3
���hpb � @ sj U d dl Z d dlmZ d dlmZ d dlmZ d dlmZ d dlZd dl Z d dl
Z
d dl m Z d dl
Z
d dlZd dlmZmZmZmZmZmZmZmZ d dlZd dlZd dlZd dl m Z m Z d d l m Z m!Z!m"Z" d d
l#m$Z$ d dl%j&j'Z(d d l)m*Z*m+Z+m,Z, d d l-m.Z.m/Z/ d d
l0m1Z1m2Z2m3Z3m4Z4m5Z5m6Z6m7Z7m8Z8m9Z9m:Z:m;Z;m<Z< d dl=m>Z> d dl?Z@d dl?mAZAmBZBmCZCmDZDmEZEmFZFmGZGmHZHmIZImJZJ d dlKmLZL d dlMmNZNmOZOmPZPmQZQmRZRmSZS d dlTmUZU d dlVmWZW d dlXmYZY e/� ZZdZ[dZ\dZ]ddddgZ^ee* ^d d d d gZ_ee* _e`d d!d"d#gZaee* ad$d%d&d'gZbee* be^ea Zce_eb Zdeed(d)gZfee* fegd*d+gZhee* heid,d-gZjee* jd.d/gZkee* kd0d1gZlee* lemd2gZneod3gZpeqd4gZrefec Zseseh ej ek el en er ep ZteuevfZwd5d6� Zxd7d8� Zyex� d9d:� Zz�dPeee+ e,d;�d<d=�Z{�dQeei d>�d?d@�Z|�dReei d>�dAdB�Z}edCdD� �Z~�dSdFdG�Zeeme`f eedH�dIdJ�Z�dKe$dLdMfeemeif eeme`f eeeedN�dOdP�Z�dQdR� Z��dTemdT�dUdV�Z�f dWfeedX�dYdZ�Z�ej�e�e
j�e
j� �ej�d[fd\�Z�ej�e�d]j�e�e�e�d^�dU���e
j� �ej�d[fd\�Z��dVdadb�Z��dWdcdd�Z�dedf� Z��dXdgdh�Z�e�dYdidj��Z�edkdl� �Z�edmdn� �Z�emdo�dpdq�Z�dKdSe$dSdSdLdMdrfeEeEeemeif emeeme`f ememeeeeeidds� dtdu�Z��dZeemeif dw�dxdy�Z��d[eieid{�d|d}�Z�d~d� Z�d�d�� Z��d\d�d��Z��d]d�d��Z��d^d�d��Z��d_d�d��Z��d`d�d��Z��dad�d��Z��dbd�d��Z�dSde$ddLdMfeeeed��d�d��Z�dSdKdSe$dSdddSdSdSdLdMd�f
d�d��Z�dSdKdKdSe$dSddddSddSdLdMd�fd�d��Z�d�d�� Z��dcd�d��Z�d�d�� Z�d�d�� Z�d�d�� Z�d�d�� Z�d�d�� Z��ddd�d��Z��ded�d��Z��dfd�d��Z��dgd�d��Z��dhd�d��Z��did�d��Z��djd�d��Z��dkd�d��Z��dld�d��Z��dmd�dÄZ��dnd�dƄZ��dod�dȄZ��dpd�dʄZ�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�d�gZ��dqd�d�Z�d�d� Z��drd�d�Z��dsd�d�Z��dtd�d�Z��dud�d�Z�d�d�� ZÐdvd�d��ZĐdwd�d��ZŐdxd�d��ZƐdyd�d��Zǐdzd�d��ZȐd �d� Zɐd�d� Zʐd�d� Zːd{�d�d�Z̐d|�d �d
�Z͐d}�d �d �Zΐd~�d
�d�Zϐd�d�d�ZАd�d� Zѐd�ZҐd�ZӐd'�d(� ZԐd��d)�d*�Z�eѐd+e]dde�e�f�d,�d-��Z�e�Z�eeؐd.dSdSf�d/�d0��Z�G �d1�d2� �d2�Z�e�d3�d4� �Z�e�d��d5�d6��Zܐd��d8�d9�Z�G �d:�d;� �d;eI�Z�G �d<�d=� �d=eC�Z�G �d>�d?� �d?eA�Z�eei�d@��dA�dB��Z�d��dC�dD�Z�eei �dE��dF�dG�Z�ee� e�dH��dI�dJ�Z�e@j�j�j�j�j� Z�dK�dL� Z�eie�dM��dN�dO�Z�dS (� � N)�Counter)�contextmanager)�datetime)�wraps)�rmtree)�Any�Callable�ContextManager�List�Optional�Type�Union�cast)�rand�randn)�can_set_locale� get_locales�
set_locale)�
no_default)�Dtype�FilePathOrBuffer�
FrameOrSeries)�_get_lzma_file� _import_lzma) �is_bool�is_categorical_dtype�is_datetime64_dtype�is_datetime64tz_dtype�is_extension_array_dtype�is_interval_dtype� is_number�is_numeric_dtype�is_period_dtype� is_sequence�is_timedelta64_dtype�needs_i8_conversion)�array_equivalent)
� Categorical�CategoricalIndex� DataFrame�
DatetimeIndex�Index�
IntervalIndex�
MultiIndex�
RangeIndex�Series� bdate_range)�take_1d)�
DatetimeArray�ExtensionArray�
IntervalArray� PeriodArray�TimedeltaArray� period_array)�DatetimeLikeArrayMixin)�urlopen)� pprint_thing� � FZuint8Zuint16Zuint32Zuint64ZUInt8ZUInt16ZUInt32ZUInt64Zint8Zint16Zint32Zint64ZInt8ZInt16ZInt32ZInt64Zfloat32Zfloat64Z complex64Z
complex128�str�Uzdatetime64[ns]zM8[ns]ztimedelta64[ns]zm8[ns]�bool�bytes�objectc C s& t jjdd�} d| kr"tjdt� d S )N�PANDAS_TESTING_MODE�None� deprecate�always)�os�environ�get�warnings� simplefilter�_testing_mode_warnings)� testing_mode� rM �1/tmp/pip-build-5_djhm0z/pandas/pandas/_testing.py�set_testing_moder s rO c C s& t jjdd�} d| kr"tjdt� d S )NrB rC rD �ignore)rF rG rH rI rJ rK )rL rM rM rN �reset_testing_modey s rQ c C s t jddd� dS )zJ
Reset the display options for printing and representing objects.
z ^display.T)ZsilentN)�pdZ reset_optionrM rM rM rN �reset_display_options� s rS )�obj�path�returnc
C sF |}|dkr dt d�� d�}t|��}tj| |� tj|�S Q R X dS )a�
Pickle an object and then read it again.
Parameters
----------
obj : any object
The object to pickle and then re-read.
path : str, path object or file-like object, default None
The path where the pickled object is written and then read.
Returns
-------
pandas object
The original object that was pickled and then re-read.
N�__�
z __.pickle)�rands� ensure_cleanrR Z to_pickleZ read_pickle)rT rU �_pathZ temp_pathrM rM rN �round_trip_pickle� s
r\ )rU c
C sP ddl }|jd�j}|dkr d}t|�� }| ||�� |||��}W dQ R X |S )a�
Write an object to file specified by a pathlib.Path and read it back
Parameters
----------
writer : callable bound to pandas object
IO writing function (e.g. DataFrame.to_csv )
reader : callable
IO reading function (e.g. pd.read_csv )
path : str, default None
The path where the object is written and then read.
Returns
-------
pandas object
The original object that was serialized and then re-read.
r NZpathlibZ
___pathlib___)�pytest� importorskip�PathrZ )�writer�readerrU r] r_ rT rM rM rN �round_trip_pathlib� s
rb c
C sP ddl }|jd�j}|dkr d}t|�� }| ||�� |||��}W dQ R X |S )a�
Write an object to file specified by a py.path LocalPath and read it back.
Parameters
----------
writer : callable bound to pandas object
IO writing function (e.g. DataFrame.to_csv )
reader : callable
IO reading function (e.g. pd.read_csv )
path : str, default None
The path where the object is written and then read.
Returns
-------
pandas object
The original object that was serialized and then re-read.
r Nzpy.pathZ___localpath___)r] r^ �localrZ )r` ra rU r] Z LocalPathrT rM rM rN �round_trip_localpath� s
rd c c s� |dkrt | d�}n�|dkr*tj | d�}n�|dkr@tj| d�}nn|dkrXtt�| d�}nV|dkr�tj| �}|j� }t |�dkr�|j |j
� �}q�t d| � d ���nt d
|� ���z
|V W d|j � |dkr�|j � X dS ) a
Open a compressed file and return a file object.
Parameters
----------
path : str
The path where the file is read from.
compression : {'gzip', 'bz2', 'zip', 'xz', None}
Name of the decompression to use
Returns
-------
file object
N�rb�gzip�bz2�xz�zip� z ZIP file z error. Only one file per ZIP.z Unrecognized compression type: )
�openrf rg �BZ2Filer �lzma�zipfile�ZipFile�namelist�len�pop�
ValueError�close)rU � compression�fZzip_fileZ zip_namesrM rM rN �decompress_file� s(
rw �testc
C s� | dkrt j}n@| dkr tj}n0| dkr0tj}n | dkrBtt�}ntd| � ���| dkrjd}||f}d}nd}|f}d }|||d
��}t ||�|� W d Q R X d S ) a�
Write data to a compressed file.
