From 3e016816c77429f2582624254a54485b13af3912 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Mon, 28 Jan 2019 18:50:31 +0100
Subject: [PATCH 01/21] API: re-enable custom label types in set_index
---
pandas/core/frame.py | 25 +++++++++++++++----------
pandas/tests/frame/test_alter_axes.py | 25 +++++++++++++++++++++++++
2 files changed, 40 insertions(+), 10 deletions(-)
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 28c6f3c23a3ce..2630a09500395 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4135,13 +4135,8 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
'array, or a list containing only valid column keys and '
'one-dimensional arrays.')
- if (is_scalar(keys) or isinstance(keys, tuple)
- or isinstance(keys, (ABCIndexClass, ABCSeries, np.ndarray))):
- # make sure we have a container of keys/arrays we can iterate over
- # tuples can appear as valid column keys!
+ if not isinstance(keys, list):
keys = [keys]
- elif not isinstance(keys, list):
- raise ValueError(err_msg)
missing = []
for col in keys:
@@ -4150,10 +4145,20 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
# tuples are always considered keys, never as list-likes
if col not in self:
missing.append(col)
- elif (not isinstance(col, (ABCIndexClass, ABCSeries,
- np.ndarray, list))
- or getattr(col, 'ndim', 1) > 1):
- raise ValueError(err_msg)
+ elif isinstance(col, (ABCIndexClass, ABCSeries,
+ np.ndarray, list)):
+ # arrays are fine as long as they are one-dimensional
+ if getattr(col, 'ndim', 1) > 1:
+ raise ValueError(err_msg)
+ else:
+ # everything else gets tried as a key; see GH 24969
+ try:
+ self[col]
+ str(col)
+ except KeyError:
+ tipo = type(col)
+ raise ValueError(err_msg,
+ 'Received column of type {}'.format(tipo))
if missing:
raise KeyError('{}'.format(missing))
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index c2355742199dc..524033da6f8c9 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -270,6 +270,31 @@ def test_set_index_raise_on_type(self, frame_of_index_cols, box,
df.set_index(['A', df['A'], box(df['A'])],
drop=drop, append=append)
+ def test_set_index_custom_label_type(self):
+ # GH 24969
+
+ class Thing:
+ def __init__(self, name, color):
+ self.name = name
+ self.color = color
+
+ def __str__(self):
+ return "<Thing %r>" % (self.name,)
+
+ thing1 = Thing('One', 'red')
+ thing2 = Thing('Two', 'blue')
+ df = DataFrame({thing1: [0, 1], thing2: [2, 3]})
+ expected = DataFrame({thing1: [0, 1]},
+ index=Index([2, 3], name=thing2))
+
+ # use custom label directly
+ result = df.set_index(thing2)
+ tm.assert_frame_equal(result, expected)
+
+ # custom label wrapped in list
+ result = df.set_index([thing2])
+ tm.assert_frame_equal(result, expected)
+
def test_construction_with_categorical_index(self):
ci = tm.makeCategoricalIndex(10)
ci.name = 'B'
From 1b71e68579c81ce9b8585d284adbf1c7273f9ec9 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Mon, 28 Jan 2019 19:26:29 +0100
Subject: [PATCH 02/21] Fix doc pattern?
---
pandas/core/frame.py | 1 -
pandas/tests/frame/test_alter_axes.py | 2 +-
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 2630a09500395..a36c850fc5633 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4154,7 +4154,6 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
# everything else gets tried as a key; see GH 24969
try:
self[col]
- str(col)
except KeyError:
tipo = type(col)
raise ValueError(err_msg,
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index 524033da6f8c9..c68910ef9d099 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -273,7 +273,7 @@ def test_set_index_raise_on_type(self, frame_of_index_cols, box,
def test_set_index_custom_label_type(self):
# GH 24969
- class Thing:
+ class Thing(object):
def __init__(self, name, color):
self.name = name
self.color = color
From caeb125ad5766ee5425095dadbfe2ce49e747968 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Wed, 30 Jan 2019 00:01:06 +0100
Subject: [PATCH 03/21] Review (TomAugspurger)
---
doc/source/whatsnew/v0.24.1.rst | 1 +
1 file changed, 1 insertion(+)
diff --git a/doc/source/whatsnew/v0.24.1.rst b/doc/source/whatsnew/v0.24.1.rst
index de33ce64c1597..f088969afde11 100644
--- a/doc/source/whatsnew/v0.24.1.rst
+++ b/doc/source/whatsnew/v0.24.1.rst
@@ -73,6 +73,7 @@ Bug Fixes
**Reshaping**
- Bug in :func:`merge` when merging by index name would sometimes result in an incorrectly numbered index (:issue:`24212`)
+- Fixed regression where custom hashable types could not be used as column keys in :meth:`DataFrame.set_index` anymore (:issue:`24969`)
**Other**
From 8bd53407a1ce0bd2ba27d809d5aaad5c014bca49 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Wed, 30 Jan 2019 00:16:59 +0100
Subject: [PATCH 04/21] Review (jreback)
---
pandas/core/frame.py | 9 ++++++-
pandas/tests/frame/test_alter_axes.py | 39 ++++++++++++++++++++++++---
2 files changed, 43 insertions(+), 5 deletions(-)
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index a36c850fc5633..22317154a791f 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4153,8 +4153,15 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
else:
# everything else gets tried as a key; see GH 24969
try:
- self[col]
+ hash(col) and self[col]
+ except TypeError:
+ # for unhashable types
+ tipo = type(col)
+ raise TypeError(err_msg,
+ 'Received column of type {}'.format(tipo))
except KeyError:
+ # for hashable types that are not keys;
+ # treat as ValueError, not missing key
tipo = type(col)
raise ValueError(err_msg,
'Received column of type {}'.format(tipo))
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index c68910ef9d099..3df35bc693319 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -261,12 +261,17 @@ def test_set_index_raise_on_type(self, frame_of_index_cols, box,
df = frame_of_index_cols
msg = 'The parameter "keys" may be a column key, .*'
- # forbidden type, e.g. set/tuple/iter
- with pytest.raises(ValueError, match=msg):
+ # forbidden type, e.g. set/iter
+
+ # iter is hashable, hence it is treated as a ValueError
+ # set is unhashable, hence it is a TypeError
