o
    ҷh                  	   @   s  d dl m Z mZ d dlmZ d dlmZ d dlZd dlmZ d dl	Z	d dl
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	lm Z m!Z! g d
Z"ej#de$dgdej#de$dgdej#de$dgdej#de$ddej#de$ddgZ%de&de'fddZ(dd Z)ej*dd e%D e&ddd Z+ej*dd Z,ej*dd Z-G dd dZ.G d d! d!Z/dS )"    )datetimetime)partial)BytesION)Path)URLError)
BadZipFile)	DataFrameIndex
MultiIndexSeries)ArrowStringArrayStringArray).xls.xlsx.xlsm.xlsb.odsxlrdmarksopenpyxlpyxlsbodfread_extreturnc                 C   s   | j d } | dkr|dkrdS | dkr|dkrdS |dkr#| dkr#dS | dkr-|dkr-dS |dkr7| dkr7dS | d	krA|dkrAdS d
S )zm
    Filter out invalid (engine, ext) pairs instead of skipping, as that
    produces 500+ pytest.skips.
    r   r   r   Fr   r   r   r   r   T)values)enginer    r   U/var/www/html/venv/lib/python3.10/site-packages/pandas/tests/io/excel/test_readers.py_is_valid_engine_ext_pair<   s   
r    c                 C   s    | j |f }tj|| jd}|S )z
    engine gives us a pytest.param object with some marks, read_ext is just
    a string.  We need to generate a new pytest.param inheriting the marks.
    r   )r   pytestparamr   )r   r   r   	new_paramr   r   r   _transfer_marksQ   s   r$   c                 C   s*   g | ]}t D ]}t||rt||qqS r   )read_ext_paramsr    r$   ).0engextr   r   r   
<listcomp>\   s    r)   )paramsidsc                 C      | j S )zS
    Fixture for Excel reader engine and read_ext, only including valid pairs.
    )r"   )requestr   r   r   engine_and_read_ext[   s   r.   c                 C   s   | \}}|S Nr   r.   r   r   r   r   r   r   k      r   c                 C   s   | \}}|S r/   r   r0   r   r   r   r   q   r1   c                   @   sx  e Zd Zejdddd Zdd Zdd Zd	d
 Zdd Z	dd Z
ejdg dg dg dg dg dg dgdd Zejdddgddggdd Zdd Zdd Zd d! Zd"d# Zd$d% Zd&d' Zejd(d)d*gd+d, Zd-d. Zd/d0 Zd1d2 Zd3d4 Zd5d6 Zd7d8 Zd9d: Zd;d< Zejd=d)eg d>g d?g d>d@dAej dBgdCfdDdEe!e!dCee"g d>dDdFe"g d?dEdFg dGdHdIej dJgdCfgdKdL Z#dMdN Z$dOdP Z%dQdR Z&ejdSi dHfdTdUidVfgdWdX Z'dYdZ Z(ejd[d\ed]d^gifd_ed`gdafgdbdc Z)ddde Z*dfdg Z+dhdi Z,djdk Z-dldm Z.dndo Z/ej0dpdqdr Z1dsdt Z2dudv Z3dwdx Z4ejdydzd{dzgdzd{gd|d}d|gd|d}ggd~d Z5dd Z6dd Z7ejj8ejj9dd Z:e;j<ejj9dd Z=ejj9dd Z>ejj?dd Z@dd ZAe;Bddd ZCdd ZDdd ZEdd ZFejddej dddgfdej gd fgdd ZGdd ZHdd ZIdd ZJdd ZKdd ZLdd ZMdd ZNdd ZOejdddd{dVgd{d)fddd)d{dVgd)fddd{dVgd{dVgd)fddd{dVgd{d)fddd)d)d{d*gfddd)d)dd fgdd ZPdd ZQdd ZRdd ZSdd ZTddĄ ZUddƄ ZVddȄ ZWddʄ ZXd)S )TestReadersTautousec                 C   2   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   pdr:   chdirsetattrselfr   datapathmonkeypatchfuncr   r   r   cd_and_set_enginex      zTestReaders.cd_and_set_enginec           	      C   s   dd }| tjd| dddddd}td	| d
}t|}W d    n1 s+w   Y  |d ur7|}n||dd   }||ksEJ d S )Nc                 _   r,   r/   r6   )r?   argskwargsr   r   r   parser   s   z,TestReaders.test_engine_used.<locals>.parserparser   r   r   r   xlsxxlsmxlsbxlsodstest1rb   )r=   r;   	ExcelFileopenr:   )	r?   r   r   rA   rG   expected_defaultsfresultexpectedr   r   r   test_engine_used   s   zTestReaders.test_engine_usedc              	   C   s   ddiddiddiddiddid}|dd  dv r t d	}n|dd  d
kr.t d}nt d}|d urbtjt|d tjd| dd||dd   d W d    d S 1 s[w   Y  d S d S )Nfooabcd{   TrueTrI   rQ   >   rM   rL   z8open_workbook() got an unexpected keyword argument 'foo'rN   z/load() got an unexpected keyword argument 'foo'z8load_workbook() got an unexpected keyword argument 'foo'matchrO   Sheet1r   )
sheet_name	index_colengine_kwargs)reescaper!   raises	TypeErrorr;   r:   )r?   r   r   rT   msgr   r   r   test_engine_kwargs   s*   
"zTestReaders.test_engine_kwargsc                 C   s   d}t jt|d tjd| dddd W d    n1 sw   Y  t jt|d tjd| dd	gddd
 W d    d S 1 sDw   Y  d S )Nz Passing an integer for `usecols`r]   rO   r_   r      r`   ra   usecolsSheet2rQ   r`   skiprowsra   rk   r!   re   
ValueErrorr;   r:   r?   r   rg   r   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	g d
d}tjd| ddgd	g d
d}tj	||dd tj	||dd d S )Nr   3Sheets containing datetimes not supported by pyxlsbreasonBCcolumnsrO   r_   r   )r      ri   rj   rl   rQ   rm   Fcheck_names
node
