STY: use pytest.raises context manager (plotting, reductions, scalar...) (#25483)

* STY: use pytest.raises context manager (plotting, reductions, scalar...)

* revert removed testing in test_timedelta.py

* remove TODO from test_frame.py

* skip py2 ci failure

Author: simonjayhawkins

Diff for: pandas/tests/plotting/test_boxplot_method.py

@@ -267,13 +267,20 @@ def test_grouped_box_return_type(self):
     def test_grouped_box_layout(self):
         df = self.hist_df
 
-        pytest.raises(ValueError, df.boxplot, column=['weight', 'height'],
-                      by=df.gender, layout=(1, 1))
-        pytest.raises(ValueError, df.boxplot,
-                      column=['height', 'weight', 'category'],
-                      layout=(2, 1), return_type='dict')
-        pytest.raises(ValueError, df.boxplot, column=['weight', 'height'],
-                      by=df.gender, layout=(-1, -1))
+        msg = "Layout of 1x1 must be larger than required size 2"
+        with pytest.raises(ValueError, match=msg):
+            df.boxplot(column=['weight', 'height'], by=df.gender,
+                       layout=(1, 1))
+
+        msg = "The 'layout' keyword is not supported when 'by' is None"
+        with pytest.raises(ValueError, match=msg):
+            df.boxplot(column=['height', 'weight', 'category'],
+                       layout=(2, 1), return_type='dict')
+
+        msg = "At least one dimension of layout must be positive"
+        with pytest.raises(ValueError, match=msg):
+            df.boxplot(column=['weight', 'height'], by=df.gender,
+                       layout=(-1, -1))
 
         # _check_plot_works adds an ax so catch warning. see GH #13188
         with tm.assert_produces_warning(UserWarning):

Diff for: pandas/tests/plotting/test_datetimelike.py

@@ -97,7 +97,9 @@ def test_nonnumeric_exclude(self):
         assert len(ax.get_lines()) == 1  # B was plotted
         self.plt.close(fig)
 
-        pytest.raises(TypeError, df['A'].plot)
+        msg = "Empty 'DataFrame': no numeric data to plot"
+        with pytest.raises(TypeError, match=msg):
+            df['A'].plot()
 
     def test_tsplot_deprecated(self):
         from pandas.tseries.plotting import tsplot
@@ -140,10 +142,15 @@ def f(*args, **kwds):
     def test_both_style_and_color(self):
 
         ts = tm.makeTimeSeries()
-        pytest.raises(ValueError, ts.plot, style='b-', color='#000099')
+        msg = ("Cannot pass 'style' string with a color symbol and 'color' "
+               "keyword argument. Please use one or the other or pass 'style'"
+               " without a color symbol")
+        with pytest.raises(ValueError, match=msg):
+            ts.plot(style='b-', color='#000099')
 
         s = ts.reset_index(drop=True)
-        pytest.raises(ValueError, s.plot, style='b-', color='#000099')
+        with pytest.raises(ValueError, match=msg):
+            s.plot(style='b-', color='#000099')
 
     @pytest.mark.slow
     def test_high_freq(self):

Diff for: pandas/tests/plotting/test_hist_method.py

@@ -332,12 +332,17 @@ def test_grouped_hist_legacy2(self):
     @pytest.mark.slow
     def test_grouped_hist_layout(self):
         df = self.hist_df
-        pytest.raises(ValueError, df.hist, column='weight', by=df.gender,
-                      layout=(1, 1))
-        pytest.raises(ValueError, df.hist, column='height', by=df.category,
-                      layout=(1, 3))
-        pytest.raises(ValueError, df.hist, column='height', by=df.category,
-                      layout=(-1, -1))
+        msg = "Layout of 1x1 must be larger than required size 2"
+        with pytest.raises(ValueError, match=msg):
+            df.hist(column='weight', by=df.gender, layout=(1, 1))
+
+        msg = "Layout of 1x3 must be larger than required size 4"
+        with pytest.raises(ValueError, match=msg):
+            df.hist(column='height', by=df.category, layout=(1, 3))
+
+        msg = "At least one dimension of layout must be positive"
+        with pytest.raises(ValueError, match=msg):
+            df.hist(column='height', by=df.category, layout=(-1, -1))
 
         with tm.assert_produces_warning(UserWarning):
             axes = _check_plot_works(df.hist, column='height', by=df.gender,

Diff for: pandas/tests/plotting/test_misc.py

@@ -278,14 +278,20 @@ def test_subplot_titles(self, iris):
         assert [p.get_title() for p in plot] == title
 