Parameters
----------
compression : {'gzip', 'bz2', 'zip', 'xz'}
The compression type to use.
path : str
The file path to write the data.
data : str
The data to write.
dest : str, default "test"
The destination file (for ZIP only)
Raises
------
ValueError : An invalid compression value was passed in.
ri rf rg rh z Unrecognized compression type: �w�writestr�wb�write)�modeN)
rn ro rf �GzipFilerg rl r rm rs �getattr) ru rU �data�destZcompress_methodr} �args�methodrv rM rM rN �write_to_compressed s$
r� )�check_less_preciserV c C s* t | t�r| rdS dS ndd| S dS )a|
Return the tolerance equivalent to the deprecated `check_less_precise`
parameter.
Parameters
----------
check_less_precise : bool or int
Returns
-------
float
Tolerance to be used as relative/absolute tolerance.
Examples
--------
>>> # Using check_less_precise as a bool:
>>> _get_tol_from_less_precise(False)
0.5e-5
>>> _get_tol_from_less_precise(True)
0.5e-3
>>> # Using check_less_precise as an int representing the decimal
>>> # tolerance intended:
>>> _get_tol_from_less_precise(2)
0.5e-2
>>> _get_tol_from_less_precise(8)
0.5e-8
g����Mb@?g�h㈵��>g �?rX N)�
isinstancer? )r� rM rM rN �_get_tol_from_less_precise4 s
r� �equivg�h㈵��>g:�0�yE>)� check_dtyper� �rtol�atolc K s |t k r$tjdtdd� t|� }}t| tj�rNt| |fd|||d�|�� n�t| tj �rxt
| |fd|||d�|�� n�t| tj �r�t | |fd|||d�|�� nt|r�t
| �r�t
|�r�nBt| �r�t|�r�n0t| tj�s�t|tj�r�d}nd}t| ||d � tj| |f|||d
�|�� d S ) a�
Check that the left and right objects are approximately equal.
By approximately equal, we refer to objects that are numbers or that
contain numbers which may be equivalent to specific levels of precision.
Parameters
----------
left : object
right : object
check_dtype : bool or {'equiv'}, default 'equiv'
Check dtype if both a and b are the same type. If 'equiv' is passed in,
then `RangeIndex` and `Int64Index` are also considered equivalent
when doing type checking.
check_less_precise : bool or int, default False
Specify comparison precision. 5 digits (False) or 3 digits (True)
after decimal points are compared. If int, then specify the number
of digits to compare.
When comparing two numbers, if the first number has magnitude less
than 1e-5, we compare the two numbers directly and check whether
they are equivalent within the specified precision. Otherwise, we
compare the **ratio** of the second number to the first number and
check whether it is equivalent to 1 within the specified precision.
.. deprecated:: 1.1.0
Use `rtol` and `atol` instead to define relative/absolute
tolerance, respectively. Similar to :func:`math.isclose`.
rtol : float, default 1e-5
Relative tolerance.
.. versionadded:: 1.1.0
atol : float, default 1e-8
Absolute tolerance.
.. versionadded:: 1.1.0
z�The 'check_less_precise' keyword in testing.assert_*_equal is deprecated and will be removed in a future version. You can stop passing 'check_less_precise' to silence this warning.� )�
stacklevelF)� check_exact�exactr� r� )r� r� r� r� z numpy array�Input)rT )r� r� r� N)r rI �warn�
FutureWarningr� r� rR r+ �assert_index_equalr/ �assert_series_equalr) �assert_frame_equalr r �np�ndarray�assert_class_equal�_testing�assert_almost_equal)�left�rightr� r� r� r� �kwargsrT rM rM rN r� ] sX .
r� c C sZ |j }t| |�s.t|� d|� dt| �� d���t||�sVt|� d|� dt|�� d���dS )a�
Helper method for our assert_* methods that ensures that
the two objects being compared have the right type before
proceeding with the comparison.
Parameters
----------
left : The first object being compared.
right : The second object being compared.
cls : The class type to check against.
Raises
------
AssertionError : Either `left` or `right` is not an instance of `cls`.
z Expected type z, found z insteadN)�__name__r� �AssertionError�type)r� r� �clsZcls_namerM rM rN �_check_isinstance� s
r� T)� compare_keysc C s t | |t� tj| ||d� d S )N)r� )r� �dictr� �assert_dict_equal)r� r� r� rM rM rN r� � s r� g �?)�pc C s t | � |kS )N)r )�sizer� rM rM rN �randbool� s r� rj )�dtype� i� � �Oc C s6 t jjt| t j|� d�jt j| f�j|�}|j|�S )z,
Generate an array of byte strings.
)r� ) r� �random�choice� RANDS_CHARS�prod�view�str_�reshape�astype)�ncharsr� r� �retvalrM rM rN � rands_array� s r� c C s6 t jjt| t j|� d�jt j| f�j|�}|j|�S )z/
Generate an array of unicode strings.
)r� ) r� r� r� � RANDU_CHARSr� r� �unicode_r� r� )r� r� r� r� rM rM rN � randu_array s r� c C s dj tjjt| ��S )zt
Generate one random byte string.
See `rands_array` if you want to create an array of random strings.
r� )�joinr� r� r� r� )r� rM rM rN rY s rY c C s> ddl m}m} | d kr2x |� D ] } || � q W n|| � d S )Nr )rt � get_fignums)�matplotlib.pyplotrt r� )Zfignum�_closer� rM rM rN rt s
rt c ) k s* | pd} d}| |d< |r<t jf |�}z
|V W d|j� X n�ttjj| ��rTtd��yt jf |�\}} W n& t k
r� ddl
}|j d� Y nX z
| V W dytj|� W n* t k
r� t
d|� d| � d �� Y nX ytjj| �r�tj| � W n2 t k
�r" } zt
d
|� �� W Y dd}~X nX X dS ) a
Gets a temporary path and agrees to remove on close.
Parameters
----------
filename : str (optional)
if None, creates a temporary file which is then removed when out of
scope. if passed, creates temporary file with filename as ending.
return_filelike : bool (default False)
if True, returns a file-like which is *always* cleaned. Necessary for
savefig and other functions which want to append extensions.
**kwargs
Additional keywords passed in for creating a temporary file.
:meth:`tempFile.TemporaryFile` is used when `return_filelike` is ``True``.
:meth:`tempfile.mkstemp` is used when `return_filelike` is ``False``.
Note that the `filename` parameter will be passed in as the `suffix`
argument to either function.
See Also
--------
tempfile.TemporaryFile
tempfile.mkstemp
r� N�suffixz-Can't pass a qualified name to ensure_clean()r z$no unicode file names on this systemz Couldn't close file descriptor: z (file: �)z Exception on removing file: )�tempfile�
TemporaryFilert rq rF rU �dirnamers �mkstemp�UnicodeEncodeErrorr] �skip�OSError�print�exists�remove)�filenameZreturn_fileliker� �fdrv r] �erM rM rN rZ ' s4
rZ c c s@ t jdd�} z
| V W dy t| � W n tk
r8 Y nX X dS )z{
Get a temporary directory path and agrees to remove on close.
Yields
------
Temporary directory path
r� )r� N)r� �mkdtempr r� )Zdirectory_namerM rM rN �ensure_clean_dirf s
r� c c s2 t tj�} z
dV W dtjj� tjj| � X dS )z�
Get a context manager to safely set environment variables
All changes will be undone on close, hence environment variables set
within this contextmanager will neither persist nor change global state.
N)r� rF rG �clear�update)Z
saved_environrM rM rN �!ensure_safe_environment_variablesy s
r� )rV c C s t | �t |�kS )zO
Checks if the set of unique elements of arr1 and arr2 are equivalent.
)� frozenset)Zarr1Zarr2rM rM rN �
equalContents� s r� r+ ) r� r� r� � check_namesr� r� �check_categoricalr� r� rT rV c
s� d}
d� �fdd� } dd� } |t k r@tjdtdd � t|� }}t| |t� | | || d
� | j|jkr�| � d �}
| j� d | � �}|j� d |� �}t| |
||� t | �t |�kr�| � d
�}
t | �� d | � �}t |�� d |� �}t| |
||� | jdk�rrt
t | �} t
t |�}xft | j�D ]X}| | |�}| ||�}d|� d�}t
||�|||||d� | | j| |j| | d
� �qW |�rֈ �r�| j|��s�tj| j|jkjt��d t | � }| � dtj|d�� d�}t| || |� n tj| j|j||�| | |d� |�r
td| || d
� t| tj��s&t|tj��r6td| || d
� t| tj ��sRt|tj ��r`t | j |j � � �r�t | j ��s~t |j ��r�t!| j |j | � d�d
� dS ) a�
Check that left and right Index are equal.
Parameters
----------
left : Index
right : Index
exact : bool or {'equiv'}, default 'equiv'
Whether to check the Index class, dtype and inferred_type
are identical. If 'equiv', then RangeIndex can be substituted for
Int64Index as well.
check_names : bool, default True
Whether to check the names attribute.
check_less_precise : bool or int, default False
Specify comparison precision. Only used when check_exact is False.
5 digits (False) or 3 digits (True) after decimal points are compared.
If int, then specify the digits to compare.
