add tests from DataFrame.info

Author: MarcoGorelli

pandas/core/frame.py

@@ -2313,13 +2313,11 @@ def to_html(
     1   column_2  1000000 non-null  object
     2   column_3  1000000 non-null  object
 dtypes: object(3)
-memory usage: 188.8 MB
-""",
-see_also_sub = """
+memory usage: 188.8 MB""",
+        see_also_sub="""
 DataFrame.describe: Generate descriptive statistics of DataFrame
     columns.
-DataFrame.memory_usage: Memory usage of DataFrame columns.
-"""
+DataFrame.memory_usage: Memory usage of DataFrame columns.""",
     )
     @Appender(NDFrame.info.__doc__)
     def info(

pandas/core/generic.py

@@ -1804,7 +1804,7 @@ def info(
             cols = self.columns
             dtypes = self.dtypes
         else:
-            cols = pd.Series([self.name])
+            cols = pd.Index([self.name])
             dtypes = pd.Series([self.dtypes])
 
         col_count = len(cols)
@@ -1900,7 +1900,8 @@ def _verbose_repr():
                 )
 
         def _non_verbose_repr():
-            lines.append(cols._summary(name="Columns"))
+            if self._typ == "dataframe":
+                lines.append(cols._summary(name="Columns"))
 
         def _sizeof_fmt(num, size_qualifier):
             # returns size in human readable format

pandas/core/series.py

@@ -4198,17 +4198,13 @@ def replace(
 ---  --------------  -----
  0   5 non-null      object
 dtypes: object(1)
-memory usage: 80.0+ bytes
-    """,
-    see_also_sub="""
-Series.describe: Generate descriptive statistics of Series
-Series.memory_usage: Memory usage of Series.
-    """
+memory usage: 80.0+ bytes""",
+        see_also_sub="""
+Series.describe: Generate descriptive statistics of Series.
+Series.memory_usage: Memory usage of Series.""",
     )
     @Appender(NDFrame.info.__doc__)
-    def info(
-        self, verbose=None, buf=None, memory_usage=None, null_counts=None
-    ) -> None:
+    def info(self, verbose=None, buf=None, memory_usage=None, null_counts=None) -> None:
         return super().info(verbose, buf, None, memory_usage, null_counts)
 
     @Appender(generic._shared_docs["shift"] % _shared_doc_kwargs)

pandas/tests/series/test_repr.py

@@ -1,8 +1,11 @@
 from datetime import datetime, timedelta
+from io import StringIO
 
 import numpy as np
 import pytest
 
+from pandas.compat import PYPY
+
 import pandas as pd
 from pandas import (
     Categorical,
@@ -488,22 +491,117 @@ def test_categorical_series_repr_timedelta_ordered(self):
 
         assert repr(s) == exp
 
-    def test_info(self, capsys):
+    @pytest.mark.parametrize("verbose", [True, False])
+    def test_info(self, verbose, capsys):
         index = MultiIndex(
             levels=[["foo", "bar", "baz", "qux"], ["one", "two", "three"]],
             codes=[[0, 0, 0, 1, 1, 2, 2, 3, 3, 3], [0, 1, 2, 0, 1, 1, 2, 0, 1, 2]],
             names=["first", "second"],
         )
         s = Series(range(len(index)), index=index, name="sth")
-        s.info()
+        s.info(verbose=verbose)
         expected = """<class 'pandas.core.series.Series'>
 MultiIndex: 10 entries, ('foo', 'one') to ('qux', 'three')
-Series name: sth
+"""
+        if verbose:
+            expected += """Series name: sth
  #   Non-Null Count  Dtype
 ---  --------------  -----
  0   10 non-null     int64
-dtypes: int64(1)
-memory usage: 505.0+ bytes
+"""
+        expected += f"""dtypes: int64(1)
+memory usage: {s.memory_usage()}.0+ bytes
 """
         result = capsys.readouterr().out
         assert result == expected
+
+    @pytest.mark.skipif(PYPY, reason="on PyPy deep=True doesn't change result")
+    def test_info_memory_usage_deep_not_pypy(self):
+        s_with_object_index = pd.Series({"a": [1]}, index=["foo"])
+        assert s_with_object_index.memory_usage(
+            index=True, deep=True
+        ) > s_with_object_index.memory_usage(index=True)
+
+        s_object = pd.Series({"a": ["a"]})
+        assert s_object.memory_usage(deep=True) > s_object.memory_usage()
+
+    @pytest.mark.skipif(not PYPY, reason="on PyPy deep=True doesn't change result")
+    def test_info_memory_usage_deep_pypy(self):
+        s_with_object_index = pd.Series({"a": [1]}, index=["foo"])
+        assert s_with_object_index.memory_usage(
+            index=True, deep=True
+        ) == s_with_object_index.memory_usage(index=True)
+
+        s_object = pd.Series({"a": ["a"]})
+        assert s_object.memory_usage(deep=True) == s_object.memory_usage()
+
+    def test_info_memory_usage_qualified(self):
+
+        buf = StringIO()
+        s = pd.Series(1, index=[1, 2, 3])
+        s.info(buf=buf)
+        assert "+" not in buf.getvalue()
+
+        buf = StringIO()
+        s = pd.Series(1, index=list("ABC"))
+        s.info(buf=buf)
+        assert "+" in buf.getvalue()
+
+        buf = StringIO()
+        s = Series(1, index=pd.MultiIndex.from_product([range(3), range(3)]),)
+        s.info(buf=buf)
+        assert "+" not in buf.getvalue()
+
+        buf = StringIO()
+        s = Series(1, index=pd.MultiIndex.from_product([range(3), ["foo", "bar"]]),)
+        s.info(buf=buf)
+        assert "+" in buf.getvalue()
+
+    def test_info_memory_usage_bug_on_multiindex(self):
+        # GH 14308
+        # memory usage introspection should not materialize .values
+
+        from string import ascii_uppercase as uppercase
+
+        def memory_usage(f):
+            return f.memory_usage(deep=True)
+
+        N = 100
+        M = len(uppercase)
+        index = pd.MultiIndex.from_product(
+            [list(uppercase), pd.date_range("20160101", periods=N)],
+            names=["id", "date"],
+        )
+        s = Series(np.random.randn(N * M), index=index)
+
+        unstacked = s.unstack("id")
+        assert s.values.nbytes == unstacked.values.nbytes
+        assert s.memory_usage(deep=True) > unstacked.memory_usage(deep=True).sum()
+
+        # high upper bound
+        assert (
+            unstacked.memory_usage(deep=True).sum() - s.memory_usage(deep=True) < 2000
+        )
+
+    def test_info_categorical(self):
+        # GH14298
+        idx = pd.CategoricalIndex(["a", "b"])
+        s = pd.Series(np.zeros((2)), index=idx)
+
+        buf = StringIO()
+        s.info(buf=buf)
+
+    def test_info_categorical_column(self):
+
+        # make sure it works
+        n = 2500
+        s = Series(
+            np.array(list("abcdefghij")).take(np.random.randint(0, 10, size=n))
+        ).astype("category")
+        s.isna()
+        buf = StringIO()
+        s.info(buf=buf)
+
+        s2 = s[s == "d"]
+        buf = StringIO()
+        s2.info(buf=buf)