From 0fdbfd34add4fe9940cd1991122e851d62824724 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 09:26:57 -0600
Subject: [PATCH 01/39] wip
---
doc/source/whatsnew/v0.24.0.txt | 1 +
pandas/core/arrays/base.py | 40 ++++++++++++++++++-
pandas/core/arrays/categorical.py | 2 +
pandas/core/arrays/integer.py | 24 +----------
pandas/core/arrays/interval.py | 10 -----
pandas/core/arrays/period.py | 8 ----
pandas/io/formats/printing.py | 37 +++++++++++++----
pandas/tests/arrays/interval/test_interval.py | 32 +++++++++++----
pandas/tests/arrays/test_integer.py | 29 +++++++-------
pandas/tests/arrays/test_period.py | 33 ++++++++++++++-
pandas/tests/extension/base/__init__.py | 1 +
pandas/tests/extension/base/interface.py | 25 ------------
pandas/tests/extension/base/printing.py | 38 ++++++++++++++++++
pandas/tests/extension/decimal/array.py | 3 --
.../tests/extension/decimal/test_decimal.py | 6 ++-
pandas/tests/extension/json/array.py | 3 --
pandas/tests/extension/json/test_json.py | 7 +++-
pandas/tests/extension/test_integer.py | 6 ++-
pandas/tests/extension/test_interval.py | 6 ++-
pandas/tests/extension/test_period.py | 6 ++-
pandas/tests/extension/test_sparse.py | 6 +++
21 files changed, 213 insertions(+), 110 deletions(-)
create mode 100644 pandas/tests/extension/base/printing.py
diff --git a/doc/source/whatsnew/v0.24.0.txt b/doc/source/whatsnew/v0.24.0.txt
index 5fefb9e3e405c..1378caeac5edb 100644
--- a/doc/source/whatsnew/v0.24.0.txt
+++ b/doc/source/whatsnew/v0.24.0.txt
@@ -857,6 +857,7 @@ update the ``ExtensionDtype._metadata`` tuple to match the signature of your
- :meth:`DataFrame.stack` no longer converts to object dtype for DataFrames where each column has the same extension dtype. The output Series will have the same dtype as the columns (:issue:`23077`).
- :meth:`Series.unstack` and :meth:`DataFrame.unstack` no longer convert extension arrays to object-dtype ndarrays. Each column in the output ``DataFrame`` will now have the same dtype as the input (:issue:`23077`).
- Bug when grouping :meth:`Dataframe.groupby()` and aggregating on ``ExtensionArray`` it was not returning the actual ``ExtensionArray`` dtype (:issue:`23227`).
+- A default repr is now provided.
.. _whatsnew_0240.api.incompatibilities:
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index f842d1237cb14..41ec0971bef7c 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -49,6 +49,13 @@ class ExtensionArray(object):
* _formatting_values
+ A default repr displaying the type, (truncated) data, length,
+ and dtype is provided. It can be customized or replaced by
+ by overriding:
+
+ * _formatter
+ * __repr__
+
Some methods require casting the ExtensionArray to an ndarray of Python
objects with ``self.astype(object)``, which may be expensive. When
performance is a concern, we highly recommend overriding the following
@@ -653,8 +660,35 @@ def copy(self, deep=False):
raise AbstractMethodError(self)
# ------------------------------------------------------------------------
- # Block-related methods
+ # Printing
# ------------------------------------------------------------------------
+ def __repr__(self):
+ from pandas.io.formats.printing import format_object_summary
+
+ template = (
+ '<{class_name}>\n'
+ '{data}\n'
+ 'Length: {length}, dtype: {dtype}'
+ )
+ # the short repr has no trailing newline, while the truncated
+ # repr does. So we include a newline in our template, and strip
+ # any trailing newlines from format_object_summary
+ data = format_object_summary(self, self._formatter, name=False,
+ trailing_comma=False).rstrip('\n')
+ name = self.__class__.__name__
+ return template.format(class_name=name, data=data,
+ length=len(self),
+ dtype=self.dtype)
+
+ @property
+ def _formatter(self):
+ # type: () -> Callable[Any]
+ """Formatting function for scalar values.
+
+ This is used in the default '__repr__'. The formatting function
+ receives instances of your scalar type.
+ """
+ return str
def _formatting_values(self):
# type: () -> np.ndarray
@@ -662,6 +696,10 @@ def _formatting_values(self):
"""An array of values to be printed in, e.g. the Series repr"""
return np.array(self)
+ # ------------------------------------------------------------------------
+ # Reshaping
+ # ------------------------------------------------------------------------
+
@classmethod
def _concat_same_type(cls, to_concat):
# type: (Sequence[ExtensionArray]) -> ExtensionArray
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 4363f3ccb14e2..b74938a9c7b18 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -1986,6 +1986,8 @@ def __unicode__(self):
return result
+ __repr__ = __unicode__
+
def _maybe_coerce_indexer(self, indexer):
""" return an indexer coerced to the codes dtype """
if isinstance(indexer, np.ndarray) and indexer.dtype.kind == 'i':
diff --git a/pandas/core/arrays/integer.py b/pandas/core/arrays/integer.py
index 9e045a7785660..27c845d666e60 100644
--- a/pandas/core/arrays/integer.py
+++ b/pandas/core/arrays/integer.py
@@ -6,7 +6,7 @@
from pandas._libs import lib
from pandas.util._decorators import cache_readonly
-from pandas.compat import u, range, string_types
+from pandas.compat import range, string_types
from pandas.compat import set_function_name
from pandas.core import nanops
@@ -24,9 +24,6 @@
from pandas.core.dtypes.dtypes import register_extension_dtype
from pandas.core.dtypes.missing import isna, notna
-from pandas.io.formats.printing import (
- format_object_summary, format_object_attrs, default_pprint)
-
class _IntegerDtype(ExtensionDtype):
"""
@@ -353,25 +350,6 @@ def __setitem__(self, key, value):
def __len__(self):
return len(self._data)
- def __repr__(self):
- """
- Return a string representation for this object.
-
- Invoked by unicode(df) in py2 only. Yields a Unicode String in both
- py2/py3.
- """
- klass = self.__class__.__name__
- data = format_object_summary(self, default_pprint, False)
- attrs = format_object_attrs(self)
- space = " "
-
- prepr = (u(",%s") %
- space).join(u("%s=%s") % (k, v) for k, v in attrs)
-
- res = u("%s(%s%s)") % (klass, data, prepr)
-
- return res
-
@property
def nbytes(self):
return self._data.nbytes + self._mask.nbytes
diff --git a/pandas/core/arrays/interval.py b/pandas/core/arrays/interval.py
index 94be29893d2b9..1b3687e555748 100644
--- a/pandas/core/arrays/interval.py
+++ b/pandas/core/arrays/interval.py
@@ -835,16 +835,6 @@ def _format_data(self):
return summary
- def __repr__(self):
- tpl = textwrap.dedent("""\
- {cls}({data},
- {lead}closed='{closed}',
- {lead}dtype='{dtype}')""")
- return tpl.format(cls=self.__class__.__name__,
- data=self._format_data(),
- lead=' ' * len(self.__class__.__name__) + ' ',
- closed=self.closed, dtype=self.dtype)
-
def _format_space(self):
space = ' ' * (len(self.__class__.__name__) + 1)
return "\n{space}".format(space=space)
diff --git a/pandas/core/arrays/period.py b/pandas/core/arrays/period.py
index 5a75f2706b218..f6996f8e62d4d 100644
--- a/pandas/core/arrays/period.py
+++ b/pandas/core/arrays/period.py
@@ -330,14 +330,6 @@ def start_time(self):
def end_time(self):
return self.to_timestamp(how='end')
- def __repr__(self):
- return '<{}>\n{}\nLength: {}, dtype: {}'.format(
- self.__class__.__name__,
- [str(s) for s in self],
- len(self),
- self.dtype
- )
-
def __setitem__(
self,
key, # type: Union[int, Sequence[int], Sequence[bool]]
diff --git a/pandas/io/formats/printing.py b/pandas/io/formats/printing.py
index f814bf965a1e9..3d56b4b222ba6 100644
--- a/pandas/io/formats/printing.py
+++ b/pandas/io/formats/printing.py
@@ -271,7 +271,9 @@ class TableSchemaFormatter(BaseFormatter):
max_seq_items=max_seq_items)
-def format_object_summary(obj, formatter, is_justify=True, name=None):
+def format_object_summary(obj, formatter, is_justify=True, name=None,
+ trailing_comma=True,
+ truncated_trailing_newline=True):
"""
Return the formatted obj as a unicode string
@@ -283,9 +285,14 @@ def format_object_summary(obj, formatter, is_justify=True, name=None):
string formatter for an element
is_justify : boolean
should justify the display
- name : name, optiona
+ name : name, optional
defaults to the class name of the obj
+ Pass ``False`` to indicate that subsequent lines should
+ not be indented to align with the name.
+ trailing_comma : bool, default True
+ Whether to include a comma after the closing ']'
+
Returns
-------
summary string
@@ -300,8 +307,13 @@ def format_object_summary(obj, formatter, is_justify=True, name=None):
if name is None:
name = obj.__class__.__name__
- space1 = "\n%s" % (' ' * (len(name) + 1))
- space2 = "\n%s" % (' ' * (len(name) + 2))
+ if name is False:
+ space1 = "\n"
+ space2 = "\n " # space for the opening '['
+ else:
+ name_len = len(name)
+ space1 = "\n%s" % (' ' * (name_len + 1))
+ space2 = "\n%s" % (' ' * (name_len + 2))
n = len(obj)
sep = ','
@@ -328,15 +340,20 @@ def best_len(values):
else:
return 0
+ if trailing_comma:
+ close = ', '
+ else:
+ close = ''
+
if n == 0:
- summary = '[], '
+ summary = '[]{}'.format(close)
elif n == 1:
first = formatter(obj[0])
- summary = '[%s], ' % first
+ summary = '[{}]{}'.format(first, close)
elif n == 2:
first = formatter(obj[0])
last = formatter(obj[-1])
- summary = '[%s, %s], ' % (first, last)
+ summary = '[{}, {}]{}'.format(first, last, close)
else:
if n > max_seq_items:
@@ -381,7 +398,11 @@ def best_len(values):
summary, line = _extend_line(summary, line, tail[-1],
display_width - 2, space2)
summary += line
- summary += '],'
+
+ # right now close is either '' or ', '
+ # Now we want to include the ']', but not the maybe space.
+ close = ']' + close.rstrip(' ')
+ summary += close
if len(summary) > (display_width):
summary += space1
diff --git a/pandas/tests/arrays/interval/test_interval.py b/pandas/tests/arrays/interval/test_interval.py
index 9a191dda3a73a..1138f64e2009a 100644
--- a/pandas/tests/arrays/interval/test_interval.py
+++ b/pandas/tests/arrays/interval/test_interval.py
@@ -1,10 +1,10 @@
# -*- coding: utf-8 -*-
import numpy as np
-import pytest
-from pandas import Index, IntervalIndex, date_range, timedelta_range
+from pandas import Index, date_range, option_context, timedelta_range
from pandas.core.arrays import IntervalArray
import pandas.util.testing as tm
+import pytest
@pytest.fixture(params=[
@@ -65,8 +65,26 @@ def test_set_na(self, left_right_dtypes):
tm.assert_extension_array_equal(result, expected)
-def test_repr_matches():
- idx = IntervalIndex.from_breaks([1, 2, 3])
- a = repr(idx)
- b = repr(idx.values)
- assert a.replace("Index", "Array") == b
+def test_repr_small():
+ arr = IntervalArray.from_breaks([1, 2, 3])
+ result = repr(arr)
+ expected = (
+ '<IntervalArray>\n'
+ '[(1, 2], (2, 3]]\n'
+ 'Length: 2, dtype: interval[int64]'
+ )
+ assert result == expected
+
+
+def test_repr_large():
+ arr = IntervalArray.from_breaks([1, 2, 3, 4, 5, 6])
+ with option_context('display.max_seq_items', 2):
+ result = repr(arr)
+ expected = (
+ '<IntervalArray>\n'
+ '[(1, 2],\n'
+ ' ...\n'
+ ' (5, 6]] \n'
+ 'Length: 5, dtype: interval[int64]'
+ )
+ assert result == expected
diff --git a/pandas/tests/arrays/test_integer.py b/pandas/tests/arrays/test_integer.py
index 0fe07caed5b85..6b99ede3436ce 100644
--- a/pandas/tests/arrays/test_integer.py
+++ b/pandas/tests/arrays/test_integer.py
@@ -1,6 +1,5 @@
# -*- coding: utf-8 -*-
import numpy as np
-import pytest
from pandas.core.dtypes.generic import ABCIndexClass
@@ -12,6 +11,7 @@
UInt32Dtype, UInt64Dtype)
from pandas.tests.extension.base import BaseOpsUtil
import pandas.util.testing as tm
+import pytest
def make_data():
@@ -57,24 +57,23 @@ def test_dtypes(dtype):
assert dtype.name is not None
-class TestInterface(object):
+def test_repr_array(data):
+ result = repr(data)
+ assert '<IntegerArray>' in result
- def test_repr_array(self, data):
- result = repr(data)
+ # not long
+ assert '...' not in result
+ assert 'Length: ' in result
+ assert 'dtype: ' in result
- # not long
- assert '...' not in result
- assert 'dtype=' in result
- assert 'IntegerArray' in result
-
- def test_repr_array_long(self, data):
- # some arrays may be able to assert a ... in the repr
- with pd.option_context('display.max_seq_items', 1):
- result = repr(data)
+def test_repr_array_long(data):
+ # some arrays may be able to assert a ... in the repr
+ with pd.option_context('display.max_seq_items', 1):
+ result = repr(data)
- assert '...' in result
- assert 'length' in result
+ assert '...' in result
+ assert 'Length' in result
class TestConstructors(object):
diff --git a/pandas/tests/arrays/test_period.py b/pandas/tests/arrays/test_period.py
index 0125729048cdd..553f187e2e8fe 100644
--- a/pandas/tests/arrays/test_period.py
+++ b/pandas/tests/arrays/test_period.py
@@ -1,5 +1,4 @@
import numpy as np
-import pytest
from pandas._libs.tslibs import iNaT
from pandas._libs.tslibs.period import IncompatibleFrequency
@@ -10,6 +9,7 @@
import pandas as pd
from pandas.core.arrays import PeriodArray, period_array
import pandas.util.testing as tm
+import pytest
# ----------------------------------------------------------------------------
# Constructors
@@ -195,3 +195,34 @@ def tet_sub_period():
other = pd.Period("2000", freq="M")
with tm.assert_raises_regex(IncompatibleFrequency, "freq"):
arr - other
+
+
+# ----------------------------------------------------------------------------
+# Printing
+
+def test_repr_small():
+ arr = period_array(['2000', '2001'], freq='D')
+ result = str(arr)
+ expected = (
+ '<PeriodArray>\n'
+ '[2000-01-01, 2001-01-01]\n'
+ 'Length: 2, dtype: period[D]'
+ )
+ assert result == expected
+
+
+def test_repr_large():
+ arr = period_array(['2000', '2001'] * 500, freq='D')
+ result = str(arr)
+ expected = (
+ '<PeriodArray>\n'
+ '[2000-01-01, 2001-01-01, 2000-01-01, 2001-01-01, 2000-01-01, '
+ '2001-01-01,\n' # continuation
+ ' 2000-01-01, 2001-01-01, 2000-01-01, 2001-01-01,\n'
+ ' ...\n'
+ ' 2000-01-01, 2001-01-01, 2000-01-01, 2001-01-01, 2000-01-01, '
+ '2001-01-01,\n' # continuation
+ ' 2000-01-01, 2001-01-01, 2000-01-01, 2001-01-01]\n'
+ 'Length: 1000, dtype: period[D]'
+ )
+ assert result == expected
diff --git a/pandas/tests/extension/base/__init__.py b/pandas/tests/extension/base/__init__.py
index d11bb8b6beb77..57704b77bb233 100644
--- a/pandas/tests/extension/base/__init__.py
+++ b/pandas/tests/extension/base/__init__.py
@@ -48,6 +48,7 @@ class TestMyDtype(BaseDtypeTests):
from .interface import BaseInterfaceTests # noqa
from .methods import BaseMethodsTests # noqa
from .ops import BaseArithmeticOpsTests, BaseComparisonOpsTests, BaseOpsUtil # noqa
+from .printing import BasePrintingTests # noqa
from .reduce import BaseNoReduceTests, BaseNumericReduceTests, BaseBooleanReduceTests # noqa
from .missing import BaseMissingTests # noqa
from .reshaping import BaseReshapingTests # noqa
diff --git a/pandas/tests/extension/base/interface.py b/pandas/tests/extension/base/interface.py
index 00a480d311b58..f8464dbac8053 100644
--- a/pandas/tests/extension/base/interface.py
+++ b/pandas/tests/extension/base/interface.py
@@ -1,7 +1,5 @@
import numpy as np
-from pandas.compat import StringIO
-
from pandas.core.dtypes.common import is_extension_array_dtype
from pandas.core.dtypes.dtypes import ExtensionDtype
@@ -35,29 +33,6 @@ def test_array_interface(self, data):
result = np.array(data)
assert result[0] == data[0]
- def test_repr(self, data):
- ser = pd.Series(data)
- assert data.dtype.name in repr(ser)
-
- df = pd.DataFrame({"A": data})
- repr(df)
-
- def test_repr_array(self, data):
- # some arrays may be able to assert
- # attributes in the repr
- repr(data)
-
- def test_repr_array_long(self, data):
- # some arrays may be able to assert a ... in the repr
- with pd.option_context('display.max_seq_items', 1):
- repr(data)
-
- def test_dtype_name_in_info(self, data):
- buf = StringIO()
- pd.DataFrame({"A": data}).info(buf=buf)
- result = buf.getvalue()
- assert data.dtype.name in result
-
def test_is_extension_array_dtype(self, data):
assert is_extension_array_dtype(data)
assert is_extension_array_dtype(data.dtype)
diff --git a/pandas/tests/extension/base/printing.py b/pandas/tests/extension/base/printing.py
new file mode 100644
index 0000000000000..3d224c2b06c0b
--- /dev/null
+++ b/pandas/tests/extension/base/printing.py
@@ -0,0 +1,38 @@
+import io
+
+import pandas as pd
+import pytest
+
+from .base import BaseExtensionTests
+
+
+class BasePrintingTests(BaseExtensionTests):
+ """Tests checking the formatting of your EA when printed."""