.. deprecated:: 1.1.0
Use `rtol` and `atol` instead to define relative/absolute
tolerance, respectively. Similar to :func:`math.isclose`.
check_exact : bool, default True
Whether to compare number exactly.
check_categorical : bool, default True
Whether to compare internal Categorical exactly.
rtol : float, default 1e-5
Relative tolerance. Only used when check_exact is False.
.. versionadded:: 1.1.0
atol : float, default 1e-8
Absolute tolerance. Only used when check_exact is False.
.. versionadded:: 1.1.0
obj : str, default 'Index'
Specify object name being compared, internally used to show appropriate
assertion message.
Tr+ c sV �rRt | |�|d� � r(td| ||d� | jdkrB|jdksRt�ntd| ||d� d S )N)r� rT r� )rT �string�
inferred_type)r� �assert_attr_equalr� r� )�l�rrT )r� r� rM rN � _check_types� s
z(assert_index_equal.<locals>._check_typesc S s> | j | }| j| }t|j||jd�}|j|| j| d�}|S )N)�
fill_value)�name)�levels�codesr1 �_valuesZ _na_valueZ
_shallow_copy�names)�index�level�uniqueZ level_codesZfilled�valuesrM rM rN �_get_ilevel_values� s
z.assert_index_equal.<locals>._get_ilevel_valuesz�The 'check_less_precise' keyword in testing.assert_*_equal is deprecated and will be removed in a future version. You can stop passing 'check_less_precise' to silence this warning.r� )r� )rT z levels are differentz, z length are differentrj zMultiIndex level [�])r� r� r� r� r� rT g Y@z values are different (� z %))r� r� r� rT �lobjZrobjr� �freqz categoryN)r+ )"r rI r� r� r� r� r+ �nlevels�raise_assert_detailrq r r- �ranger� r� �equalsr� �sumr� r� �int�roundr� r� r� r� rR Z PeriodIndexr, �assert_interval_array_equalr� r r� �assert_categorical_equal)r� r� r� r� r� r� r� r� r� rT �__tracebackhide__r� r� �msg1�msg2Zmsg3r� ZllevelZrlevelr� �diff�msgrM )r� r� rN r� � sz 1
$ r� r� )r� c C s� d}dd� }|dkrjt | �t |�kr�t | �jt |�jh}t|ddh �r�|� d�}t|||| �||�� n4|r�t | �t |�kr�|� d�}t|||| �||�� d S )
z#
Checks classes are equal.
Tc S s t | t�r| S t| �jS )N)r� r+ r� r� )�xrM rM rN �
repr_class8 s
z&assert_class_equal.<locals>.repr_classr� Z
Int64Indexr. z classes are not equivalentz classes are differentN)r� r� rq r� )r� r� r� rT r� r� �typesr� rM rM rN r� 2 s
r� �
Attributes)�attrrT c
C s� d}t || �}t || �}||kr$dS t|�rLtj|�rLt|�rLtj|�rLdS y ||k}W n tk
rp d}Y nX t|t�s�|j� }|r�dS d| � d�}t||||� dS )aO
Check attributes are equal. Both objects must have attribute.
Parameters
----------
attr : str
Attribute name being compared.
left : object
right : object
obj : str, default 'Attributes'
Specify object name being compared, internally used to show appropriate
assertion message
TFz Attribute "z" are differentN) r r r� �isnan� TypeErrorr� r? �allr� ) r� r� r� rT r� Z left_attrZ
right_attr�resultr� rM rM rN r� L s(
r� c C s� dd l j}t| tjtjf�r\xl| j� D ]0}dtt |�j
�� �}t||j t f�s&t
|��q&W n.dtt | �j
�� �}t| |jtt f�s�t
|��d S )Nr zBone of 'objs' is not a matplotlib Axes instance, type encountered zoobjs is neither an ndarray of Artist instances nor a single ArtistArtist instance, tuple, or dict, 'objs' is a )r� Zpyplotr� rR r/ r� r� Zravel�reprr� r� ZAxesr� r� ZArtist�tuple)�objsZplt�elr� rM rM rN �"assert_is_valid_plot_return_objecty s
r c C s. t | ttf�r| j} t| tjtj| ��� dS )z#Assert that the sequence is sorted.N)r� r+ r/ r� �assert_numpy_array_equalr� �sort�array)�seqrM rM rN �assert_is_sorted� s r r' c C s� t | |t� |rDt| j|j|� d�d� t| j|j||� d�d� nxy| jj� }|jj� }W n" tk
r~ | j|j }}Y nX t|||� d�d� t| jj| j�|jj|j�|� d�d� t d| ||d� dS )a{
Test that Categoricals are equivalent.
Parameters
----------
left : Categorical
right : Categorical
check_dtype : bool, default True
Check that integer dtype of the codes are the same
check_category_order : bool, default True
Whether the order of the categories should be compared, which
implies identical integer codes. If False, only the resulting
values are compared. The ordered attribute is
checked regardless.
obj : str, default 'Categorical'
Specify object name being compared, internally used to show appropriate
assertion message
z .categories)rT z.codes)r� rT z.values�orderedN)
r� r' r� �
categoriesr r� Z sort_valuesr Ztaker� )r� r� r� �check_category_orderrT �lc�rcrM rM rN r� � s"
r� r4 c C sT t | |t� t| j|j||� d�d� t| j|j||� d�d� td| ||d� dS )a
Test that two IntervalArrays are equivalent.
Parameters
----------
left, right : IntervalArray
The IntervalArrays to compare.
exact : bool or {'equiv'}, default 'equiv'
Whether to check the Index class, dtype and inferred_type
are identical. If 'equiv', then RangeIndex can be substituted for
Int64Index as well.
obj : str, default 'IntervalArray'
Specify object name being compared, internally used to show appropriate
assertion message
z.left)r� rT �closed)rT N)r� r4 r� r� r� r� )r� r� r� rT rM rM rN r� � s r� r5 c C s8 t | |t� t| j|j|� d�d� td| ||d� d S )Nz._data)rT r� )r� r5 r �_datar� )r� r� rT rM rM rN �assert_period_array_equal� s r r2 c C sL d}t | |t� t| j|j|� d�d� td| ||d� td| ||d� d S )NTz._data)rT r� �tz)r� r2 r r r� )r� r� rT r� rM rM rN �assert_datetime_array_equal� s
r r6 c C s< d}t | |t� t| j|j|� d�d� td| ||d� d S )NTz._data)rT r� )r� r6 r r r� )r� r� rT r� rM rM rN � assert_timedelta_array_equal� s r c C s� d}| � d|� �}t |tj�r0|dt|�� �7 }t |tj�rFt|�}nt|�rVt|�}t |tj�rlt|�}nt|�r|t|�}|d|� d|� �7 }|d k r�|d|� �7 }t|��d S )NTz are different
z
[index]: z
[left]: z
[right]: z
[diff]: )r� r� r� r: r r r� )rT �messager� r� r� � index_valuesr� r� rM rM rN r� � s"
r� � numpy arrayc
s� d}t | |�d� t| |tj� dd� } | | �}
| |�} |dkrd|
| k r�tt|
�� dt| �� ���n*|dkr�|
| kr�tt|
�� dt| �� ���� ��fd d
�} t| |�d �s�| | ||� |r�t| tj�r�t|tj�r�td | |�d� d
S )a+
Check that 'np.ndarray' is equivalent.
Parameters
----------
left, right : numpy.ndarray or iterable
The two arrays to be compared.
strict_nan : bool, default False
If True, consider NaN and None to be different.
check_dtype : bool, default True
Check dtype if both a and b are np.ndarray.
err_msg : str, default None
If provided, used as assertion message.
check_same : None|'copy'|'same', default None
Ensure left and right refer/do not refer to the same memory area.
obj : str, default 'numpy array'
Specify object name being compared, internally used to show appropriate
assertion message.
index_values : numpy.ndarray, default None
optional index (shared by both left and right), used in output.
T)rT c S s t | dd �d k r| jS | S )N�base)r r )rT rM rM rN � _get_base6 s z+assert_numpy_array_equal.<locals>._get_baseZsamez is not �copyz is c s� |d kr�| j |j kr,t��� d�| j |j � d}x,t| |�D ] \}}t||�d�s<|d7 }q<W |d | j }�� dtj|d�� d�}t�|| |� d � t|��d S )
Nz shapes are differentr )�
strict_nanrj g Y@z values are different (r� z %))r )�shaper� ri r& r� r� r� r� )r� r� �err_msgr� r� r� r� )r rT r rM rN �_raiseC s z(assert_numpy_array_equal.<locals>._raise)r r� N) r� r� r� r� r� r r& r� r� )
r� r� r r� r Z
check_samerT r r� r Z left_baseZ
right_baser! rM )r rT r rN r s$ r )r� r� c C s. |t k r$tjdtdd� t|� }}t| t�s6td��t|t�sHtd��|r\td| |dd� t| t �r�t|t �r�t
|�t
| �kr�t t j
| j�t j
|j�|d � d
S t j
| j� �}t j
|j� �} t || d |d � t j
| | jt��}
t j
|| jt��} |�rt |
| d|d � ntj|
| |||d|d
� d
S )ac
Check that left and right ExtensionArrays are equal.