+ exc = ValueError if box == iter else TypeError
+
+ with pytest.raises(exc, match=msg):
df.set_index(box(df['A']), drop=drop, append=append)
- # forbidden type in list, e.g. set/tuple/iter
- with pytest.raises(ValueError, match=msg):
+ # forbidden type in list, e.g. set/iter
+ with pytest.raises(exc, match=msg):
df.set_index(['A', df['A'], box(df['A'])],
drop=drop, append=append)
@@ -295,6 +300,32 @@ def __str__(self):
result = df.set_index([thing2])
tm.assert_frame_equal(result, expected)
+ def test_set_index_custom_label_type_raises(self):
+ # GH 24969
+
+ # purposefully inherit from something unhashable
+ class Thing(set):
+ def __init__(self, name, color):
+ self.name = name
+ self.color = color
+
+ def __str__(self):
+ return "<Thing %r>" % (self.name,)
+
+ thing1 = Thing('One', 'red')
+ thing2 = Thing('Two', 'blue')
+ df = DataFrame([[0, 2], [1, 3]], columns=[thing1, thing2])
+
+ msg = 'The parameter "keys" may be a column key, .*'
+
+ with pytest.raises(TypeError, match=msg):
+ # use custom label directly
+ df.set_index(thing2)
+
+ with pytest.raises(TypeError, match=msg):
+ # custom label wrapped in list
+ df.set_index([thing2])
+
def test_construction_with_categorical_index(self):
ci = tm.makeCategoricalIndex(10)
ci.name = 'B'
From d76ecfb9c304bdaede34ddd8e2c28f13db54f8ba Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Thu, 31 Jan 2019 00:40:28 +0100
Subject: [PATCH 05/21] Review (jreback & jorisvandenbossche)
---
doc/source/whatsnew/v0.24.1.rst | 3 ++-
pandas/core/frame.py | 19 ++++++++++++-------
2 files changed, 14 insertions(+), 8 deletions(-)
diff --git a/doc/source/whatsnew/v0.24.1.rst b/doc/source/whatsnew/v0.24.1.rst
index f35b4cdb936c5..4eb1a6ec83845 100644
--- a/doc/source/whatsnew/v0.24.1.rst
+++ b/doc/source/whatsnew/v0.24.1.rst
@@ -75,12 +75,13 @@ Bug Fixes
**Reshaping**
- Bug in :meth:`DataFrame.groupby` with :class:`Grouper` when there is a time change (DST) and grouping frequency is ``'1d'`` (:issue:`24972`)
-- Fixed regression where custom hashable types could not be used as column keys in :meth:`DataFrame.set_index` anymore (:issue:`24969`)
+- Fixed regression where custom hashable types could not be used as column keys in :meth:`DataFrame.set_index` (:issue:`24969`)
**Visualization**
- Fixed the warning for implicitly registered matplotlib converters not showing. See :ref:`whatsnew_0211.converters` for more (:issue:`24963`).
+
**Other**
-
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index b0a03932a9ac4..6f934f2c38a0b 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4154,18 +4154,23 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
else:
# everything else gets tried as a key; see GH 24969
try:
- hash(col) and self[col]
+ found = col in self.columns
+ if not found:
+ # check if this raises KeyError (e.g. iter is hashable
+ # but never a valid key); will be caught below
+ self[col]
+ # otherwise the key is missing
+ missing.append(col)
except TypeError:
# for unhashable types
tipo = type(col)
- raise TypeError(err_msg,
- 'Received column of type {}'.format(tipo))
+ raise TypeError(err_msg + 'Received column of '
+ 'type {}'.format(tipo))
except KeyError:
- # for hashable types that are not keys;
- # treat as ValueError, not missing key
+ # for hashable types that are not keys
tipo = type(col)
- raise ValueError(err_msg,
- 'Received column of type {}'.format(tipo))
+ raise ValueError(err_msg + ' Received column of '
+ 'type {}'.format(tipo))
if missing:
raise KeyError('{}'.format(missing))
From d2ffb8125610146398c846555caa0bef20cdbbc2 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Thu, 31 Jan 2019 00:42:08 +0100
Subject: [PATCH 06/21] Revert last two commits
---
pandas/core/frame.py | 18 +++----------
pandas/tests/frame/test_alter_axes.py | 39 +++------------------------
2 files changed, 7 insertions(+), 50 deletions(-)
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 6f934f2c38a0b..722d3339f578a 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4154,23 +4154,11 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
else:
# everything else gets tried as a key; see GH 24969
try:
- found = col in self.columns
- if not found:
- # check if this raises KeyError (e.g. iter is hashable
- # but never a valid key); will be caught below
- self[col]
- # otherwise the key is missing
- missing.append(col)
- except TypeError:
- # for unhashable types
- tipo = type(col)
- raise TypeError(err_msg + 'Received column of '
- 'type {}'.format(tipo))
+ self[col]
except KeyError:
- # for hashable types that are not keys
tipo = type(col)
- raise ValueError(err_msg + ' Received column of '
- 'type {}'.format(tipo))
+ raise ValueError(err_msg,
+ 'Received column of type {}'.format(tipo))
if missing:
raise KeyError('{}'.format(missing))
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index 3df35bc693319..c68910ef9d099 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -261,17 +261,12 @@ def test_set_index_raise_on_type(self, frame_of_index_cols, box,
df = frame_of_index_cols
msg = 'The parameter "keys" may be a column key, .*'
- # forbidden type, e.g. set/iter
-
- # iter is hashable, hence it is treated as a ValueError
- # set is unhashable, hence it is a TypeError
- exc = ValueError if box == iter else TypeError
-
- with pytest.raises(exc, match=msg):
+ # forbidden type, e.g. set/tuple/iter
+ with pytest.raises(ValueError, match=msg):
df.set_index(box(df['A']), drop=drop, append=append)
- # forbidden type in list, e.g. set/iter
- with pytest.raises(exc, match=msg):
+ # forbidden type in list, e.g. set/tuple/iter
+ with pytest.raises(ValueError, match=msg):
df.set_index(['A', df['A'], box(df['A'])],
drop=drop, append=append)
@@ -300,32 +295,6 @@ def __str__(self):
result = df.set_index([thing2])
tm.assert_frame_equal(result, expected)
- def test_set_index_custom_label_type_raises(self):
- # GH 24969
-
- # purposefully inherit from something unhashable
- class Thing(set):
- def __init__(self, name, color):
- self.name = name
- self.color = color
-
- def __str__(self):
- return "<Thing %r>" % (self.name,)
