U
    sVc                  	   @   s  d dl m Z mZ d dlmZ d dlZd dlmZ d dlmZ d dl	m
Z
 d dlZd dlZd dlm  mZ d dlZd dlmZmZmZmZ d dlmZ dd	d
ddgZejdedgdejdedejdgdejdedgdejdeddejdeddgZ e!e"dddZ#dd Z$ej%dd e D e!ddd Z&ej%dd Z'ej%dd  Z(G d!d" d"Z)G d#d$ d$Z*dS )%    )datetimetime)partialNPath)URLError)
BadZipFile)	DataFrameIndex
MultiIndexSeries.xls.xlsx.xlsm.xlsb.odsxlrdmarksopenpyxlzignore:.*html argumentpyxlsbodf)read_extreturnc                 C   s   | j d } | dkr|dkrdS | dkr2|dkr2dS |dkrF| dkrFdS | dkrZ|dkrZdS |dkrn| dkrndS | d	kr|dkrdS d
S )zm
    Filter out invalid (engine, ext) pairs instead of skipping, as that
    produces 500+ pytest.skips.
    r   r   r   Fr   r   r   r   r   T)values)enginer    r   F/tmp/pip-unpacked-wheel-xj8nt62q/pandas/tests/io/excel/test_readers.py_is_valid_engine_ext_pair6   s    
r   c                 C   s    | j |f }tj|| jd}|S )z
    engine gives us a pytest.param object with some marks, read_ext is just
    a string.  We need to generate a new pytest.param inheriting the marks.
    r   )r   pytestparamr   )r   r   r   	new_paramr   r   r   _transfer_marksK   s    r"   c                 C   s*   g | ]"}t D ]}t||rt||qqS r   )read_ext_paramsr   r"   ).0Zengextr   r   r   
<listcomp>V   s
    