+
+ @pytest.mark.parametrize("size", ["big", "small"])
+ def test_array_repr(self, data, size):
+ if size == "small":
+ data = data[:5]
+ else:
+ data = type(data)._concat_same_type([data] * 5)
+
+ result = repr(data)
+ assert data.__class__.__name__ in result
+ assert 'Length: {}'.format(len(data)) in result
+ assert str(data.dtype) in result
+ if size == 'big':
+ assert '...' in result
+
+ def test_series_repr(self, data):
+ ser = pd.Series(data)
+ assert data.dtype.name in repr(ser)
+
+ def test_dataframe_repr(self, data):
+ df = pd.DataFrame({"A": data})
+ repr(df)
+
+ def test_dtype_name_in_info(self, data):
+ buf = io.StringIO()
+ pd.DataFrame({"A": data}).info(buf=buf)
+ result = buf.getvalue()
+ assert data.dtype.name in result
diff --git a/pandas/tests/extension/decimal/array.py b/pandas/tests/extension/decimal/array.py
index 3c8905c578c4f..79e81f1034c6d 100644
--- a/pandas/tests/extension/decimal/array.py
+++ b/pandas/tests/extension/decimal/array.py
@@ -114,9 +114,6 @@ def __setitem__(self, key, value):
def __len__(self):
return len(self._data)
- def __repr__(self):
- return 'DecimalArray({!r})'.format(self._data)
-
@property
def nbytes(self):
n = len(self)
diff --git a/pandas/tests/extension/decimal/test_decimal.py b/pandas/tests/extension/decimal/test_decimal.py
index af5f6bf0a2f65..1a37b781dfcda 100644
--- a/pandas/tests/extension/decimal/test_decimal.py
+++ b/pandas/tests/extension/decimal/test_decimal.py
@@ -3,12 +3,12 @@
import operator
import numpy as np
-import pytest
import pandas as pd
from pandas import compat
from pandas.tests.extension import base
import pandas.util.testing as tm
+import pytest
from .array import DecimalArray, DecimalDtype, make_data, to_decimal
@@ -200,6 +200,10 @@ class TestSetitem(BaseDecimal, base.BaseSetitemTests):
pass
+class TestPrinting(BaseDecimal, base.BasePrintingTests):
+ pass
+
+
# TODO(extension)
@pytest.mark.xfail(reason=(
"raising AssertionError as this is not implemented, "
diff --git a/pandas/tests/extension/json/array.py b/pandas/tests/extension/json/array.py
index 2c6e74fda8a0e..d58b7ddf29123 100644
--- a/pandas/tests/extension/json/array.py
+++ b/pandas/tests/extension/json/array.py
@@ -115,9 +115,6 @@ def __setitem__(self, key, value):
def __len__(self):
return len(self.data)
- def __repr__(self):
- return 'JSONArary({!r})'.format(self.data)
-
@property
def nbytes(self):
return sys.getsizeof(self.data)
diff --git a/pandas/tests/extension/json/test_json.py b/pandas/tests/extension/json/test_json.py
index b7c61496f0bf0..6934b9affb39a 100644
--- a/pandas/tests/extension/json/test_json.py
+++ b/pandas/tests/extension/json/test_json.py
@@ -1,13 +1,12 @@
import collections
import operator
-import pytest
-
from pandas.compat import PY2, PY36
import pandas as pd
from pandas.tests.extension import base
import pandas.util.testing as tm
+import pytest
from .array import JSONArray, JSONDtype, make_data
@@ -283,3 +282,7 @@ def _check_divmod_op(self, s, op, other, exc=NotImplementedError):
class TestComparisonOps(BaseJSON, base.BaseComparisonOpsTests):
pass
+
+
+class TestPrinting(BaseJSON, base.BasePrintingTests):
+ pass
diff --git a/pandas/tests/extension/test_integer.py b/pandas/tests/extension/test_integer.py
index efee647389884..4d5aa54b65320 100644
--- a/pandas/tests/extension/test_integer.py
+++ b/pandas/tests/extension/test_integer.py
@@ -14,7 +14,6 @@
"""
import numpy as np
-import pytest
from pandas.core.dtypes.common import is_extension_array_dtype
@@ -24,6 +23,7 @@
Int8Dtype, Int16Dtype, Int32Dtype, Int64Dtype, UInt8Dtype, UInt16Dtype,
UInt32Dtype, UInt64Dtype)
from pandas.tests.extension import base
+import pytest
def make_data():
@@ -210,3 +210,7 @@ class TestNumericReduce(base.BaseNumericReduceTests):
class TestBooleanReduce(base.BaseBooleanReduceTests):
pass
+
+
+class TestPrinting(base.BasePrintingTests):
+ pass
diff --git a/pandas/tests/extension/test_interval.py b/pandas/tests/extension/test_interval.py
index 2c7bc79c324b4..8924c87200cae 100644
--- a/pandas/tests/extension/test_interval.py
+++ b/pandas/tests/extension/test_interval.py
@@ -14,7 +14,6 @@
"""
import numpy as np
-import pytest
from pandas.core.dtypes.dtypes import IntervalDtype
@@ -22,6 +21,7 @@
from pandas.core.arrays import IntervalArray
from pandas.tests.extension import base
import pandas.util.testing as tm
+import pytest
def make_data():
@@ -147,3 +147,7 @@ class TestReshaping(BaseInterval, base.BaseReshapingTests):
class TestSetitem(BaseInterval, base.BaseSetitemTests):
pass
+
+
+class TestPrinting(BaseInterval, base.BasePrintingTests):
+ pass
diff --git a/pandas/tests/extension/test_period.py b/pandas/tests/extension/test_period.py
index 83f30aed88e65..d6d6a6366f49f 100644
--- a/pandas/tests/extension/test_period.py
+++ b/pandas/tests/extension/test_period.py
@@ -1,5 +1,4 @@
import numpy as np
-import pytest
from pandas._libs.tslib import iNaT
@@ -9,6 +8,7 @@
from pandas.core.arrays import PeriodArray
from pandas.tests.extension import base
import pandas.util.testing as tm
+import pytest
@pytest.fixture
@@ -155,3 +155,7 @@ class TestSetitem(BasePeriodTests, base.BaseSetitemTests):
class TestGroupby(BasePeriodTests, base.BaseGroupbyTests):
pass
+
+
+class TestPrinting(BasePeriodTests, base.BasePrintingTests):
+ pass
diff --git a/pandas/tests/extension/test_sparse.py b/pandas/tests/extension/test_sparse.py
index 4f67a13215cfd..891e5f4dd9a95 100644
--- a/pandas/tests/extension/test_sparse.py
+++ b/pandas/tests/extension/test_sparse.py
@@ -316,3 +316,9 @@ def _compare_other(self, s, data, op_name, other):
s = pd.Series(data)
result = op(s, other)
tm.assert_series_equal(result, expected)
+
+
+class TestPrinting(BaseSparseTests, base.BasePrintingTests):
+ @pytest.mark.xfail(reason='Different repr', strict=True)
+ def test_array_repr(self, data, size):
+ super(TestPrinting, self).test_array_repr(data, size)
From ace62aaedf9e877de8bd937eda37d04398a196d7 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 12:07:47 -0600
Subject: [PATCH 02/39] Deprecate formatting_values
---
doc/source/whatsnew/v0.24.0.txt | 3 ++-
pandas/core/arrays/base.py | 13 +++++++++----
pandas/core/arrays/categorical.py | 3 ---
pandas/core/arrays/integer.py | 4 ----
pandas/core/arrays/interval.py | 3 ---
pandas/core/arrays/period.py | 5 +++++
pandas/core/internals/blocks.py | 18 ++++++++++++++++--
pandas/io/formats/printing.py | 10 +++++-----
pandas/tests/arrays/test_period.py | 6 +++---
pandas/tests/extension/decimal/test_decimal.py | 3 ++-
10 files changed, 42 insertions(+), 26 deletions(-)
diff --git a/doc/source/whatsnew/v0.24.0.txt b/doc/source/whatsnew/v0.24.0.txt
index 1378caeac5edb..ac985e11c1203 100644
--- a/doc/source/whatsnew/v0.24.0.txt
+++ b/doc/source/whatsnew/v0.24.0.txt
@@ -857,7 +857,7 @@ update the ``ExtensionDtype._metadata`` tuple to match the signature of your
- :meth:`DataFrame.stack` no longer converts to object dtype for DataFrames where each column has the same extension dtype. The output Series will have the same dtype as the columns (:issue:`23077`).
- :meth:`Series.unstack` and :meth:`DataFrame.unstack` no longer convert extension arrays to object-dtype ndarrays. Each column in the output ``DataFrame`` will now have the same dtype as the input (:issue:`23077`).
- Bug when grouping :meth:`Dataframe.groupby()` and aggregating on ``ExtensionArray`` it was not returning the actual ``ExtensionArray`` dtype (:issue:`23227`).
-- A default repr is now provided.
+- A default repr for ExtensionArrays is now provided (:issue:`23601`).
.. _whatsnew_0240.api.incompatibilities:
@@ -967,6 +967,7 @@ Deprecations
- The class ``FrozenNDArray`` has been deprecated. When unpickling, ``FrozenNDArray`` will be unpickled to ``np.ndarray`` once this class is removed (:issue:`9031`)
- Deprecated the `nthreads` keyword of :func:`pandas.read_feather` in favor of
`use_threads` to reflect the changes in pyarrow 0.11.0. (:issue:`23053`)
+- :meth:`ExtensionArray._formatting_values` is deprecated. Use `ExtensionArray._formatter` instead. (:issue:`23601`)
.. _whatsnew_0240.deprecations.datetimelike_int_ops:
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 41ec0971bef7c..238dbdd5576b2 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -673,20 +673,25 @@ def __repr__(self):
# the short repr has no trailing newline, while the truncated
# repr does. So we include a newline in our template, and strip
# any trailing newlines from format_object_summary
- data = format_object_summary(self, self._formatter, name=False,
+ data = format_object_summary(self, self._formatter(), name=False,
trailing_comma=False).rstrip('\n')
name = self.__class__.__name__
return template.format(class_name=name, data=data,
length=len(self),
dtype=self.dtype)
- @property
- def _formatter(self):
- # type: () -> Callable[Any]
+ def _formatter(self, boxed=False):
+ # type: (bool) -> Callable[Any]
"""Formatting function for scalar values.
This is used in the default '__repr__'. The formatting function
receives instances of your scalar type.
+
+ Parameters
+ ----------
+ boxed: bool, default False
+ Whether the formatter is to be used by pandas inside a Series
+ or DataFrame repr.