         # Case len(title) > len(df)
-        pytest.raises(ValueError, df.plot, subplots=True,
-                      title=title + ["kittens > puppies"])
+        msg = ("The length of `title` must equal the number of columns if"
+               " using `title` of type `list` and `subplots=True`")
+        with pytest.raises(ValueError, match=msg):
+            df.plot(subplots=True, title=title + ["kittens > puppies"])
 
         # Case len(title) < len(df)
-        pytest.raises(ValueError, df.plot, subplots=True, title=title[:2])
+        with pytest.raises(ValueError, match=msg):
+            df.plot(subplots=True, title=title[:2])
 
         # Case subplots=False and title is of type list
-        pytest.raises(ValueError, df.plot, subplots=False, title=title)
+        msg = ("Using `title` of type `list` is not supported unless"
+               " `subplots=True` is passed")
+        with pytest.raises(ValueError, match=msg):
+            df.plot(subplots=False, title=title)
 
         # Case df with 3 numeric columns but layout of (2,2)
         plot = df.drop('SepalWidth', axis=1).plot(subplots=True, layout=(2, 2),

Diff for: pandas/tests/reductions/test_reductions.py

@@ -276,7 +276,9 @@ def test_timedelta_ops(self):
 
         # invalid ops
         for op in ['skew', 'kurt', 'sem', 'prod']:
-            pytest.raises(TypeError, getattr(td, op))
+            msg = "reduction operation '{}' not allowed for this dtype"
+            with pytest.raises(TypeError, match=msg.format(op)):
+                getattr(td, op)()
 
         # GH#10040
         # make sure NaT is properly handled by median()

Diff for: pandas/tests/scalar/period/test_period.py

@@ -8,6 +8,7 @@
 from pandas._libs.tslibs.ccalendar import DAYS, MONTHS
 from pandas._libs.tslibs.frequencies import INVALID_FREQ_ERR_MSG
 from pandas._libs.tslibs.parsing import DateParseError
+from pandas._libs.tslibs.period import IncompatibleFrequency
 from pandas._libs.tslibs.timezones import dateutil_gettz, maybe_get_tz
 from pandas.compat import iteritems, text_type
 from pandas.compat.numpy import np_datetime64_compat
@@ -35,7 +36,9 @@ def test_construction(self):
         i4 = Period('2005', freq='M')
         i5 = Period('2005', freq='m')
 
-        pytest.raises(ValueError, i1.__ne__, i4)
+        msg = r"Input has different freq=M from Period\(freq=A-DEC\)"
+        with pytest.raises(IncompatibleFrequency, match=msg):
+            i1 != i4
         assert i4 == i5
 
         i1 = Period.now('Q')
@@ -74,9 +77,12 @@ def test_construction(self):
                           freq='U')
         assert i1 == expected
 
-        pytest.raises(ValueError, Period, ordinal=200701)
+        msg = "Must supply freq for ordinal value"
+        with pytest.raises(ValueError, match=msg):
+            Period(ordinal=200701)
 
-        pytest.raises(ValueError, Period, '2007-1-1', freq='X')
+        with pytest.raises(ValueError, match="Invalid frequency: X"):
+            Period('2007-1-1', freq='X')
 
     def test_construction_bday(self):
 
@@ -233,10 +239,6 @@ def test_period_constructor_offsets(self):
                           freq='U')
         assert i1 == expected
 
-        pytest.raises(ValueError, Period, ordinal=200701)
-
-        pytest.raises(ValueError, Period, '2007-1-1', freq='X')
-
     def test_invalid_arguments(self):
         with pytest.raises(ValueError):
             Period(datetime.now())
@@ -925,8 +927,9 @@ def test_properties_secondly(self):
 class TestPeriodField(object):
 
     def test_get_period_field_array_raises_on_out_of_range(self):
-        pytest.raises(ValueError, libperiod.get_period_field_arr, -1,
-                      np.empty(1), 0)
+        msg = "Buffer dtype mismatch, expected 'int64_t' but got 'double'"
+        with pytest.raises(ValueError, match=msg):
+            libperiod.get_period_field_arr(-1, np.empty(1), 0)
 
 
 class TestComparisons(object):

Diff for: pandas/tests/scalar/timedelta/test_timedelta.py

@@ -250,9 +250,13 @@ def check(value):
         assert rng.microseconds == 0
         assert rng.nanoseconds == 0
 
-        pytest.raises(AttributeError, lambda: rng.hours)
-        pytest.raises(AttributeError, lambda: rng.minutes)
-        pytest.raises(AttributeError, lambda: rng.milliseconds)
+        msg = "'Timedelta' object has no attribute '{}'"
+        with pytest.raises(AttributeError, match=msg.format('hours')):
+            rng.hours
+        with pytest.raises(AttributeError, match=msg.format('minutes')):
+            rng.minutes
+        with pytest.raises(AttributeError, match=msg.format('milliseconds')):
+            rng.milliseconds
 