Parameters
----------
left, right : ExtensionArray
The two arrays to compare.
check_dtype : bool, default True
Whether to check if the ExtensionArray dtypes are identical.
index_values : numpy.ndarray, default None
Optional index (shared by both left and right), used in output.
check_less_precise : bool or int, default False
Specify comparison precision. Only used when check_exact is False.
5 digits (False) or 3 digits (True) after decimal points are compared.
If int, then specify the digits to compare.
.. deprecated:: 1.1.0
Use `rtol` and `atol` instead to define relative/absolute
tolerance, respectively. Similar to :func:`math.isclose`.
check_exact : bool, default False
Whether to compare number exactly.
rtol : float, default 1e-5
Relative tolerance. Only used when check_exact is False.
.. versionadded:: 1.1.0
atol : float, default 1e-8
Absolute tolerance. Only used when check_exact is False.
.. versionadded:: 1.1.0
Notes
-----
Missing values are checked separately from valid values.
A mask of missing values is computed for each and checked to match.
The remaining all-valid values are cast to object dtype and checked.
z�The 'check_less_precise' keyword in testing.assert_*_equal is deprecated and will be removed in a future version. You can stop passing 'check_less_precise' to silence this warning.r� )r� z left is not an ExtensionArrayz right is not an ExtensionArrayr� r3 )rT )r NzExtensionArray NA mask)rT r )r� r� r� rT r )r rI r� r� r� r� r3 r� r� r8 r� r r� �asarrayZasi8Zisnar� rA r� r� ) r� r� r� r r� r� r� r� Zleft_naZright_naZ
left_validZ right_validrM rM rN � assert_extension_array_equal_ sB .
r# r/ c C s d}|t k r(tjdtdd� t|� } }
t| |t� |rFt| ||d� t| �t|�kr�t| �� d| j � �}t|�� d|j � �}t
|d||� t | j |j |||| | |
|� d�d � | r�t | j t
jt
jf�r�| j }|j }|j|jks�t|j|jf��|�r.t| j��rt|j��r| �rntd
| |d |� �d� |�rrt| j��rrt|j��rrt| j|j|t|�tj| j �d � �nT|�r�t| j��s�t|j��r�t | j�j t |j���s�d
| j� d|j� d�}t|��n�t | j��r�t |j��r�t | j |j � n�t| j��st|j��r2t!j"| j|j| |
|t|�tj| j �d� n�t#| j��rht#|j��rht$| j|j|tj| j �d� n^t| j��s�t|j��r�t$| j|j|tj| j �d� n(t!j"| j|j| |
|t|�tj| j �d� |�r�td| ||d� | �rt| j��s�t|j��rt%| j|j|� d�|
d� dS )a
Check that left and right Series are equal.
Parameters
----------
left : Series
right : Series
check_dtype : bool, default True
Whether to check the Series dtype is identical.
check_index_type : bool or {'equiv'}, default 'equiv'
Whether to check the Index class, dtype and inferred_type
are identical.
check_series_type : bool, default True
Whether to check the Series class is identical.
check_less_precise : bool or int, default False
Specify comparison precision. Only used when check_exact is False.
5 digits (False) or 3 digits (True) after decimal points are compared.
If int, then specify the digits to compare.
When comparing two numbers, if the first number has magnitude less
than 1e-5, we compare the two numbers directly and check whether
they are equivalent within the specified precision. Otherwise, we
compare the **ratio** of the second number to the first number and
check whether it is equivalent to 1 within the specified precision.
.. deprecated:: 1.1.0
Use `rtol` and `atol` instead to define relative/absolute
tolerance, respectively. Similar to :func:`math.isclose`.
check_names : bool, default True
Whether to check the Series and Index names attribute.
check_exact : bool, default False
Whether to compare number exactly.
check_datetimelike_compat : bool, default False
Compare datetime-like which is comparable ignoring dtype.
check_categorical : bool, default True
Whether to compare internal Categorical exactly.
check_category_order : bool, default True
Whether to compare category order of internal Categoricals.
.. versionadded:: 1.0.2
check_freq : bool, default True
Whether to check the `freq` attribute on a DatetimeIndex or TimedeltaIndex.
rtol : float, default 1e-5
Relative tolerance. Only used when check_exact is False.
.. versionadded:: 1.1.0
atol : float, default 1e-8
Absolute tolerance. Only used when check_exact is False.
.. versionadded:: 1.1.0
obj : str, default 'Series'
Specify object name being compared, internally used to show appropriate
assertion message.
Tz�The 'check_less_precise' keyword in testing.assert_*_equal is deprecated and will be removed in a future version. You can stop passing 'check_less_precise' to silence this warning.r� )r� )rT z, zSeries length are differentz.index)r� r� r� r� r� r� rT r� zAttributes of )r� rT r z[datetimelike_compat=True] z is not equal to �.)r� r� r� rT r )r� r r� z category)rT r N)&r rI r� r� r� r� r/ r� rq r� r� r� r� rR r* �TimedeltaIndexr� r� r r� r� r! r r� r= r� r" r% r+ r� r r� r
r� r� r r# r� )r� r� r� �check_index_typeZcheck_series_typer� r� r� �check_datetimelike_compatr� r �
check_freqr� r� rT r� r� r� ZlidxZridxr� rM rM rN r� � s� G
r� r) c C s� d}|t k r(tjdtdd� t|� }}t| |t� |rJt| t|��sJt �| j
|j
krzt ||� d�t | j
�� t |j
�� � | r�| j
|�| } }t| j|j||| | |||� d�d� t| j|j||| | |||� d�d� |�rN|j� }| j� }x�ttt|j� �t|j� � ��D ]:}||k�s t �||k�s.t �t|| || ||d � �qW n~x|t| j�D ]n\}}||k�spt �| jd
d
�|f }|jd
d
�|f }t||||| ||
| |
|� d |� d |� d
�||d� �qZW d
S )a
Check that left and right DataFrame are equal.
This function is intended to compare two DataFrames and output any
differences. Is is mostly intended for use in unit tests.
Additional parameters allow varying the strictness of the
equality checks performed.
Parameters
----------
left : DataFrame
First DataFrame to compare.
right : DataFrame
Second DataFrame to compare.
check_dtype : bool, default True
Whether to check the DataFrame dtype is identical.
check_index_type : bool or {'equiv'}, default 'equiv'
Whether to check the Index class, dtype and inferred_type
are identical.
check_column_type : bool or {'equiv'}, default 'equiv'
Whether to check the columns class, dtype and inferred_type
are identical. Is passed as the ``exact`` argument of
:func:`assert_index_equal`.
check_frame_type : bool, default True
Whether to check the DataFrame class is identical.
check_less_precise : bool or int, default False
Specify comparison precision. Only used when check_exact is False.
5 digits (False) or 3 digits (True) after decimal points are compared.
If int, then specify the digits to compare.
When comparing two numbers, if the first number has magnitude less
than 1e-5, we compare the two numbers directly and check whether
they are equivalent within the specified precision. Otherwise, we
compare the **ratio** of the second number to the first number and
check whether it is equivalent to 1 within the specified precision.
.. deprecated:: 1.1.0
Use `rtol` and `atol` instead to define relative/absolute
tolerance, respectively. Similar to :func:`math.isclose`.
check_names : bool, default True
Whether to check that the `names` attribute for both the `index`
and `column` attributes of the DataFrame is identical.
by_blocks : bool, default False
Specify how to compare internal data. If False, compare by columns.
If True, compare by blocks.
check_exact : bool, default False
Whether to compare number exactly.
check_datetimelike_compat : bool, default False
Compare datetime-like which is comparable ignoring dtype.
check_categorical : bool, default True
Whether to compare internal Categorical exactly.
check_like : bool, default False
If True, ignore the order of index & columns.
Note: index labels must match their respective rows
(same as in columns) - same labels must be with the same data.
check_freq : bool, default True
Whether to check the `freq` attribute on a DatetimeIndex or TimedeltaIndex.
rtol : float, default 1e-5
Relative tolerance. Only used when check_exact is False.
.. versionadded:: 1.1.0
atol : float, default 1e-8
Absolute tolerance. Only used when check_exact is False.
.. versionadded:: 1.1.0
obj : str, default 'DataFrame'
Specify object name being compared, internally used to show appropriate
assertion message.
See Also
--------
assert_series_equal : Equivalent method for asserting Series equality.
DataFrame.equals : Check DataFrame equality.
Examples
--------
This example shows comparing two DataFrames that are equal
but with columns of differing dtypes.
>>> from pandas._testing import assert_frame_equal
>>> df1 = pd.DataFrame({'a': [1, 2], 'b': [3, 4]})
>>> df2 = pd.DataFrame({'a': [1, 2], 'b': [3.0, 4.0]})
df1 equals itself.
>>> assert_frame_equal(df1, df1)
df1 differs from df2 as column 'b' is of a different type.
>>> assert_frame_equal(df1, df2)
Traceback (most recent call last):
...
AssertionError: Attributes of DataFrame.iloc[:, 1] (column name="b") are different
Attribute "dtype" are different
[left]: int64
[right]: float64
Ignore differing dtypes in columns with check_dtype.