-
- thing1 = Thing('One', 'red')
- thing2 = Thing('Two', 'blue')
- df = DataFrame([[0, 2], [1, 3]], columns=[thing1, thing2])
-
- msg = 'The parameter "keys" may be a column key, .*'
-
- with pytest.raises(TypeError, match=msg):
- # use custom label directly
- df.set_index(thing2)
-
- with pytest.raises(TypeError, match=msg):
- # custom label wrapped in list
- df.set_index([thing2])
-
def test_construction_with_categorical_index(self):
ci = tm.makeCategoricalIndex(10)
ci.name = 'B'
From 5863678045ce09f8f3ddca9321420a2080bbd8bd Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Thu, 31 Jan 2019 09:07:07 +0100
Subject: [PATCH 07/21] Review (jorisvandenbossche)
---
pandas/core/frame.py | 17 +++++--
pandas/tests/frame/test_alter_axes.py | 70 +++++++++++++++++++++++++--
2 files changed, 78 insertions(+), 9 deletions(-)
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 722d3339f578a..3ba4f39367b56 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4151,14 +4151,23 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
# arrays are fine as long as they are one-dimensional
if getattr(col, 'ndim', 1) > 1:
raise ValueError(err_msg)
+ elif is_list_like(col, allow_sets=False):
+ # various iterators/generators are hashable, but should not
+ # raise a KeyError
+ tipo = type(col)
+ raise ValueError(err_msg + ' Received column of '
+ 'type {}'.format(tipo))
else:
# everything else gets tried as a key; see GH 24969
try:
- self[col]
- except KeyError:
+ found = col in self.columns
+ except TypeError:
tipo = type(col)
- raise ValueError(err_msg,
- 'Received column of type {}'.format(tipo))
+ raise TypeError(err_msg + ' Received column of '
+ 'type {}'.format(tipo))
+ else:
+ if not found:
+ missing.append(col)
if missing:
raise KeyError('{}'.format(missing))
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index c68910ef9d099..8ca038ab38b63 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -255,21 +255,40 @@ def test_set_index_raise_keys(self, frame_of_index_cols, drop, append):
@pytest.mark.parametrize('append', [True, False])
@pytest.mark.parametrize('drop', [True, False])
- @pytest.mark.parametrize('box', [set, iter])
- def test_set_index_raise_on_type(self, frame_of_index_cols, box,
- drop, append):
+ @pytest.mark.parametrize('box', [iter, lambda x: (y for y in x)],
+ ids=['iter', 'generator'])
+ def test_set_index_raise_on_type_iter(self, frame_of_index_cols, box,
+ drop, append):
df = frame_of_index_cols
msg = 'The parameter "keys" may be a column key, .*'
- # forbidden type, e.g. set/tuple/iter
+ # forbidden type, e.g. iter/generator
with pytest.raises(ValueError, match=msg):
df.set_index(box(df['A']), drop=drop, append=append)
- # forbidden type in list, e.g. set/tuple/iter
+ # forbidden type in list, e.g. iter/generator
with pytest.raises(ValueError, match=msg):
df.set_index(['A', df['A'], box(df['A'])],
drop=drop, append=append)
+ @pytest.mark.parametrize('append', [True, False])
+ @pytest.mark.parametrize('drop', [True, False])
+ @pytest.mark.parametrize('box', [set, lambda x: dict(zip(x, x)).keys()],
+ ids=['set', 'dict-view'])
+ def test_set_index_raise_on_type_unhashable(self, frame_of_index_cols, box,
+ drop, append):
+ df = frame_of_index_cols
+
+ msg = 'The parameter "keys" may be a column key, .*'
+ # forbidden type that is unhashable, e.g. set/dict-view
+ with pytest.raises(TypeError, match=msg):
+ df.set_index(box(df['A']), drop=drop, append=append)
+
+ # forbidden type in list that is unhashable, e.g. set/dict-view
+ with pytest.raises(TypeError, match=msg):
+ df.set_index(['A', df['A'], box(df['A'])],
+ drop=drop, append=append)
+
def test_set_index_custom_label_type(self):
# GH 24969
@@ -281,6 +300,10 @@ def __init__(self, name, color):
def __str__(self):
return "<Thing %r>" % (self.name,)
+ def __repr__(self):
+ # necessary for pretty KeyError
+ return self.__str__()
+
thing1 = Thing('One', 'red')
thing2 = Thing('Two', 'blue')
df = DataFrame({thing1: [0, 1], thing2: [2, 3]})
@@ -295,6 +318,43 @@ def __str__(self):
result = df.set_index([thing2])
tm.assert_frame_equal(result, expected)
+ # missing key
+ thing3 = Thing('Three', 'pink')
+ msg = "<Thing 'Three'>"
+ with pytest.raises(KeyError, match=msg):
+ # missing label directly
+ df.set_index(thing3)
+
+ with pytest.raises(KeyError, match=msg):
+ # missing label in list
+ df.set_index([thing3])
+
+ def test_set_index_custom_label_type_raises(self):
+ # GH 24969
+
+ # purposefully inherit from something unhashable
+ class Thing(set):
+ def __init__(self, name, color):
+ self.name = name
+ self.color = color
+
+ def __str__(self):
+ return "<Thing %r>" % (self.name,)
+
+ thing1 = Thing('One', 'red')
+ thing2 = Thing('Two', 'blue')
+ df = DataFrame([[0, 2], [1, 3]], columns=[thing1, thing2])
+
+ msg = 'The parameter "keys" may be a column key, .*'
+
+ with pytest.raises(TypeError, match=msg):
+ # use custom label directly
+ df.set_index(thing2)
+
+ with pytest.raises(TypeError, match=msg):
+ # custom label wrapped in list
+ df.set_index([thing2])
+
def test_construction_with_categorical_index(self):
ci = tm.makeCategoricalIndex(10)
ci.name = 'B'
From 0a7d783c116965e1dc1a3a638ade7f1fbfa33930 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Fri, 1 Feb 2019 00:23:29 +0100
Subject: [PATCH 08/21] Fix hashable listlikes (review jorisvandenbossche)
---
pandas/core/frame.py | 9 ++--
pandas/tests/frame/test_alter_axes.py | 62 ++++++++++++++++-----------
2 files changed, 43 insertions(+), 28 deletions(-)
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 3ba4f39367b56..0a9dd88ca1024 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4151,12 +4151,13 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
# arrays are fine as long as they are one-dimensional
if getattr(col, 'ndim', 1) > 1:
raise ValueError(err_msg)
- elif is_list_like(col, allow_sets=False):
+ elif is_list_like(col) and not hasattr(col, '__len__'):
# various iterators/generators are hashable, but should not
- # raise a KeyError
+ # raise a KeyError; we identify them by their lack of __len__.