add_markerr!   markxfailreindexr;   r:   tmassert_frame_equal)r?   r-   r   df_refdf1df2r   r   r   test_usecols_list   s&   	zTestReaders.test_usecols_listc                 C   sB  |dkr|j tjjdd |jg d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   rs   rt   Arv   rw   rx   rO   r_   r   zA:Drj   rl   rQ   rm   Fr{   rv   rw   zA,C,DzA,C:Dr}   r?   r-   r   r   r   r   df3r   r   r   test_usecols_str   sZ   	zTestReaders.test_usecols_strrk   )r   rQ   ri   )r   ri   rQ   )rQ   r   ri   )rQ   ri   r   )ri   r   rQ   )ri   rQ   r   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   rs   rt   r   rw   rO   r_   r   rj   Fr{   	r~   r   r!   r   r   r;   r:   r   r   )r?   r-   r   rk   r   rW   rV   r   r   r   .test_usecols_diff_positional_int_columns_order  s   z:TestReaders.test_usecols_diff_positional_int_columns_orderrv   Dc                 C   sB   |ddg }t t||_tjd| d|d}tj||dd d S )Nrv   r   rO   r_   r`   rk   Fr{   )rangelenindexr;   r:   r   r   )r?   r   rk   r   rW   rV   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   rs   rt   rO   r_   r   r`   ra   Fr{   r   r?   r-   r   r   rW   rV   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   rs   rt   rw   r   rO   r_   r   zA,D:Erj   Fr{   r   r   r   r   r   test_usecols_excel_range_str9  s   z(TestReaders.test_usecols_excel_range_strc                 C   sL   d}t jt|d tjd| ddd W d    d S 1 sw   Y  d S )NzInvalid column name: E1r]   rO   r_   zD:E1r   ro   rq   r   r   r   $test_usecols_excel_range_str_invalidG  s   "z0TestReaders.test_usecols_excel_range_str_invalidc                 C   sT   d}t jt|d tjd| ddgddgd W d    d S 1 s#w   Y  d S )Nz(list indices must be integers.*, not strr]   rO   r_   r   rw   rj   r!   re   rf   r;   r:   rq   r   r   r   test_index_col_label_errorM  s   "z&TestReaders.test_index_col_label_errorc                 C   s<   t jd| ddd}tg dtg ddd}t|| d S )NrO   Sheet3r   r   )rv   rw   r   EFnamery   r   )r;   r:   r	   r
   r   r   r?   r   rV   rW   r   r   r   test_index_col_strX  s
   zTestReaders.test_index_col_strc                 C   sR   t jd| dg dd}tg dtg gd g gd g ddd}t|| d S )	NrO   r   r   r   )r   r   r   ri   levelscodesnamesr   )r;   r:   r	   r   r   r   r   r   r   r   test_index_col_empty`  s   z TestReaders.test_index_col_emptyra   Nrz   c                 C   sT   t jd| d|d}tg dg dgg dd}|r"||j| }t|| d S )NrO   Sheet4r   )i1ax)i2by)z
Unnamed: 0col1col2rx   )r;   r:   r	   	set_indexry   r   r   )r?   r   ra   rV   rW   r   r   r   test_index_col_with_unnamedk  s   
z'TestReaders.test_index_col_with_unnamedc                 C   sL   d}t jt|d tjd| dgd W d    d S 1 sw   Y  d S )NzEUsecols do not match columns, columns expected but not found: \['E'\]r]   rO   r   rk   ro   rq   r   r   r   %test_usecols_pass_non_existent_columny  s
   "z1TestReaders.test_usecols_pass_non_existent_columnc                 C   sN   d}t jt|d tjd| ddgd W d    d S 1 s w   Y  d S )Nz['usecols' must either be list-like of all strings, all unicode, all integers or a callable.r]   rO   E1r   r   ro   rq   r   r   r   test_usecols_wrong_type  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 )	Ntest2r_   r`   aaaabbbbbTestTest1rx   r;   r:   r	   r   r   )r?   r   parsedrW   r   r   r   test_excel_stop_iterator  s   z$TestReaders.test_excel_stop_iteratorc                 C   sT   |dkr|j tjjdd tjd| dd}ttj	ggdgd}t
|| d S )	Nr   rs   rt   test3r_   r   r   rx   )r~   r   r!   r   r   r;   r:   r	   npnanr   r   )r?   r-   r   r   rW   r   r   r   test_excel_cell_error_na  s   z$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   rs   rt   rO   r_   r   r   rl   rQ   r`   rn   ra   Fr{   r`   ra   
skipfooter)
r~   r   r!   r   r   r;   r:   r   r   ilocr   r   r   r   test_excel_table  s    zTestReaders.test_excel_tablec           
      C   sP  |dkr|j tjjdd tg dg dg dg dg d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|j|jd df< tj	|| dd}t
|| t|jD ]\}}tj	|| d|d}||}	t
||	 qq|d t|d< tj	|| ddtid}t
|| d S )Nr   rs   rt   )rQ   rz      r   )g      ?g      @gHzG?gQ?g&.>)TFTTF)rQ   rz   ri   r      )r   ri   cdei  
         iq  rQ         i  ri   )IntColFloatColBoolColStrColStr2ColDateCol
test_typesr_   r   g      @r   r   r   r`   
converters)r~   r   r!   r   r   r	   	from_dictr   r;   r:   r   r   copylocr   	enumeratery   r   applystr)
r?   r-   r   rW   basenameactualfloat_expectedicolr   expr   r   r   test_reader_special_dtypes  sL   