r&   )paramsidsc                 C   s   | j S )zS
    Fixture for Excel reader engine and read_ext, only including valid pairs.
    )r    )requestr   r   r   engine_and_read_extU   s    r*   c                 C   s   | \}}|S Nr   r*   r   r   r   r   r   r   e   s    r   c                 C   s   | \}}|S r+   r   r,   r   r   r   r   k   s    r   c                   @   s  e Zd Zejdddd Zdd Zdd Zd	d
 Zdd Z	ej
ddddgdddgdddgdddgdddgdddggdd Zej
dddgddggdd Zdd Zdd Zdd Zdd Zdd  Zej
d!d"d#gd$d% Zd&d' Zd(d) Zd*d+ Zd,d- Zd.d/ Zd0d1 Zd2d3 Zd4d5 Zej
d6d"edd#dd7gd8d9d:d;gdd#dd7gd<d=ejd>gd?fd@dAeed?ee dd#dd7gd@dBe d8d9d:d;gdAdBdCdDdEdFgdGdHejdIgd?fgdJdK Z!ej
dLi dGfdMdNidfgdOdP Z"dQdR Z#ej
dSdTedUdVgifdWedXgdYfgdZd[ Z$d\d] Z%d^d_ Z&d`da Z'dbdc Z(ddde Z)ej
*dfdgdh Z+didj Z,dkdl Z-dmdn Z.ej
dodddgddgdpdqdpgdpdqggdrds Z/dtdu Z0dvdw Z1ej
j2e3j2dxddydzd{ Z4e5j6ej
j7d|d} Z8ej
j7d~d Z9ej
j:dd Z;dd Z<e5=de5j>dd Z?e5j>dd Z@dd ZAdd ZBej
ddejdddgfdejgd7 fgdd ZCdd ZDdd ZEdd ZFdd ZGdd ZHdd ZIdd ZJdd ZKej
dddddgdd"fddd"ddgd"fddddgddgd"fddddgdd"fddd"d"dd#gfddd"d"dd fgdd ZLdd ZMdd ZNdd ZOdd ZPdd ZQdd ZRdd ZSdd ZTdd ZUd"S )TestReadersTZautousec                 C   s2   t tj|d}||ddd |td| dS )zG
        Change directory and set engine for read_excel calls.
        r   iodataexcel
read_excelN)r   pdr3   chdirsetattrselfr   datapathmonkeypatchfuncr   r   r   cd_and_set_enginer   s    zTestReaders.cd_and_set_enginec           	   	   C   sz   dd }| tjd| dddddd}td	| d
}t|}W 5 Q R X |d k	rZ|}n||dd   }||ksvtd S )Nc                 _   s   | j S r+   r/   )r8   argskwargsr   r   r   parser}   s    z,TestReaders.test_engine_used.<locals>.parserparser   r   r   r   xlsxZxlsmZxlsbxlsZodstest1rb   )r6   r4   	ExcelFileopenr3   AssertionError)	r8   r   r   r:   r?   expected_defaultsfresultexpectedr   r   r   test_engine_used{   s    zTestReaders.test_engine_usedc              	   C   sl   d}t jt|d tjd| dddd W 5 Q R X t jt|d  tjd| dd	gddd
 W 5 Q R X d S )Nz Passing an integer for `usecols`matchrD   Sheet1r      
sheet_name	index_colusecolsSheet2rF   rT   skiprowsrU   rV   r   raises
ValueErrorr4   r3   r8   r   msgr   r   r   test_usecols_int   s        zTestReaders.test_usecols_intc                 C   s   |dkr|j tjjdd |jddgd}tjd| dd	d	d
dgd}tjd| ddgd	d	d
dgd}tj	||dd tj	||dd d S )Nr   3Sheets containing datetimes not supported by pyxlsbreasonBCcolumnsrD   rQ   r      rR   rS   rW   rF   rX   FZcheck_names
node
add_markerr   markxfailreindexr4   r3   tmassert_frame_equal)r8   r)   r   df_refdf1df2r   r   r   test_usecols_list   s,       	zTestReaders.test_usecols_listc                 C   sD  |dkr|j tjjdd |jdddgd}tjd| d	d
dd}tjd| ddgd
dd}tj	||dd tj	||dd |jddgd}tjd| d	d
dd}tjd| ddgd
dd}tj	||dd tj	||dd |jddgd}tjd| d	d
dd}tjd| ddgd
dd}tj	||dd tj	||dd d S )Nr   r`   ra   Arc   rd   re   rD   rQ   r   zA:DrS   rW   rF   rX   Frh   zA,C,DzA,C:Dri   r8   r)   r   rq   rr   rs   df3r   r   r   test_usecols_str   sl       	      zTestReaders.test_usecols_strrV   r   rF   rR   c                 C   sT   |dkr|j tjjdd |ddg }tjd| dd|d	}tj||d
d d S )Nr   r`   ra   ru   rd   rD   rQ   r   rS   Frh   	rj   rk   r   rl   rm   r4   r3   ro   rp   )r8   r)   r   rV   rq   rM   rL   r   r   r   .test_usecols_diff_positional_int_columns_order   s       z:TestReaders.test_usecols_diff_positional_int_columns_orderrc   Dc                 C   sB   |ddg }t t||_tjd| d|d}tj||dd d S )Nrc   r{   rD   rQ   rT   rV   Frh   )rangelenindexr4   r3   ro   rp   )r8   r   rV   rq   rM   rL   r   r   r   .test_usecols_diff_positional_str_columns_order  s    z:TestReaders.test_usecols_diff_positional_str_columns_orderc                 C   sJ   |dkr|j tjjdd |}tjd| ddd}tj||dd	 d S )
Nr   r`   ra   rD   rQ   r   rT   rU   Frh   ry   r8   r)   r   rq   rM   rL   r   r   r   test_read_excel_without_slicing  s    z+TestReaders.test_read_excel_without_slicingc                 C   sT   |dkr|j tjjdd |ddg }tjd| ddd	d
}tj||dd d S )Nr   r`   ra   rd   r{   rD   rQ   r   zA,D:ErS   Frh   ry   r   r   r   r   test_usecols_excel_range_str  s       z(TestReaders.test_usecols_excel_range_strc              	   C   s6   d}t jt|d tjd| ddd W 5 Q R X d S )NzInvalid column name: E1rO   rD   rQ   zD:E1r|   rZ   r]   r   r   r   $test_usecols_excel_range_str_invalid'  s    z0TestReaders.test_usecols_excel_range_str_invalidc              	   C   s>   d}t jt|d" tjd| ddgddgd W 5 Q R X d S )Nz(list indices must be integers.*, not strrO   rD   rQ   ru   rd   rS   r   r[   	TypeErrorr4   r3   r]   r   r   r   test_index_col_label_error-  s    z&TestReaders.test_index_col_label_errorc                 C   sX   t jd| ddddgd}tddd	gtg gd
 g gd
 dddgdd}t|| d S )NrD   Sheet3ru   rc   rd   r   r{   EFrR   levelscodesnamesrf   r   )r4   r3   r	   r   ro   rp   r8   r   rL   rM   r   r   r   test_index_col_empty8  s      z TestReaders.test_index_col_emptyrU   Nrg   c                 C   sZ   t jd| d|d}tdddgddd	ggd
ddgd}|rJ||j| }t|| d S )NrD   Sheet4r   i1axi2byz
Unnamed: 0Zcol1Zcol2re   )r4   r3   r	   	set_indexrf   ro   rp   )r8   r   rU   rL   rM   r   r   r   test_index_col_with_unnamedC  s       z'TestReaders.test_index_col_with_unnamedc              	   C   s6   d}t jt|d tjd| dgd W 5 Q R X d S )NzEUsecols do not match columns, columns expected but not found: \['E'\]rO   rD   r   rV   rZ   r]   r   r   r   %test_usecols_pass_non_existent_columnQ  s    z1TestReaders.test_usecols_pass_non_existent_columnc              	   C   s8   d}t jt|d tjd| ddgd W 5 Q R X d S )Nz['usecols' must either be list-like of all strings, all unicode, all integers or a callable.rO   rD   ZE1r   r   rZ   r]   r   r   r   test_usecols_wrong_typeZ  s    z#TestReaders.test_usecols_wrong_typec                 C   s8   t jd| dd}tddggddgd}t|| d S )	NZtest2rQ   rT   ZaaaaZbbbbbTestZTest1re   r4   r3   r	   ro   rp   )r8   r   parsedrM   r   r   r   test_excel_stop_iteratorc  s    z$TestReaders.test_excel_stop_iteratorc                 C   sT   |dkr|j tjjdd tjd| dd}ttj	ggdgd}t
|| d S )	Nr   r`   ra   Ztest3rQ   r   r   re   )rj   rk   r   rl   rm   r4   r3   r	   npnanro   rp   )r8   r)   r   r   rM   r   r   r   test_excel_cell_error_nai  s    z$TestReaders.test_excel_cell_error_nac                 C   s   |dkr|j tjjdd tjd| ddd}tjd| dd	gdd
}tj||dd tj||dd tjd| ddd	d}t||j	d d  d S )Nr   r`   ra   rD   rQ   r   r   rW   rF   rT   rY   rU   Frh   rT   rU   
skipfooter)
rj   rk   r   rl   rm   r4   r3   ro   rp   ilocrv   r   r   r   test_excel_tableu  s,          zTestReaders.test_excel_tablec                 C   s  |dkr|j tjjdd tdddddgd	d
dddgdddddgdddddgdddddgtdddtdddtdddtdddtdddgd}d}tj	|| d d!}t
|| | }|d" t|d"< d#|j|jd d$f< t
jtd%dd& tj	|| d dd'}W 5 Q R X t
|| t|jD ]4\}}tj	|| d |d(}||}	t
||	 q0|d) t|d)< tj	|| d d)tid*}t
|| | }
|
d) t|
d)< t
jtd%dd&  tj	|| d dd)tid+}W 5 Q R X t
||
 d S ),Nr   r`   ra   rF   rg      r   g      ?g      @gHzG?gQ?g&.>TFrR      r   cdei  
         iq        i  )IntColFloatColBoolColStrColStr2ColZDateColZ
test_typesrQ   r   r   g      @r   zconvert_float is deprecated)rP   raise_on_extra_warnings)rT   convert_floatr   r   rT   
converters)rT   r   r   )rj   rk   r   rl   rm   r	   	from_dictr   r4   r3   ro   rp   copyastypefloatlocr   assert_produces_warningFutureWarning	enumeraterf   r   applystr)r8   r)   r   rM   basenameactualZfloat_expectedZicolnameexpZno_convert_floatr   r   r   test_reader_special_dtypes  s    