+ # hashable listlikes with __len__ get tested as keys below.
tipo = type(col)
- raise ValueError(err_msg + ' Received column of '
- 'type {}'.format(tipo))
+ raise TypeError(err_msg + ' Received column of '
+ 'type {}'.format(tipo))
else:
# everything else gets tried as a key; see GH 24969
try:
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index 8ca038ab38b63..cbdb181554981 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -255,36 +255,18 @@ def test_set_index_raise_keys(self, frame_of_index_cols, drop, append):
@pytest.mark.parametrize('append', [True, False])
@pytest.mark.parametrize('drop', [True, False])
- @pytest.mark.parametrize('box', [iter, lambda x: (y for y in x)],
- ids=['iter', 'generator'])
- def test_set_index_raise_on_type_iter(self, frame_of_index_cols, box,
- drop, append):
+ @pytest.mark.parametrize('box', [set, iter, lambda x: (y for y in x)],
+ ids=['set', 'iter', 'generator'])
+ def test_set_index_raise_on_type(self, frame_of_index_cols, box,
+ drop, append):
df = frame_of_index_cols
msg = 'The parameter "keys" may be a column key, .*'
- # forbidden type, e.g. iter/generator
- with pytest.raises(ValueError, match=msg):
- df.set_index(box(df['A']), drop=drop, append=append)
-
- # forbidden type in list, e.g. iter/generator
- with pytest.raises(ValueError, match=msg):
- df.set_index(['A', df['A'], box(df['A'])],
- drop=drop, append=append)
-
- @pytest.mark.parametrize('append', [True, False])
- @pytest.mark.parametrize('drop', [True, False])
- @pytest.mark.parametrize('box', [set, lambda x: dict(zip(x, x)).keys()],
- ids=['set', 'dict-view'])
- def test_set_index_raise_on_type_unhashable(self, frame_of_index_cols, box,
- drop, append):
- df = frame_of_index_cols
-
- msg = 'The parameter "keys" may be a column key, .*'
- # forbidden type that is unhashable, e.g. set/dict-view
+ # forbidden type, e.g. set/iter/generator
with pytest.raises(TypeError, match=msg):
df.set_index(box(df['A']), drop=drop, append=append)
- # forbidden type in list that is unhashable, e.g. set/dict-view
+ # forbidden type in list, e.g. set/iter/generator
with pytest.raises(TypeError, match=msg):
df.set_index(['A', df['A'], box(df['A'])],
drop=drop, append=append)
@@ -329,6 +311,38 @@ def __repr__(self):
# missing label in list
df.set_index([thing3])
+ def test_set_index_custom_label_hashable_iterable(self):
+ # GH 24969
+
+ # actual example discussed in GH 24984 was e.g. for shapely.geometry
+ # objects (e.g. a collection of Points) that can be both hashable and
+ # iterable; using frozenset as a stand-in for testing here
+
+ thing1 = frozenset(['One', 'red'])
+ thing2 = frozenset(['Two', 'blue'])
+ df = DataFrame({thing1: [0, 1], thing2: [2, 3]})
+ expected = DataFrame({thing1: [0, 1]},
+ index=Index([2, 3], name=thing2))
+
+ # use custom label directly
+ result = df.set_index(thing2)
+ tm.assert_frame_equal(result, expected)
+
+ # custom label wrapped in list
+ result = df.set_index([thing2])
+ tm.assert_frame_equal(result, expected)
+
+ # missing key
+ thing3 = frozenset(['Three', 'pink'])
+ msg = r"frozenset\(\{'Three', 'pink'\}\)"
+ with pytest.raises(KeyError, match=msg):
+ # missing label directly
+ df.set_index(thing3)
+
+ with pytest.raises(KeyError, match=msg):
+ # missing label in list
+ df.set_index([thing3])
+
def test_set_index_custom_label_type_raises(self):
# GH 24969
From 794f61d4d77d075b37e40144ed3adbda669ace64 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Fri, 1 Feb 2019 01:20:16 +0100
Subject: [PATCH 09/21] Stabilize repr of frozenset
---
pandas/tests/frame/test_alter_axes.py | 18 ++++++++++++------
1 file changed, 12 insertions(+), 6 deletions(-)
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index cbdb181554981..de131e35a7447 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -282,9 +282,8 @@ def __init__(self, name, color):
def __str__(self):
return "<Thing %r>" % (self.name,)
- def __repr__(self):
- # necessary for pretty KeyError
- return self.__str__()
+ # necessary for pretty KeyError
+ __repr__ = __str__
thing1 = Thing('One', 'red')
thing2 = Thing('Two', 'blue')
@@ -318,8 +317,15 @@ def test_set_index_custom_label_hashable_iterable(self):
# objects (e.g. a collection of Points) that can be both hashable and
# iterable; using frozenset as a stand-in for testing here
- thing1 = frozenset(['One', 'red'])
- thing2 = frozenset(['Two', 'blue'])
+ class Thing(frozenset):
+ # need to stabilize repr for KeyError (due to random order in sets)
+ def __repr__(self):
+ tmp = sorted(list(self))
+ # double curly brace prints one brace in format string
+ return "frozenset({{{}}})".format(', '.join(map(repr, tmp)))
+
+ thing1 = Thing(['One', 'red'])