z&TestReaders.test_reader_special_dtypesc              
   C   sz   d}t g ddtjdddgg ddtjd	d
dgd}dd dd dd dd d}tj|| d|d}t|| d S )Ntest_converters)rQ   rz   r   r   g      )@gL2@g3333333@g:0y5>)Foundr   r   	Not foundr   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>  s    z4TestReaders.test_reader_converters.<locals>.<lambda>c                 S   s   | rd|  S t jS )Nr   )r   r   r   r   r   r   r     s    c                 S   s   | dkrdS dS )Nr   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   rz   ri   r_   r   )r	   r   r   r   r;   r:   r   r   )r?   r   r   rW   r   r   r   r   r   test_reader_converters  s"   
	
z"TestReaders.test_reader_convertersc                 C   s   d}t || }tg dg dg dddtjdgdjg dd}t|| t j|| d	d
tdd}|d 	d	|d< |d 	d
|d< g d|d< t|| d}t
jt|d t j|| ddid W d    d S 1 stw   Y  d S )N	testdtyperQ   rz   ri   r         @      @      @      @      ?       @      @r   r   r   r   rx   float64float32)r   r   r   dtyper   r   001002003004r   z(Unable to convert column d to type int64r]   r   int64)r;   r:   r	   r   r   r   r   r   r   astyper!   re   rp   )r?   r   r   r   rW   rg   r   r   r   test_reader_dtype  s.   	"zTestReaders.test_reader_dtypezdtype,expectedr   r   r  r  r  r  r  r  r  r	  r   2r   c                 C   s&   d}t j|| |d}t|| d S )Nr   r  r;   r:   r   r   )r?   r   r  rW   r   r   r   r   r   test_reader_dtype_str  s   z!TestReaders.test_reader_dtype_strc                    s  |dv rt d| d ttddgddtdd	gd
dtddgddtddgddttjdgddttjdgd
dttjdgddttjdgddttdgd ttjtjgddd
t|}j	|ddd tj
|d|d}W d    n1 sw   Y  |dkrdd lddlm  t fddjD } |d jjjdd |d<  d d g|d!< n}t|| d S )"Nr   r   No engine for filetype: ''rQ   ri   Int64r  r   r   Float64TFbooleanr   r   string   g      @z
2019-12-31rz   )
r   r   r   r   r   rU   ghijtestr   r`   dtype_backendpyarrowr   )ArrowExtensionArrayc                    s$   i | ]}| j | d dqS )T)from_pandas)array)r&   colr%  dfpar   r   
<dictcomp>\  s    z2TestReaders.test_dtype_backend.<locals>.<dictcomp>r  us)unitr  )r!   skipr	   r   r;   NA	Timestampr   ensure_cleanto_excelr:   r$  pandas.arraysr%  ry   r'  	_pa_arraycast	timestampr   )r?   r   r#  	file_pathrV   rW   r   r)  r   test_dtype_backend>  sF   zTestReaders.test_dtype_backendc                 C   s   |dv rt d| d ttjdgdtjgd}t|}|j|ddd	 tj	|dd
dd}W d    n1 s;w   Y  t
|| d S )Nr  r  r  r  r   r   r   r   Fr!  numpy_nullabler  )r`   r#  r  )r!   r/  r	   r   r   r   r2  r3  r;   r:   r   )r?   r   r*  r8  rV   r   r   r   test_dtype_backend_and_dtypek  s   z(TestReaders.test_dtype_backend_and_dtypec              	   C   sH  |dv rt d| d t d}td| ttjddgtjdtjd	tj	gtjdd
}t
|}|j|ddd tj|ddd}W d    n1 sQw   Y  |dkrwtttjddgtjdttjd	tj	gtjdd
}ntt|ddgt|d	d gd
}t
|| W d    d S 1 sw   Y  d S )Nr  r  r  r$  zmode.string_storager   r   r  r   r:  r   Fr!  r;  r"  python)r!   r/  importorskipr;   option_contextr	   r   r'  object_r0  r   r2  r3  r:   r   r   r   )r?   r   string_storager+  r*  r8  rV   rW   r   r   r   test_dtype_backend_string{  s8   
"z%TestReaders.test_dtype_backend_stringzdtypes, exp_valuea.1r  rQ   c           	      C   s\   d}dt i|}| }tj|| |d}tdg|gd}||ks&J dt|| d S )Ndf_mangle_dup_col_dtypesr   r  r   )r   rC  zdtype dict changed)r   r   r;   r:   r	   r   r   )	r?   r   dtypes	exp_valuer   
dtype_dictdtype_dict_copyrV   rW   r   r   r   test_dtype_mangle_dup_cols  s   z&TestReaders.test_dtype_mangle_dup_colsc                 C   s2   d}t || }tdg di}t|| d S )Ntest_spacestestcol)zthis is greatz4    spacesz1 trailing z
 1 leadingz2  spaces  multiple  timesr   )r?   r   r   r   rW   r   r   r   test_reader_spaces  s   zTestReaders.test_reader_spaceszbasename,expectedzgh-35802COLUMNzTest (1)zgh-36122z
got 2nd sarx   c                 C   s6   |dkrt d|  t|| }t|| d S )Nr   zSkipped for engine: )r!   r/  r;   r:   r   r   )r?   r   r   r   rW   r   r   r   r   test_read_excel_ods_nested_xml  s   	z*TestReaders.test_read_excel_ods_nested_xmlc                 C   sF   d}t j|| d d}g d}t||  |t| ks!J d S )Ntest_multisheetr   )CharlieAlphaBeta)r;   r:   r   assert_contains_allkeyslistr?   r   r   dfsexpected_keysr   r   r   test_reading_all_sheets  s
   z#TestReaders.test_reading_all_sheetsc                 C   sV   d}g d}t j|| |d}tt|}t||  t|t| ks)J d S )NrO  )rz   rP  rP  r   )r;   r:   rU  setr   rS  rT  r   )r?   r   r   rX  rW  r   r   r   %test_reading_multiple_specific_sheets  s   z1TestReaders.test_reading_multiple_specific_sheetsc                 C   s2   d}t j|| d d}g d}t||  d S )Nblank_with_headerr   )r_   rl   r   )r;   r:   r   rS  rT  rV  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blankr_   r   )r;   r:   r   r   r	   )r?   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 )Ncol_1col_2rx   r\  r_   r   )r	   r;   r:   r   r   )r?   r   rW   r   r   r   r   !test_read_excel_blank_with_header  s   z-TestReaders.test_read_excel_blank_with_headerc                 C   s   t jtdd tjd| dgd d W d    n1 sw   Y  t jtdd tjd| dd d d	 W d    d S 1 sAw   Y  d S )