         # GH 10050
         check(rng.days)
@@ -272,9 +276,13 @@ def check(value):
         assert rng.seconds == 10 * 3600 + 11 * 60 + 12
         assert rng.microseconds == 100 * 1000 + 123
         assert rng.nanoseconds == 456
-        pytest.raises(AttributeError, lambda: rng.hours)
-        pytest.raises(AttributeError, lambda: rng.minutes)
-        pytest.raises(AttributeError, lambda: rng.milliseconds)
+        msg = "'Timedelta' object has no attribute '{}'"
+        with pytest.raises(AttributeError, match=msg.format('hours')):
+            rng.hours
+        with pytest.raises(AttributeError, match=msg.format('minutes')):
+            rng.minutes
+        with pytest.raises(AttributeError, match=msg.format('milliseconds')):
+            rng.milliseconds
 
         # components
         tup = pd.to_timedelta(-1, 'us').components
@@ -449,8 +457,12 @@ def test_round(self):
             assert r2 == s2
 
         # invalid
-        for freq in ['Y', 'M', 'foobar']:
-            pytest.raises(ValueError, lambda: t1.round(freq))
+        for freq, msg in [
+            ('Y', '<YearEnd: month=12> is a non-fixed frequency'),
+            ('M', '<MonthEnd> is a non-fixed frequency'),
+                ('foobar', 'Invalid frequency: foobar')]:
+            with pytest.raises(ValueError, match=msg):
+                t1.round(freq)
 
         t1 = timedelta_range('1 days', periods=3, freq='1 min 2 s 3 us')
         t2 = -1 * t1
@@ -495,11 +507,15 @@ def test_round(self):
             r1 = t1.round(freq)
             tm.assert_index_equal(r1, s1)
             r2 = t2.round(freq)
-        tm.assert_index_equal(r2, s2)
+            tm.assert_index_equal(r2, s2)
 
         # invalid
-        for freq in ['Y', 'M', 'foobar']:
-            pytest.raises(ValueError, lambda: t1.round(freq))
+        for freq, msg in [
+            ('Y', '<YearEnd: month=12> is a non-fixed frequency'),
+            ('M', '<MonthEnd> is a non-fixed frequency'),
+                ('foobar', 'Invalid frequency: foobar')]:
+            with pytest.raises(ValueError, match=msg):
+                t1.round(freq)
 
     def test_contains(self):
         # Checking for any NaT-like objects
@@ -609,9 +625,12 @@ def test_overflow(self):
         assert np.allclose(result.value / 1000, expected.value / 1000)
 
         # sum
-        pytest.raises(ValueError, lambda: (s - s.min()).sum())
+        msg = "overflow in timedelta operation"
+        with pytest.raises(ValueError, match=msg):
+            (s - s.min()).sum()
         s1 = s[0:10000]
-        pytest.raises(ValueError, lambda: (s1 - s1.min()).sum())
+        with pytest.raises(ValueError, match=msg):
+            (s1 - s1.min()).sum()
         s2 = s[0:1000]
         result = (s2 - s2.min()).sum()
 

Diff for: pandas/tests/scalar/timestamp/test_timestamp.py

@@ -60,7 +60,9 @@ def check(value, equal):
         check(ts.hour, 9)
         check(ts.minute, 6)
         check(ts.second, 3)
-        pytest.raises(AttributeError, lambda: ts.millisecond)
+        msg = "'Timestamp' object has no attribute 'millisecond'"
+        with pytest.raises(AttributeError, match=msg):
+            ts.millisecond
         check(ts.microsecond, 100)
         check(ts.nanosecond, 1)
         check(ts.dayofweek, 6)
@@ -78,7 +80,9 @@ def check(value, equal):
         check(ts.hour, 23)
         check(ts.minute, 59)
         check(ts.second, 0)
-        pytest.raises(AttributeError, lambda: ts.millisecond)
+        msg = "'Timestamp' object has no attribute 'millisecond'"
+        with pytest.raises(AttributeError, match=msg):
+            ts.millisecond
         check(ts.microsecond, 0)
         check(ts.nanosecond, 0)
         check(ts.dayofweek, 2)

Diff for: pandas/tests/sparse/frame/test_frame.py

@@ -7,7 +7,7 @@
 import pytest
 
 from pandas._libs.sparse import BlockIndex, IntIndex
-from pandas.compat import lrange
+from pandas.compat import PY2, lrange
 from pandas.errors import PerformanceWarning
 
 import pandas as pd
@@ -145,8 +145,9 @@ def test_constructor_ndarray(self, float_frame):
         tm.assert_sp_frame_equal(sp, float_frame.reindex(columns=['A']))
 