>>> assert_frame_equal(df1, df2, check_dtype=False)
Tz�The 'check_less_precise' keyword in testing.assert_*_equal is deprecated and will be removed in a future version. You can stop passing 'check_less_precise' to silence this warning.r� )r� z shape mismatchz.index)r� r� r� r� r� r� rT z.columns)r� rT Nz .iloc[:, z] (column name="z"))
r� r& r� r� r' r� r( rT r� r� )r rI r� r� r� r� r) r� r� r� r r� r Z reindex_liker� r� �columnsZ_to_dict_of_blocks�list�set�keysr� � enumerate�ilocr� )r� r� r� r& Zcheck_column_typeZcheck_frame_typer� r� Z by_blocksr� r' r� Z
check_liker( r� r� rT r� ZrblocksZlblocksr� �i�colZlcolZrcolrM rM rN r� � sx y "& r� c K sd d}t | tj�rPt| |f|� t | tjtjf�rL| j|jksLt| j|jf���nt | tj�rlt | |f|� n�t | tj
�r�t | |f|� n�t | t �r�t
| |f|� n�t | t�r�t| |f|� n�t | t�r�t| |f|� n�t | t�r�t| |f|� npt | t��r t| |f|� nTt | tj��r*t| |f|� n6t | t��rT|i k�sDt�| |k�s`t�n tt| ���dS )aC
Wrapper for tm.assert_*_equal to dispatch to the appropriate test function.
Parameters
----------
left, right : Index, Series, DataFrame, ExtensionArray, or np.ndarray
The two items to be compared.
**kwargs
All keyword arguments are passed through to the underlying assert method.
TN) r� rR r+ r� r* r% r� r� r/ r� r) r� r4 r� r5 r r2 r r6 r r3 r# r� r� r r= �NotImplementedErrorr� )r� r� r� r� rM rM rN � assert_equal[ s2
r2 c C s� |t jkrt j| �} n�|t jkr,t j| �} n�|t jkrBt j| �} n�|t jkrft j| �j� } |r�| j} nf|tkrxt| �} nT|t kr�t | �} nB|t
kr�t
| �} n0|t j kr�t j| �} n|t
kr�t
| �} nt|��| S )z�
Helper function to wrap the expected output of a test in a given box_class.
Parameters
----------
expected : np.ndarray, Index, Series
box_cls : {Index, Series, DataFrame}
Returns
-------
subclass of box_cls
)rR r
r+ r/ r) Zto_frame�Tr5 r7 r2 r6 r� r� �to_arrayr1 )�expectedZbox_clsZ transposerM rM rN � box_expected� s,
r6 c C sV t | dd �}t|�r t| �S t|�s,t|�r6tj| �S t|�rHtj| �S t j
| �S d S )Nr� ) r r" r7 r r r2 Z_from_sequencer$ r6 r� r
)rT r� rM rM rN r4 � s
r4 c C s� t | |tjj� t| j|j� t| jtjj j
�s4t �t|jtjj j
�sJt �| j}|j}|j |�spt
dd||� n td| |� td| |� t| j� |j� � dS )z�
Check that the left and right SparseArray are equal.
Parameters
----------
left : SparseArray
right : SparseArray
zSparseArray.indexzindex are not equalr� r� N)r� rR ZarraysZ SparseArrayr Z sp_valuesr� Zsp_index�_libs�sparseZ SparseIndexr� r� r� r� Zto_dense)r� r� Z
left_indexZ right_indexrM rM rN �assert_sp_array_equal� s
r9 c C s, x&| D ] }||kst dt|�� ���qW d S )NzDid not contain item: )r� r )�iterableZdic�krM rM rN �assert_contains_all� s
r< c K s\ xVt | |�D ]H\}}t||f|� dtt|��� dtt|��� d�}||k s t|��q W dS )z�
iter1, iter2: iterables that produce elements
comparable with assert_almost_equal
Checks that the elements are equal, but not
the same object. (Does not check that items
in sequences are also not the same object)
zExpected object z and object z8 to be different objects, but they were the same object.N)ri r� r r� r� )Ziter1Ziter2Z
eql_kwargsZelem1Zelem2r� rM rM rN � assert_copy� s "r= c C s t jd | � S )N)r� �ascii_uppercase)r; rM rM rN �getCols� s r? rX c C s t td| d�|d�S )NrX )r� r� )r� )r+ r� )r; r� rM rM rN �makeStringIndex s r@ c C s t td| d�|d�S )NrX )r� r� )r� )r+ r� )r; r� rM rM rN �makeUnicodeIndex s rA � c K s4 t d|d�}ttjtj| �| |d�fd|i|��S )z' make a length k index or n categories r< )r� r� )r r� )r� r( r' Z
from_codesr� Zarange)r; �nr� r� r� rM rM rN �makeCategoricalIndex s rD c K s* t jdd| d d�}tj|fd|i|��S )z make a length k IntervalIndex r �d rj )�numr� )r� Zlinspacer, Z from_breaks)r; r� r� r� rM rM rN �makeIntervalIndex s rG c C sL | dkrt dg|d�S | dkr.t ddg|d�S t ddgdg| d |d�S )Nrj T)r� r� F)r+ )r; r� rM rM rN �
makeBoolIndex s
rH c C s t tt| ��|d�S )N)r� )r+ r* r� )r; r� rM rM rN � makeIntIndex s rI c C s t dd� t| �D �|d�S )Nc S s g | ] }d| �qS )r� �? l rM )�.0r/ rM rM rN �
<listcomp># s z!makeUIntIndex.<locals>.<listcomp>)r� )r+ r� )r; r� rM rM rN �
makeUIntIndex" s rM c K s t d| dfd|i|��S )Nr rj r� )r. )r; r� r� rM rM rN �makeRangeIndex& s rN c C s: t tjj| ��tjjd� }t|dtjjdd� |d�S )Nrj rX r � )r� )�sortedr� r� Z
random_sampler+ �randint)r; r� r� rM rM rN �makeFloatIndex* s rR �Bc K s0 t ddd�}t|| ||d�}t|fd|i|��S )Ni� rj )�periodsr� r� r� )r r0 r* )r; r� r� r� �dt�drrM rM rN �
makeDateIndex/ s rW �Dc K s t jf d| ||d�|��S )Nz1 day)�startrT r� r� )rR Ztimedelta_range)r; r� r� r� rM rM rN �makeTimedeltaIndex5 s rZ c K s* t ddd�}tjf || d|d�|��}|S )Ni� rj rS )rY rT r� r� )r rR Z period_range)r; r� r� rU rV rM rM rN �makePeriodIndex9 s r[ c K s t jd d|i|��S )
N�foo�barrj r� r� �r\ r] �rj r� �r^ r_ )r` )r- Z from_product)r; r� r� rM rM rN �makeMultiIndex? s ra ZAliceZBobZCharlieZDanZEdithZFrankZGeorgeZHannahZIngridZJerryZKevinZLauraZMichaelZNorbertZOliverZPatriciaZQuinnZRayZSarahZTimZUrsulaZVictorZWendyZXavierZYvonneZZelda�
2000-01-01�
2000-12-31�1Dc C s� t j| ||dd�}t|�}tjj|�}|jt|d�|jd|d�|j |�d d |j |�d d d�}t j
||t |�d�}|j d
|kr�|j
d d � }|S ) a�
Make a DataFrame with a DatetimeIndex
Parameters
----------
start : str or Timestamp, default "2000-01-01"
The start of the index. Passed to date_range with `freq`.
end : str or Timestamp, default "2000-12-31"
The end of the index. Passed to date_range with `freq`.
freq : str or Freq
The frequency to use for the DatetimeIndex
seed : int, optional
The random state seed.
* name : object dtype with string names
* id : int dtype with
* x, y : float dtype
Examples
--------
>>> _make_timeseries()
id name x y
timestamp
2000-01-01 982 Frank 0.031261 0.986727
2000-01-02 1025 Edith -0.086358 -0.032920
2000-01-03 982 Edith 0.473177 0.298654
2000-01-04 1009 Sarah 0.534344 -0.750377
2000-01-05 963 Zelda -0.271573 0.054424
... ... ... ... ...
2000-12-27 980 Ingrid -0.132333 -0.422195
2000-12-28 972 Frank -0.376007 -0.298687
2000-12-29 1009 Ursula -0.865047 -0.503133
2000-12-30 1000 Hannah -0.063757 -0.507336
2000-12-31 972 Tim -0.869120 0.531685
� timestamp)rY �endr� r� )r� i� r� rj )r� �idr� �y)r� r) N���ri )rR Z
date_rangerq r� r� � RandomStater� �_namesZpoissonr r) rP r� r. ) rY rf r� �seedr� rC �stater) �dfrM rM rN �_make_timeseriesa s $ ro c c s* t ttttttg} x| D ]
}|V qW d S )N)rW r[ rZ rN rG rD ra )�make_index_funcs�make_index_funcrM rM rN � index_subclass_makers_generator� s
rr c c s( t ttg}x|D ]}|| d�V qW dS )z�
Generator which can be iterated over to get instances of all the classes
which represent time-series.