Nz \(sheet: Sheet1\)$r]   r\  rQ   )headerr`   rO   c                 S   s   dd S )NrQ   r   r   r   r   r   r   r          zHTestReaders.test_exception_message_includes_sheet_name.<locals>.<lambda>)rk   r`   )r!   re   rp   r;   r:   ZeroDivisionErrorr?   r   r   r   r   *test_exception_message_includes_sheet_name  s   "z6TestReaders.test_exception_message_includes_sheet_namez-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r5|j tjjdd |d u rH|dv rH|j tjjdd td| }t	
|| d S )Nr   rs   rt   z
2016-03-12zMarc Johnsonz
2016-03-16z
Jack Blackg@xDzTimothy BrownDateColWithBigInt	StringColrx   r   zMaybe not supported by openpyxl)r   r   z)Defaults to openpyxl, maybe not supportedtestdateoverflow)r~   r   r!   r   r   r	   r;   r1  r:   r   r   )r?   r-   r   r   rW   rV   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   rs   rt   rO   r_   r   r   )ra   r`   Fr{   r   )r?   r-   r   r   filenamer`   r   r   r   r   r   test_sheet_name#  s   
zTestReaders.test_sheet_namec                 C   sd   d| }t j|ddd}t|d}t j|ddd}t|| W d    d S 1 s+w   Y  d S )NrO   r_   r   r   rP   )r;   r:   rS   r   r   )r?   r   pthrW   rU   r   r   r   r   test_excel_read_buffer5  s   "z"TestReaders.test_excel_read_bufferc                 C   sF   d}t jtdd tjd|d W d    d S 1 sw   Y  d S )NrY   zUnknown engine: foor]   r   r6   ro   )r?   
bad_enginer   r   r   test_bad_engine_raises<  s   "z"TestReaders.test_bad_engine_raisesr`   ri   r   r   r_   c                 C   sJ   d}t jt|d tjd| |d W d    d S 1 sw   Y  d S Nz?Worksheet index 3 is invalid|Worksheet named 'Sheet4' not foundr]   r^  r   ro   )r?   r   r`   rg   r   r   r   test_bad_sheetname_raisesA  s   "z%TestReaders.test_bad_sheetname_raisesc                 C   sV   d| }d g d}tjt|d t| W d    d S 1 s$w   Y  d S )NrY   |)z(No such file or directoryu   没有那个文件或目录zFile o directory non esistente)r]   )joinr!   re   FileNotFoundErrorr;   r:   )r?   r   bad_filer^   r   r   r   test_missing_file_raisesK  s   
"z$TestReaders.test_missing_file_raisesc                 C   s~   d}|d u rt }d}n|dkrddlm} |}d}nt}d}tj||d tt| W d    d S 1 s8w   Y  d S )	Ns   foozLExcel file format cannot be determined, you must specify an engine manually.r   r   )	XLRDErrorzFUnsupported format, or corrupt file: Expected BOF record; found b'foo'zFile is not a zip filer]   )	rp   r   rz  r   r!   re   r;   r:   r   )r?   r   
bad_streamerrorrg   rz  r   r   r   test_corrupt_bytes_raisesX  s   "z%TestReaders.test_corrupt_bytes_raisesc                 C   sh   t d| d}|j| d W d    n1 sw   Y  t|j}td| }t|| d S )NrO   rP   )content)rS   serve_contentreadr;   r:   urlr   r   )r?   
httpserverr   rU   	url_tablelocal_tabler   r   r   test_read_from_http_urln  s   z#TestReaders.test_read_from_http_urlc                 C   s~   t d| d}|jd| |d W d    n1 sw   Y  d|j d| }tj||d}td| }t|| d S )NrO   rP   KeyBodys3:///test1)storage_options)rS   
put_objectr   r;   r:   r   r   )r?   r   s3_public_buckets3sorU   r  r  r  r   r   r   test_read_from_s3_urlw  s   z!TestReaders.test_read_from_s3_urlc           	      C   s   t d| d}|jd| |d W d    n1 sw   Y  dd l}|jdi |}| d|j d| }t|}W d    n1 sHw   Y  td| }t|| d S )NrO   rP   r  r   r  r  r   )	rS   r  s3fsS3FileSystemr   r;   r:   r   r   )	r?   r   r  r  rU   r  s3r  r  r   r   r   test_read_from_s3_object  s   z$TestReaders.test_read_from_s3_objectc                 C   s|   t j|dddd| }t|}z	td| }W n ty5   dt  }t	
d|  Y nw t|| d S )Nr7   r8   r9   rO   zfile://localhost/ zfailing on )ospathrv  r;   r:   r   platformunamestripr!   r/  r   r   )r?   r   r@   
localtabler  r  platform_infor   r   r   test_read_from_file_url  s   
z#TestReaders.test_read_from_file_urlc                 C   sD   d| }t j|ddd}td| }t j|ddd}t|| d S )NrO   r_   r   r   )r;   r:   r   r   r   )r?   r   str_pathrW   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   )localrO   r_   r   )	py.pathr  r  r  rv  r;   r:   r   r   )r?   r   	LocalPathr  rW   r  r   r   r   r   test_read_from_py_localpath  s   z'TestReaders.test_read_from_py_localpathc                 C   sZ   t jd| }t|d}tj|ddd}~|  W d    d S 1 s&w   Y  d S )NrO   rP   r_   r   r   )r  r  rv  rS   r;   r:   r  )r?   r   r  rU   r   r   r   r   test_close_from_py_localpath  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$d%d&td'd(d)gi}tj	d*| d+d,}t
|| tj	d-| d+d,}t
|| d S ).Nr   rs   rt   TimerQ   rz   ri   -   8   i r      1   i@ r     *   i    9   #   i 	   )      i           i'	 r   i`
 5   i 5    %   r   i       6   
times_1900r_   r   
times_1904)r~   r   r!   r   r   r	   r   r   r;   r:   r   r   )r?   r-   r   r   rW   r   r   r   r   test_reader_seconds  s4   