"""
return str
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index b74938a9c7b18..f25009ba72cc9 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -2344,9 +2344,6 @@ def _concat_same_type(self, to_concat):
return _concat_categorical(to_concat)
- def _formatting_values(self):
- return self
-
def isin(self, values):
"""
Check whether `values` are contained in Categorical.
diff --git a/pandas/core/arrays/integer.py b/pandas/core/arrays/integer.py
index 27c845d666e60..3224284788dc1 100644
--- a/pandas/core/arrays/integer.py
+++ b/pandas/core/arrays/integer.py
@@ -297,10 +297,6 @@ def __iter__(self):
else:
yield self._data[i]
- def _formatting_values(self):
- # type: () -> np.ndarray
- return self._coerce_to_ndarray()
-
def take(self, indexer, allow_fill=False, fill_value=None):
from pandas.api.extensions import take
diff --git a/pandas/core/arrays/interval.py b/pandas/core/arrays/interval.py
index 1b3687e555748..51b406214ed85 100644
--- a/pandas/core/arrays/interval.py
+++ b/pandas/core/arrays/interval.py
@@ -689,9 +689,6 @@ def copy(self, deep=False):
# TODO: Could skip verify_integrity here.
return type(self).from_arrays(left, right, closed=closed)
- def _formatting_values(self):
- return np.asarray(self)
-
def isna(self):
return isna(self.left)
diff --git a/pandas/core/arrays/period.py b/pandas/core/arrays/period.py
index f6996f8e62d4d..b3cfe8860fccb 100644
--- a/pandas/core/arrays/period.py
+++ b/pandas/core/arrays/period.py
@@ -330,6 +330,11 @@ def start_time(self):
def end_time(self):
return self.to_timestamp(how='end')
+ def _formatter(self, boxed=False):
+ if boxed:
+ return str
+ return "'{}'".format
+
def __setitem__(
self,
key, # type: Union[int, Sequence[int], Sequence[bool]]
diff --git a/pandas/core/internals/blocks.py b/pandas/core/internals/blocks.py
index 1f2a1ee52159e..600b2d18a7554 100644
--- a/pandas/core/internals/blocks.py
+++ b/pandas/core/internals/blocks.py
@@ -66,7 +66,7 @@
import pandas.core.missing as missing
from pandas.core.base import PandasObject
-from pandas.core.arrays import Categorical
+from pandas.core.arrays import Categorical, ExtensionArray
from pandas.core.indexes.datetimes import DatetimeIndex
from pandas.core.indexes.timedeltas import TimedeltaIndex
@@ -1951,7 +1951,21 @@ def _slice(self, slicer):
return self.values[slicer]
def formatting_values(self):
- return self.values._formatting_values()
+ # Deprecating the ability to override _formatting_values.
+ # Do the warning here, it's only user in pandas, since we
+ # have to check if the subclass overrode it.
+ fv = getattr(type(self.values), '_formatting_values', None)
+ if fv is not ExtensionArray._formatting_values:
+ msg = (
+ "'ExtensionArray._formatting_values' is deprecated. "
+ "Specify 'ExtensionArray._formatter' instead."
+ )
+ warnings.warn(msg, FutureWarning)
+ return self.values._formatting_values()
+
+ # the future implementation (and current, if not overrode)
+ formatter = self.values._formatter(boxed=True)
+ return np.array([formatter(x) for x in self.values], dtype=object)
def concat_same_type(self, to_concat, placement=None):
"""
diff --git a/pandas/io/formats/printing.py b/pandas/io/formats/printing.py
index 3d56b4b222ba6..baeab662db9f1 100644
--- a/pandas/io/formats/printing.py
+++ b/pandas/io/formats/printing.py
@@ -341,19 +341,19 @@ def best_len(values):
return 0
if trailing_comma:
- close = ', '
+ close = u', '
else:
- close = ''
+ close = u''
if n == 0:
- summary = '[]{}'.format(close)
+ summary = u'[]{}'.format(close)
elif n == 1:
first = formatter(obj[0])
- summary = '[{}]{}'.format(first, close)
+ summary = u'[{}]{}'.format(first, close)
elif n == 2:
first = formatter(obj[0])
last = formatter(obj[-1])
- summary = '[{}, {}]{}'.format(first, last, close)
+ summary = u'[{}, {}]{}'.format(first, last, close)
else:
if n > max_seq_items:
diff --git a/pandas/tests/arrays/test_period.py b/pandas/tests/arrays/test_period.py
index 553f187e2e8fe..cc9fef90d959f 100644
--- a/pandas/tests/arrays/test_period.py
+++ b/pandas/tests/arrays/test_period.py
@@ -204,9 +204,9 @@ def test_repr_small():
arr = period_array(['2000', '2001'], freq='D')
result = str(arr)
expected = (
- '<PeriodArray>\n'
- '[2000-01-01, 2001-01-01]\n'
- 'Length: 2, dtype: period[D]'
+ "<PeriodArray>\n"
+ "['2000-01-01', '2001-01-01']\n"
+ "Length: 2, dtype: period[D]"
)
assert result == expected
diff --git a/pandas/tests/extension/decimal/test_decimal.py b/pandas/tests/extension/decimal/test_decimal.py
index 1a37b781dfcda..2e7afc378fa44 100644
--- a/pandas/tests/extension/decimal/test_decimal.py
+++ b/pandas/tests/extension/decimal/test_decimal.py
@@ -188,7 +188,8 @@ def test_value_counts(self, all_data, dropna):
class TestCasting(BaseDecimal, base.BaseCastingTests):
- pass
+ pytestmark = pytest.mark.skipif(compat.PY2,
+ reason="Unhashble dtype in Py2.")
class TestGroupby(BaseDecimal, base.BaseGroupbyTests):
From 6e76b51dced8477d6e63c175854180838aa6283e Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 12:16:03 -0600
Subject: [PATCH 03/39] test for warning
---
pandas/core/internals/blocks.py | 2 +-
pandas/tests/extension/decimal/test_decimal.py | 14 +++++++++++++-
2 files changed, 14 insertions(+), 2 deletions(-)
diff --git a/pandas/core/internals/blocks.py b/pandas/core/internals/blocks.py
index 600b2d18a7554..b9c13196aea1d 100644
--- a/pandas/core/internals/blocks.py
+++ b/pandas/core/internals/blocks.py
@@ -1960,7 +1960,7 @@ def formatting_values(self):
"'ExtensionArray._formatting_values' is deprecated. "
"Specify 'ExtensionArray._formatter' instead."
)
- warnings.warn(msg, FutureWarning)
+ warnings.warn(msg, FutureWarning, stacklevel=10)
return self.values._formatting_values()
# the future implementation (and current, if not overrode)
diff --git a/pandas/tests/extension/decimal/test_decimal.py b/pandas/tests/extension/decimal/test_decimal.py
index 2e7afc378fa44..5dd53439df47f 100644
--- a/pandas/tests/extension/decimal/test_decimal.py
+++ b/pandas/tests/extension/decimal/test_decimal.py
@@ -202,7 +202,8 @@ class TestSetitem(BaseDecimal, base.BaseSetitemTests):
class TestPrinting(BaseDecimal, base.BasePrintingTests):
- pass
+ pytestmark = pytest.mark.skipif(compat.PY2,
+ reason="Unhashble dtype in Py2.")
# TODO(extension)
@@ -384,3 +385,14 @@ def test_divmod_array(reverse, expected_div, expected_mod):
tm.assert_extension_array_equal(div, expected_div)
tm.assert_extension_array_equal(mod, expected_mod)
+
+
+def test_formatting_values_deprecated():
+ class DecimalArray2(DecimalArray):
+ def _formatting_values(self):
+ return np.array(self)
+
+ ser = pd.Series(DecimalArray2([decimal.Decimal('1.0')]))
+
+ with tm.assert_produces_warning(FutureWarning):
+ repr(ser)
From fef04e64927e749c4788a3a2df71b666aecbca5a Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 12:19:58 -0600
Subject: [PATCH 04/39] compat
---
pandas/tests/extension/decimal/test_decimal.py | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/pandas/tests/extension/decimal/test_decimal.py b/pandas/tests/extension/decimal/test_decimal.py
index 5dd53439df47f..3f7581484ee7e 100644
--- a/pandas/tests/extension/decimal/test_decimal.py
+++ b/pandas/tests/extension/decimal/test_decimal.py
@@ -393,6 +393,9 @@ def _formatting_values(self):
return np.array(self)
ser = pd.Series(DecimalArray2([decimal.Decimal('1.0')]))
+ # different levels for 2 vs. 3
+ check_stacklevel = compat.PY3
- with tm.assert_produces_warning(FutureWarning):
+ with tm.assert_produces_warning(FutureWarning,
+ check_stacklevel=check_stacklevel):
repr(ser)
From 1885a97203ecba615657b197d9a2f178aa105065 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 13:22:29 -0600
Subject: [PATCH 05/39] na formatter
---
pandas/core/arrays/integer.py | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/pandas/core/arrays/integer.py b/pandas/core/arrays/integer.py
index 3224284788dc1..984ef4c8c673c 100644
--- a/pandas/core/arrays/integer.py
+++ b/pandas/core/arrays/integer.py
@@ -264,6 +264,13 @@ def _from_sequence(cls, scalars, dtype=None, copy=False):
def _from_factorized(cls, values, original):
return integer_array(values, dtype=original.dtype)
+ def _formatter(self, boxed=False):
+ def fmt(x):
+ if isna(x):
+ return 'NaN'
+ return str(x)
+ return fmt
+
def __getitem__(self, item):
if is_integer(item):
if self._mask[item]:
From ecfcd72043f1d5e2e742f652f991f42006a4cf19 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 13:33:25 -0600
Subject: [PATCH 06/39] clean
---
pandas/io/formats/printing.py | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/pandas/io/formats/printing.py b/pandas/io/formats/printing.py
index baeab662db9f1..61c62d49cc6cc 100644
--- a/pandas/io/formats/printing.py
+++ b/pandas/io/formats/printing.py
@@ -272,8 +272,7 @@ class TableSchemaFormatter(BaseFormatter):
def format_object_summary(obj, formatter, is_justify=True, name=None,
- trailing_comma=True,
- truncated_trailing_newline=True):
+ trailing_comma=True):
"""
Return the formatted obj as a unicode string
From 37638cc9a820ce7981350e2ceefcea9317a9e2f8 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 14:29:25 -0600
Subject: [PATCH 07/39] wip
---
pandas/core/arrays/base.py | 6 +--
pandas/core/internals/blocks.py | 4 +-
pandas/io/formats/format.py | 60 +++++++++++++-----------------
pandas/tests/arrays/test_period.py | 20 +++++-----
4 files changed, 40 insertions(+), 50 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 238dbdd5576b2..e15eb74119578 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -680,8 +680,8 @@ def __repr__(self):
length=len(self),
dtype=self.dtype)
- def _formatter(self, boxed=False):
- # type: (bool) -> Callable[Any]
+ def _formatter(self, formatter):
+ # type: (ExtensionArrayFormatter) -> Callable[Any]
"""Formatting function for scalar values.
This is used in the default '__repr__'. The formatting function
@@ -693,7 +693,7 @@ def _formatter(self, boxed=False):
Whether the formatter is to be used by pandas inside a Series
or DataFrame repr.
"""
- return str
+ return formatter.formatter or str
def _formatting_values(self):
# type: () -> np.ndarray
diff --git a/pandas/core/internals/blocks.py b/pandas/core/internals/blocks.py
index b9c13196aea1d..267b769ce7a65 100644
--- a/pandas/core/internals/blocks.py
+++ b/pandas/core/internals/blocks.py
@@ -1963,9 +1963,7 @@ def formatting_values(self):
warnings.warn(msg, FutureWarning, stacklevel=10)
return self.values._formatting_values()
- # the future implementation (and current, if not overrode)
- formatter = self.values._formatter(boxed=True)
- return np.array([formatter(x) for x in self.values], dtype=object)
+ return self.values
def concat_same_type(self, to_concat, placement=None):
"""
diff --git a/pandas/io/formats/format.py b/pandas/io/formats/format.py
index 6f64605bcf175..22d91a47c4082 100644
--- a/pandas/io/formats/format.py
+++ b/pandas/io/formats/format.py
@@ -16,11 +16,12 @@
from pandas.compat import StringIO, lzip, map, u, zip
from pandas.core.dtypes.common import (
- is_categorical_dtype, is_datetime64_dtype, is_datetimetz, is_float,
- is_float_dtype, is_integer, is_integer_dtype, is_interval_dtype,
- is_list_like, is_numeric_dtype, is_period_arraylike, is_scalar,
- is_timedelta64_dtype)
-from pandas.core.dtypes.generic import ABCMultiIndex, ABCSparseArray
+ is_categorical_dtype, is_datetime64_dtype, is_datetimetz,
+ is_extension_array_dtype, is_float, is_float_dtype, is_integer,
+ is_integer_dtype, is_list_like, is_numeric_dtype, is_period_arraylike,
+ is_scalar, is_timedelta64_dtype)
+from pandas.core.dtypes.generic import (
+ ABCIndex, ABCMultiIndex, ABCSeries, ABCSparseArray)
from pandas.core.dtypes.missing import isna, notna
from pandas import compat
@@ -849,22 +850,20 @@ def _get_column_name_list(self):
def format_array(values, formatter, float_format=None, na_rep='NaN',
digits=None, space=None, justify='right', decimal='.'):
- if is_categorical_dtype(values):
- fmt_klass = CategoricalArrayFormatter
- elif is_interval_dtype(values):
- fmt_klass = IntervalArrayFormatter
+ if is_period_arraylike(values):
+ fmt_klass = PeriodArrayFormatter
+ elif is_datetime64_dtype(values.dtype):
+ fmt_klass = Datetime64Formatter
+ elif is_timedelta64_dtype(values.dtype):
+ fmt_klass = Timedelta64Formatter
+ elif is_extension_array_dtype(values.dtype):
+ fmt_klass = ExtensionArrayFormatter
elif is_float_dtype(values.dtype):
fmt_klass = FloatArrayFormatter
- elif is_period_arraylike(values):
- fmt_klass = PeriodArrayFormatter
elif is_integer_dtype(values.dtype):
fmt_klass = IntArrayFormatter
elif is_datetimetz(values):
fmt_klass = Datetime64TZFormatter
- elif is_datetime64_dtype(values.dtype):
- fmt_klass = Datetime64Formatter
- elif is_timedelta64_dtype(values.dtype):
- fmt_klass = Timedelta64Formatter
else:
fmt_klass = GenericArrayFormatter
@@ -1126,14 +1125,18 @@ def _format_strings(self):
return fmt_values.tolist()
-class IntervalArrayFormatter(GenericArrayFormatter):
-
- def __init__(self, values, *args, **kwargs):
- GenericArrayFormatter.__init__(self, values, *args, **kwargs)
-
+class ExtensionArrayFormatter(GenericArrayFormatter):
def _format_strings(self):
- formatter = self.formatter or str
- fmt_values = np.array([formatter(x) for x in self.values])
+ values = self.values
+ if isinstance(values, (ABCIndex, ABCSeries)):
+ values = values._values
+
+ formatter = self.values._formatter(self)
+ fmt_values = format_array(np.asarray(self.values),
+ formatter,
+ float_format=self.float_format,
+ na_rep=self.na_rep, digits=self.digits,
+ space=self.space, justify=self.justify)
return fmt_values
@@ -1152,19 +1155,6 @@ def _format_strings(self):
return fmt_values
-class CategoricalArrayFormatter(GenericArrayFormatter):
-
- def __init__(self, values, *args, **kwargs):
- GenericArrayFormatter.__init__(self, values, *args, **kwargs)
-
- def _format_strings(self):
- fmt_values = format_array(self.values.get_values(), self.formatter,
- float_format=self.float_format,
- na_rep=self.na_rep, digits=self.digits,
- space=self.space, justify=self.justify)
- return fmt_values
-
-
def format_percentiles(percentiles):
"""
Outputs rounded and formatted percentiles.