+ thing2 = Thing(['Two', 'blue'])
df = DataFrame({thing1: [0, 1], thing2: [2, 3]})
expected = DataFrame({thing1: [0, 1]},
index=Index([2, 3], name=thing2))
@@ -333,7 +339,7 @@ def test_set_index_custom_label_hashable_iterable(self):
tm.assert_frame_equal(result, expected)
# missing key
- thing3 = frozenset(['Three', 'pink'])
+ thing3 = Thing(['Three', 'pink'])
msg = r"frozenset\(\{'Three', 'pink'\}\)"
with pytest.raises(KeyError, match=msg):
# missing label directly
From c58e8b64a0d016a8551998b0fefa5d0bcea0b22c Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Fri, 1 Feb 2019 20:05:10 +0100
Subject: [PATCH 10/21] Review (WillAyd)
---
pandas/core/frame.py | 8 +-------
1 file changed, 1 insertion(+), 7 deletions(-)
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 988da53d7e811..9dd5b0750c200 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4148,13 +4148,7 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
missing = []
for col in keys:
- if (is_scalar(col) or isinstance(col, tuple)):
- # if col is a valid column key, everything is fine
- # tuples are always considered keys, never as list-likes
- if col not in self:
- missing.append(col)
- elif isinstance(col, (ABCIndexClass, ABCSeries,
- np.ndarray, list)):
+ if isinstance(col, (ABCIndexClass, ABCSeries, np.ndarray, list)):
# arrays are fine as long as they are one-dimensional
if getattr(col, 'ndim', 1) > 1:
raise ValueError(err_msg)
From 29fa8c005ad1a5d3663e1b29006bb99b045e560f Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Fri, 1 Feb 2019 20:25:39 +0100
Subject: [PATCH 11/21] Unambiguous KeyError message
---
pandas/core/frame.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 9dd5b0750c200..3799d9b91d652 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4172,7 +4172,7 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
missing.append(col)
if missing:
- raise KeyError('{}'.format(missing))
+ raise KeyError('None of {} are in the columns'.format(missing))
if inplace:
frame = self
From 5590433a434bdc3f613a0919227a448a316cd5f8 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Sun, 3 Feb 2019 21:56:45 +0100
Subject: [PATCH 12/21] Remove redundant whatsnew
---
doc/source/whatsnew/v0.24.1.rst | 1 -
1 file changed, 1 deletion(-)
diff --git a/doc/source/whatsnew/v0.24.1.rst b/doc/source/whatsnew/v0.24.1.rst
index bd1f124a415b4..be0a2eb682e87 100644
--- a/doc/source/whatsnew/v0.24.1.rst
+++ b/doc/source/whatsnew/v0.24.1.rst
@@ -68,7 +68,6 @@ Bug Fixes
**Reshaping**
- Bug in :meth:`DataFrame.groupby` with :class:`Grouper` when there is a time change (DST) and grouping frequency is ``'1d'`` (:issue:`24972`)
-- Fixed regression where custom hashable types could not be used as column keys in :meth:`DataFrame.set_index` (:issue:`24969`)
**Visualization**
From 2c4eaeaabe81e8287b0dc7e410fb7cc66f6c88be Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Sat, 9 Feb 2019 21:01:48 +0100
Subject: [PATCH 13/21] Review (jorisvandenbossche)
---
pandas/compat/__init__.py | 2 ++
pandas/core/frame.py | 7 +++----
2 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/pandas/compat/__init__.py b/pandas/compat/__init__.py
index d7ca7f8963f70..4036af85b7212 100644
--- a/pandas/compat/__init__.py
+++ b/pandas/compat/__init__.py
@@ -137,6 +137,7 @@ def lfilter(*args, **kwargs):
reload = reload
Hashable = collections.abc.Hashable
Iterable = collections.abc.Iterable
+ Iterator = collections.abc.Iterator
Mapping = collections.abc.Mapping
MutableMapping = collections.abc.MutableMapping
Sequence = collections.abc.Sequence
@@ -199,6 +200,7 @@ def get_range_parameters(data):
Hashable = collections.Hashable
Iterable = collections.Iterable
+ Iterator = collections.Iterator
Mapping = collections.Mapping
MutableMapping = collections.MutableMapping
Sequence = collections.Sequence
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index c7252834b76c4..33b62826c5075 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -34,7 +34,7 @@
from pandas import compat
from pandas.compat import (range, map, zip, lmap, lzip, StringIO, u,
- PY36, raise_with_traceback,
+ PY36, raise_with_traceback, Iterator,
string_and_binary_types)
from pandas.compat.numpy import function as nv
from pandas.core.dtypes.cast import (
@@ -4130,10 +4130,9 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
# arrays are fine as long as they are one-dimensional
if getattr(col, 'ndim', 1) > 1:
raise ValueError(err_msg)
- elif is_list_like(col) and not hasattr(col, '__len__'):
+ elif isinstance(col, Iterator):
# various iterators/generators are hashable, but should not
- # raise a KeyError; we identify them by their lack of __len__.
- # hashable listlikes with __len__ get tested as keys below.