zTestReaders.test_reader_secondsc              	   C   s:  |dkr|j tjjdd tddgddgg}d| }td	d
t	ddgddt	ddgddt	ddgddt	ddgg|d}tj
|ddd	gdd}t|| ||_g d|_tj
|ddd	gd}tj||dd ||_tj
|ddd	gdd	gd }tj||dd g d|_|d!d"g|_tj
|d#dd	gd}t|| ttd|_|d$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   rs   rt   rY   barr   r   testmultiindexrQ   r   
2015-01-01Trz   r   
2015-01-02Fri   r   
2015-01-03r   r   
2015-01-04rx   	mi_columnr   )r`   rc  ra   r  mi_indexr   r{   bothr`   ra   rc  ilvl1ilvl2mi_index_namec1c2mi_column_name)levelname_with_int	both_nameboth_name_skiprows)r`   ra   rc  rn   )r~   r   r!   r   r   r   from_productr	   r;   r1  r:   r   r   r   ry   	set_namesrU  r   
set_levels)r?   r-   r   mimi_filerW   r   r   r   r   test_read_excel_multiindex  st   


z&TestReaders.test_read_excel_multiindexzsheet_name,idx_lvl2both_name_blank_after_mi_namer   r   both_name_multiple_blanksr   c           	   	   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
g d|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 (GH4679rt   r  rY   r  r   r   r  r  r   rQ   r   r  Trz   r   r  Fri   r   r  r   r   r  )rY   rY   r  r  r  r  r   r   r  )r~   r   r!   r   r   r   r  r	   r;   r1  from_arraysr:   r   r   )	r?   r-   r   r`   idx_lvl2r  r  rW   rV   r   r   r   +test_read_excel_multiindex_blank_after_name;  s6   
z7TestReaders.test_read_excel_multiindex_blank_after_namec                 C   sP   d| }t j|dddgd}tddg}tg dgd	 |d
}t|| d S )Nr  index_col_noner   rQ   r`   rc  r   rv   keyvalr   rz   rx   )r;   r:   r   r  r	   r   r   )r?   r   r  rV   exp_columnsrW   r   r   r   &test_read_excel_multiindex_header_onlyd  s
   z2TestReaders.test_read_excel_multiindex_header_onlyc           	      C   s  d| }t jt jt jt jt jt jgg dg dg dg dg dgtd}g d}tg d	g d
gg dg dgd d gd}tg d	d d}t|||d}tj|ddd}t	
|| ||_tj|dddgd}t	
|| t g dg dg dg dg dg}g d}tg dg dgg dg dgd d gd}tg dd d}t|||d}tj|ddd}t	
|| ||_tj|dddgd}t	j
||dd d S )Ntest_index_name_pre17)R0C0R0C1R0C2R0C3R0C4)R1C0R1C1R1C2R1C3R1C4)R2C0R2C1R2C2R2C3R2C4)R3C0R3C1R3C2R3C3R3C4)R4C0R4C1R4C2R4C3R4C4r  )C_l0_g0C_l0_g1C_l0_g2C_l0_g3C_l0_g4)R0R_l0_g0R_l0_g1R_l0_g2R_l0_g3R_l0_g4)R1R_l1_g0R_l1_g1R_l1_g2R_l1_g3R_l1_g4)r   rQ   rz   ri   r   r   r   r   r   ry   single_namesr   r   multi_namesrQ   )r  r  r  r	  r
  )r  r  r  r  r  )r   rQ   rz   ri   r   single_no_namesmulti_no_namesFr{   )r   r'  r   objectr   r
   r	   r;   r:   r   r   r   )	r?   r   rm  r8   ry   r  sirW   r   r   r   r   test_excel_old_index_formato  sf   	z'TestReaders.test_excel_old_index_formatc              	   C   sR   d}dD ]"}t jt|d tjd| |d W d    n1 s!w   Y  qd S )Nz#Passing a bool to header is invalid)TFr]   rO   rc  r   )r?   r   rg   argr   r   r   test_read_excel_bool_header_arg  s   z+TestReaders.test_read_excel_bool_header_argc              	   C   sJ  |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g d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g dd}tddtddgddtddgddtddggg dd}t	
|| d S )Nr   rs   rt   testskiprowsskiprows_listr   rz   r`   rn   rQ   r   r  Tr   r  Fri   r   r  r   r   r  r  rx   c                 S      | dv S N)r   rz   r   r   r   r   r   r     rd  z6TestReaders.test_read_excel_skiprows.<locals>.<lambda>)r`   rn   r   )r~   r   r!   r   r   r;   r:   r	   r1  r   r   r   r'  r?   r-   r   r   rW   r   r   r   test_read_excel_skiprows  sX   		z$TestReaders.test_read_excel_skiprowsc                 C   sv   |dkr|j tjjdd tjd| ddd d}td	d
tddgddtddggg dd}t	
|| d S )Nr   rs   rt   r  r  c                 S   s   | dvS )N)rQ   ri   r   r   r   r   r   r   r     rd  zFTestReaders.test_read_excel_skiprows_callable_not_in.<locals>.<lambda>r  rQ   r   r  Tri   r   r  Fr  rx   )r~   r   r!   r   r   r;   r:   r	   r1  r   r   r!  r   r   r   (test_read_excel_skiprows_callable_not_in  s$   	z4TestReaders.test_read_excel_skiprows_callable_not_inc                 C   s@   d}t jd| |d}t d| }|d | }t|| d S )Nr   rO   nrowsr  )r?   r   num_rows_to_pullr   rW   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 )NrO   r   r$  )r;   r:   r   r   r   )r?   r   rW   num_records_in_filer&  r   r   r   r   0test_read_excel_nrows_greater_than_nrows_in_file  s
   z<TestReaders.test_read_excel_nrows_greater_than_nrows_in_filec                 C   sJ   d}t jt|d tjd| dd W d    d S 1 sw   Y  d S )Nz'nrows' must be an integer >=0r]   rO   r   r$  ro   rq   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,skiprowsr  r  r  r  r  r  r  c                 C   r  r   r   r   r   r   r   r   -  rd  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]).
        )r`   rc  ra   rn   Nri   )r`   rc  ra   rn   r%  )r;   r:   r   r   r   )	r?   r   rm  r`   rc  ra   rn   rW   r   r   r   r   test_read_excel_nrows_params%  s$   z(TestReaders.test_read_excel_nrows_paramsc                 C   sF   t jtdd td| dd W d    d S 1 sw   Y  d S )Nzbut 3 positional argumentsr]   rO   r_   r   r   rf  r   r   r   test_deprecated_kwargsI  s   "z"TestReaders.test_deprecated_kwargsc                 C   sV   d| }g d}t jg ddd}t||dd}tj|dd	d
gd d}t|| d S )Nr  ))rv   rv   r  ri   r   r-  ))r   r   r  rQ   rz   r.  )r   rQ   r  )rz   ri   r  r  r   rQ   r  )r   from_tuplesr	   r;   r:   r   r   )r?   r   	file_namer8   idxrW   rV   r   r   r   "test_no_header_with_list_index_colM  s   z.TestReaders.test_no_header_with_list_index_colc                 C   s>   d| }dt jddg}t|dgd}t|}t|| d S )None_col_blank_lineg      ?rQ   rz   numbersrx   )r   r   r	   r;   r:   r   r   )r?   r   r0  r8   rW   rV   r   r   r   test_one_col_noskip_blank_lineZ  s
   
