TST: Add skip_on_exception decorator (#326)

We have tests that might fail for reasons
beyond our control. This decorator catches
those exceptions and skips those tests when
such an error is raised.

Author: gfyoung

pandas_datareader/_testing.py

@@ -0,0 +1,32 @@
+"""
+Utilities for testing purposes.
+"""
+
+from functools import wraps
+
+
+def skip_on_exception(exp):
+    """
+    Skip a test if a specific Exception is raised. This is because
+    the Exception is raised for reasons beyond our control (e.g.
+    flakey 3rd-party API).
+
+    Parameters
+    ----------
+    exp : The Exception under which to execute try-except.
+    """
+
+    from pytest import skip
+
+    def outer_wrapper(f):
+
+        @wraps(f)
+        def wrapper(*args, **kwargs):
+            try:
+                f(*args, **kwargs)
+            except exp as e:
+                skip(e)
+
+        return wrapper
+
+    return outer_wrapper

pandas_datareader/tests/google/test_options.py

@@ -8,6 +8,7 @@
 
 import pandas_datareader.data as web
 from pandas_datareader._utils import RemoteDataError
+from pandas_datareader._testing import skip_on_exception
 
 
 class TestGoogleOptions(object):
@@ -17,12 +18,9 @@ def setup_class(cls):
         # GOOG has monthlies
         cls.goog = web.Options('GOOG', 'google')
 
+    @skip_on_exception(RemoteDataError)
     def test_get_options_data(self):
-        try:
-            options = self.goog.get_options_data(
-                expiry=self.goog.expiry_dates[0])
-        except RemoteDataError as e:  # pragma: no cover
-            raise pytest.skip(e)
+        options = self.goog.get_options_data(expiry=self.goog.expiry_dates[0])
 
         assert isinstance(options, pd.DataFrame)
         assert len(options) > 10
@@ -48,11 +46,9 @@ def test_get_options_data_yearmonth(self):
         with pytest.raises(NotImplementedError):
             self.goog.get_options_data(month=1, year=2016)
 
+    @skip_on_exception(RemoteDataError)
     def test_expiry_dates(self):
-        try:
-            dates = self.goog.expiry_dates
-        except RemoteDataError as e:  # pragma: no cover
-            pytest.skip(e)
+        dates = self.goog.expiry_dates
 
         assert len(dates) == 2
         assert isinstance(dates, list)

pandas_datareader/tests/test_edgar.py

@@ -5,6 +5,7 @@
 import pandas_datareader.data as web
 
 from pandas_datareader._utils import RemoteDataError
+from pandas_datareader._testing import skip_on_exception
 
 
 class TestEdgarIndex(object):
@@ -15,58 +16,47 @@ def setup_class(cls):
         # Disabling tests until re-write.
         pytest.skip("Disabling tests until re-write.")
 
+    @skip_on_exception(RemoteDataError)
     def test_get_full_index(self):
-        try:
-            ed = web.DataReader('full', 'edgar-index')
-            assert len(ed) > 1000
+        ed = web.DataReader('full', 'edgar-index')
+        assert len(ed) > 1000
 
-            exp_columns = pd.Index(['cik', 'company_name', 'form_type',
-                                    'date_filed', 'filename'], dtype='object')
-            tm.assert_index_equal(ed.columns, exp_columns)
-
-        except RemoteDataError as e:
-            pytest.skip(e)
+        exp_columns = pd.Index(['cik', 'company_name', 'form_type',
+                                'date_filed', 'filename'], dtype='object')
+        tm.assert_index_equal(ed.columns, exp_columns)
 
+    @skip_on_exception(RemoteDataError)
     def test_get_nonzip_index_and_low_date(self):
-        try:
-            ed = web.DataReader('daily', 'edgar-index', '1994-06-30',
-                                '1994-07-02')
-            assert len(ed) > 200
-            assert ed.index.nlevels == 2
-
-            dti = ed.index.get_level_values(0)
-            assert isinstance(dti, pd.DatetimeIndex)
-            exp_columns = pd.Index(['company_name', 'form_type',
-                                    'filename'], dtype='object')
-            tm.assert_index_equal(ed.columns, exp_columns)
-
-        except RemoteDataError as e:
-            pytest.skip(e)
-
+        ed = web.DataReader('daily', 'edgar-index', '1994-06-30',
+                            '1994-07-02')
+        assert len(ed) > 200
+        assert ed.index.nlevels == 2
+
+        dti = ed.index.get_level_values(0)
+        assert isinstance(dti, pd.DatetimeIndex)
+        exp_columns = pd.Index(['company_name', 'form_type',
+                                'filename'], dtype='object')
+        tm.assert_index_equal(ed.columns, exp_columns)
+
+    @skip_on_exception(RemoteDataError)
     def test_get_gz_index_and_no_date(self):
         # TODO: Rewrite, as this test causes Travis to timeout.
 