diff --git a/pandas/tests/arrays/test_period.py b/pandas/tests/arrays/test_period.py
index cc9fef90d959f..245d932bb139e 100644
--- a/pandas/tests/arrays/test_period.py
+++ b/pandas/tests/arrays/test_period.py
@@ -215,14 +215,16 @@ def test_repr_large():
arr = period_array(['2000', '2001'] * 500, freq='D')
result = str(arr)
expected = (
- '<PeriodArray>\n'
- '[2000-01-01, 2001-01-01, 2000-01-01, 2001-01-01, 2000-01-01, '
- '2001-01-01,\n' # continuation
- ' 2000-01-01, 2001-01-01, 2000-01-01, 2001-01-01,\n'
- ' ...\n'
- ' 2000-01-01, 2001-01-01, 2000-01-01, 2001-01-01, 2000-01-01, '
- '2001-01-01,\n' # continuation
- ' 2000-01-01, 2001-01-01, 2000-01-01, 2001-01-01]\n'
- 'Length: 1000, dtype: period[D]'
+ "<PeriodArray>\n"
+ "['2000-01-01', '2001-01-01', '2000-01-01', '2001-01-01', "
+ "'2000-01-01',\n"
+ " '2001-01-01', '2000-01-01', '2001-01-01', '2000-01-01', "
+ "'2001-01-01',\n"
+ " ...\n"
+ " '2000-01-01', '2001-01-01', '2000-01-01', '2001-01-01', "
+ "'2000-01-01',\n"
+ " '2001-01-01', '2000-01-01', '2001-01-01', '2000-01-01', "
+ "'2001-01-01']\n"
+ "Length: 1000, dtype: period[D]"
)
assert result == expected
From 6e64b7bd87d8fb6e79151863e367e55466cffda4 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 14:47:45 -0600
Subject: [PATCH 08/39] more cleanup
---
pandas/core/arrays/base.py | 4 ++--
pandas/core/arrays/categorical.py | 4 ++++
pandas/io/formats/format.py | 32 +++++++------------------------
3 files changed, 13 insertions(+), 27 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index e15eb74119578..d4a8da03913c3 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -680,7 +680,7 @@ def __repr__(self):
length=len(self),
dtype=self.dtype)
- def _formatter(self, formatter):
+ def _formatter(self, formatter=None):
# type: (ExtensionArrayFormatter) -> Callable[Any]
"""Formatting function for scalar values.
@@ -693,7 +693,7 @@ def _formatter(self, formatter):
Whether the formatter is to be used by pandas inside a Series
or DataFrame repr.
"""
- return formatter.formatter or str
+ return getattr(formatter, 'formatter', None) or str
def _formatting_values(self):
# type: () -> np.ndarray
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index f25009ba72cc9..82edddb4c2414 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -499,6 +499,10 @@ def _constructor(self):
def _from_sequence(cls, scalars, dtype=None, copy=False):
return Categorical(scalars, dtype=dtype)
+ def _formatter(self, formatter):
+ # backwards compat with old printing.
+ return None
+
def copy(self):
""" Copy constructor. """
return self._constructor(values=self._codes.copy(),
diff --git a/pandas/io/formats/format.py b/pandas/io/formats/format.py
index 22d91a47c4082..42e3a633b2f42 100644
--- a/pandas/io/formats/format.py
+++ b/pandas/io/formats/format.py
@@ -18,10 +18,10 @@
from pandas.core.dtypes.common import (
is_categorical_dtype, is_datetime64_dtype, is_datetimetz,
is_extension_array_dtype, is_float, is_float_dtype, is_integer,
- is_integer_dtype, is_list_like, is_numeric_dtype, is_period_arraylike,
- is_scalar, is_timedelta64_dtype)
+ is_integer_dtype, is_list_like, is_numeric_dtype, is_scalar,
+ is_timedelta64_dtype)
from pandas.core.dtypes.generic import (
- ABCIndex, ABCMultiIndex, ABCSeries, ABCSparseArray)
+ ABCIndexClass, ABCMultiIndex, ABCSeries, ABCSparseArray)
from pandas.core.dtypes.missing import isna, notna
from pandas import compat
@@ -30,7 +30,6 @@
from pandas.core.config import get_option, set_option
from pandas.core.index import Index, ensure_index
from pandas.core.indexes.datetimes import DatetimeIndex
-from pandas.core.indexes.period import PeriodIndex
from pandas.io.common import _expand_user, _stringify_path
from pandas.io.formats.printing import adjoin, justify, pprint_thing
@@ -850,9 +849,7 @@ def _get_column_name_list(self):
def format_array(values, formatter, float_format=None, na_rep='NaN',
digits=None, space=None, justify='right', decimal='.'):
- if is_period_arraylike(values):
- fmt_klass = PeriodArrayFormatter
- elif is_datetime64_dtype(values.dtype):
+ if is_datetime64_dtype(values.dtype):
fmt_klass = Datetime64Formatter
elif is_timedelta64_dtype(values.dtype):
fmt_klass = Timedelta64Formatter
@@ -1128,11 +1125,11 @@ def _format_strings(self):
class ExtensionArrayFormatter(GenericArrayFormatter):
def _format_strings(self):
values = self.values
- if isinstance(values, (ABCIndex, ABCSeries)):
+ if isinstance(values, (ABCIndexClass, ABCSeries)):
values = values._values
- formatter = self.values._formatter(self)
- fmt_values = format_array(np.asarray(self.values),
+ formatter = values._formatter(self)
+ fmt_values = format_array(np.asarray(values),
formatter,
float_format=self.float_format,
na_rep=self.na_rep, digits=self.digits,
@@ -1140,21 +1137,6 @@ def _format_strings(self):
return fmt_values
-class PeriodArrayFormatter(IntArrayFormatter):
-
- def _format_strings(self):
- from pandas.core.indexes.period import IncompatibleFrequency
- try:
- values = PeriodIndex(self.values).to_native_types()
- except IncompatibleFrequency:
- # periods may contains different freq
- values = Index(self.values, dtype='object').to_native_types()
-
- formatter = self.formatter or (lambda x: '{x}'.format(x=x))
- fmt_values = [formatter(x) for x in values]
- return fmt_values
-
-
def format_percentiles(percentiles):
"""
Outputs rounded and formatted percentiles.
From 193747e9c3b5d09315e8848495e9e6f91437c733 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 14:56:09 -0600
Subject: [PATCH 09/39] update docs, type
---
pandas/core/arrays/base.py | 16 ++++++++++++----
pandas/core/arrays/integer.py | 14 ++++++++------
pandas/core/arrays/period.py | 4 ++--
pandas/tests/arrays/test_integer.py | 5 +++++
4 files changed, 27 insertions(+), 12 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index d4a8da03913c3..be0b272f2816f 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -681,7 +681,7 @@ def __repr__(self):
dtype=self.dtype)
def _formatter(self, formatter=None):
- # type: (ExtensionArrayFormatter) -> Callable[Any]
+ # type: (Optional[ExtensionArrayFormatter]) -> Callable[[Any], str]
"""Formatting function for scalar values.
This is used in the default '__repr__'. The formatting function
@@ -689,9 +689,17 @@ def _formatter(self, formatter=None):
Parameters
----------
- boxed: bool, default False
- Whether the formatter is to be used by pandas inside a Series
- or DataFrame repr.
+ formatter: GenericArrayFormatter, optional
+ The formatter this array is being rendered with. The formatter
+ may have a `.formatter` method already defined. By default, this
+ will be used if a `formatter` is passed, otherwise the formatter
+ is ``str``.
+
+ Returns
+ -------
+ Callable[[Any], str]
+ A callable that gets instances of the scalar type and
+ returns a string.
"""
return getattr(formatter, 'formatter', None) or str
diff --git a/pandas/core/arrays/integer.py b/pandas/core/arrays/integer.py
index 984ef4c8c673c..98eb9df98328f 100644
--- a/pandas/core/arrays/integer.py
+++ b/pandas/core/arrays/integer.py
@@ -264,12 +264,14 @@ def _from_sequence(cls, scalars, dtype=None, copy=False):
def _from_factorized(cls, values, original):
return integer_array(values, dtype=original.dtype)
- def _formatter(self, boxed=False):
- def fmt(x):
- if isna(x):
- return 'NaN'
- return str(x)
- return fmt
+ def _formatter(self, formatter=None):
+ if formatter is None:
+ def fmt(x):
+ if isna(x):
+ return 'NaN'
+ return str(x)
+ return fmt
+ return formatter.formatter
def __getitem__(self, item):
if is_integer(item):
diff --git a/pandas/core/arrays/period.py b/pandas/core/arrays/period.py
index 4f2848b3f5591..36a00db7625ce 100644
--- a/pandas/core/arrays/period.py
+++ b/pandas/core/arrays/period.py
@@ -330,8 +330,8 @@ def start_time(self):
def end_time(self):
return self.to_timestamp(how='end')
- def _formatter(self, boxed=False):
- if boxed:
+ def _formatter(self, formatter=None):
+ if formatter:
return str
return "'{}'".format
diff --git a/pandas/tests/arrays/test_integer.py b/pandas/tests/arrays/test_integer.py
index 6b99ede3436ce..94bf3dfd938a1 100644
--- a/pandas/tests/arrays/test_integer.py
+++ b/pandas/tests/arrays/test_integer.py
@@ -67,6 +67,11 @@ def test_repr_array(data):
assert 'dtype: ' in result
+def test_na_repr(data):
+ result = repr(integer_array([1, None]))
+ assert 'NaN' in result
+
+
def test_repr_array_long(data):
# some arrays may be able to assert a ... in the repr
with pd.option_context('display.max_seq_items', 1):
From 5a2e1e4bc5f03f84499f143861c547e9e248e692 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 15:02:39 -0600
Subject: [PATCH 10/39] format
---
pandas/core/arrays/base.py | 1 +
1 file changed, 1 insertion(+)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index be0b272f2816f..7bdb54bec02fb 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -662,6 +662,7 @@ def copy(self, deep=False):
# ------------------------------------------------------------------------
# Printing
# ------------------------------------------------------------------------
+
def __repr__(self):
from pandas.io.formats.printing import format_object_summary
From 1635b7319264829a2b1136220f4780e2965ad907 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 15:35:53 -0600
Subject: [PATCH 11/39] try this
---
pandas/io/formats/format.py | 9 ++++++++-
pandas/tests/frame/test_repr_info.py | 12 ++++++------
2 files changed, 14 insertions(+), 7 deletions(-)
diff --git a/pandas/io/formats/format.py b/pandas/io/formats/format.py
index 42e3a633b2f42..a7567e52f3077 100644
--- a/pandas/io/formats/format.py
+++ b/pandas/io/formats/format.py
@@ -1129,7 +1129,14 @@ def _format_strings(self):
values = values._values
formatter = values._formatter(self)
- fmt_values = format_array(np.asarray(values),
+
+ if is_categorical_dtype(values.dtype):
+ # Categorical is special for now, so that we can preserve tzinfo
+ array = values.get_values()
+ else:
+ array = np.asarray(values)
+
+ fmt_values = format_array(array,
formatter,
float_format=self.float_format,
na_rep=self.na_rep, digits=self.digits,
diff --git a/pandas/tests/frame/test_repr_info.py b/pandas/tests/frame/test_repr_info.py
index 668613c494a47..96c523b10197f 100644
--- a/pandas/tests/frame/test_repr_info.py
+++ b/pandas/tests/frame/test_repr_info.py
@@ -514,12 +514,12 @@ def test_repr_categorical_dates_periods(self):
tz='US/Eastern')
p = period_range('2011-01', freq='M', periods=5)
df = DataFrame({'dt': dt, 'p': p})
- exp = """ dt p
-0 2011-01-01 09:00:00-05:00 2011-01
-1 2011-01-01 10:00:00-05:00 2011-02
-2 2011-01-01 11:00:00-05:00 2011-03
-3 2011-01-01 12:00:00-05:00 2011-04
-4 2011-01-01 13:00:00-05:00 2011-05"""
+ exp = """ dt p
+0 2011-01-01 09:00:00-05:00 2011-01
+1 2011-01-01 10:00:00-05:00 2011-02
+2 2011-01-01 11:00:00-05:00 2011-03
+3 2011-01-01 12:00:00-05:00 2011-04
+4 2011-01-01 13:00:00-05:00 2011-05"""
df = DataFrame({'dt': Categorical(dt), 'p': Categorical(p)})
assert repr(df) == exp
From e2b1941f72ace652af8d125e6ef4aa85d18f6112 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 16:05:33 -0600
Subject: [PATCH 12/39] updates
---
doc/source/whatsnew/v0.24.0.txt | 1 +
pandas/core/arrays/period.py | 2 +-
pandas/core/arrays/sparse.py | 10 ++++++++
pandas/tests/series/test_repr.py | 40 ++++++++++++++++----------------
4 files changed, 32 insertions(+), 21 deletions(-)
diff --git a/doc/source/whatsnew/v0.24.0.txt b/doc/source/whatsnew/v0.24.0.txt
index ac985e11c1203..1b839866ba190 100644
--- a/doc/source/whatsnew/v0.24.0.txt
+++ b/doc/source/whatsnew/v0.24.0.txt
@@ -1120,6 +1120,7 @@ Datetimelike
- Bug in rounding methods of :class:`DatetimeIndex` (:meth:`~DatetimeIndex.round`, :meth:`~DatetimeIndex.ceil`, :meth:`~DatetimeIndex.floor`) and :class:`Timestamp` (:meth:`~Timestamp.round`, :meth:`~Timestamp.ceil`, :meth:`~Timestamp.floor`) could give rise to loss of precision (:issue:`22591`)
- Bug in :func:`to_datetime` with an :class:`Index` argument that would drop the ``name`` from the result (:issue:`21697`)
- Bug in :class:`PeriodIndex` where adding or subtracting a :class:`timedelta` or :class:`Tick` object produced incorrect results (:issue:`22988`)
+- Bug in the :class:`Series` repr with Period data missing a space before the data (:issue:`23601`)
- Bug in :func:`date_range` when decrementing a start date to a past end date by a negative frequency (:issue:`23270`)
- Bug in :meth:`Series.min` which would return ``NaN`` instead of ``NaT`` when called on a series of ``NaT`` (:issue:`23282`)