z*TestReaders.test_one_col_noskip_blank_linec                 C   sj   d| }t ddg}tjtjgtjtjgddgddgg}t||d}tj|d	d
dgd}t|| d S )Nr  )r   r   )r   rv   rQ   ri   rz   r   rx   mi_column_empty_rowsr   r  )	r   r/  r   r   r	   r;   r:   r   r   )r?   r   r0  ry   r8   rW   rV   r   r   r    test_multiheader_two_blank_linesb  s   $
z,TestReaders.test_multiheader_two_blank_linesc                 C   s$   d| }t |}|jdksJ dS )z
        Sheets can contain blank cells with no data. Some of our readers
        were including those cells, creating many empty rows and columns
        trailing_blanks)ri   ri   N)r;   r:   shape)r?   r   r0  rV   r   r   r   test_trailing_blanksm  s   
z TestReaders.test_trailing_blanksc                 C   v   |dkr	t d |dkr|jt jjdd t jtdd tj	d| d	d
 W d    d S 1 s4w   Y  d S )Nr   *chartsheets do not exist in the ODF formatr   4pyxlsb can't distinguish chartsheets from worksheetsrt   z"Worksheet named 'Chart1' not foundr]   
chartsheetChart1r   
r!   r/  r~   r   r   r   re   rp   r;   r:   r?   r-   r   r   r   r   r   test_ignore_chartsheets_by_strv  s   
"z*TestReaders.test_ignore_chartsheets_by_strc                 C   r;  )Nr   r<  r   r=  rt   z0Worksheet index 1 is invalid, 1 worksheets foundr]   r>  rQ   r   r@  rA  r   r   r   test_ignore_chartsheets_by_int  s   
"z*TestReaders.test_ignore_chartsheets_by_intc                 C   sF   t jd| ddd}tg dg dg dgg dd	}t|| d S )
Ntest_decimal,rQ   )decimalrn   )rQ   gAc̝ė@g	hAABCpoig2[j@)rz   gHzG^@g{G@DEFuytgUq&?)ri   g%Cq@g^@GHIrezg)@)IdNumber1Number2Text1Text2Number3rx   r   r   r   r   r   test_euro_decimal_format  s   z$TestReaders.test_euro_decimal_format)Y__name__
__module____qualname__r!   fixturerC   rX   rh   rr   r   r   r   parametrizer   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r  r	   r   r   r   r   r  r9  r<  rB  rI  rL  rN  rY  r[  r]  r_  rb  rg  filterwarningsrk  rn  rp  rr  rt  ry  r}  network
single_cpur  tdskip_if_not_us_localer  r  slowr  r  
skip_if_nor  r  r  r  r  r  r  r  r"  r#  r'  r)  r*  r+  r,  r2  r5  r7  r:  rB  rC  rS  r   r   r   r   r2   w   s   