    
  z&TestReaders.test_reader_special_dtypesc              
   C   s   d}t dddddgdtjdd	d
gdddddgdtjdddgd}dd dd dd dd d}tj|| d|d}t|| d S )NZtest_convertersrF   rg   r   r   g      )@gL2@g3333333@g:0y5>Found	Not found1345)r   r   r   r   c                 S   s   | dkrt | S dS )N r   )intr   r   r   r   <lambda>      z4TestReaders.test_reader_converters.<locals>.<lambda>c                 S   s   | rd|  S t jS )Nr   )r   r   r   r   r   r   r     r   c                 S   s   | dkrdS dS )Nr   r   r   r   r   r   r   r   r     r   c                 S   s   | rt | S dS )Nr   )r   r   r   r   r   r     r   )r   r   rg   rR   rQ   r   )r	   r   r   r   r4   r3   ro   rp   )r8   r   r   rM   r   r   r   r   r   test_reader_converters  s&    
	  z"TestReaders.test_reader_convertersc              	   C   s   d}t || }tddddgdddd	gddddgd
dtjdgdjddddgd}t|| t j|| ddtdd}|d 	d|d< |d 	d|d< ddddg|d< t|| d}t
jt|d t j|| ddid W 5 Q R X d S )N	testdtyperF   rg   rR   r         @      @      @      @      ?       @      @r   r   r   r   r   r   r   r   re   float64float32)r   r   r   dtype001002003004z(Unable to convert column d to type int64rO   int64)r4   r3   r	   r   r   rn   ro   rp   r   r   r   r[   r\   )r8   r   r   r   rM   r^   r   r   r   test_reader_dtype  s.    



	 
zTestReaders.test_reader_dtypezdtype,expectedr   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   2r   c                 C   s&   d}t j|| |d}t|| d S )Nr   r   r4   r3   ro   rp   )r8   r   r   rM   r   r   r   r   r   test_reader_dtype_str  s    z!TestReaders.test_reader_dtype_strzdtypes, exp_valuea.1r   c           	      C   s\   d}dt i|}| }tj|| |d}tdg|gd}||ksLtdt|| d S )NZdf_mangle_dup_col_dtypesr   r   r   )r   r   zdtype dict changed)r   r   r4   r3   r	   rI   ro   rp   )	r8   r   ZdtypesZ	exp_valuer   Z
dtype_dictZdtype_dict_copyrL   rM   r   r   r   test_dtype_mangle_dup_cols6  s    z&TestReaders.test_dtype_mangle_dup_colsc                 C   s8   d}t || }tddddddgi}t|| d S )NZtest_spacesZtestcolzthis is greatz4    spacesz1 trailing z
 1 leadingz2  spaces  multiple  timesr   )r8   r   r   r   rM   r   r   r   test_reader_spacesB  s    zTestReaders.test_reader_spaceszbasename,expectedzgh-35802ZCOLUMNzTest (1)zgh-36122z
got 2nd sare   c                 C   s6   |dkrt d|  t|| }t|| d S )Nr   zSkipped for engine: )r   skipr4   r3   ro   rp   )r8   r   r   r   rM   r   r   r   r   test_read_excel_ods_nested_xmlU  s    	z*TestReaders.test_read_excel_ods_nested_xmlc                 C   sH   d}t j|| d d}dddg}t||  |t| ksDtd S )Ntest_multisheetr   CharlieAlphaBeta)r4   r3   ro   assert_contains_allkeyslistrI   r8   r   r   dfsexpected_keysr   r   r   test_reading_all_sheetsd  s
    
