test_oecd.py

@@ -1,3 +1,5 @@
+import nose
+
 from datetime import datetime
 import numpy as np
 import pandas as pd
@@ -46,21 +48,29 @@ def test_get_un_den(self):
               11.96023574, 11.48458378, 11.56435375, 11.91022276, 11.79401904,
               11.38345975, 11.32948829, 11.07884758]
 
-        index = pd.date_range('1960-01-01', '2012-01-01', freq='AS', name='Time')
-        for label, values in [('Australia', au), ('Japan', jp), ('United States', us)]:
+        index = pd.date_range('1960-01-01', '2012-01-01', freq='AS',
+                              name='Time')
+        for label, values in [('Australia', au), ('Japan', jp),
+                              ('United States', us)]:
             expected = pd.Series(values, index=index, name=label)
             tm.assert_series_equal(df[label], expected)
 
     def test_get_tourism(self):
-        df = web.DataReader('TOURISM_INBOUND', 'oecd', start=datetime(2005, 1, 1),
+        df = web.DataReader('TOURISM_INBOUND', 'oecd',
+                            start=datetime(2008, 1, 1),
                             end=datetime(2012, 1, 1))
 
-        jp = np.array([6728.0, 7334.0, 8347.0, 8351.0, 6790.0, 8611.0, 6219.0, 8368.0], dtype=float)
-        us = np.array([np.nan, 183178.0, 175142.0, 175632.0, 160359.0, 162269.0, 164672.0, 171630.0], dtype=float)
-        index = pd.date_range('2005-01-01', '2012-01-01', freq='AS', name='Year')
+        jp = np.array([8351000, 6790000, 8611000, 6219000,
+                      8368000], dtype=float)
+        us = np.array([175702309, 160507417, 164079732, 167600277,
+                       171320408], dtype=float)
+        index = pd.date_range('2008-01-01', '2012-01-01', freq='AS',
+                              name='Year')
         for label, values in [('Japan', jp), ('United States', us)]:
-            expected = pd.Series(values, index=index, name='Total international arrivals')
-            tm.assert_series_equal(df[label]['Total international arrivals'], expected)
+            expected = pd.Series(values, index=index,
+                                 name='Total international arrivals')
+            tm.assert_series_equal(df[label]['Total international arrivals'],
+                                   expected)
 
     def test_oecd_invalid_symbol(self):
         with tm.assertRaises(RemoteDataError):
@@ -71,5 +81,5 @@ def test_oecd_invalid_symbol(self):
 
 
 if __name__ == '__main__':
-    import nose
-    nose.runmodule(argv=[__file__, '-vvs', '-x', '--pdb', '--pdb-failure'], exit=False)
+    nose.runmodule(argv=[__file__, '-vvs', '-x', '--pdb', '--pdb-failure'],
+                   exit=False)