5(



	3
-$




 




	!V
"I7
	r2   c                
   @   s   e Zd Zdd Zejdddd Zdd Zd	d
 Zej	
dg d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/d0 Zd1d2 Zd3S )4TestExcelFileReadc              	   C   s   d}t jt|d1 td| d}tj| |d W d    n1 s%w   Y  W d    d S W d    d S 1 s=w   Y  d S )NzPassing bytes to 'read_excel' is deprecated and will be removed in a future version. To read from a byte string, wrap it in a `BytesIO` object.r]   rO   rP   r6   )r   assert_produces_warningFutureWarningrS   r;   r:   r  )r?   r   r   rg   rU   r   r   r   test_deprecate_bytes_input  s   "z,TestExcelFileRead.test_deprecate_bytes_inputTr3   c                 C   r5   )zH
        Change directory and set engine for ExcelFile objects.
        r6   r7   r8   r9   rR   N)r   r;   rR   r<   r=   r>   r   r   r   rC     rD   z#TestExcelFileRead.cd_and_set_enginec                 C   sr   dddddd}t d| }|j}W d    n1 sw   Y  |d ur)|}n||dd   }||ks7J d S )Nr   r   r   r   rI   rO   rQ   )r;   rR   r   )r?   r   r   rT   r9   rV   rW   r   r   r   rX     s   z"TestExcelFileRead.test_engine_usedc                 C   s  t d| }t j|dddgd}W d    n1 sw   Y  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 d    n1 sUw   Y  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 d    n1 sw   Y  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 d    n1 sw   Y  ttjgdgtjgtjgdggd	gd
}t|| d S )Ntest4r_   Fappler`   keep_default_na	na_valuesr0  rQ   rabbitr   rx   Ttest51.#QNANr   )r;   rR   r:   r	   r   r   r   r   )r?   r   r9   r   rW   r   r   r   test_excel_passes_na  sH   

 