z#TestReaders.test_reading_all_sheetsc                 C   sX   d}dddg}t j|| |d}tt|}t||  t|t| ksTtd S )Nr   rg   r   r   )	r4   r3   r   setro   r   r   r~   rI   )r8   r   r   r  r   r   r   r   %test_reading_multiple_specific_sheetsq  s    
z1TestReaders.test_reading_multiple_specific_sheetsc                 C   s4   d}t j|| d d}dddg}t||  d S )Nblank_with_headerr   rQ   rW   r   )r4   r3   ro   r   r   r   r   r   r   "test_reading_all_sheets_with_blank  s    
z.TestReaders.test_reading_all_sheets_with_blankc                 C   s$   t jd| dd}t|t  d S )NblankrQ   r   )r4   r3   ro   rp   r	   )r8   r   r   r   r   r   test_read_excel_blank  s    z!TestReaders.test_read_excel_blankc                 C   s0   t ddgd}tjd| dd}t|| d S )NZcol_1Zcol_2re   r  rQ   r   )r	   r4   r3   ro   rp   )r8   r   rM   r   r   r   r   !test_read_excel_blank_with_header  s    z-TestReaders.test_read_excel_blank_with_headerz-ignore:Cell A4 is marked:UserWarning:openpyxlc                 C   s   |dkr|j tjjdd ttddgtddgdd	ggd
dgd}|dkrj|j tjjdd |d kr|dkr|j tjjdd td| }t	
|| d S )Nr   r`   ra   z
2016-03-12zMarc Johnsonz
2016-03-16z
Jack Blackg@xDzTimothy BrownZDateColWithBigIntZ	StringColre   r   zMaybe not supported by openpyxl)r   r   z)Defaults to openpyxl, maybe not supportedZtestdateoverflow)rj   rk   r   rl   rm   r	   r4   	Timestampr3   ro   rp   )r8   r)   r   r   rM   rL   r   r   r   test_date_conversion_overflow  s.    	z)TestReaders.test_date_conversion_overflowc                 C   sr   |dkr|j tjjdd d}d}tj|| |dd}tj|| d|d}tj||d	d
 tj||d	d
 d S Nr   r`   ra   rD   rQ   r   r   )rU   rT   Frh   ry   )r8   r)   r   rq   filenamerT   rr   rs   r   r   r   test_sheet_name  s       zTestReaders.test_sheet_namec              	   C   sN   d| }t j|ddd}t|d"}t j|ddd}t|| W 5 Q R X d S )NrD   rQ   r   r   rE   )r4   r3   rH   ro   rp   )r8   r   pthrM   rK   r   r   r   r   test_excel_read_buffer  s
    z"TestReaders.test_excel_read_bufferc              	   C   s0   d}t jtdd tjd|d W 5 Q R X d S )NfoozUnknown engine: foorO   r   r/   rZ   )r8   Z
bad_enginer   r   r   test_bad_engine_raises  s    z"TestReaders.test_bad_engine_raisesrT   r   rQ   c              	   C   s4   d}t jt|d tjd| |d W 5 Q R X d S Nz?Worksheet index 3 is invalid|Worksheet named 'Sheet4' not foundrO   r  r   rZ   )r8   r   rT   r^   r   r   r   test_bad_sheetname_raises  s    z%TestReaders.test_bad_sheetname_raisesc              	   C   s6   d| }d}t jt|d t| W 5 Q R X d S )Nr  uV   (No such file or directory|没有那个文件或目录|File o directory non esistente)rO   )r   r[   FileNotFoundErrorr4   r3   )r8   r   Zbad_filerP   r   r   r   test_missing_file_raises  s    
z$TestReaders.test_missing_file_raisesc              	   C   sd   d}|d krt }d}n&|dkr4ddlm} |}d}nt}d}tj||d t| W 5 Q R X d S )	Ns   foozLExcel file format cannot be determined, you must specify an engine manually.r   r   )	XLRDErrorzFUnsupported format, or corrupt file: Expected BOF record; found b'foo'zFile is not a zip filerO   )r\   r   r  r   r   r[   r4   r3   )r8   r   Z
bad_streamerrorr^   r  r   r   r   test_corrupt_bytes_raises  s    z%TestReaders.test_corrupt_bytes_raisesz^https://raw.githubusercontent.com/pandas-dev/pandas/main/pandas/tests/io/data/excel/test1.xlsx)urlZcheck_before_testc                 C   s0   d| }t |}t d| }t|| d S )NzYhttps://raw.githubusercontent.com/pandas-dev/pandas/main/pandas/tests/io/data/excel/test1rD   r   )r8   r   r  	url_tablelocal_tabler   r   r   test_read_from_http_url  s    

z#TestReaders.test_read_from_http_urlc              	   C   sf   t d| d}|djd| |d W 5 Q R X d| }tj||d}td| }t|| d S )NrD   rE   pandas-testZKeyZBodys3://pandas-test/test1)Zstorage_options)rH   Bucket
put_objectr4   r3   ro   rp   )r8   r   s3_resources3sorK   r  r  r  r   r   r   test_read_from_s3_url
  s    "z!TestReaders.test_read_from_s3_urlc           	   	   C   s   t d| d}|djd| |d W 5 Q R X dd l}|jf |}| d| }t|}W 5 Q R X td| }t|| d S )NrD   rE   r  r  r   r   )	rH   r!  r"  s3fsZS3FileSystemr4   r3   ro   rp   )	r8   r   r#  r$  rK   r&  Zs3r  r  r   r   r   test_read_from_s3_object  s    "z$TestReaders.test_read_from_s3_objectc                 C   s   t j|dddd| }t|}ztd| }W n> tk
rt   dd l}d|  }t	
d|  Y nX t|| d S )	Nr0   r1   r2   rD   zfile://localhost/r    zfailing on )ospathjoinr4   r3   r   platformunamestripr   r   ro   rp   )r8   r   r9   Z
localtabler  r  r,  Zplatform_infor   r   r   test_read_from_file_url(  s    
z#TestReaders.test_read_from_file_urlc                 C   sP   ddl m} d| }tj|ddd}|d| }tj|ddd}t|| d S )Nr   r   rD   rQ   r   )pathlibr   r4   r3   ro   rp   )r8   r   r   str_pathrM   path_objr   r   r   r   test_read_from_pathlib_path:  s    z'TestReaders.test_read_from_pathlib_pathzpy.pathc                 C   s\   ddl m} tjd| }tj|ddd}| d| }tj|ddd}t|| d S )Nr   )localrD   rQ   r   )	Zpy.pathr4  r)  r*  r+  r4   r3   ro   rp   )r8   r   Z	LocalPathr1  rM   r2  r   r   r   r   test_read_from_py_localpathG  s    z'TestReaders.test_read_from_py_localpathc              	   C   sD   t jd| }t|d }tj|ddd}~|  W 5 Q R X d S )NrD   rE   rQ   r   r   )r)  r*  r+  rH   r4   r3   read)r8   r   r1  rK   r   r   r   r   test_close_from_py_localpathV  s
    z(TestReaders.test_close_from_py_localpathc                 C   s   |dkr|j tjjdd tdtdddtddd	d
tddddtddddtddddtddddtddddtdddd tdd!dd"td#d$d%d&td'd(d)gi}tj	d*| d+d,}t
|| tj	d-| d+d,}t
|| d S ).Nr   r`   ra   ZTimerF   rg   rR   -   8   i r      1   i@       *   i    9   #   i 	   )      i           i'	 r   i`
 5   i 5    %   r   i       6   Z
times_1900rQ   r   Z
times_1904)rj   rk   r   rl   rm   r	   r   r   r4   r3   ro   rp   )r8   r)   r   r   rM   r   r   r   r   test_reader_secondsa  s4    