+ # raise a KeyError; other list-likes get tested as keys below.
tipo = type(col)
raise TypeError(err_msg + ' Received column of '
'type {}'.format(tipo))
From ea10359f00ebafd3bef71d24fd342ef5db459b47 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Sun, 17 Feb 2019 14:06:49 +0100
Subject: [PATCH 14/21] Review (jreback)
---
doc/source/whatsnew/v0.25.0.rst | 1 +
pandas/core/frame.py | 18 ++++++++----------
pandas/tests/frame/test_alter_axes.py | 14 ++++++++------
3 files changed, 17 insertions(+), 16 deletions(-)
diff --git a/doc/source/whatsnew/v0.25.0.rst b/doc/source/whatsnew/v0.25.0.rst
index 686c5ad0165e7..b9b7461838dcc 100644
--- a/doc/source/whatsnew/v0.25.0.rst
+++ b/doc/source/whatsnew/v0.25.0.rst
@@ -21,6 +21,7 @@ Other Enhancements
- :meth:`Timestamp.replace` now supports the ``fold`` argument to disambiguate DST transition times (:issue:`25017`)
- :meth:`DataFrame.at_time` and :meth:`Series.at_time` now support :meth:`datetime.time` objects with timezones (:issue:`24043`)
+- :meth:`DataFrame.set_index` now works for instances of ``abc.Iterator``, provided their output is of the same length as the calling frame (:issue:`22484`)
-
.. _whatsnew_0250.api_breaking:
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 59c4f2b2a9568..c8056091e1dac 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4109,24 +4109,19 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
missing = []
for col in keys:
- if isinstance(col, (ABCIndexClass, ABCSeries, np.ndarray, list)):
+ if isinstance(col, (ABCIndexClass, ABCSeries, np.ndarray,
+ list, Iterator)):
# arrays are fine as long as they are one-dimensional
- if getattr(col, 'ndim', 1) > 1:
+ # iterators get converted to list below
+ if getattr(col, 'ndim', 1) != 1:
raise ValueError(err_msg)
- elif isinstance(col, Iterator):
- # various iterators/generators are hashable, but should not
- # raise a KeyError; other list-likes get tested as keys below.
- tipo = type(col)
- raise TypeError(err_msg + ' Received column of '
- 'type {}'.format(tipo))
else:
# everything else gets tried as a key; see GH 24969
try:
found = col in self.columns
except TypeError:
- tipo = type(col)
raise TypeError(err_msg + ' Received column of '
- 'type {}'.format(tipo))
+ 'type {}'.format(type(col)))
else:
if not found:
missing.append(col)
@@ -4162,6 +4157,9 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
elif isinstance(col, (list, np.ndarray)):
arrays.append(col)
names.append(None)
+ elif isinstance(col, Iterator):
+ arrays.append(list(col))
+ names.append(None)
# from here, col can only be a column label
else:
arrays.append(frame[col]._values)
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index aee2384718602..aabf666ea0e26 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -178,10 +178,10 @@ def test_set_index_pass_arrays(self, frame_of_index_cols,
# MultiIndex constructor does not work directly on Series -> lambda
# We also emulate a "constructor" for the label -> lambda
# also test index name if append=True (name is duplicate here for A)
- @pytest.mark.parametrize('box2', [Series, Index, np.array, list,
+ @pytest.mark.parametrize('box2', [Series, Index, np.array, list, iter,
lambda x: MultiIndex.from_arrays([x]),
lambda x: x.name])
- @pytest.mark.parametrize('box1', [Series, Index, np.array, list,
+ @pytest.mark.parametrize('box1', [Series, Index, np.array, list, iter,
lambda x: MultiIndex.from_arrays([x]),
lambda x: x.name])
@pytest.mark.parametrize('append, index_name', [(True, None),
@@ -195,6 +195,9 @@ def test_set_index_pass_arrays_duplicate(self, frame_of_index_cols, drop,
keys = [box1(df['A']), box2(df['A'])]
result = df.set_index(keys, drop=drop, append=append)
+ # if either box is iter, it has been consumed; re-read
+ keys = [box1(df['A']), box2(df['A'])]
+
# need to adapt first drop for case that both keys are 'A' --
# cannot drop the same column twice;
# use "is" because == would give ambiguous Boolean error for containers
@@ -255,18 +258,17 @@ def test_set_index_raise_keys(self, frame_of_index_cols, drop, append):
@pytest.mark.parametrize('append', [True, False])
@pytest.mark.parametrize('drop', [True, False])
- @pytest.mark.parametrize('box', [set, iter, lambda x: (y for y in x)],
- ids=['set', 'iter', 'generator'])
+ @pytest.mark.parametrize('box', [set], ids=['set'])
def test_set_index_raise_on_type(self, frame_of_index_cols, box,
drop, append):
df = frame_of_index_cols
msg = 'The parameter "keys" may be a column key, .*'
- # forbidden type, e.g. set/iter/generator
+ # forbidden type, e.g. set
with pytest.raises(TypeError, match=msg):
df.set_index(box(df['A']), drop=drop, append=append)
- # forbidden type in list, e.g. set/iter/generator
+ # forbidden type in list, e.g. set
with pytest.raises(TypeError, match=msg):
df.set_index(['A', df['A'], box(df['A'])],
drop=drop, append=append)
From ca178954b2ee4fd05196140674e1e8c0b1eb95c8 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Sun, 17 Feb 2019 16:13:26 +0100
Subject: [PATCH 15/21] Retrigger after connectivity issues
From f4deacce184a5ea5a89825a4e5f50d5655acc580 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Mon, 18 Feb 2019 20:30:16 +0100
Subject: [PATCH 16/21] Review (jorisvandenbossche)
---
pandas/core/frame.py | 9 +++++++++
pandas/tests/frame/test_alter_axes.py | 24 ++++++++++++++++++++++++
2 files changed, 33 insertions(+)