 z&TestExcelFileRead.test_excel_passes_na	na_filter)NTFc                 C   s   i }|d ur
||d< t d| }t j|fdddgd|}W d    n1 s*w   Y  |du r@d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 )Nrm  rj  r_   Tre  rf  Frk  rQ   r   ri  r   rx   )r;   rR   r:   r   r   r	   r   r   )r?   r   rm  rF   r9   r   rW   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 d    n1 s3w   Y  tj	||d	d
 tj	||d	d
 td| }|j
ddd}|j
ddgdd}W d    n1 sjw   Y  tj	||d	d
 tj	||d	d
 td| }tj|dddd}W d    n1 sw   Y  t	||jd d  td| }|j
dddd}W d    n1 sw   Y  t	||jd d  d S )Nr   rs   rt   rO   r   r   rQ   r   Fr{   )ra   )rn   ra   r   r   )ra   r   )r~   r   r!   r   r   r;   rR   r:   r   r   rH   r   )r?   r-   r   r   r9   r   r   r   r   r   r   test_excel_table_sheet_by_index  s4   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 d    n1 s,w   Y  t|| }|jd|d}W d    n1 sJw   Y  tj	||d	d
 tj	||d	d
 d S rl  )
r~   r   r!   r   r   r;   rR   rH   r   r   )	r?   r-   r   r   rm  r`   r9   	df1_parse	df2_parser   r   r   rn  $  s    z!TestExcelFileRead.test_sheet_namer`   ri   r   r   r_   c              	   C   s   d}t jt|d. td| }|j|d W d    n1 s"w   Y  W d    d S W d    d S 1 s:w   Y  d S rs  )r!   re   rp   r;   rR   rH   )r?   r   r`   rg   r9   r   r   r   rt  8  s   "z+TestExcelFileRead.test_bad_sheetname_raisesc              	   C   s   d| }t j|dd|d}t|d%}t |}t j|ddd}W d    n1 s+w   Y  W d    n1 s:w   Y  t|| d S )NrO   r_   r   r`   ra   r   rP   r   )r;   r:   rS   rR   r   r   )r?   r   r   ro  rW   rU   rM   r   r   r   r   rp  C  s   z(TestExcelFileRead.test_excel_read_bufferc              	   C   sx   t d| d&}t|}tj|dd|d W d    n1 s!w   Y  W d    n1 s0w   Y  |js:J d S )NrO   rP   r_   r   rr  )rS   r;   rR   r:   closed)r?   r   r   rU   rJ   r   r   r   test_reader_closes_fileM  s   z)TestExcelFileRead.test_reader_closes_filec              	   C   s   d}t d| /}tjt|d t j|dd W d    n1 s#w   Y  W d    d S W d    d S 1 s;w   Y  d S )Nz8Engine should not be specified when passing an ExcelFilerO   r]   rY   r6   )r;   rR   r!   re   rp   r:   )r?   r   rg   xlr   r   r   test_conflicting_excel_enginesU  s   "z0TestExcelFileRead.test_conflicting_excel_enginesc                 C   sj   t jd| |d}td| d}| }W d    n1 sw   Y  t jt||d}t|| d S )NrO   r6   rP   )r;   r:   rS   r  r   r   r   )r?   r   r   rW   rU   r8   r   r   r   r   test_excel_read_binary]  s   
z(TestExcelFileRead.test_excel_read_binaryc                 C   sZ   t d| d}t|}W d    n1 sw   Y  tjd| |d}t|| d S )NrO   rP   r6   )rS   r;   r:   r   r   )r?   r   r   rU   rV   rW   r   r   r   %test_excel_read_binary_via_read_excelg  s
   z7TestExcelFileRead.test_excel_read_binary_via_read_excelc              	   C   s   t dd1}tjtdd tj|ddgd W d    n1 s!w   Y  W d    d S W d    d S 1 s9w   Y  d S )Nzdf_header_oob.xlsxrP   zexceeds maximumr]   r   rQ   r  )rS   r!   re   rp   r;   r:   )r?   r   rU   r   r   r   )test_read_excel_header_index_out_of_rangen  s   "z;TestExcelFileRead.test_read_excel_header_index_out_of_rangerm  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ZI2r   r  )r   zB.1I11I12r  rQ   ri   r  )r   ry   r  r_   r   r  )r
   r   r/  r	   r;   r:   r   r   )r?   rm  r1  colsrW   rV   r   r   r   test_header_with_index_colt  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 d    n1 s.w   Y  tj	t
dt
d	fgt
d t
d	 gd
}tg g |d}t|| d S )Nr   rs   rt   test_datetime_mir   rQ   )rc  ra   r   z
02/29/2020z
03/01/2020r  r  )r~   r   r!   r   r   r;   rR   r:   r   r/  to_datetimeto_pydatetimer	   r   r   )	r?   r-   r   r   rU   r9   r   expected_column_indexrW   r   r   r   test_read_datetime_multiindex  s$   z/TestExcelFileRead.test_read_datetime_multiindexc              	   C   sx   t jtdd+ td| dd W d    n1 sw   Y  W d    d S W d    d S 1 s5w   Y  d S )NzValue must be one of *r]   zio.excelz.readerabc)r!   re   rp   r;   r?  rf  r   r   r   test_engine_invalid_option  s   "z,TestExcelFileRead.test_engine_invalid_optionc                 C   st   |dkr	t d |dkr|jt jjdd td| }|jdgks(J W d    d S 1 s3w   Y  d S )Nr   r<  r   r=  rt   r>  r_   )	r!   r/  r~   r   r   r   r;   rR   sheet_names)r?   r-   r   r   r9   r   r   r   test_ignore_chartsheets  s   
"z)TestExcelFileRead.test_ignore_chartsheetsc              
   C   s   t f}|d u rtd|  n|dkrdd l}t |jjf}td| B}t|j	ddd t
d z	tj||d W n	 |yH   Y nw W d    n1 sSw   Y  W d    d S W d    d S 1 skw   Y  d S )	NzInvalid test for engine=r   r   corruptzutf-8)encodingFr6   )r   r!   r/  r   biffhrz  r   r2  r   
write_textra  r;   rR   )r?   r   r   errorsr   filer   r   r   test_corrupt_files_closed  s&   "z+TestExcelFileRead.test_corrupt_files_closedN)rT  rU  rV  rc  r!   rW  rC   rX   rl  r   rX  rn  ro  rn  rt  rp  rt  rv  rw  rx  ry  r  r  r  r  r  r   r   r   r   r`    s6    

&





r`  )0r   r   	functoolsr   r7   r   r  pathlibr   r  rc   urllib.errorr   zipfiler   numpyr   r!   pandas.util._test_decoratorsutil_test_decoratorsr\  pandasr;   r	   r
   r   r   pandas._testing_testingr   pandas.core.arraysr   r   r%   r"   r_  engine_paramsr   boolr    r$   rW  r.   r   r   r2   r`  r   r   r   r   <module>   sx    

	

          3