zTestReaders.test_reader_secondsc              	   C   sB  |dkr|j tjjdd tddgddgg}d| }td	d
t	ddgddt	ddgddt	ddgddt	ddgg|d}tj
|ddd	gdd}t|| ||_ddddg|_tj
|ddd	gd}tj||dd ||_tj
|d dd	gdd	gd!}tj||dd ddddg|_|d"d#g|_tj
|d$dd	gd}t|| ttd|_|d%d&g|_tj
|d'dd	gdd}t|| |jd	dgd	d(d%d&g|_tj
|d)ddd	gd!}t|| |d%d&g|_|d"d#g|_tj
|d*dd	gdd	gd!}t|| tj
|d+dd	gdd	gdd,}t|| d S )-Nr   r`   ra   r  barr   r   testmultiindexrF   r   
2015-01-01Trg   r   
2015-01-02FrR   r   
2015-01-03r   r   
2015-01-04re   	mi_columnr   )rT   headerrU   r   r   mi_indexr   rh   bothrT   rU   rV  ilvl1ilvl2Zmi_index_namec1c2mi_column_name)levelZname_with_intZ	both_nameZboth_name_skiprows)rT   rU   rV  rY   )rj   rk   r   rl   rm   r   from_productr	   r4   r
  r3   ro   rp   r   rf   Z	set_namesr   r}   Z
set_levels)r8   r)   r   mimi_filerM   r   r   r   r   test_read_excel_multiindex  s    
               z&TestReaders.test_read_excel_multiindexzsheet_name,idx_lvl2Zboth_name_blank_after_mi_namer   r   Zboth_name_multiple_blanksc           	   	   C   s   |dkr|j tjjdd d| }tjddgddggd	d
gd}tddt	ddgddt	ddgddt	ddgddt	ddgg|tj
ddddg|fddgdd}tj||ddgddgd}t|| d S )Nr   z;Sheets containing datetimes not supported by pyxlsb (GH4679ra   rP  r  rO  r   r   r\  r]  r   rF   r   rQ  Trg   r   rR  FrR   r   rS  r   r   rT  rZ  r[  r   r   rY  )rj   rk   r   rl   rm   r   r`  r	   r4   r
  Zfrom_arraysr3   ro   rp   )	r8   r)   r   rT   Zidx_lvl2rb  ra  rM   rL   r   r   r   +test_read_excel_multiindex_blank_after_name  s6    z7TestReaders.test_read_excel_multiindex_blank_after_namec                 C   sT   d| }t j|dddgd}tddg}tddd	d
ggd |d}t|| d S )NrP  index_col_noner   rF   rT   rV  ru   rc   keyvalrg   rR   r   re   )r4   r3   r   r`  r	   ro   rp   )r8   r   rb  rL   Zexp_columnsrM   r   r   r   &test_read_excel_multiindex_header_only  s
    z2TestReaders.test_read_excel_multiindex_header_onlyc           	      C   s  d| }t d d d d d gdddddgddd	d
dgdddddgdddddgdddddgg}dddddg}td d!d"d#d$d%gd&d'd(d)d*d+ggd,d-d.d/d0d1gd,d-d.d/d0d1ggd d gd2}td d!d"d#d$d%gd d3}t|||d4}tj|d5d,d6}t|| ||_	tj|d7d,d-gd6}t|| t dddddgddd	d
dgdddddgdddddgdddddgg}dddddg}td!d"d#d$d%gd'd(d)d*d+ggd,d-d.d/d0gd,d-d.d/d0ggd d gd2}td!d"d#d$d%gd d3}t|||d4}tj|d8d,d6}t|| ||_	tj|d9d,d-gd6}tj||d:d; d S )<NZtest_index_name_pre17ZR0C0ZR0C1ZR0C2ZR0C3ZR0C4ZR1C0ZR1C1ZR1C2ZR1C3ZR1C4ZR2C0ZR2C1ZR2C2ZR2C3ZR2C4ZR3C0ZR3C1ZR3C2ZR3C3ZR3C4ZR4C0ZR4C1ZR4C2ZR4C3ZR4C4ZC_l0_g0ZC_l0_g1ZC_l0_g2ZC_l0_g3ZC_l0_g4ZR0ZR_l0_g0ZR_l0_g1ZR_l0_g2ZR_l0_g3ZR_l0_g4ZR1ZR_l1_g0ZR_l1_g1ZR_l1_g2ZR_l1_g3ZR_l1_g4r   rF   rg   rR   r   r   r   r   r   rf   Zsingle_namesr   Zmulti_namesZsingle_no_namesZmulti_no_namesFrh   )
r   arrayr   r
   r	   r4   r3   ro   rp   r   )	r8   r   r  r1   rf   ra  sirM   r   r   r   r   test_excel_old_index_format  sf    
 	z'TestReaders.test_excel_old_index_formatc              
   C   s>   d}dD ]0}t jt|d tjd| |d W 5 Q R X qd S )Nz#Passing a bool to header is invalid)TFrO   rD   rV  r   )r8   r   r^   argr   r   r   test_read_excel_bool_header_argT  s    z+TestReaders.test_read_excel_bool_header_argc              	   C   sV  |dkr|j tjjdd tjd| dddgd}td	d
tddgddtddgddtddgddtddggddddgd}t	
|| tjd| dtddgd}t	
|| tjd| ddd d}t	
|| tjd| ddddddgd}tddtddgddtddgddtddggddddgd}t	
|| d S )Nr   r`   ra   testskiprowsskiprows_listr   rg   rT   rY   rF   r   rQ  Tr   rR  FrR   r   rS  r   r   rT  r   r   r   r   re   c                 S   s   | dkS )N)r   rg   r   r   r   r   r   r   }  r   z6TestReaders.test_read_excel_skiprows.<locals>.<lambda>)rT   rY   r   )rj   rk   r   rl   rm   r4   r3   r	   r
  ro   rp   r   ro  r8   r)   r   r   rM   r   r   r   test_read_excel_skiprows[  s\      
	