- Bug in :func:`DataFrame.combine` with datetimelike values raising a TypeError (:issue:`23079`)
diff --git a/pandas/core/arrays/period.py b/pandas/core/arrays/period.py
index 36a00db7625ce..6d7ab6e6176d1 100644
--- a/pandas/core/arrays/period.py
+++ b/pandas/core/arrays/period.py
@@ -332,7 +332,7 @@ def end_time(self):
def _formatter(self, formatter=None):
if formatter:
- return str
+ return formatter.formatter or str
return "'{}'".format
def __setitem__(
diff --git a/pandas/core/arrays/sparse.py b/pandas/core/arrays/sparse.py
index a63b3fb53625f..dd0d4ba587c6c 100644
--- a/pandas/core/arrays/sparse.py
+++ b/pandas/core/arrays/sparse.py
@@ -1661,6 +1661,16 @@ def __unicode__(self):
fill=printing.pprint_thing(self.fill_value),
index=printing.pprint_thing(self.sp_index))
+ def _formatter(self, formatter=None):
+ if formatter is None:
+ def fmt(x):
+ if isna(x) and isinstance(x, float):
+ return 'NaN'
+ return str(x)
+
+ return fmt
+ return formatter.formatter
+
SparseArray._add_arithmetic_ops()
SparseArray._add_comparison_ops()
diff --git a/pandas/tests/series/test_repr.py b/pandas/tests/series/test_repr.py
index ef96274746655..c4a0496f7fb27 100644
--- a/pandas/tests/series/test_repr.py
+++ b/pandas/tests/series/test_repr.py
@@ -364,11 +364,11 @@ def test_categorical_series_repr_datetime_ordered(self):
def test_categorical_series_repr_period(self):
idx = period_range('2011-01-01 09:00', freq='H', periods=5)
s = Series(Categorical(idx))
- exp = """0 2011-01-01 09:00
-1 2011-01-01 10:00
-2 2011-01-01 11:00
-3 2011-01-01 12:00
-4 2011-01-01 13:00
+ exp = """0 2011-01-01 09:00
+1 2011-01-01 10:00
+2 2011-01-01 11:00
+3 2011-01-01 12:00
+4 2011-01-01 13:00
dtype: category
Categories (5, period[H]): [2011-01-01 09:00, 2011-01-01 10:00, 2011-01-01 11:00, 2011-01-01 12:00,
2011-01-01 13:00]""" # noqa
@@ -377,11 +377,11 @@ def test_categorical_series_repr_period(self):
idx = period_range('2011-01', freq='M', periods=5)
s = Series(Categorical(idx))
- exp = """0 2011-01
-1 2011-02
-2 2011-03
-3 2011-04
-4 2011-05
+ exp = """0 2011-01
+1 2011-02
+2 2011-03
+3 2011-04
+4 2011-05
dtype: category
Categories (5, period[M]): [2011-01, 2011-02, 2011-03, 2011-04, 2011-05]"""
@@ -390,11 +390,11 @@ def test_categorical_series_repr_period(self):
def test_categorical_series_repr_period_ordered(self):
idx = period_range('2011-01-01 09:00', freq='H', periods=5)
s = Series(Categorical(idx, ordered=True))
- exp = """0 2011-01-01 09:00
-1 2011-01-01 10:00
-2 2011-01-01 11:00
-3 2011-01-01 12:00
-4 2011-01-01 13:00
+ exp = """0 2011-01-01 09:00
+1 2011-01-01 10:00
+2 2011-01-01 11:00
+3 2011-01-01 12:00
+4 2011-01-01 13:00
dtype: category
Categories (5, period[H]): [2011-01-01 09:00 < 2011-01-01 10:00 < 2011-01-01 11:00 < 2011-01-01 12:00 <
2011-01-01 13:00]""" # noqa
@@ -403,11 +403,11 @@ def test_categorical_series_repr_period_ordered(self):
idx = period_range('2011-01', freq='M', periods=5)
s = Series(Categorical(idx, ordered=True))
- exp = """0 2011-01
-1 2011-02
-2 2011-03
-3 2011-04
-4 2011-05
+ exp = """0 2011-01
+1 2011-02
+2 2011-03
+3 2011-04
+4 2011-05
dtype: category
Categories (5, period[M]): [2011-01 < 2011-02 < 2011-03 < 2011-04 < 2011-05]"""
From 48e55ccaa06908a6dab1b27cfef63f00549771ff Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 20:52:28 -0600
Subject: [PATCH 13/39] fixup interval
---
pandas/core/arrays/interval.py | 10 ++++++++++
pandas/tests/extension/test_interval.py | 4 +++-
2 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/pandas/core/arrays/interval.py b/pandas/core/arrays/interval.py
index 51b406214ed85..244952e8e884c 100644
--- a/pandas/core/arrays/interval.py
+++ b/pandas/core/arrays/interval.py
@@ -832,6 +832,16 @@ def _format_data(self):
return summary
+ def __repr__(self):
+ tpl = textwrap.dedent("""\
+ {cls}({data},
+ {lead}closed='{closed}',
+ {lead}dtype='{dtype}')""")
+ return tpl.format(cls=self.__class__.__name__,
+ data=self._format_data(),
+ lead=' ' * len(self.__class__.__name__) + ' ',
+ closed=self.closed, dtype=self.dtype)
+
def _format_space(self):
space = ' ' * (len(self.__class__.__name__) + 1)
return "\n{space}".format(space=space)
diff --git a/pandas/tests/extension/test_interval.py b/pandas/tests/extension/test_interval.py
index 8924c87200cae..281a61cb55463 100644
--- a/pandas/tests/extension/test_interval.py
+++ b/pandas/tests/extension/test_interval.py
@@ -150,4 +150,6 @@ class TestSetitem(BaseInterval, base.BaseSetitemTests):
class TestPrinting(BaseInterval, base.BasePrintingTests):
- pass
+ @pytest.mark.skip(reason="custom repr")
+ def test_array_repr(self, data, size):
+ pass
From d8e7ba475b358877d3fcd5c93a7ba607de4cc2af Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Fri, 9 Nov 2018 21:08:41 -0600
Subject: [PATCH 14/39] py2 compat
---
pandas/core/internals/blocks.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/pandas/core/internals/blocks.py b/pandas/core/internals/blocks.py
index 267b769ce7a65..2a8d49feb0bad 100644
--- a/pandas/core/internals/blocks.py
+++ b/pandas/core/internals/blocks.py
@@ -1955,7 +1955,7 @@ def formatting_values(self):
# Do the warning here, it's only user in pandas, since we
# have to check if the subclass overrode it.
fv = getattr(type(self.values), '_formatting_values', None)
- if fv is not ExtensionArray._formatting_values:
+ if fv and fv != ExtensionArray._formatting_values:
msg = (
"'ExtensionArray._formatting_values' is deprecated. "
"Specify 'ExtensionArray._formatter' instead."
From b312fe410bbcc67fd9db0ca3218a85df2af22f80 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sat, 10 Nov 2018 06:22:33 -0600
Subject: [PATCH 15/39] revert interval
---
pandas/tests/arrays/interval/test_interval.py | 32 ++++---------------
1 file changed, 7 insertions(+), 25 deletions(-)
diff --git a/pandas/tests/arrays/interval/test_interval.py b/pandas/tests/arrays/interval/test_interval.py
index 1138f64e2009a..9a191dda3a73a 100644
--- a/pandas/tests/arrays/interval/test_interval.py
+++ b/pandas/tests/arrays/interval/test_interval.py
@@ -1,10 +1,10 @@
# -*- coding: utf-8 -*-
import numpy as np
+import pytest
-from pandas import Index, date_range, option_context, timedelta_range
+from pandas import Index, IntervalIndex, date_range, timedelta_range
from pandas.core.arrays import IntervalArray
import pandas.util.testing as tm
-import pytest
@pytest.fixture(params=[
@@ -65,26 +65,8 @@ def test_set_na(self, left_right_dtypes):
tm.assert_extension_array_equal(result, expected)
-def test_repr_small():
- arr = IntervalArray.from_breaks([1, 2, 3])
- result = repr(arr)
- expected = (
- '<IntervalArray>\n'
- '[(1, 2], (2, 3]]\n'
- 'Length: 2, dtype: interval[int64]'
- )
- assert result == expected
-
-
-def test_repr_large():
- arr = IntervalArray.from_breaks([1, 2, 3, 4, 5, 6])
- with option_context('display.max_seq_items', 2):
- result = repr(arr)
- expected = (
- '<IntervalArray>\n'
- '[(1, 2],\n'
- ' ...\n'
- ' (5, 6]] \n'
- 'Length: 5, dtype: interval[int64]'
- )
- assert result == expected
+def test_repr_matches():
+ idx = IntervalIndex.from_breaks([1, 2, 3])
+ a = repr(idx)
+ b = repr(idx.values)
+ assert a.replace("Index", "Array") == b
From 445736d939dc1ce8615767ccb79f84c8bfdad947 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sat, 10 Nov 2018 06:30:57 -0600
Subject: [PATCH 16/39] unicode, bytes
---
pandas/core/arrays/base.py | 15 +++++++++++++++
pandas/tests/extension/base/printing.py | 12 ++++++++++++
2 files changed, 27 insertions(+)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 7bdb54bec02fb..cfcf2e98c2bc6 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -9,6 +9,8 @@
import operator
+from pandas import compat
+from pandas.core.config import get_option
from pandas.core.dtypes.generic import ABCSeries, ABCIndexClass
from pandas.errors import AbstractMethodError
from pandas.compat.numpy import function as nv
@@ -662,6 +664,19 @@ def copy(self, deep=False):
# ------------------------------------------------------------------------
# Printing
# ------------------------------------------------------------------------
+ def __unicode__(self):
+ result = str(self)
+ if compat.PY2:
+ encoding = get_option("display.encoding")
+ result = result.decode(encoding)
+ return result
+
+ def __bytes__(self):
+ result = str(self)
+ if compat.PY3:
+ encoding = get_option("display.encoding")
+ result = result.encode(encoding)
+ return result
def __repr__(self):
from pandas.io.formats.printing import format_object_summary
diff --git a/pandas/tests/extension/base/printing.py b/pandas/tests/extension/base/printing.py
index 3d224c2b06c0b..512ef23f4c247 100644
--- a/pandas/tests/extension/base/printing.py
+++ b/pandas/tests/extension/base/printing.py
@@ -1,6 +1,7 @@
import io
import pandas as pd
+from pandas import compat
import pytest
from .base import BaseExtensionTests
@@ -23,6 +24,17 @@ def test_array_repr(self, data, size):
if size == 'big':
assert '...' in result
+ def test_array_repr_bytes(self, data):
+ result = bytes(data)
+ if compat.PY2:
+ assert isinstance(result, str)
+ else:
+ assert isinstance(result, bytes)
+
+ def test_array_repr_unicode(self, data):
+ result = compat.u(data)
+ assert isinstance(result, compat.text_type)
+
def test_series_repr(self, data):
ser = pd.Series(data)
assert data.dtype.name in repr(ser)
From 60e0d028b53d99a0ea52ed0943e70971081e342f Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sat, 10 Nov 2018 06:36:39 -0600
Subject: [PATCH 17/39] isort
---
pandas/tests/extension/base/printing.py | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/pandas/tests/extension/base/printing.py b/pandas/tests/extension/base/printing.py
index 512ef23f4c247..f13a3c244ccb8 100644
--- a/pandas/tests/extension/base/printing.py
+++ b/pandas/tests/extension/base/printing.py
@@ -1,8 +1,9 @@
import io
+import pytest
+
import pandas as pd
from pandas import compat
-import pytest
from .base import BaseExtensionTests
From 5b07906cf28879ba174d3f356eab10df1fd8a858 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sat, 10 Nov 2018 06:52:16 -0600
Subject: [PATCH 18/39] py3 fixup
---
pandas/tests/extension/base/printing.py | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/pandas/tests/extension/base/printing.py b/pandas/tests/extension/base/printing.py
index f13a3c244ccb8..a04863b7a63b1 100644
--- a/pandas/tests/extension/base/printing.py
+++ b/pandas/tests/extension/base/printing.py
@@ -33,7 +33,10 @@ def test_array_repr_bytes(self, data):
assert isinstance(result, bytes)
def test_array_repr_unicode(self, data):
- result = compat.u(data)
+ if compat.PY2:
+ result = compat.u(data)
+ else:
+ result = str(data)
assert isinstance(result, compat.text_type)
def test_series_repr(self, data):
From ff0c9981986cad9877ddcdf593e0c424b0b3f25d Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sat, 10 Nov 2018 14:41:43 -0600
Subject: [PATCH 19/39] fixup
---
pandas/core/arrays/base.py | 32 +++++++++++++++----------------
pandas/core/arrays/categorical.py | 2 --
2 files changed, 15 insertions(+), 19 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index cfcf2e98c2bc6..054ff1b360e8f 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -665,26 +665,12 @@ def copy(self, deep=False):
# Printing
# ------------------------------------------------------------------------
def __unicode__(self):
- result = str(self)
- if compat.PY2:
- encoding = get_option("display.encoding")
- result = result.decode(encoding)
- return result
-
- def __bytes__(self):
- result = str(self)
- if compat.PY3:
- encoding = get_option("display.encoding")
- result = result.encode(encoding)
- return result
-
- def __repr__(self):
from pandas.io.formats.printing import format_object_summary
template = (
- '<{class_name}>\n'
- '{data}\n'
- 'Length: {length}, dtype: {dtype}'
+ u'<{class_name}>\n'
+ u'{data}\n'
+ u'Length: {length}, dtype: {dtype}'
)
# the short repr has no trailing newline, while the truncated
# repr does. So we include a newline in our template, and strip
@@ -696,6 +682,18 @@ def __repr__(self):
length=len(self),
dtype=self.dtype)
+ def __str__(self):
+ if compat.PY3:
+ return self.__unicode__()
+ return self.__bytes__()
+
+ def __bytes__(self):
+ encoding = get_option("display.encoding")
+ return self.__unicode__().encode(encoding, 'replace')
+
+ def __repr__(self):
+ return str(self)
+
def _formatter(self, formatter=None):
# type: (Optional[ExtensionArrayFormatter]) -> Callable[[Any], str]
"""Formatting function for scalar values.