Parameters
----------
k: length of each of the index instances
)r; N)rW r[ rZ )r; rp rq rM rM rN � all_timeseries_index_generator� s
rs c C s t t�}ttt�|| d�S )N)r� r� )r@ �_Nr/ r )r� r� rM rM rN �makeFloatSeries� s ru c C s t t�}ttt�|| d�S )N)r� r� )r@ rt r/ r )r� r� rM rM rN �makeStringSeries� s rv c C s* t t�}t|td�}t t�}t||| d�S )N)r� )r� r� )r@ rt r+ rA r/ )r� r� r� rM rM rN �makeObjectSeries� s rw c s t t�� � fdd�tt�D �S )Nc s i | ]}t tt�� d �|�qS ))r� )r/ r rt )rK �c)r� rM rN �
<dictcomp>� s z!getSeriesData.<locals>.<dictcomp>)r@ rt r? �_KrM rM )r� rN �
getSeriesData� s r{ c C s& | d kr t } tt| �t| |d�|d�S )N)r� )r� r� )rt r/ r rW )�nperr� r� rM rM rN �makeTimeSeries� s r} c C s" | d kr t } tt| �t| �|d�S )N)r� r� )rt r/ r r[ )r| r� rM rM rN �makePeriodSeries� s r~ c s � �fdd�t t�D �S )Nc s i | ]}t �� �|�qS rM )r} )rK rx )r� r| rM rN ry � s z%getTimeSeriesData.<locals>.<dictcomp>)r? rz )r| r� rM )r� r| rN �getTimeSeriesData� s r c s � fdd�t t�D �S )Nc s i | ]}t � �|�qS rM )r~ )rK rx )r| rM rN ry � s z!getPeriodData.<locals>.<dictcomp>)r? rz )r| rM )r| rN �
getPeriodData� s r� c C s t | |�}t|�S )N)r r) )r| r� r� rM rM rN �makeTimeDataFrame� s
r� c C s t � } t| �S )N)r{ r) )r� rM rM rN �
makeDataFrame� s r� c C sN t dddddg�} dddd d
gdddddgd d d
ddgtddd�d�}| |fS )N�a�brx �dr� g g �?g @g @g @Zfoo1Zfoo2Zfoo3Zfoo4Zfoo5z1/1/2009r� )rT )�ArS �CrX )r+ r0 )r� r� rM rM rN �getMixedTypeDict� s r� c C s t t� d �S )Nrj )r) r� rM rM rM rN �makeMixedDataFrame� s r� c C s t | �}t|�S )N)r� r) )r| r� rM rM rN �makePeriodFrame� s r� �#c sX |dkrdg| }t |�r&t|�|ks*t�|dksR|dksR|dksRt|�|ksRt�|dksn|dkrj|dksnt�|dkr�� fd d
�t|�D �}|dkr�d}t|t�r�|dkr�|g}tttt t
t t t
d�j|�}|r�|| �}|r�|d |_|S |dk �rtt|�� d���t|�|k �r,|jdg|t|� � t|�|k�s>t�tdd� |D ���sVt�g }x�t|�D ]z} dd� }
| || d } t� } x2t| �D ]&}
� � d| � d|
� �}|| | |< �q�W t| j� |
d�d| � }|j|� �qdW tt|� �}| dk�rt|d |d d�}nF|dk�rF|dk�r&dn|d }tdd� |D �|d�}ntj ||d�}|S ) a�
Create an index/multindex with given dimensions, levels, names, etc'
nentries - number of entries in index
nlevels - number of levels (> 1 produces multindex)
prefix - a string prefix for labels
names - (Optional), bool or list of strings. if True will use default
names, if false will use no names, if a list is given, the name of
each level in the index will be taken from the list.
ndupe_l - (Optional), list of ints, the number of rows for which the
label will repeated at the corresponding level, you can specify just
the first few, the rest will use the default ndupe_l of 1.
len(ndupe_l) <= nlevels.
idx_type - "i"/"f"/"s"/"u"/"dt"/"p"/"td".
If idx_type is not None, `idx_nlevels` must be 1.
"i"/"f" creates an integer/float index,
"s"/"u" creates a string/unicode index
"dt" create a datetime index.
"td" create a datetime index.
if unspecified, string labels will be generated.
Nrj FTr/ rv �s�urU r� �tdc s g | ]}� t |� �qS rM )r= )rK r/ )�prefixrM rN rL s z#makeCustomIndex.<locals>.<listcomp>)r/ rv r� r� rU r� r� r zI is not a legal value for `idx_type`, use 'i'/'f'/'s'/'u'/'dt'/'p'/'td'.c s s | ]}|d kV qdS )r NrM )rK r� rM rM rN � <genexpr>B s z"makeCustomIndex.<locals>.<genexpr>c S s* dd l }|jdd| �jd�}dd� |D �S )Nr z[^\d_]_?r� �_c S s g | ] }t |��qS rM )r� )rK rF rM rM rN rL K s z4makeCustomIndex.<locals>.keyfunc.<locals>.<listcomp>)�re�sub�split)r� r� Z
numeric_tuplerM rM rN �keyfuncG s z makeCustomIndex.<locals>.keyfuncZ_lZ_g)�key)r� c s s | ]}|d V qdS )r NrM )rK r� rM rM rN r� _ s )r� )r/ rv r� r� rU r� r� ) r# rq r� r� r� r= r� rI rR r@ rA rW rZ r[ rH r� rs r �extendr r rP �elements�appendr* ri r+ r- � from_tuples)Znentriesr� r� r� �ndupe_l�idx_typeZidx_func�idxZtuplesr/ r� Z
div_factorZcnt�j�labelr r� r� rM )r� rN �makeCustomIndex� sf
(
r� c s� |dks t �|dkst �| dks4| dkr0|d
ks4t �|
dksP|
dkrL|d
ksPt �t�|d |||
d �} t| |d
||| d �}
� dkr�dd� � � �fdd�t| �D �}t||
| | d�S )a�
Create a DataFrame using supplied parameters.
Parameters
----------
nrows, ncols - number of data rows/cols
c_idx_names, idx_names - False/True/list of strings, yields No names ,
default names or uses the provided names for the levels of the
corresponding index. You can provide a single string when
c_idx_nlevels ==1.
c_idx_nlevels - number of levels in columns index. > 1 will yield MultiIndex
r_idx_nlevels - number of levels in rows index. > 1 will yield MultiIndex
data_gen_f - a function f(row,col) which return the data value
at that position, the default generator used yields values of the form
"RxCy" based on position.
c_ndupe_l, r_ndupe_l - list of integers, determines the number
of duplicates for each label at a given level of the corresponding
index. The default `None` value produces a multiplicity of 1 across
all levels, i.e. a unique index. Will accept a partial list of length
N < idx_nlevels, for just the first N levels. If ndupe doesn't divide
nrows/ncol, the last label might have lower multiplicity.
dtype - passed to the DataFrame constructor as is, in case you wish to
have more control in conjunction with a custom `data_gen_f`
r_idx_type, c_idx_type - "i"/"f"/"s"/"u"/"dt"/"td".
If idx_type is not None, `idx_nlevels` must be 1.
"i"/"f" creates an integer/float index,
"s"/"u" creates a string/unicode index
"dt" create a datetime index.
"td" create a timedelta index.
if unspecified, string labels will be generated.
Examples
--------
# 5 row, 3 columns, default names on both, single index on both axis
>> makeCustomDataframe(5,3)
# make the data a random int between 1 and 100
>> mkdf(5,3,data_gen_f=lambda r,c:randint(1,100))
# 2-level multiindex on rows with each label duplicated
# twice on first level, default names on both axis, single
# index on both axis
>> a=makeCustomDataframe(5,3,r_idx_nlevels=2,r_ndupe_l=[2])
# DatetimeIndex on row, index with unicode labels on columns
# no names on either axis
>> a=makeCustomDataframe(5,3,c_idx_names=False,r_idx_names=False,
r_idx_type="dt",c_idx_type="u")
# 4-level multindex on rows with names provided, 2-level multindex
# on columns with default labels and default names.
>> a=makeCustomDataframe(5,3,r_idx_nlevels=4,
r_idx_names=["FEE","FI","FO","FAM"],
c_idx_nlevels=2)
>> a=mkdf(5,3,r_idx_nlevels=2,c_idx_nlevels=4)
r Nr/ rv r� r� rU r� r� rj r� )r� r� r� r� r� �Rc S s d| � d|� �S )Nr� r� rM )r� rx rM rM rN �<lambda>� s z%makeCustomDataframe.<locals>.<lambda>c s$ g | ] � �� fd d�t ��D ��qS )c s g | ]}� �|��qS rM rM )rK rx )�
data_gen_fr� rM rN rL � s z2makeCustomDataframe.<locals>.<listcomp>.<listcomp>)r� )rK )r� �ncols)r� rN rL � s z'makeCustomDataframe.<locals>.<listcomp>)r� )r/ rv r� r� rU r� r� )r/ rv r� r� rU r� r� )r� r� r� r) )�nrowsr� Z c_idx_namesZ r_idx_namesZ
c_idx_nlevelsZ
r_idx_nlevelsr� Z c_ndupe_lZ r_ndupe_lr� Z
c_idx_typeZ
r_idx_typer) r� r� rM )r� r� rN �makeCustomDataframee s0 H r� c s� |d krt j}n t jj|�}tt jd| � � ���d}d}t�| |� �}� ��fdd�}|||�}x |j�k r�|d9 }|||�}qlW t j|d � �jt�} || � jt�}
|
j � | j � fS )Nrj r� gR� ��Q�?c s. | j t|��}tjtj|� � ��d �� S )N)r r� r� r� �floor)�rngZ _extra_size�ind)r� r� r� rM rN �_gen_unique_rand� s z-_create_missing_idx.<locals>._gen_unique_randg�������?g �?)
r� r� rj r� r� �minr� r� r� �tolist) r� r� �density� random_stateZmin_rowsZfacZ
extra_sizer� r� r� r/ rM )r� r� r� rN �_create_missing_idx� s
r� ��������?c C s0 t � }t|j| |d��\}}tj|j||f< |S )N)r� r� )r� r� r r� �nanr� )r� r� rn r/ r� rM rM rN �makeMissingDataframe� s r� c s t � �� fdd��}|S )aB
allows a decorator to take optional positional and keyword arguments.