	z$TestReaders.test_read_excel_skiprowsc                 C   sz   |dkr|j tjjdd tjd| ddd d}td	d
tddgddtddggddddgd}t	
|| d S )Nr   r`   ra   ru  rv  c                 S   s   | dkS )N)rF   rR   r   r   r   r   r   r   r     r   zFTestReaders.test_read_excel_skiprows_callable_not_in.<locals>.<lambda>rw  rF   r   rQ  TrR   r   rS  Fr   r   r   r   re   )rj   rk   r   rl   rm   r4   r3   r	   r
  ro   rp   rx  r   r   r   (test_read_excel_skiprows_callable_not_in  s$    
	z4TestReaders.test_read_excel_skiprows_callable_not_inc                 C   s@   d}t jd| |d}t d| }|d | }t|| d S )Nr   rD   nrowsr   )r8   r   num_rows_to_pullr   rM   r   r   r   test_read_excel_nrows  s
    z!TestReaders.test_read_excel_nrowsc                 C   s@   t d| }t|}|d }t jd| |d}t|| d S )NrD   r   r{  )r4   r3   r~   ro   rp   )r8   r   rM   Znum_records_in_filer}  r   r   r   r   0test_read_excel_nrows_greater_than_nrows_in_file  s
    z<TestReaders.test_read_excel_nrows_greater_than_nrows_in_filec              	   C   s4   d}t jt|d tjd| dd W 5 Q R X d S )Nz'nrows' must be an integer >=0rO   rD   r   r{  rZ   r]   r   r   r   +test_read_excel_nrows_non_integer_parameter  s    z7TestReaders.test_read_excel_nrows_non_integer_parameterz-filename,sheet_name,header,index_col,skiprowsrP  rU  rW  rX  r^  ru  rv  c                 C   s   | dkp| dkS )Nr   rg   r   r   r   r   r   r     r   zTestReaders.<lambda>c           	      C   sL   t j|| ||||djdd }t j|| ||||dd}t|| dS )z
        For various parameters, we should get the same result whether we
        limit the rows during load (nrows=3) or after (df.iloc[:3]).
        )rT   rV  rU   rY   NrR   )rT   rV  rU   rY   r|  )r4   r3   r   ro   rp   )	r8   r   r  rT   rV  rU   rY   rM   r   r   r   r   test_read_excel_nrows_params  s$    z(TestReaders.test_read_excel_nrows_paramsc              	   C   s   d| }t jtdd tj|dddd}tdd	d
gd
ddgdd}d|j_t || tj|ddd}t	d
ddgdd	d
gd}t 
|| tj|ddd}tddd	gdd}t || W 5 Q R X d S )NZtest_squeezezThe squeeze argument has been deprecated and will be removed in a future version. Append .squeeze\("columns"\) to the call to squeeze.

rO   Ztwo_columnsr   T)rT   rU   squeezerg   rR   r   r   r<  r   rm  r   )rT   r  )r   r   Z
one_columnrF   )ro   r   r   r4   r3   r   r   r   Zassert_series_equalr	   rp   )r8   r   rK   r   rM   r   r   r   test_read_excel_squeeze  s(       z#TestReaders.test_read_excel_squeezec              	   C   s>   t jtdd td| dd W 5 Q R X td|  d S )NF)r   rD   rQ   r   )ro   r   r   r4   r3   r8   r   r   r   r   test_deprecated_kwargs  s    z"TestReaders.test_deprecated_kwargsc                 C   s^   d| }ddddg}t jddddgdd}t||d	d
}tj|dddgd d}t|| d S )NrP  )rc   rc   ri  )rR   r   )ru   ru   )rF   rg   )r   rF   rd  )rg   rR   rn  rf  r   rF   rY  )r   from_tuplesr	   r4   r3   ro   rp   )r8   r   	file_namer1   idxrM   rL   r   r   r   "test_no_header_with_list_index_col  s    
    z.TestReaders.test_no_header_with_list_index_colc                 C   s>   d| }dt jddg}t|dgd}t|}t|| d S )NZone_col_blank_lineg      ?rF   rg   Znumbersre   )r   r   r	   r4   r3   ro   rp   )r8   r   r  r1   rM   rL   r   r   r   test_one_col_noskip_blank_line  s
    