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 82edddb4c2414..3593f83770114 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -1990,8 +1990,6 @@ def __unicode__(self):
return result
- __repr__ = __unicode__
-
def _maybe_coerce_indexer(self, indexer):
""" return an indexer coerced to the codes dtype """
if isinstance(indexer, np.ndarray) and indexer.dtype.kind == 'i':
From 2fd3d5d0dba6369503ad0d71fda3155b1ffb8bcf Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sat, 10 Nov 2018 14:45:22 -0600
Subject: [PATCH 20/39] unicode
---
pandas/core/arrays/base.py | 17 ++---------------
1 file changed, 2 insertions(+), 15 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 054ff1b360e8f..3e529cc0a4c11 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -9,8 +9,7 @@
import operator
-from pandas import compat
-from pandas.core.config import get_option
+from pandas.core.base import StringMixin
from pandas.core.dtypes.generic import ABCSeries, ABCIndexClass
from pandas.errors import AbstractMethodError
from pandas.compat.numpy import function as nv
@@ -21,7 +20,7 @@
_not_implemented_message = "{} does not implement {}."
-class ExtensionArray(object):
+class ExtensionArray(StringMixin):
"""Abstract base class for custom 1-D array types.
pandas will recognize instances of this class as proper arrays
@@ -682,18 +681,6 @@ def __unicode__(self):
length=len(self),
dtype=self.dtype)
- def __str__(self):
- if compat.PY3:
- return self.__unicode__()
- return self.__bytes__()
-
- def __bytes__(self):
- encoding = get_option("display.encoding")
- return self.__unicode__().encode(encoding, 'replace')
-
- def __repr__(self):
- return str(self)
-
def _formatter(self, formatter=None):
# type: (Optional[ExtensionArrayFormatter]) -> Callable[[Any], str]
"""Formatting function for scalar values.
From 5d8d2fc9c7e7d0541d991e2f3fb107e57bd23a09 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sat, 10 Nov 2018 14:45:55 -0600
Subject: [PATCH 21/39] unicode
---
pandas/core/arrays/base.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 3e529cc0a4c11..b6eba6a17359a 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -55,7 +55,7 @@ class ExtensionArray(StringMixin):
by overriding:
* _formatter
- * __repr__
+ * __unicode__
Some methods require casting the ExtensionArray to an ndarray of Python
objects with ``self.astype(object)``, which may be expensive. When
From 4d343eafead092d5d8a6895f4df9c7206b5510fc Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sat, 10 Nov 2018 16:30:55 -0600
Subject: [PATCH 22/39] unicode
---
pandas/tests/extension/base/printing.py | 5 +----
setup.cfg | 2 +-
2 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/pandas/tests/extension/base/printing.py b/pandas/tests/extension/base/printing.py
index a04863b7a63b1..7a41e7689f4a6 100644
--- a/pandas/tests/extension/base/printing.py
+++ b/pandas/tests/extension/base/printing.py
@@ -33,10 +33,7 @@ def test_array_repr_bytes(self, data):
assert isinstance(result, bytes)
def test_array_repr_unicode(self, data):
- if compat.PY2:
- result = compat.u(data)
- else:
- result = str(data)
+ result = compat.text_type(data)
assert isinstance(result, compat.text_type)
def test_series_repr(self, data):
diff --git a/setup.cfg b/setup.cfg
index 4726a0ddb2fb2..2e07182196d5b 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -90,7 +90,7 @@ known_post_core=pandas.tseries,pandas.io,pandas.plotting
sections=FUTURE,STDLIB,THIRDPARTY,PRE_CORE,DTYPES,FIRSTPARTY,POST_CORE,LOCALFOLDER
known_first_party=pandas
-known_third_party=Cython,numpy,python-dateutil,pytz,pyarrow
+known_third_party=Cython,numpy,python-dateutil,pytz,pyarrow,pytest
multi_line_output=4
force_grid_wrap=0
combine_as_imports=True
From 5b291d50e8722a9ef72ffa90f0476ff85a7edb66 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sat, 10 Nov 2018 16:33:20 -0600
Subject: [PATCH 23/39] lint
---
pandas/tests/arrays/test_integer.py | 2 +-
pandas/tests/arrays/test_period.py | 2 +-
pandas/tests/extension/decimal/test_decimal.py | 2 +-
pandas/tests/extension/json/test_json.py | 3 ++-
pandas/tests/extension/test_integer.py | 2 +-
pandas/tests/extension/test_interval.py | 2 +-
pandas/tests/extension/test_period.py | 2 +-
7 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/pandas/tests/arrays/test_integer.py b/pandas/tests/arrays/test_integer.py
index 206cfab3ab6a4..ebae1a2797bdc 100644
--- a/pandas/tests/arrays/test_integer.py
+++ b/pandas/tests/arrays/test_integer.py
@@ -1,5 +1,6 @@
# -*- coding: utf-8 -*-
import numpy as np
+import pytest
from pandas.core.dtypes.generic import ABCIndexClass
@@ -11,7 +12,6 @@
UInt32Dtype, UInt64Dtype)
from pandas.tests.extension.base import BaseOpsUtil
import pandas.util.testing as tm
-import pytest
def make_data():
diff --git a/pandas/tests/arrays/test_period.py b/pandas/tests/arrays/test_period.py
index 95c80ba61cc9a..f4fb8f2814a70 100644
--- a/pandas/tests/arrays/test_period.py
+++ b/pandas/tests/arrays/test_period.py
@@ -1,4 +1,5 @@
import numpy as np
+import pytest
from pandas._libs.tslibs import iNaT
from pandas._libs.tslibs.period import IncompatibleFrequency
@@ -9,7 +10,6 @@
import pandas as pd
from pandas.core.arrays import PeriodArray, period_array
import pandas.util.testing as tm
-import pytest
# ----------------------------------------------------------------------------
# Constructors
diff --git a/pandas/tests/extension/decimal/test_decimal.py b/pandas/tests/extension/decimal/test_decimal.py
index ed9f1acc10a6a..0ddb09c9b5617 100644
--- a/pandas/tests/extension/decimal/test_decimal.py
+++ b/pandas/tests/extension/decimal/test_decimal.py
@@ -3,12 +3,12 @@
import operator
import numpy as np
+import pytest
import pandas as pd
from pandas import compat
from pandas.tests.extension import base
import pandas.util.testing as tm
-import pytest
from .array import DecimalArray, DecimalDtype, make_data, to_decimal
diff --git a/pandas/tests/extension/json/test_json.py b/pandas/tests/extension/json/test_json.py
index a9244a1ab8436..29e4289226c68 100644
--- a/pandas/tests/extension/json/test_json.py
+++ b/pandas/tests/extension/json/test_json.py
@@ -1,12 +1,13 @@
import collections
import operator
+import pytest
+
from pandas.compat import PY2, PY36
import pandas as pd
from pandas.tests.extension import base
import pandas.util.testing as tm
-import pytest
from .array import JSONArray, JSONDtype, make_data
diff --git a/pandas/tests/extension/test_integer.py b/pandas/tests/extension/test_integer.py
index 4d5aa54b65320..0abae56ef8723 100644
--- a/pandas/tests/extension/test_integer.py
+++ b/pandas/tests/extension/test_integer.py
@@ -14,6 +14,7 @@
"""
import numpy as np
+import pytest
from pandas.core.dtypes.common import is_extension_array_dtype
@@ -23,7 +24,6 @@
Int8Dtype, Int16Dtype, Int32Dtype, Int64Dtype, UInt8Dtype, UInt16Dtype,
UInt32Dtype, UInt64Dtype)
from pandas.tests.extension import base
-import pytest
def make_data():
diff --git a/pandas/tests/extension/test_interval.py b/pandas/tests/extension/test_interval.py
index 9f49c2c920a95..644f3ef94f40b 100644
--- a/pandas/tests/extension/test_interval.py
+++ b/pandas/tests/extension/test_interval.py
@@ -14,13 +14,13 @@
"""
import numpy as np
+import pytest
from pandas.core.dtypes.dtypes import IntervalDtype
from pandas import Interval
from pandas.core.arrays import IntervalArray
from pandas.tests.extension import base
-import pytest
def make_data():
diff --git a/pandas/tests/extension/test_period.py b/pandas/tests/extension/test_period.py
index be74e7b330934..08e21fc30ad10 100644
--- a/pandas/tests/extension/test_period.py
+++ b/pandas/tests/extension/test_period.py
@@ -1,4 +1,5 @@
import numpy as np
+import pytest
from pandas._libs.tslib import iNaT
@@ -7,7 +8,6 @@
import pandas as pd
from pandas.core.arrays import PeriodArray
from pandas.tests.extension import base
-import pytest
@pytest.fixture
From 1b93bf0cc06a5d5fc5b54e9f8f295391997f7bbf Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sun, 11 Nov 2018 14:52:27 -0600
Subject: [PATCH 24/39] update repr tests
---
doc/source/whatsnew/v0.24.0.txt | 2 +-
pandas/core/arrays/base.py | 2 +-
pandas/tests/arrays/test_integer.py | 39 ++++++++++++++---------------
3 files changed, 21 insertions(+), 22 deletions(-)
diff --git a/doc/source/whatsnew/v0.24.0.txt b/doc/source/whatsnew/v0.24.0.txt
index 4d19a2eeb09f6..c6e23ff6e15a4 100644
--- a/doc/source/whatsnew/v0.24.0.txt
+++ b/doc/source/whatsnew/v0.24.0.txt
@@ -858,7 +858,7 @@ update the ``ExtensionDtype._metadata`` tuple to match the signature of your
- :meth:`DataFrame.stack` no longer converts to object dtype for DataFrames where each column has the same extension dtype. The output Series will have the same dtype as the columns (:issue:`23077`).
- :meth:`Series.unstack` and :meth:`DataFrame.unstack` no longer convert extension arrays to object-dtype ndarrays. Each column in the output ``DataFrame`` will now have the same dtype as the input (:issue:`23077`).
- Bug when grouping :meth:`Dataframe.groupby()` and aggregating on ``ExtensionArray`` it was not returning the actual ``ExtensionArray`` dtype (:issue:`23227`).
-- A default repr for ExtensionArrays is now provided (:issue:`23601`).
+- A default repr for :class:`ExtensionArray` is now provided (:issue:`23601`).
.. _whatsnew_0240.api.incompatibilities:
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index b6eba6a17359a..bcfd7f2d1c281 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -675,7 +675,7 @@ def __unicode__(self):
# repr does. So we include a newline in our template, and strip
# any trailing newlines from format_object_summary
data = format_object_summary(self, self._formatter(), name=False,
- trailing_comma=False).rstrip('\n')
+ trailing_comma=False).rstrip()
name = self.__class__.__name__
return template.format(class_name=name, data=data,
length=len(self),
diff --git a/pandas/tests/arrays/test_integer.py b/pandas/tests/arrays/test_integer.py
index ebae1a2797bdc..c0b3ab3592188 100644
--- a/pandas/tests/arrays/test_integer.py
+++ b/pandas/tests/arrays/test_integer.py
@@ -57,28 +57,27 @@ def test_dtypes(dtype):
assert dtype.name is not None
-def test_repr_array(data):
- result = repr(data)
- assert '<IntegerArray>' in result
-
- # not long
- assert '...' not in result
- assert 'Length: ' in result
- assert 'dtype: ' in result
-
-
-def test_na_repr(data):
- result = repr(integer_array([1, None]))
- assert 'NaN' in result
-
+def test_repr_array():
+ result = repr(integer_array([1, None, 3]))
+ expected = (
+ '<IntegerArray>\n'
+ '[1, NaN, 3]\n'
+ 'Length: 3, dtype: Int64'
+ )
+ assert result == expected
-def test_repr_array_long(data):
- # some arrays may be able to assert a ... in the repr
- with pd.option_context('display.max_seq_items', 1):
- result = repr(data)
- assert '...' in result
- assert 'Length' in result
+def test_repr_array_long():
+ data = integer_array([1, 2, None] * 1000)
+ expected = (
+ "<IntegerArray>\n"
+ "[ 1, 2, NaN, 1, 2, NaN, 1, 2, NaN, 1,\n"
+ " ...\n"
+ " NaN, 1, 2, NaN, 1, 2, NaN, 1, 2, NaN]\n"
+ "Length: 3000, dtype: Int64"
+ )
+ result = repr(data)
+ assert result == expected
class TestConstructors(object):
From 0f4083e2329ce89e34ec94b16bc11c0afd7f5811 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 12 Nov 2018 06:59:20 -0600
Subject: [PATCH 25/39] remove periodarray
---
pandas/core/arrays/period.py | 8 --------
1 file changed, 8 deletions(-)
diff --git a/pandas/core/arrays/period.py b/pandas/core/arrays/period.py
index 1a6c10ee05261..09bfed93612aa 100644
--- a/pandas/core/arrays/period.py
+++ b/pandas/core/arrays/period.py
@@ -366,14 +366,6 @@ def to_timestamp(self, freq=None, how='start'):
# --------------------------------------------------------------------
# Array-like / EA-Interface Methods
- def __repr__(self):
- return '<{}>\n{}\nLength: {}, dtype: {}'.format(
- self.__class__.__name__,
- [str(s) for s in self],
- len(self),
- self.dtype
- )
-
def _formatter(self, formatter=None):
if formatter:
return formatter.formatter or str
From ebadf6f211aa0cfcfbe083cb3505a14c63718a1b Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 12 Nov 2018 08:11:12 -0600
Subject: [PATCH 26/39] FutureWarning -> DeprecationWarning
---
pandas/core/arrays/base.py | 5 -----
pandas/core/internals/blocks.py | 2 +-
pandas/tests/extension/decimal/test_decimal.py | 2 +-
3 files changed, 2 insertions(+), 7 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index bcfd7f2d1c281..a5b351ee4fa3e 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -45,11 +45,6 @@ class ExtensionArray(StringMixin):
* copy
* _concat_same_type
- An additional method is available to satisfy pandas' internal,
- private block API.