-        try:
-            ed = web.DataReader('daily', 'edgar-index')
-            assert len(ed) > 2000
-        except RemoteDataError as e:
-            pytest.skip(e)
+        ed = web.DataReader('daily', 'edgar-index')
+        assert len(ed) > 2000
 
+    @skip_on_exception(RemoteDataError)
     def test_6_digit_date(self):
-        try:
-            ed = web.DataReader('daily', 'edgar-index', start='1998-05-18',
-                                end='1998-05-18')
-            assert len(ed) < 1200
-            assert ed.index.nlevels == 2
-
-            dti = ed.index.get_level_values(0)
-            assert isinstance(dti, pd.DatetimeIndex)
-            assert dti[0] == pd.Timestamp('1998-05-18')
-            assert dti[-1] == pd.Timestamp('1998-05-18')
-
-            exp_columns = pd.Index(['company_name', 'form_type',
-                                    'filename'], dtype='object')
-            tm.assert_index_equal(ed.columns, exp_columns)
-
-        except RemoteDataError as e:
-            pytest.skip(e)
+        ed = web.DataReader('daily', 'edgar-index', start='1998-05-18',
+                            end='1998-05-18')
+        assert len(ed) < 1200
+        assert ed.index.nlevels == 2
+
+        dti = ed.index.get_level_values(0)
+        assert isinstance(dti, pd.DatetimeIndex)
+        assert dti[0] == pd.Timestamp('1998-05-18')
+        assert dti[-1] == pd.Timestamp('1998-05-18')
+
+        exp_columns = pd.Index(['company_name', 'form_type',
+                                'filename'], dtype='object')
+        tm.assert_index_equal(ed.columns, exp_columns)

pandas_datareader/tests/test_enigma.py

@@ -5,6 +5,7 @@
 
 import pandas_datareader as pdr
 import pandas_datareader.data as web
+from pandas_datareader._testing import skip_on_exception
 
 TEST_API_KEY = os.getenv('ENIGMA_API_KEY')
 
@@ -15,21 +16,17 @@ class TestEnigma(object):
     def setup_class(cls):
         pytest.importorskip("lxml")
 
+    @skip_on_exception(HTTPError)
     def test_enigma_datareader(self):
-        try:
-            df = web.DataReader('enigma.inspections.restaurants.fl',
-                                'enigma', access_key=TEST_API_KEY)
-            assert 'serialid' in df.columns
-        except HTTPError as e:  # pragma: no cover
-            pytest.skip(e)
+        df = web.DataReader('enigma.inspections.restaurants.fl',
+                            'enigma', access_key=TEST_API_KEY)
+        assert 'serialid' in df.columns
 
+    @skip_on_exception(HTTPError)
     def test_enigma_get_data_enigma(self):
-        try:
-            df = pdr.get_data_enigma(
-                'enigma.inspections.restaurants.fl', TEST_API_KEY)
-            assert 'serialid' in df.columns
-        except HTTPError as e:  # pragma: no cover
-            pytest.skip(e)
+        df = pdr.get_data_enigma(
+            'enigma.inspections.restaurants.fl', TEST_API_KEY)
+        assert 'serialid' in df.columns
 
     def test_bad_key(self):
         with pytest.raises(HTTPError):

pandas_datareader/tests/test_eurostat.py

@@ -3,11 +3,14 @@
 import pandas.util.testing as tm
 import pandas_datareader.data as web
 
+from pandas_datareader._utils import RemoteDataError
 from pandas_datareader.compat import assert_raises_regex
+from pandas_datareader._testing import skip_on_exception
 
 
 class TestEurostat(object):
 