Assumes that taking a single, callable, positional argument means that
it is decorating a function, i.e. something like this::
@my_decorator
def function(): pass
Calls decorator with decorator(f, *args, **kwargs)
c sN � ��fdd�}� o,t � �dko,t� d �}|rF� d }g � ||�S |S d S )Nc s �| f� ���S )NrM )rv )r� � decoratorr� rM rN �dec s z+optional_args.<locals>.wrapper.<locals>.decrj r )rq �callable)r� r� r� Z
is_decoratingrv )r� )r� r� rN �wrapper� s z optional_args.<locals>.wrapper)r )r� r� rM )r� rN �
optional_args� s
r� � timed out�
Server Hangup�#HTTP Error 503: Service Unavailable�502: Proxy Error� HTTP Error 502: internal error�HTTP Error 502�HTTP Error 503�HTTP Error 403�HTTP Error 400�$Temporary failure in name resolution�Name or service not known�Connection refused�certificate verify�e �o �n �h �6 �< c C s dd l } t| jjtfS )Nr )� http.client�IOError�client�
HTTPException� TimeoutError)�httprM rM rN �_get_default_network_errors4 s r� c C sD |dkrt � }yt| �� W dQ R X W n |k
r: dS X dS dS )a@
Try to connect to the given url. True if succeeds, False if IOError
raised
Parameters
----------
url : basestring
The URL to try to connect to
Returns
-------
connectable : bool
Return True if no IOError (unable to connect) or URLError (bad url) was
raised
NFT)r� r9 )�url�
error_classesrM rM rN � can_connect; s
r� zhttp://www.google.comc sF ddl m� �dkrt� �d�_t��� �������fdd��}|S )aZ
Label a test as requiring network connection and, if an error is
encountered, only raise if it does not find a network connection.
In comparison to ``network``, this assumes an added contract to your test:
you must assert that, under normal conditions, your test will ONLY fail if
it does not have network connectivity.
You can call this in 3 ways: as a standard decorator, with keyword
arguments, or with a positional argument that is the url to check.
Parameters
----------
t : callable
The test requiring network connectivity.
url : path
The url to test via ``pandas.io.common.urlopen`` to check
for connectivity. Defaults to 'http://www.google.com'.
raise_on_error : bool
If True, never catches errors.
check_before_test : bool
If True, checks connectivity before running the test case.
error_classes : tuple or Exception
error classes to ignore. If not in ``error_classes``, raises the error.
defaults to IOError. Be careful about changing the error classes here.
skip_errnos : iterable of int
Any exception that has .errno or .reason.erno set to one
of these values will be skipped with an appropriate
message.
_skip_on_messages: iterable of string
any exception e for which one of the strings is
a substring of str(e) will be skipped with an appropriate
message. Intended to suppress errors where an errno isn't available.
Notes
-----
* ``raise_on_error`` supersedes ``check_before_test``
Returns
-------
t : callable
The decorated test ``t``, with checks for connectivity errors.
Example
-------
Tests decorated with @network will fail if it's possible to make a network
connection to another URL (defaults to google.com)::
>>> from pandas._testing import network
>>> from pandas.io.common import urlopen
>>> @network
... def test_network():
... with urlopen("rabbit://bonanza.com"):
... pass
Traceback
...
URLError: <urlopen error unknown url type: rabit>
You can specify alternative URLs::
>>> @network("http://www.yahoo.com")
... def test_something_with_yahoo():
... raise IOError("Failure Message")
>>> test_something_with_yahoo()
Traceback (most recent call last):
...
IOError: Failure Message
If you set check_before_test, it will check the url first and not run the
test on failure::
>>> @network("failing://url.blaher", check_before_test=True)
... def test_something():
... print("I ran!")
... raise ValueError("Failure")
>>> test_something()
Traceback (most recent call last):
...
Errors not related to networking will always be raised.
r )r� NTc s� �r� rt ���s�� y
�| |�S tk
r� } z�t|dd �}| r`t|d�r`t|jdd �}|�krv�d|� �� t|�� t� fdd��D ��r��d|� �� t|��s�� �s�t ���r�� n�d|� �� W Y d d }~X nX d S )N�errno�reasonz+Skipping test due to known errno and error c 3 s | ]}|j � � j � kV qd S )N)�lower)rK �m)�e_strrM rN r� � s z+network.<locals>.wrapper.<locals>.<genexpr>z;Skipping test because exception message is known and error z4Skipping test due to lack of connectivity and error )r� � Exceptionr �hasattrr� r= �anyr� )r� r� �errr� )�_skip_on_messages�check_before_testr� �raise_on_errorr� � skip_errnos�tr� )r� rN r� � s(
znetwork.<locals>.wrapper)r] r� r� �networkr )r� r� r� r� r� r� r� r� rM )r� r� r� r� r� r� r� r� rN r� W s \ " r� rE c
c s d}t jdd��� }d}t j|� |V g }x�|D ]�}| r�t|j| �r�d}|r�t|jttf�r�ddlm} m }
| |
� d d �} d|j
� d| j
� d |j � �} |j
| j
ks�t | ��q2|j
|jj|j |j
|jf� q2W | r�d
t| j�� �} |s�t | ��|o�|�rt d t|�� ���W d Q R X d S )
ar
Context manager for running code expected to either raise a specific
warning, or not raise any warnings. Verifies that the code raises the
expected warning, and that it does not raise any other unexpected
warnings. It is basically a wrapper around ``warnings.catch_warnings``.
Parameters
----------
expected_warning : {Warning, False, None}, default Warning
The type of Exception raised. ``exception.Warning`` is the base
class for all warnings. To check that no warning is returned,
specify ``False`` or ``None``.
filter_level : str or None, default "always"
Specifies whether warnings are ignored, displayed, or turned
into errors.
Valid values are:
* "error" - turns matching warnings into exceptions
* "ignore" - discard the warning
* "always" - always emit a warning
* "default" - print the warning the first time it is generated
from each location
* "module" - print the warning the first time it is generated
from each module
* "once" - print the warning the first time it is generated
check_stacklevel : bool, default True
If True, displays the line that called the function containing
the warning to show were the function is called. Otherwise, the
line that implements the function is displayed.
raise_on_extra_warnings : bool, default True
Whether extra warnings not of the type `expected_warning` should
cause the test to fail.
Examples
--------
>>> import warnings
>>> with assert_produces_warning():
... warnings.warn(UserWarning())
...
>>> with assert_produces_warning(False):
... warnings.warn(RuntimeWarning())
...
Traceback (most recent call last):
...
AssertionError: Caused unexpected warning(s): ['RuntimeWarning'].
>>> with assert_produces_warning(UserWarning):
... warnings.warn(RuntimeWarning())
Traceback (most recent call last):
...
AssertionError: Did not see expected warning of class 'UserWarning'.
..warn:: This is *not* thread-safe.
T)�recordFr )� getframeinfo�stackr� zGWarning not set with correct stacklevel. File where warning is raised: z != z. Warning message: z&Did not see expected warning of class z Caused unexpected warning(s): N)rI �catch_warningsrJ �
issubclass�categoryr� �DeprecationWarning�inspectr� r� r� r r� r� r� �linenor )
Zexpected_warningZ filter_levelZcheck_stacklevelZraise_on_extra_warningsr� ry Z saw_warningZextra_warningsZactual_warningr� r� Zcallerr� rM rM rN �assert_produces_warning� s6 =
r� c @ s( e Zd ZdZdd� Zdd� Zdd� ZdS ) �
RNGContexta,
Context manager to set the numpy random number generator speed. Returns
to the original value upon exiting the context manager.
Parameters
----------
seed : int
Seed for numpy.random.seed
Examples
--------
with RNGContext(42):
np.random.randn()
c C s
|| _ d S )N)rl )�selfrl rM rM rN �__init__[
s zRNGContext.__init__c C s t jj� | _t jj| j� d S )N)r� r� Z get_state� start_staterl )r� rM rM rN � __enter__^
s zRNGContext.__enter__c C s t jj| j� d S )N)r� r� Z set_stater� )r� �exc_type� exc_value� tracebackrM rM rN �__exit__c
s zRNGContext.__exit__N)r� �
__module__� __qualname__�__doc__r� r� r� rM rM rM rN r� K
s r� c k sD ddl }dddh}| |kr"td��|j| f|� dV |j| � dS )au
Context manager to temporarily register a CSV dialect for parsing CSV.
Parameters
----------
name : str
The name of the dialect.
kwargs : mapping
The parameters for the dialect.
Raises
------
ValueError : the name of the dialect conflicts with a builtin one.