-
- * _formatting_values
-
A default repr displaying the type, (truncated) data, length,
and dtype is provided. It can be customized or replaced by
by overriding:
diff --git a/pandas/core/internals/blocks.py b/pandas/core/internals/blocks.py
index 2a8d49feb0bad..835039e86d60f 100644
--- a/pandas/core/internals/blocks.py
+++ b/pandas/core/internals/blocks.py
@@ -1960,7 +1960,7 @@ def formatting_values(self):
"'ExtensionArray._formatting_values' is deprecated. "
"Specify 'ExtensionArray._formatter' instead."
)
- warnings.warn(msg, FutureWarning, stacklevel=10)
+ warnings.warn(msg, DeprecationWarning, stacklevel=10)
return self.values._formatting_values()
return self.values
diff --git a/pandas/tests/extension/decimal/test_decimal.py b/pandas/tests/extension/decimal/test_decimal.py
index 0ddb09c9b5617..6281c5360cd03 100644
--- a/pandas/tests/extension/decimal/test_decimal.py
+++ b/pandas/tests/extension/decimal/test_decimal.py
@@ -396,6 +396,6 @@ def _formatting_values(self):
# different levels for 2 vs. 3
check_stacklevel = compat.PY3
- with tm.assert_produces_warning(FutureWarning,
+ with tm.assert_produces_warning(DeprecationWarning,
check_stacklevel=check_stacklevel):
repr(ser)
From e5f6976bfd3077299149febe80a3a76e4fee1090 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 12 Nov 2018 08:19:01 -0600
Subject: [PATCH 27/39] wip
---
pandas/core/arrays/base.py | 21 ++++++++++++---------
1 file changed, 12 insertions(+), 9 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index a5b351ee4fa3e..26f676186b919 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -9,7 +9,6 @@
import operator
-from pandas.core.base import StringMixin
from pandas.core.dtypes.generic import ABCSeries, ABCIndexClass
from pandas.errors import AbstractMethodError
from pandas.compat.numpy import function as nv
@@ -20,7 +19,7 @@
_not_implemented_message = "{} does not implement {}."
-class ExtensionArray(StringMixin):
+class ExtensionArray(object):
"""Abstract base class for custom 1-D array types.
pandas will recognize instances of this class as proper arrays
@@ -50,7 +49,7 @@ class ExtensionArray(StringMixin):
by overriding:
* _formatter
- * __unicode__
+ * __repr__
Some methods require casting the ExtensionArray to an ndarray of Python
objects with ``self.astype(object)``, which may be expensive. When
@@ -658,7 +657,7 @@ def copy(self, deep=False):
# ------------------------------------------------------------------------
# Printing
# ------------------------------------------------------------------------
- def __unicode__(self):
+ def __repr__(self):
from pandas.io.formats.printing import format_object_summary
template = (
@@ -686,10 +685,14 @@ def _formatter(self, formatter=None):
Parameters
----------
formatter: GenericArrayFormatter, optional
- The formatter this array is being rendered with. The formatter
- may have a `.formatter` method already defined. By default, this
- will be used if a `formatter` is passed, otherwise the formatter
- is ``str``.
+ The formatter this array is being rendered with. This will be
+ passed when the ExtensionArray is being rendered inside of a
+ Series, Index, or DataFrame. This will be ``None`` when called
+ from a top-level ``repr(array)``.
+
+ By default, when ``formatter`` is passed, the return value
+ is ``formatter.formatter``. Otherwise, the default formatter
+ is ``repr``.
Returns
-------
@@ -697,7 +700,7 @@ def _formatter(self, formatter=None):
A callable that gets instances of the scalar type and
returns a string.
"""
- return getattr(formatter, 'formatter', None) or str
+ return getattr(formatter, 'formatter', None) or repr
def _formatting_values(self):
# type: () -> np.ndarray
From 221cee9799be6f2608fcbf962b58d6b56aa0ff4f Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 12 Nov 2018 09:04:31 -0600
Subject: [PATCH 28/39] use repr
---
pandas/core/arrays/base.py | 19 +++++++++++++++----
pandas/core/arrays/categorical.py | 5 ++++-
2 files changed, 19 insertions(+), 5 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 26f676186b919..f2af678d74d0d 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -661,7 +661,7 @@ def __repr__(self):
from pandas.io.formats.printing import format_object_summary
template = (
- u'<{class_name}>\n'
+ u'{class_name}'
u'{data}\n'
u'Length: {length}, dtype: {dtype}'
)
@@ -670,11 +670,17 @@ def __repr__(self):
# any trailing newlines from format_object_summary
data = format_object_summary(self, self._formatter(), name=False,
trailing_comma=False).rstrip()
- name = self.__class__.__name__
- return template.format(class_name=name, data=data,
+ class_name = u'<{}>\n'.format(self.__class__.__name__)
+ return template.format(class_name=class_name, data=data,
length=len(self),
dtype=self.dtype)
+ def __bytes__(self):
+ from pandas.core.config import get_option
+
+ encoding = get_option("display.encoding")
+ return str(self).encode(encoding, 'replace')
+
def _formatter(self, formatter=None):
# type: (Optional[ExtensionArrayFormatter]) -> Callable[[Any], str]
"""Formatting function for scalar values.
@@ -705,7 +711,12 @@ def _formatter(self, formatter=None):
def _formatting_values(self):
# type: () -> np.ndarray
# At the moment, this has to be an array since we use result.dtype
- """An array of values to be printed in, e.g. the Series repr"""
+ """An array of values to be printed in, e.g. the Series repr
+
+ .. deprecated:: 0.24.0
+
+ Use :meth:`ExtensionArray._formatter` instead.
+ """
return np.array(self)
# ------------------------------------------------------------------------
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 3593f83770114..02ab3471c6be1 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -499,7 +499,7 @@ def _constructor(self):
def _from_sequence(cls, scalars, dtype=None, copy=False):
return Categorical(scalars, dtype=dtype)
- def _formatter(self, formatter):
+ def _formatter(self, formatter=None):
# backwards compat with old printing.
return None
@@ -1990,6 +1990,9 @@ def __unicode__(self):
return result
+ def __repr__(self):
+ return super(ExtensionArray, self).__repr__()
+
def _maybe_coerce_indexer(self, indexer):
""" return an indexer coerced to the codes dtype """
if isinstance(indexer, np.ndarray) and indexer.dtype.kind == 'i':
From 439f2f896cd16ff7d80d708b51ffe265a4d3bbfc Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 12 Nov 2018 09:05:50 -0600
Subject: [PATCH 29/39] fixup! use repr
---
pandas/core/arrays/categorical.py | 3 +++
1 file changed, 3 insertions(+)
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 02ab3471c6be1..70c1a8b878f03 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -1993,6 +1993,9 @@ def __unicode__(self):
def __repr__(self):
return super(ExtensionArray, self).__repr__()
+ def __bytes__(self):
+ return super(ExtensionArray, self).__bytes__()
+
def _maybe_coerce_indexer(self, indexer):
""" return an indexer coerced to the codes dtype """
if isinstance(indexer, np.ndarray) and indexer.dtype.kind == 'i':
From 23645464899846d730ab5f13ecb5dbee5488b37c Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 12 Nov 2018 09:11:27 -0600
Subject: [PATCH 30/39] fixup! fixup! use repr
---
pandas/core/arrays/base.py | 21 +++++++++++++--------
1 file changed, 13 insertions(+), 8 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index f2af678d74d0d..8481502af4e2f 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -691,14 +691,19 @@ def _formatter(self, formatter=None):
Parameters
----------
formatter: GenericArrayFormatter, optional
- The formatter this array is being rendered with. This will be
- passed when the ExtensionArray is being rendered inside of a
- Series, Index, or DataFrame. This will be ``None`` when called
- from a top-level ``repr(array)``.
-
- By default, when ``formatter`` is passed, the return value
- is ``formatter.formatter``. Otherwise, the default formatter
- is ``repr``.
+ The formatter this array is being rendered with. When the array
+ is being rendered inside an Index, Series, or DataFrame, a
+ formatter will be provided. So if you want your objects to
+ render differently inside a Series from on its own, checking
+ with ``formatter is None`` is an option.
+
+ The default behavior depends on whether `formatter` is passed.
+
+ * When `formatter` is None, :func:`repr` is returned.
+ * When `formatter` is passed, ``formatter.formatter`` is used,
+ which falls back to :func:`repr` if that isn't specified.
+
+ In general, just returning :func:`repr` should be fine.
Returns
-------
From 62b1e2f0325ee6d83a96f4d4ca853ac7923da01f Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 12 Nov 2018 11:19:07 -0600
Subject: [PATCH 31/39] remove bytes
---
pandas/core/arrays/base.py | 6 ------
pandas/core/arrays/categorical.py | 4 +---
pandas/tests/extension/base/printing.py | 7 -------
3 files changed, 1 insertion(+), 16 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 8481502af4e2f..ee9e47baddcf3 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -675,12 +675,6 @@ def __repr__(self):
length=len(self),
dtype=self.dtype)
- def __bytes__(self):
- from pandas.core.config import get_option
-
- encoding = get_option("display.encoding")
- return str(self).encode(encoding, 'replace')
-
def _formatter(self, formatter=None):
# type: (Optional[ExtensionArrayFormatter]) -> Callable[[Any], str]
"""Formatting function for scalar values.
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 70c1a8b878f03..e3f46e8a9226b 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -1991,11 +1991,9 @@ def __unicode__(self):
return result
def __repr__(self):
+ # We want PandasObject.__repr__, which dispatches to __unicode__
return super(ExtensionArray, self).__repr__()
- def __bytes__(self):
- return super(ExtensionArray, self).__bytes__()
-
def _maybe_coerce_indexer(self, indexer):
""" return an indexer coerced to the codes dtype """
if isinstance(indexer, np.ndarray) and indexer.dtype.kind == 'i':
diff --git a/pandas/tests/extension/base/printing.py b/pandas/tests/extension/base/printing.py
index 7a41e7689f4a6..b2ba1d95cf33e 100644
--- a/pandas/tests/extension/base/printing.py
+++ b/pandas/tests/extension/base/printing.py
@@ -25,13 +25,6 @@ def test_array_repr(self, data, size):
if size == 'big':
assert '...' in result
- def test_array_repr_bytes(self, data):
- result = bytes(data)
- if compat.PY2:
- assert isinstance(result, str)
- else:
- assert isinstance(result, bytes)
-
def test_array_repr_unicode(self, data):
result = compat.text_type(data)
assert isinstance(result, compat.text_type)
From 27db397efca6840f285778b92aeefe5327a0b231 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Thu, 15 Nov 2018 09:05:47 -0600
Subject: [PATCH 32/39] simplify formatter
---
pandas/core/arrays/base.py | 32 ++++++++++++-------------------
pandas/core/arrays/categorical.py | 2 +-
pandas/core/arrays/integer.py | 14 ++++++--------
pandas/core/arrays/period.py | 6 +++---
pandas/core/arrays/sparse.py | 16 +++++++---------
5 files changed, 29 insertions(+), 41 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index ee9e47baddcf3..edc3ccc934690 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -675,37 +675,29 @@ def __repr__(self):
length=len(self),
dtype=self.dtype)
- def _formatter(self, formatter=None):
- # type: (Optional[ExtensionArrayFormatter]) -> Callable[[Any], str]
+ def _formatter(self, boxed=False):
+ # type: (bool) -> Callable[[Any], str]
"""Formatting function for scalar values.
- This is used in the default '__repr__'. The formatting function
- receives instances of your scalar type.
+ This is used in the default '__repr__'. The returned formatting
+ function receives instances of your scalar type.
Parameters
----------
- formatter: GenericArrayFormatter, optional
- The formatter this array is being rendered with. When the array
- is being rendered inside an Index, Series, or DataFrame, a
- formatter will be provided. So if you want your objects to
- render differently inside a Series from on its own, checking
- with ``formatter is None`` is an option.
-
- The default behavior depends on whether `formatter` is passed.
-
- * When `formatter` is None, :func:`repr` is returned.
- * When `formatter` is passed, ``formatter.formatter`` is used,
- which falls back to :func:`repr` if that isn't specified.
-
- In general, just returning :func:`repr` should be fine.
+ boxed: bool, default False
+ An indicated for whether or not your array is being printed
+ within a Series, DataFrame, or Index (True), or just by
+ itself (False). This may be useful if you want scalar values
+ to appear differently within a Series versus on its own (e.g.
+ quoted or not).
Returns
-------
Callable[[Any], str]
A callable that gets instances of the scalar type and
- returns a string.
+ returns a string. By defult, :func:`repr` is used.