         # raise on level argument
-        pytest.raises(TypeError, float_frame.reindex, columns=['A'],
-                      level=1)
+        msg = "Reindex by level not supported for sparse"
+        with pytest.raises(TypeError, match=msg):
+            float_frame.reindex(columns=['A'], level=1)
 
         # wrong length index / columns
         with pytest.raises(ValueError, match="^Index length"):
@@ -433,7 +434,8 @@ def test_getitem(self):
         exp = sdf.reindex(columns=['a', 'b'])
         tm.assert_sp_frame_equal(result, exp)
 
-        pytest.raises(Exception, sdf.__getitem__, ['a', 'd'])
+        with pytest.raises(KeyError, match=r"\['d'\] not in index"):
+            sdf[['a', 'd']]
 
     def test_iloc(self, float_frame):
 
@@ -504,7 +506,9 @@ def test_getitem_overload(self, float_frame):
         subframe = float_frame[indexer]
 
         tm.assert_index_equal(subindex, subframe.index)
-        pytest.raises(Exception, float_frame.__getitem__, indexer[:-1])
+        msg = "Item wrong length 9 instead of 10"
+        with pytest.raises(ValueError, match=msg):
+            float_frame[indexer[:-1]]
 
     def test_setitem(self, float_frame, float_frame_int_kind,
                      float_frame_dense,
@@ -551,8 +555,9 @@ def _check_frame(frame, orig):
             assert len(frame['I'].sp_values) == N // 2
 
             # insert ndarray wrong size
-            pytest.raises(Exception, frame.__setitem__, 'foo',
-                          np.random.randn(N - 1))
+            msg = "Length of values does not match length of index"
+            with pytest.raises(AssertionError, match=msg):
+                frame['foo'] = np.random.randn(N - 1)
 
             # scalar value
             frame['J'] = 5
@@ -625,17 +630,22 @@ def test_delitem(self, float_frame):
 
     def test_set_columns(self, float_frame):
         float_frame.columns = float_frame.columns
-        pytest.raises(Exception, setattr, float_frame, 'columns',
-                      float_frame.columns[:-1])
+        msg = ("Length mismatch: Expected axis has 4 elements, new values have"
+               " 3 elements")
+        with pytest.raises(ValueError, match=msg):
+            float_frame.columns = float_frame.columns[:-1]
 
     def test_set_index(self, float_frame):
         float_frame.index = float_frame.index
-        pytest.raises(Exception, setattr, float_frame, 'index',
-                      float_frame.index[:-1])
+        msg = ("Length mismatch: Expected axis has 10 elements, new values"
+               " have 9 elements")
+        with pytest.raises(ValueError, match=msg):
+            float_frame.index = float_frame.index[:-1]
 
     def test_ctor_reindex(self):
         idx = pd.Index([0, 1, 2, 3])
-        with pytest.raises(ValueError, match=''):
+        msg = "Length of passed values is 2, index implies 4"
+        with pytest.raises(ValueError, match=msg):
             pd.SparseDataFrame({"A": [1, 2]}, index=idx)
 
     def test_append(self, float_frame):
@@ -858,14 +868,18 @@ def test_describe(self, float_frame):
         str(float_frame)
         desc = float_frame.describe()  # noqa
 
+    @pytest.mark.skipif(PY2, reason="pytest.raises match regex fails")
     def test_join(self, float_frame):
         left = float_frame.loc[:, ['A', 'B']]
         right = float_frame.loc[:, ['C', 'D']]
         joined = left.join(right)
         tm.assert_sp_frame_equal(joined, float_frame, exact_indices=False)
 
         right = float_frame.loc[:, ['B', 'D']]
-        pytest.raises(Exception, left.join, right)
+        msg = (r"columns overlap but no suffix specified: Index\(\['B'\],"
+               r" dtype='object'\)")
+        with pytest.raises(ValueError, match=msg):
+            left.join(right)
 
         with pytest.raises(ValueError, match='Other Series must have a name'):
             float_frame.join(Series(
@@ -1046,8 +1060,11 @@ def _check(frame):
         _check(float_frame_int_kind)
 
         # for now
-        pytest.raises(Exception, _check, float_frame_fill0)
-        pytest.raises(Exception, _check, float_frame_fill2)
+        msg = "This routine assumes NaN fill value"
+        with pytest.raises(TypeError, match=msg):
+            _check(float_frame_fill0)
+        with pytest.raises(TypeError, match=msg):
+            _check(float_frame_fill2)
 
     def test_transpose(self, float_frame, float_frame_int_kind,
                        float_frame_dense,