z*TestReaders.test_one_col_noskip_blank_linec                 C   sj   d| }t ddg}tjtjgtjtjgddgddgg}t||d}tj|d	d
dgd}t|| d S )NrP  )r   ru   )r   rc   rF   rR   rg   r   re   Zmi_column_empty_rowsr   rg  )	r   r  r   r   r	   r4   r3   ro   rp   )r8   r   r  rf   r1   rM   rL   r   r   r    test_multiheader_two_blank_lines  s    $  z,TestReaders.test_multiheader_two_blank_linesc                 C   s$   d| }t |}|jdks tdS )z
        Sheets can contain blank cells with no data. Some of our readers
        were including those cells, creating many empty rows and columns
        Ztrailing_blanks)rR   rR   N)r4   r3   shaperI   )r8   r   r  rL   r   r   r   test_trailing_blanks$  s    
z TestReaders.test_trailing_blanksc              	   C   s`   |dkrt d |dkr0|jt jjdd t jtdd tj	d| d	d
 W 5 Q R X d S )Nr   *chartsheets do not exist in the ODF formatr   4pyxlsb can't distinguish chartsheets from worksheetsra   z"Worksheet named 'Chart1' not foundrO   
chartsheetZChart1r   
r   r   rj   rk   rl   rm   r[   r\   r4   r3   r8   r)   r   r   r   r   r   test_ignore_chartsheets_by_str-  s    
z*TestReaders.test_ignore_chartsheets_by_strc              	   C   s`   |dkrt d |dkr0|jt jjdd t jtdd tj	d| d	d
 W 5 Q R X d S )Nr   r  r   r  ra   z0Worksheet index 1 is invalid, 1 worksheets foundrO   r  rF   r   r  r  r   r   r   test_ignore_chartsheets_by_int:  s    
 z*TestReaders.test_ignore_chartsheets_by_intc              	   C   sf   t jd| ddd}tdddddd	gd
dddddgddddddggddddddgd}t|| d S )NZtest_decimal,rF   )decimalrY   gAc̝ė@g	hAABCZpoig2[j@rg   gHzG^@g{G@ZDEFZuytgUq&?rR   g%Cq@g^@ZGHIZrezg)@ZIdZNumber1ZNumber2ZText1ZText2ZNumber3re   r   r   r   r   r   test_euro_decimal_formatI  s    z$TestReaders.test_euro_decimal_format)V__name__
__module____qualname__r   fixturer<   rN   r_   rt   rx   rl   parametrizerz   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r	   r   r   r   r   r   r   r   r   r  r  r  r  r	  filterwarningsr  r  r  r  r  r  r  networkro   r  tdZskip_if_not_us_localeZ
single_cpur%  r'  Zslowr/  r3  
skip_if_noZcheck_file_leaksr5  r7  rN  rc  re  rl  rq  rt  ry  rz  r~  r  r  r  r  r  r  r  r  r  r  r  r  r   r   r   r   r-   q   s  

5 2


		R









 
	