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index c8056091e1dac..dea81c9077e74 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4167,6 +4167,15 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
if drop:
to_remove.append(col)
+ if len(arrays[-1]) != len(self):
+ # check newest element against length of calling frame;
+ # ensure_index_from_sequences does not raise if append is False
+ raise ValueError('Length mismatch: Expected {len_self} rows, '
+ 'received array of length {len_col}'.format(
+ len_self=len(self),
+ len_col=len(arrays[-1])
+ ))
+
index = ensure_index_from_sequences(arrays, names)
if verify_integrity and not index.is_unique:
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index aabf666ea0e26..61e4768e930fe 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -256,6 +256,30 @@ def test_set_index_raise_keys(self, frame_of_index_cols, drop, append):
df.set_index(['A', df['A'], tuple(df['A'])],
drop=drop, append=append)
+ # MultiIndex constructor does not work directly on Series -> lambda
+ @pytest.mark.parametrize('box', [Series, Index, np.array, iter,
+ lambda x: MultiIndex.from_arrays([x])],
+ ids=['Series', 'Index', 'np.array',
+ 'iter', 'MultiIndex'])
+ @pytest.mark.parametrize('length', [4, 6], ids=['too_short', 'too_long'])
+ @pytest.mark.parametrize('append', [True, False])
+ @pytest.mark.parametrize('drop', [True, False])
+ def test_set_index_raise_on_len(self, frame_of_index_cols, box, length,
+ drop, append):
+ df = frame_of_index_cols # has length 5
+
+ values = np.random.randint(0, 10, (length,))
+
+ msg = 'Length mismatch: Expected 5 rows, received array of length.*'
+
+ # wrong length directly
+ with pytest.raises(ValueError, match=msg):
+ df.set_index(box(values), drop=drop, append=append)
+
+ # wrong length in list
+ with pytest.raises(ValueError, match=msg):
+ df.set_index(['A', df.A, box(values)], drop=drop, append=append)
+
@pytest.mark.parametrize('append', [True, False])
@pytest.mark.parametrize('drop', [True, False])
@pytest.mark.parametrize('box', [set], ids=['set'])
From 9bfcfde175f6c78c2f7415139a561e2972cbe89b Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Mon, 18 Feb 2019 20:39:48 +0100
Subject: [PATCH 17/21] move test for easier diff
---
pandas/tests/frame/test_alter_axes.py | 34 +++++++++++++--------------
1 file changed, 17 insertions(+), 17 deletions(-)
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index 61e4768e930fe..f31fe6fee9949 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -256,6 +256,23 @@ def test_set_index_raise_keys(self, frame_of_index_cols, drop, append):
df.set_index(['A', df['A'], tuple(df['A'])],
drop=drop, append=append)
+ @pytest.mark.parametrize('append', [True, False])
+ @pytest.mark.parametrize('drop', [True, False])
+ @pytest.mark.parametrize('box', [set], ids=['set'])
+ def test_set_index_raise_on_type(self, frame_of_index_cols, box,
+ drop, append):
+ df = frame_of_index_cols
+
+ msg = 'The parameter "keys" may be a column key, .*'
+ # forbidden type, e.g. set
+ with pytest.raises(TypeError, match=msg):
+ df.set_index(box(df['A']), drop=drop, append=append)
+
+ # forbidden type in list, e.g. set
+ with pytest.raises(TypeError, match=msg):
+ df.set_index(['A', df['A'], box(df['A'])],
+ drop=drop, append=append)
+
# MultiIndex constructor does not work directly on Series -> lambda
@pytest.mark.parametrize('box', [Series, Index, np.array, iter,
lambda x: MultiIndex.from_arrays([x])],
@@ -280,23 +297,6 @@ def test_set_index_raise_on_len(self, frame_of_index_cols, box, length,
with pytest.raises(ValueError, match=msg):
df.set_index(['A', df.A, box(values)], drop=drop, append=append)
- @pytest.mark.parametrize('append', [True, False])
- @pytest.mark.parametrize('drop', [True, False])
- @pytest.mark.parametrize('box', [set], ids=['set'])
- def test_set_index_raise_on_type(self, frame_of_index_cols, box,
- drop, append):
- df = frame_of_index_cols
-
- msg = 'The parameter "keys" may be a column key, .*'
- # forbidden type, e.g. set
- with pytest.raises(TypeError, match=msg):
- df.set_index(box(df['A']), drop=drop, append=append)
-
- # forbidden type in list, e.g. set
- with pytest.raises(TypeError, match=msg):
- df.set_index(['A', df['A'], box(df['A'])],
- drop=drop, append=append)
-
def test_set_index_custom_label_type(self):
# GH 24969
From ca2ac6078e528a6a1b6a72da1a1cbe2a08d70f9c Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Tue, 19 Feb 2019 20:19:41 +0100
Subject: [PATCH 18/21] Review (jreback)
---
doc/source/whatsnew/v0.25.0.rst | 2 +-
pandas/tests/frame/test_alter_axes.py | 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/doc/source/whatsnew/v0.25.0.rst b/doc/source/whatsnew/v0.25.0.rst
index 70defbc070f73..797ec2b6c3a83 100644
--- a/doc/source/whatsnew/v0.25.0.rst
+++ b/doc/source/whatsnew/v0.25.0.rst
@@ -21,7 +21,7 @@ Other Enhancements
- :meth:`Timestamp.replace` now supports the ``fold`` argument to disambiguate DST transition times (:issue:`25017`)
- :meth:`DataFrame.at_time` and :meth:`Series.at_time` now support :meth:`datetime.time` objects with timezones (:issue:`24043`)
-- :meth:`DataFrame.set_index` now works for instances of ``abc.Iterator``, provided their output is of the same length as the calling frame (:issue:`22484`)
+- :meth:`DataFrame.set_index` now works for instances of ``abc.Iterator``, provided their output is of the same length as the calling frame (:issue:`22484`, :issue:`24984`)
-