See Also
--------
csv : Python's CSV library.
r N�excelz excel-tab�unixz Cannot override builtin dialect.)�csvrs �register_dialect�unregister_dialect)r� r� r� Z_BUILTIN_DIALECTSrM rM rN �with_csv_dialecth
s
r� c c sP ddl m} |d kr|j}|j}|j}|j| � ||_d V ||_|j|� d S )Nr )� expressions)Zpandas.core.computationr Z
_MIN_ELEMENTSZ _USE_NUMEXPRZset_use_numexpr)ZuseZ min_elements�exprZolduseZoldminrM rM rN � use_numexpr�
s
r r� c sF �dks t ��dk � � r(t���ks(t �ddl�� ���fdd�}|S )a
Decorator to run the same function multiple times in parallel.
Parameters
----------
num_threads : int, optional
The number of times the function is run in parallel.
kwargs_list : list of dicts, optional
The list of kwargs to update original
function kwargs on different threads.
Notes
-----
This decorator does not pass the return value of the decorated function.
Original from scikit-image:
https://github.com/scikit-image/scikit-image/pull/1519
r Nc s t � �� ����fdd��}|S )Nc s� �r� �fdd�}n � fdd�}g }x2t ��D ]&}||�}�j�| |d�}|j|� q.W x|D ] }|j� q^W x|D ] }|j� qtW d S )Nc s t � f�| �S )N)r� )r/ )r� � kwargs_listrM rN r� �
s z?test_parallel.<locals>.wrapper.<locals>.inner.<locals>.<lambda>c s � S )NrM )r/ )r� rM rN r� �
s )�targetr� r� )r� �Threadr� rY r� )r� r� Z
update_kwargs�threadsr/ Zupdated_kwargs�thread)�func�has_kwargs_listr � num_threads� threading)r� rN �inner�
s
z-test_parallel.<locals>.wrapper.<locals>.inner)r )r r )r r r
r )r rN r� �
s z test_parallel.<locals>.wrapper)r� rq r )r
r r� rM )r r r
r rN �
test_parallel�
s r
c @ s, e Zd ZddgZedd� �Zedd� �ZdS )�SubclassedSeries�testattrr� c C s t S )N)r )r� rM rM rN � _constructor�
s z SubclassedSeries._constructorc C s t S )N)�SubclassedDataFrame)r� rM rM rN �_constructor_expanddim�
s z'SubclassedSeries._constructor_expanddimN)r� r� r� � _metadata�propertyr r rM rM rM rN r �
s r c @ s* e Zd ZdgZedd� �Zedd� �ZdS )r r c C s t S )N)r )r� rM rM rN r �
s z SubclassedDataFrame._constructorc C s t S )N)r )r� rM rM rN �_constructor_sliced�
s z'SubclassedDataFrame._constructor_slicedN)r� r� r� r r r r rM rM rM rN r �
s r c @ s e Zd Zedd� �ZdS )�SubclassedCategoricalc C s t S )N)r )r� rM rM rN r �
s z"SubclassedCategorical._constructorN)r� r� r� r r rM rM rM rN r �
s r )r c # sL ddl � ddl�� �fdd�}� jjd�}|| � z
dV W d||� X dS )a�
Context manager for temporarily setting a timezone.
Parameters
----------
tz : str
A string representing a valid timezone.
Examples
--------
>>> from datetime import datetime
>>> from dateutil.tz import tzlocal
>>> tzlocal().tzname(datetime.now())
'IST'
>>> with set_timezone('US/Eastern'):
... tzlocal().tzname(datetime.now())
...
'EDT'
r Nc sB | d kr,y � j d= W q> tk
r( Y q>X n| � j d< �j� d S )N�TZ)rG �KeyError�tzset)r )rF �timerM rN �setTZ s
zset_timezone.<locals>.setTZr )rF r rG rH )r r Zorig_tzrM )rF r rN � set_timezone�
s
r c s" �r�fdd�}n � fdd�}|S )a`
Create a function for calling on an array.
Parameters
----------
alternative : function
The function to be called on the array with no NaNs.
Only used when 'skipna_alternative' is None.
skipna_alternative : function
The function to be called on the original array
Returns
-------
function
c s
� | j �S )N)r� )r� )�skipna_alternativerM rN �skipna_wrapper$ s z,_make_skipna_wrapper.<locals>.skipna_wrapperc s" | j � }t|�dkrtjS � |�S )Nr )Zdropnarq r� r� )r� Znona)� alternativerM rN r ) s rM )r r r rM )r r rN �_make_skipna_wrapper s r )� rows_listc C s t j}|j| �| }|S )aW
Convert list of CSV rows to single CSV-formatted string for current OS.
This method is used for creating expected value of to_csv() method.
Parameters
----------
rows_list : List[str]
Each element represents the row of csv.
Returns
-------
str
Expected output of to_csv() in current OS.
)rF �linesepr� )r! �sepr5 rM rM rN � convert_rows_list_to_csv_str2 s r$ )�expected_exceptionrV c C s ddl }|j| dd�S )a$
Helper function to mark pytest.raises that have an external error message.
Parameters
----------
expected_exception : Exception
Expected error to raise.
Returns
-------
Callable
Regular `pytest.raises` function with `match` equal to `None`.
r N)�match)r] Zraises)r% r] rM rM rN �external_error_raisedG s r' c sD g }x:|D ]2\�� |j ��� f� |� ��fdd�tD �7 }q
W |S )a�
Combine frame, functions from SelectionMixin._cython_table
keys and expected result.
Parameters
----------
ndframe : DataFrame or Series
func_names_and_expected : Sequence of two items
The first item is a name of a NDFrame method ('sum', 'prod') etc.
The second item is the expected return value.
Returns
-------
list
List of three items (DataFrame, function, expected result)
c s" g | ]\}}|�kr�|� f�qS rM rM )rK r r� )r5 � func_name�ndframerM rN rL r s z+get_cython_table_params.<locals>.<listcomp>)r� � cython_table)r) Zfunc_names_and_expected�resultsrM )r5 r( r) rN �get_cython_table_params] s r, )�op_namerV c sP | j d�}ytt|�}W n2 tk
rJ tt|dd� �� � fdd�}Y nX |S )z�
The operator function for a given op name.
Parameters
----------
op_name : string
The op name, in form of "add" or "__add__".
Returns
-------
function
A function performing the operation.
r� rj Nc s
� || �S )NrM )r� rh )�roprM rN r� � s z"get_op_from_name.<locals>.<lambda>)�stripr �operator�AttributeError)r- Z short_opname�oprM )r. rN �get_op_from_namey s
r3 )N)N)N)rx )Ti� )r� )r� )N)NF)Tr� )r� )TTr' )r� r4 )r5 )r2 )r6 )NN)FTNNr N)T)rX N)rX N)rX rB N)rX N)rX N)rX N)rX N)rX N)rX N)rX rS N)rX rX N)rX N)rX N)rb rc rd N)rX )N)N)N)NrS N)NN)NrS )N)NrS )N)r� FNN)
TTrj rj NNNNNN)N)r� N)
r� r� r� r� r� r� r� r� r� r� r� r� r� )r� r� r� r� r� r� )N)N)r� N)N)�rg � collectionsr �
contextlibr r � functoolsr rf r0 rF �shutilr r� r� Ztypingr r r r
r r r
r rI rn Znumpyr� Z numpy.randomr r Zpandas._config.localizationr r r Zpandas._libs.libr Zpandas._libs.testingr7 Ztestingr� Zpandas._typingr r r Z
pandas.compatr r Zpandas.core.dtypes.commonr r r r r r r r! r" r# r$ r% Zpandas.core.dtypes.missingr& ZpandasrR r' r( r) r* r+ r, r- r. r/ r0 Zpandas.core.algorithmsr1 Zpandas.core.arraysr2 r3 r4 r5 r6 r7 Z pandas.core.arrays.datetimeliker8 Zpandas.io.commonr9 Zpandas.io.formats.printingr: rm rt rz Z _RAISE_NETWORK_ERROR_DEFAULTZUNSIGNED_INT_DTYPESZUNSIGNED_EA_INT_DTYPESr� ZSIGNED_INT_DTYPESZSIGNED_EA_INT_DTYPESZALL_INT_DTYPESZALL_EA_INT_DTYPES�floatZ FLOAT_DTYPES�complexZCOMPLEX_DTYPESr= Z
STRING_DTYPESZDATETIME64_DTYPESZTIMEDELTA64_DTYPESr? Z BOOL_DTYPESr@ Z BYTES_DTYPESrA Z
OBJECT_DTYPESZALL_REAL_DTYPESZALL_NUMPY_DTYPESr� �ResourceWarningrK rO rQ rS r\ rb rd rw r� r� r� r� r� r� r
r* �
ascii_letters�digitsr� r� r� �map�chrr� r� r� r� r� rY rt rZ r� r� r� r� r� r� r r r� r� r r r r� r r# r� r� r2 r6 r4 r9 r< r= r? r@ rA rD rG rH rI rM rN rR rW rZ r[ ra rk ro rr rs ru rv rw r{ r} r~ r r� r� r� r� r� r� r� r� r� r� r� Z_network_error_messagesZ_network_errno_valsr� r� r� Zwith_connectivity_check�Warningr� r� r� r r
r r r r r r$ r� r' �corer ZSelectionMixinZ
_cython_table�itemsr* r, r3 rM rM rM rN �<module> s ( 8 0 + +,(g
>
( - . ! K] ? @( *% 3
k _ ! |h 1 * |