"""
- return getattr(formatter, 'formatter', None) or repr
+ return repr
def _formatting_values(self):
# type: () -> np.ndarray
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index cdd925a8568c7..09eb1b0231c31 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -499,7 +499,7 @@ def _constructor(self):
def _from_sequence(cls, scalars, dtype=None, copy=False):
return Categorical(scalars, dtype=dtype)
- def _formatter(self, formatter=None):
+ def _formatter(self, boxed=False):
# backwards compat with old printing.
return None
diff --git a/pandas/core/arrays/integer.py b/pandas/core/arrays/integer.py
index 98eb9df98328f..984ef4c8c673c 100644
--- a/pandas/core/arrays/integer.py
+++ b/pandas/core/arrays/integer.py
@@ -264,14 +264,12 @@ def _from_sequence(cls, scalars, dtype=None, copy=False):
def _from_factorized(cls, values, original):
return integer_array(values, dtype=original.dtype)
- def _formatter(self, formatter=None):
- if formatter is None:
- def fmt(x):
- if isna(x):
- return 'NaN'
- return str(x)
- return fmt
- return formatter.formatter
+ def _formatter(self, boxed=False):
+ def fmt(x):
+ if isna(x):
+ return 'NaN'
+ return str(x)
+ return fmt
def __getitem__(self, item):
if is_integer(item):
diff --git a/pandas/core/arrays/period.py b/pandas/core/arrays/period.py
index 64779a9644515..d1728e7b43cb7 100644
--- a/pandas/core/arrays/period.py
+++ b/pandas/core/arrays/period.py
@@ -350,9 +350,9 @@ def to_timestamp(self, freq=None, how='start'):
# --------------------------------------------------------------------
# Array-like / EA-Interface Methods
- def _formatter(self, formatter=None):
- if formatter:
- return formatter.formatter or str
+ def _formatter(self, boxed=False):
+ if boxed:
+ return str
return "'{}'".format
def __setitem__(
diff --git a/pandas/core/arrays/sparse.py b/pandas/core/arrays/sparse.py
index b4b33b71b2383..68ae39f54ea74 100644
--- a/pandas/core/arrays/sparse.py
+++ b/pandas/core/arrays/sparse.py
@@ -1738,15 +1738,13 @@ def __unicode__(self):
fill=printing.pprint_thing(self.fill_value),
index=printing.pprint_thing(self.sp_index))
- def _formatter(self, formatter=None):
- if formatter is None:
- def fmt(x):
- if isna(x) and isinstance(x, float):
- return 'NaN'
- return str(x)
-
- return fmt
- return formatter.formatter
+ def _formatter(self, boxed=False):
+ def fmt(x):
+ if isna(x) and isinstance(x, float):
+ return 'NaN'
+ return str(x)
+
+ return fmt
SparseArray._add_arithmetic_ops()
From ef390fcc03e8b924d6da0b6e540bc124638047cf Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 19 Nov 2018 06:29:42 -0600
Subject: [PATCH 33/39] Updates: misc
* whatsnew
* docstrings
---
doc/source/whatsnew/v0.24.0.rst | 2 +-
pandas/core/arrays/base.py | 6 +++---
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/doc/source/whatsnew/v0.24.0.rst b/doc/source/whatsnew/v0.24.0.rst
index b21c492cd554b..57754e9ebaf1b 100644
--- a/doc/source/whatsnew/v0.24.0.rst
+++ b/doc/source/whatsnew/v0.24.0.rst
@@ -1197,7 +1197,7 @@ Datetimelike
- Bug in rounding methods of :class:`DatetimeIndex` (:meth:`~DatetimeIndex.round`, :meth:`~DatetimeIndex.ceil`, :meth:`~DatetimeIndex.floor`) and :class:`Timestamp` (:meth:`~Timestamp.round`, :meth:`~Timestamp.ceil`, :meth:`~Timestamp.floor`) could give rise to loss of precision (:issue:`22591`)
- Bug in :func:`to_datetime` with an :class:`Index` argument that would drop the ``name`` from the result (:issue:`21697`)
- Bug in :class:`PeriodIndex` where adding or subtracting a :class:`timedelta` or :class:`Tick` object produced incorrect results (:issue:`22988`)
-- Bug in the :class:`Series` repr with Period data missing a space before the data (:issue:`23601`)
+- Bug in the :class:`Series` repr with period-dtype data missing a space before the data (:issue:`23601`)
- Bug in :func:`date_range` when decrementing a start date to a past end date by a negative frequency (:issue:`23270`)
- Bug in :meth:`Series.min` which would return ``NaN`` instead of ``NaT`` when called on a series of ``NaT`` (:issue:`23282`)
- Bug in :func:`DataFrame.combine` with datetimelike values raising a TypeError (:issue:`23079`)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index edc3ccc934690..c22660a784dee 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -48,8 +48,8 @@ class ExtensionArray(object):
and dtype is provided. It can be customized or replaced by
by overriding:
- * _formatter
- * __repr__
+ * __repr__ : A default repr for the ExtensionArray.
+ * _formatter : Print scalars inside a Series or DataFrame.
Some methods require casting the ExtensionArray to an ndarray of Python
objects with ``self.astype(object)``, which may be expensive. When
@@ -695,7 +695,7 @@ def _formatter(self, boxed=False):
-------
Callable[[Any], str]
A callable that gets instances of the scalar type and
- returns a string. By defult, :func:`repr` is used.
+ returns a string. By default, :func:`repr` is used.
"""
return repr
From 2b5fe251b67deb45eaf5fd28d7b13beda11c267c Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 19 Nov 2018 06:38:03 -0600
Subject: [PATCH 34/39] BUG: Fixed SparseArray formatter
We want to fall back to the implementation in formats.
---
pandas/core/arrays/base.py | 2 +-
pandas/core/arrays/sparse.py | 7 +------
pandas/io/formats/format.py | 5 ++++-
3 files changed, 6 insertions(+), 8 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index c22660a784dee..0303dea84248b 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -676,7 +676,7 @@ def __repr__(self):
dtype=self.dtype)
def _formatter(self, boxed=False):
- # type: (bool) -> Callable[[Any], str]
+ # type: (bool) -> Callable[[Any], Optional[str]]
"""Formatting function for scalar values.
This is used in the default '__repr__'. The returned formatting
diff --git a/pandas/core/arrays/sparse.py b/pandas/core/arrays/sparse.py
index 68ae39f54ea74..a951c8a5469f6 100644
--- a/pandas/core/arrays/sparse.py
+++ b/pandas/core/arrays/sparse.py
@@ -1739,12 +1739,7 @@ def __unicode__(self):
index=printing.pprint_thing(self.sp_index))
def _formatter(self, boxed=False):
- def fmt(x):
- if isna(x) and isinstance(x, float):
- return 'NaN'
- return str(x)
-
- return fmt
+ return None
SparseArray._add_arithmetic_ops()
diff --git a/pandas/io/formats/format.py b/pandas/io/formats/format.py
index 406be5cc1b8bc..8ada7b92b2268 100644
--- a/pandas/io/formats/format.py
+++ b/pandas/io/formats/format.py
@@ -1123,9 +1123,12 @@ class ExtensionArrayFormatter(GenericArrayFormatter):
def _format_strings(self):
values = self.values
if isinstance(values, (ABCIndexClass, ABCSeries)):
+ boxed = True
values = values._values
+ else:
+ boxed = False
- formatter = values._formatter(self)
+ formatter = values._formatter(boxed=boxed)
if is_categorical_dtype(values.dtype):
# Categorical is special for now, so that we can preserve tzinfo
From d9df6bf6c743f50ec68e4895893f82f34c4c7ed0 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 19 Nov 2018 19:04:56 -0600
Subject: [PATCH 35/39] correct boxing
---
pandas/io/formats/format.py | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/pandas/io/formats/format.py b/pandas/io/formats/format.py
index 8ada7b92b2268..eda7167dfcbe1 100644
--- a/pandas/io/formats/format.py
+++ b/pandas/io/formats/format.py
@@ -1123,12 +1123,9 @@ class ExtensionArrayFormatter(GenericArrayFormatter):
def _format_strings(self):
values = self.values
if isinstance(values, (ABCIndexClass, ABCSeries)):
- boxed = True
values = values._values
- else:
- boxed = False
- formatter = values._formatter(boxed=boxed)
+ formatter = values._formatter(boxed=True)
if is_categorical_dtype(values.dtype):
# Categorical is special for now, so that we can preserve tzinfo
From 3825aeb7f61b4d350888882689bce521b4a603ce Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sun, 2 Dec 2018 07:13:14 -0600
Subject: [PATCH 36/39] Use Array formatter in PeriodIndex
---
pandas/core/indexes/period.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/pandas/core/indexes/period.py b/pandas/core/indexes/period.py
index 56df454bddf1c..d5e5cba7a59dc 100644
--- a/pandas/core/indexes/period.py
+++ b/pandas/core/indexes/period.py
@@ -529,7 +529,7 @@ def __array_wrap__(self, result, context=None):
@property
def _formatter_func(self):
- return lambda x: "'%s'" % x
+ return self.array._formatter(boxed=False)
def asof_locs(self, where, mask):
"""
From 2a60c15e755bb64b582875c4548cd38b6a92cf70 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Sun, 2 Dec 2018 07:15:55 -0600
Subject: [PATCH 37/39] Use repr / str
---
pandas/core/arrays/base.py | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 3a8ed4eb97ccc..717178410c1ed 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -697,8 +697,12 @@ def _formatter(self, boxed=False):
-------
Callable[[Any], str]
A callable that gets instances of the scalar type and
- returns a string. By default, :func:`repr` is used.
+ returns a string. By default, :func:`repr` is used
+ when ``boxed=False`` and :func:`str` is used when
+ ``boxed=True``.
"""
+ if boxed:
+ return str
return repr
def _formatting_values(self):
From a7ef10406be10e21605e32b05fe196711f46bed4 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 3 Dec 2018 08:37:01 -0600
Subject: [PATCH 38/39] Update for review
* docs
* removed overloading of name=False
* added indent_for_name
---
doc/source/whatsnew/v0.24.0.rst | 2 +-
pandas/core/arrays/base.py | 3 ++-
pandas/core/arrays/categorical.py | 2 +-
pandas/core/arrays/sparse.py | 2 ++
pandas/io/formats/printing.py | 15 ++++++++-------
5 files changed, 14 insertions(+), 10 deletions(-)
diff --git a/doc/source/whatsnew/v0.24.0.rst b/doc/source/whatsnew/v0.24.0.rst
index e35a2c1118204..02ffd07e81ff9 100644
--- a/doc/source/whatsnew/v0.24.0.rst
+++ b/doc/source/whatsnew/v0.24.0.rst
@@ -1117,7 +1117,7 @@ Deprecations
- The methods :meth:`Series.str.partition` and :meth:`Series.str.rpartition` have deprecated the ``pat`` keyword in favor of ``sep`` (:issue:`22676`)
- Deprecated the `nthreads` keyword of :func:`pandas.read_feather` in favor of
`use_threads` to reflect the changes in pyarrow 0.11.0. (:issue:`23053`)
-- :meth:`ExtensionArray._formatting_values` is deprecated. Use `ExtensionArray._formatter` instead. (:issue:`23601`)
+- :meth:`ExtensionArray._formatting_values` is deprecated. Use :attr:`ExtensionArray._formatter` instead. (:issue:`23601`)
- :func:`pandas.read_excel` has deprecated accepting ``usecols`` as an integer. Please pass in a list of ints from 0 to ``usecols`` inclusive instead (:issue:`23527`)
- Constructing a :class:`TimedeltaIndex` from data with ``datetime64``-dtyped data is deprecated, will raise ``TypeError`` in a future version (:issue:`23539`)
- Constructing a :class:`DatetimeIndex` from data with ``timedelta64``-dtyped data is deprecated, will raise ``TypeError`` in a future version (:issue:`23675`)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 717178410c1ed..9c5696499817c 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -670,7 +670,8 @@ def __repr__(self):
# the short repr has no trailing newline, while the truncated
# repr does. So we include a newline in our template, and strip
# any trailing newlines from format_object_summary
- data = format_object_summary(self, self._formatter(), name=False,
+ data = format_object_summary(self, self._formatter(),
+ indent_for_name=False,
trailing_comma=False).rstrip()
class_name = u'<{}>\n'.format(self.__class__.__name__)
return template.format(class_name=class_name, data=data,
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 94f18db642176..5ddfdd094e216 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -495,7 +495,7 @@ def _from_sequence(cls, scalars, dtype=None, copy=False):
return Categorical(scalars, dtype=dtype)
def _formatter(self, boxed=False):
- # backwards compat with old printing.
+ # Defer to CategoricalFormatter's formatter.
return None
def copy(self):
diff --git a/pandas/core/arrays/sparse.py b/pandas/core/arrays/sparse.py
index 7ef47b73b67de..6285fd47dd41f 100644
--- a/pandas/core/arrays/sparse.py
+++ b/pandas/core/arrays/sparse.py
@@ -1736,6 +1736,8 @@ def __unicode__(self):
index=printing.pprint_thing(self.sp_index))
def _formatter(self, boxed=False):
+ # Defer to the formatter from the GenericArrayFormatter calling us.
+ # This will infer the correct formatter from the dtype of the values.
return None
diff --git a/pandas/io/formats/printing.py b/pandas/io/formats/printing.py
index 842985aec6145..2e61c7a591280 100644
--- a/pandas/io/formats/printing.py
+++ b/pandas/io/formats/printing.py
@@ -272,6 +272,7 @@ class TableSchemaFormatter(BaseFormatter):
def format_object_summary(obj, formatter, is_justify=True, name=None,
+ indent_for_name=True,
trailing_comma=True):
"""
Return the formatted obj as a unicode string
@@ -286,9 +287,9 @@ def format_object_summary(obj, formatter, is_justify=True, name=None,
should justify the display
name : name, optional
defaults to the class name of the obj
-
- Pass ``False`` to indicate that subsequent lines should
- not be indented to align with the name.
+ indent_for_name : bool, default True
+ Whether subsequent lines should be be indented to
+ align with the name.
trailing_comma : bool, default True
Whether to include a comma after the closing ']'
@@ -306,13 +307,13 @@ def format_object_summary(obj, formatter, is_justify=True, name=None,
if name is None:
name = obj.__class__.__name__
- if name is False:
- space1 = "\n"
- space2 = "\n " # space for the opening '['
- else:
+ if indent_for_name:
name_len = len(name)
space1 = "\n%s" % (' ' * (name_len + 1))
space2 = "\n%s" % (' ' * (name_len + 2))
+ else:
+ space1 = "\n"
+ space2 = "\n " # space for the opening '['
n = len(obj)
sep = ','
From a3b1c92347b469740a7d7ad20f9cb97b357086dd Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 3 Dec 2018 08:38:49 -0600
Subject: [PATCH 39/39] REF: removed trailing_comma argument
---
pandas/core/arrays/base.py | 3 +--
pandas/io/formats/printing.py | 10 ++--------
2 files changed, 3 insertions(+), 10 deletions(-)
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 9c5696499817c..9a8b1de95218a 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -671,8 +671,7 @@ def __repr__(self):
# repr does. So we include a newline in our template, and strip
# any trailing newlines from format_object_summary
data = format_object_summary(self, self._formatter(),
- indent_for_name=False,
- trailing_comma=False).rstrip()
+ indent_for_name=False).rstrip(', \n')
class_name = u'<{}>\n'.format(self.__class__.__name__)
return template.format(class_name=class_name, data=data,
length=len(self),
diff --git a/pandas/io/formats/printing.py b/pandas/io/formats/printing.py
index 2e61c7a591280..6d45d1e5dfcee 100644
--- a/pandas/io/formats/printing.py
+++ b/pandas/io/formats/printing.py
@@ -272,8 +272,7 @@ class TableSchemaFormatter(BaseFormatter):
def format_object_summary(obj, formatter, is_justify=True, name=None,
- indent_for_name=True,
- trailing_comma=True):
+ indent_for_name=True):
"""
Return the formatted obj as a unicode string
@@ -290,8 +289,6 @@ def format_object_summary(obj, formatter, is_justify=True, name=None,
indent_for_name : bool, default True
Whether subsequent lines should be be indented to
align with the name.
- trailing_comma : bool, default True
- Whether to include a comma after the closing ']'
Returns
-------
@@ -340,10 +337,7 @@ def best_len(values):
else:
return 0
- if trailing_comma:
- close = u', '
- else:
- close = u''
+ close = u', '
if n == 0:
summary = u'[]{}'.format(close)