.. _whatsnew_0250.api_breaking:
diff --git a/pandas/tests/frame/test_alter_axes.py b/pandas/tests/frame/test_alter_axes.py
index f31fe6fee9949..a25e893e08900 100644
--- a/pandas/tests/frame/test_alter_axes.py
+++ b/pandas/tests/frame/test_alter_axes.py
@@ -283,6 +283,7 @@ def test_set_index_raise_on_type(self, frame_of_index_cols, box,
@pytest.mark.parametrize('drop', [True, False])
def test_set_index_raise_on_len(self, frame_of_index_cols, box, length,
drop, append):
+ # GH 24984
df = frame_of_index_cols # has length 5
values = np.random.randint(0, 10, (length,))
From 87bd0a6f7b9377b24b2a185f8a658f9c23f5beac Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Tue, 19 Feb 2019 22:20:52 +0100
Subject: [PATCH 19/21] Add 'conda list' for azure/posix after activate
'pandas-dev'
---
ci/incremental/setup_conda_environment.cmd | 1 +
ci/incremental/setup_conda_environment.sh | 3 +++
ci/install_travis.sh | 1 +
3 files changed, 5 insertions(+)
diff --git a/ci/incremental/setup_conda_environment.cmd b/ci/incremental/setup_conda_environment.cmd
index c104d78591384..1893954c570c8 100644
--- a/ci/incremental/setup_conda_environment.cmd
+++ b/ci/incremental/setup_conda_environment.cmd
@@ -16,6 +16,7 @@ conda remove --all -q -y -n pandas-dev
conda env create --file=ci\deps\azure-windows-%CONDA_PY%.yaml
call activate pandas-dev
+@rem Display pandas-dev environment (for debugging)
conda list
if %errorlevel% neq 0 exit /b %errorlevel%
diff --git a/ci/incremental/setup_conda_environment.sh b/ci/incremental/setup_conda_environment.sh
index f174c17a614d8..6903553abcfe9 100755
--- a/ci/incremental/setup_conda_environment.sh
+++ b/ci/incremental/setup_conda_environment.sh
@@ -23,6 +23,9 @@ set +v
source activate pandas-dev
set -v
+# Display pandas-dev environment (for debugging)
+conda list
+
# remove any installed pandas package
# w/o removing anything else
echo
diff --git a/ci/install_travis.sh b/ci/install_travis.sh
index d1a940f119228..5169a55ae6e91 100755
--- a/ci/install_travis.sh
+++ b/ci/install_travis.sh
@@ -100,6 +100,7 @@ pip list --format columns |grep pandas
echo "[running setup.py develop]"
python setup.py develop || exit 1
+# Display pandas-dev environment (for debugging)
echo
echo "[show environment]"
conda list
From 759b369a8a27db1cb21da454ea718aa33b6431c3 Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Wed, 20 Feb 2019 18:24:34 +0100
Subject: [PATCH 20/21] Revert "Add 'conda list' for azure/posix after activate
'pandas-dev'"
This reverts commit 87bd0a6f7b9377b24b2a185f8a658f9c23f5beac.
---
ci/incremental/setup_conda_environment.cmd | 1 -
ci/incremental/setup_conda_environment.sh | 3 ---
ci/install_travis.sh | 1 -
3 files changed, 5 deletions(-)
diff --git a/ci/incremental/setup_conda_environment.cmd b/ci/incremental/setup_conda_environment.cmd
index 1893954c570c8..c104d78591384 100644
--- a/ci/incremental/setup_conda_environment.cmd
+++ b/ci/incremental/setup_conda_environment.cmd
@@ -16,7 +16,6 @@ conda remove --all -q -y -n pandas-dev
conda env create --file=ci\deps\azure-windows-%CONDA_PY%.yaml
call activate pandas-dev
-@rem Display pandas-dev environment (for debugging)
conda list
if %errorlevel% neq 0 exit /b %errorlevel%
diff --git a/ci/incremental/setup_conda_environment.sh b/ci/incremental/setup_conda_environment.sh
index 6903553abcfe9..f174c17a614d8 100755
--- a/ci/incremental/setup_conda_environment.sh
+++ b/ci/incremental/setup_conda_environment.sh
@@ -23,9 +23,6 @@ set +v
source activate pandas-dev
set -v
-# Display pandas-dev environment (for debugging)
-conda list
-
# remove any installed pandas package
# w/o removing anything else
echo
diff --git a/ci/install_travis.sh b/ci/install_travis.sh
index 5169a55ae6e91..d1a940f119228 100755
--- a/ci/install_travis.sh
+++ b/ci/install_travis.sh
@@ -100,7 +100,6 @@ pip list --format columns |grep pandas
echo "[running setup.py develop]"
python setup.py develop || exit 1
-# Display pandas-dev environment (for debugging)
echo
echo "[show environment]"
conda list
From ecc7d03973975d4af91927e2117fefa02891fcbb Mon Sep 17 00:00:00 2001
From: "H. Vetinari" <h.vetinari@gmx.com>
Date: Thu, 21 Feb 2019 08:19:17 +0100
Subject: [PATCH 21/21] Reflect change in docstring
---
pandas/core/frame.py | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index dace327126aa2..608e5c53ec094 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4025,7 +4025,8 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
This parameter can be either a single column key, a single array of
the same length as the calling DataFrame, or a list containing an
arbitrary combination of column keys and arrays. Here, "array"
- encompasses :class:`Series`, :class:`Index` and ``np.ndarray``.
+ encompasses :class:`Series`, :class:`Index`, ``np.ndarray``, and
+ instances of :class:`abc.Iterator`.
drop : bool, default True
Delete columns to be used as the new index.
append : bool, default False
@@ -4169,8 +4170,8 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
to_remove.append(col)
if len(arrays[-1]) != len(self):
- # check newest element against length of calling frame;
- # ensure_index_from_sequences does not raise if append is False
+ # check newest element against length of calling frame, since
+ # ensure_index_from_sequences would not raise for append=False.
raise ValueError('Length mismatch: Expected {len_self} rows, '
'received array of length {len_col}'.format(
len_self=len(self),