+    @skip_on_exception(RemoteDataError)
     def test_get_cdh_e_fos(self):
         # Employed doctorate holders in non managerial and non professional
         # occupations by fields of science (%)
@@ -32,6 +35,7 @@ def test_get_cdh_e_fos(self):
         expected = pd.DataFrame(values, index=exp_idx, columns=exp_col)
         tm.assert_frame_equal(df, expected)
 
+    @skip_on_exception(RemoteDataError)
     def test_get_sts_cobp_a(self):
         # Building permits - annual data (2010 = 100)
         df = web.DataReader('sts_cobp_a', 'eurostat',
@@ -64,6 +68,7 @@ def test_get_sts_cobp_a(self):
             result = df[expected.name]
             tm.assert_series_equal(result, expected)
 
+    @skip_on_exception(RemoteDataError)
     def test_get_nrg_pc_202(self):
         # see gh-149
 
@@ -86,6 +91,7 @@ def test_get_nrg_pc_202(self):
 
         tm.assert_series_equal(df[name], exp)
 
+    @skip_on_exception(RemoteDataError)
     def test_get_prc_hicp_manr_exceeds_limit(self):
         # see gh-149
         msg = 'Query size exceeds maximum limit'

pandas_datareader/tests/test_nasdaq.py

@@ -1,15 +1,12 @@
-import pytest
 import pandas_datareader.data as web
 
 from pandas_datareader._utils import RemoteDataError
+from pandas_datareader._testing import skip_on_exception
 
 
 class TestNasdaqSymbols(object):
 
+    @skip_on_exception(RemoteDataError)
     def test_get_symbols(self):
-        try:
-            symbols = web.DataReader('symbols', 'nasdaq')
-        except RemoteDataError as e:
-            pytest.skip(e)
-
+        symbols = web.DataReader('symbols', 'nasdaq')
         assert 'IBM' in symbols.index

pandas_datareader/tests/test_wb.py

@@ -8,6 +8,7 @@
 import pandas.util.testing as tm
 from pandas_datareader.wb import (search, download, get_countries,
                                   get_indicators, WorldBankReader)
+from pandas_datareader._testing import skip_on_exception
 from pandas_datareader.compat import assert_raises_regex
 
 
@@ -140,6 +141,7 @@ def test_wdi_download_error_handling(self):
             assert isinstance(result, pd.DataFrame)
             assert len(result) == 2
 
+    @skip_on_exception(ValueError)
     def test_wdi_download_w_retired_indicator(self):
 
         cntry_codes = ['CA', 'MX', 'US']
@@ -156,32 +158,25 @@ def test_wdi_download_w_retired_indicator(self):
 
         inds = ['GDPPCKD']
 
-        try:
-            result = download(country=cntry_codes, indicator=inds,
-                              start=2003, end=2004, errors='ignore')
-        # If for some reason result actually ever has data, it's cause WB
-        # fixed the issue with this ticker.  Find another bad one.
-        except ValueError as e:
-            pytest.skip("No indicators returned data: {0}".format(e))
+        result = download(country=cntry_codes, indicator=inds,
+                          start=2003, end=2004, errors='ignore')
 
-        # if it ever gets here, it means WB unretired the indicator.
+        # If it ever gets here, it means WB unretired the indicator.
         # even if they dropped it completely, it would still get caught above
         # or the WB API changed somehow in a really unexpected way.
         if len(result) > 0:  # pragma: no cover
             pytest.skip("Invalid results")
 
+    @skip_on_exception(ValueError)
     def test_wdi_download_w_crash_inducing_countrycode(self):
 
         cntry_codes = ['CA', 'MX', 'US', 'XXX']
         inds = ['NY.GDP.PCAP.CD']
 
-        try:
-            result = download(country=cntry_codes, indicator=inds,
-                              start=2003, end=2004, errors='ignore')
-        except ValueError as e:
-            pytest.skip("No indicators returned data: {0}".format(e))
+        result = download(country=cntry_codes, indicator=inds,
+                          start=2003, end=2004, errors='ignore')
 
-        # if it ever gets here, it means the country code XXX got used by WB
+        # If it ever gets here, it means the country code XXX got used by WB
         # or the WB API changed somehow in a really unexpected way.
         if len(result) > 0:  # pragma: no cover
             pytest.skip("Invalid results")