!V
"H7
	r-   c                
   @   s   e Zd Zejdddd Zdd Zdd Zej	d	d
ddgdd Z
dd Zdd Zej	ddddgddgdddgddggdd Zdd Zdd Zdd Zdd  Zd!d" Zd#d$ Zej	d%d&d'gd(d) Zd*d+ Zd,d- Zd.d/ Zd0d1 Zd
S )2TestExcelFileReadTr.   c                 C   s2   t tj|d}||ddd |td| dS )zH
        Change directory and set engine for ExcelFile objects.
        r/   r0   r1   r2   rG   N)r   r4   rG   r5   r6   r7   r   r   r   r<   X  s    z#TestExcelFileRead.cd_and_set_enginec              	   C   s^   dddddd}t d| }|j}W 5 Q R X |d k	r>|}n||dd   }||ksZtd S )Nr   r   r   r   rA   rD   rF   )r4   rG   r   rI   )r8   r   r   rJ   r2   rL   rM   r   r   r   rN   a  s    z"TestExcelFileRead.test_engine_usedc              	   C   s  t d| }t j|dddgd}W 5 Q R X tdgdgdgtjgdggd	gd
}t|| t d| }t j|dddgd}W 5 Q R X ttjgdgtjgtjgdggd	gd
}t|| t d| }t j|dddgd}W 5 Q R X tdgdgdgtjgdggd	gd
}t|| t d| }t j|dddgd}W 5 Q R X ttjgdgtjgtjgdggd	gd
}t|| d S )NZtest4rQ   FapplerT   Zkeep_default_naZ	na_valuesZNArF   rabbitr   re   Ttest51.#QNANr   )r4   rG   r3   r	   r   r   ro   rp   )r8   r   r2   r   rM   r   r   r   test_excel_passes_nas  s`                    z&TestExcelFileRead.test_excel_passes_na	na_filterNFc              	   C   s   i }|d k	r||d< t d| "}t j|fdddgd|}W 5 Q R X |dkrldgd	gd
gdgdgg}ntjgd	gtjgtjgdgg}t|dgd}t|| d S )Nr  r  rQ   Tr  r  Fr  rF   r   r  r   re   )r4   rG   r3   r   r   r	   ro   rp   )r8   r   r  r>   r2   r   rM   r   r   r   test_excel_passes_na_filter  s$    z-TestExcelFileRead.test_excel_passes_na_filterc              	   C   s\  |dkr|j tjjdd td| *}tj|ddd}tj|ddgdd}W 5 Q R X tj	||d	d
 tj	||d	d
 td| &}|j
ddd}|j
ddgdd}W 5 Q R X tj	||d	d
 tj	||d	d
 td| }tj|dddd}W 5 Q R X t	||jd d  td| }|j
dddd}W 5 Q R X t	||jd d  d S )Nr   r`   ra   rD   r   r   rF   r   Frh   )rU   )rY   rU   r   r   )rU   r   )rj   rk   r   rl   rm   r4   rG   r3   ro   rp   r@   r   )r8   r)   r   rq   r2   rr   rs   rw   r   r   r   test_excel_table_sheet_by_index  s,    z1TestExcelFileRead.test_excel_table_sheet_by_indexc           	   	   C   s   |dkr|j tjjdd d}d}t|| }|j|dd}W 5 Q R X t|| }|jd|d}W 5 Q R X tj	||d	d
 tj	||d	d
 d S r  )
rj   rk   r   rl   rm   r4   rG   r@   ro   rp   )	r8   r)   r   rq   r  rT   r2   Z	df1_parseZ	df2_parser   r   r   r    s    z!TestExcelFileRead.test_sheet_namerT   rR   r   r   rQ   c              
   C   sH   d}t jt|d, td| }|j|d W 5 Q R X W 5 Q R X d S r  )r   r[   r\   r4   rG   r@   )r8   r   rT   r^   r2   r   r   r   r    s    z+TestExcelFileRead.test_bad_sheetname_raisesc              
   C   sf   d| }t j|dd|d}t|d,}t |}t j|ddd}W 5 Q R X W 5 Q R X t|| d S )NrD   rQ   r   rT   rU   r   rE   r   )r4   r3   rH   rG   ro   rp   )r8   r   r   r  rM   rK   rC   r   r   r   r   r    s    $z(TestExcelFileRead.test_excel_read_bufferc              
   C   sP   t d| d.}t|}tj|dd|d W 5 Q R X W 5 Q R X |jsLtd S )NrD   rE   rQ   r   r  )rH   r4   rG   r3   closedrI   )r8   r   r   rK   rB   r   r   r   test_reader_closes_file  s    &z)TestExcelFileRead.test_reader_closes_filec              
   C   sJ   d}t d| .}tjt|d t j|dd W 5 Q R X W 5 Q R X d S )Nz8Engine should not be specified when passing an ExcelFilerD   rO   r  r/   )r4   rG   r   r[   r\   r3   )r8   r   r^   Zxlr   r   r   test_conflicting_excel_engines   s    z0TestExcelFileRead.test_conflicting_excel_enginesc              	   C   sR   t jd| |d}td| d}| }W 5 Q R X t j||d}t|| d S )NrD   r/   rE   )r4   r3   rH   r6  ro   rp   )r8   r   r   rM   rK   r1   r   r   r   r   test_excel_read_binary  s
    z(TestExcelFileRead.test_excel_read_binaryc              	   C   sF   t d| d}t|}W 5 Q R X tjd| |d}t|| d S )NrD   rE   r/   )rH   r4   r3   ro   rp   )r8   r   r   rK   rL   rM   r   r   r   %test_excel_read_binary_via_read_excel  s    z7TestExcelFileRead.test_excel_read_binary_via_read_excelc              
   C   sF   t dd2}tjtdd tj|ddgd W 5 Q R X W 5 Q R X d S )Nzdf_header_oob.xlsxrE   zexceeds maximumrO   r   rF   rr  )rH   r   r[   r\   r4   r3   )r8   r   rK   r   r   r   )test_read_excel_header_index_out_of_range  s    z;TestExcelFileRead.test_read_excel_header_index_out_of_ranger  zdf_empty.xlsxzdf_equals.xlsxc                 C   s`   t dgdd}tjddgddgd}td	d
gg||dd}tj|dddd	gd}t|| d S )NZZI2rm  rh  )ru   zB.1ZI11ZI12rd  rF   rR   r   )r   rf   r   rQ   r   rY  )r
   r   r  r	   r4   r3   ro   rp   )r8   r  r  colsrM   rL   r   r   r   test_header_with_index_col  s       z,TestExcelFileRead.test_header_with_index_colc           	   	   C   s   |dkr|j tjjdd d| }t|}tj|ddgd|d}W 5 Q R X tj	t
dt
d	fgt
d t
d	 gd
}tg |d}t|| d S )Nr   r`   ra   Ztest_datetime_mir   rF   )rV  rU   r   z
02/29/2020z
03/01/2020rd  re   )rj   rk   r   rl   rm   r4   rG   r3   r   r  Zto_datetimeZto_pydatetimer	   ro   rp   )	r8   r)   r   r   rK   r2   r   Zexpected_column_indexrM   r   r   r   test_read_datetime_multiindex*  s"     z/TestExcelFileRead.test_read_datetime_multiindexc              
   C   s>   t jtdd& td| dd W 5 Q R X W 5 Q R X d S )NzValue must be one of *rO   zio.excelz.readerabc)r   r[   r\   r4   Zoption_contextr  r   r   r   test_engine_invalid_optionA  s    z,TestExcelFileRead.test_engine_invalid_optionc              	   C   s^   |dkrt d |dkr0|jt jjdd td| }|jdgksPt	W 5 Q R X d S )Nr   r  r   r  ra   r  rQ   )
r   r   rj   rk   rl   rm   r4   rG   Zsheet_namesrI   )r8   r)   r   r   r2   r   r   r   test_ignore_chartsheetsG  s    
z)TestExcelFileRead.test_ignore_chartsheetsc              
   C   s   t f}|d kr td|  n|dkr<dd l}t |jjf}td| R}t|	d t
d. ztj||d W n |k
r   Y nX W 5 Q R X W 5 Q R X d S )NzInvalid test for engine=r   r   ZcorruptFr/   )r   r   r   r   Zbiffhr  ro   Zensure_cleanr   
write_textr   r4   rG   )r8   r   r   errorsr   filer   r   r   test_corrupt_files_closedT  s    z+TestExcelFileRead.test_corrupt_files_closed)r  r  r  r   r  r<   rN   r  rl   r  r  r  r  r  r  r  r  r  r  r  r  r  r  r  r  r   r   r   r   r  W  s2   

&





r  )+r   r   	functoolsr   r)  r0  r   urllib.errorr   zipfiler   Znumpyr   r   Zpandas.util._test_decoratorsutilZ_test_decoratorsr  Zpandasr4   r	   r
   r   r   Zpandas._testingZ_testingro   r#   r    r  rl   r  Zengine_paramsr   boolr   r"   r  r*   r   r   r-   r  r   r   r   r   <module>   sn   

	


         o