U
    _{fs                  	   @   s  d dl mZ d dlmZ d dlmZ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Z d dlm  mZ d dlZd d	lmZmZmZmZmZm Z m!Z!m"Z" d dl#m$Z% d d
l&m'Z'm(Z( d dl)m*Z* d dl+Zd dl+m,Z, ej-ddddgddd Z.dd Z/e0de0ddd Z1dd Z2e0de0de0ddd Z3ej45dej6de0de0dgd ej6de0dd gG d!d" d"Z7dS )#    )partial)reload)BytesIOStringION)Path)Iterator)URLError)is_platform_windows)NA	DataFrame
MultiIndexSeries	Timestamp
date_rangeread_csvto_datetime)ArrowStringArrayStringArray)file_path_to_url	read_htmlzchinese_utf-16.htmlzchinese_utf-32.htmlzchinese_utf-8.htmlzletz_latin1.html)paramsc                 C   s   |ddd| j S )z6Parametrized fixture for HTML encoding test filenames.iodataZhtml_encoding)param)requestdatapath r   X/var/www/html/Darija-Ai-API/env/lib/python3.8/site-packages/pandas/tests/io/test_html.pyhtml_encoding_file*   s    
r   c                 O   s   t | t |ks,tdt |  dt | d}ttdd | |}|sPt|t| |D ]*\}}tj||f|| |jrZtdqZd S )Nz*lists are not of equal size len(list1) == z, len(list2) == z$not all list elements are DataFramesc                 S   s   t | tot |tS N)
isinstancer   )xyr   r   r   <lambda>@       z(assert_framelist_equal.<locals>.<lambda>zframes are both empty)lenAssertionErrorallmapziptmassert_frame_equalempty)Zlist1Zlist2argskwargsmsgZboth_framesZframe_iZframe_jr   r   r   assert_framelist_equal7   s    r1   bs4html5libc              	   C   sJ   dd l }| |dd tjtdd t|dddd	d
d W 5 Q R X d S )Nr   __version__z4.2zPandas requires versionmatchr   r   html	spam.htmlr2   flavor)r2   setattrpytestraisesImportErrorr   )Zmonkeypatchr   r2   r   r   r   test_bs4_version_failsK   s    r?   c               	   C   s@   d} d}d| d }t jt|d t| d|d W 5 Q R X d S )Nz
google.comzinvalid flavorz\{z \} is not a valid set of flavorsr5   Zgoogler6   r:   r<   r=   
ValueErrorr   )urlr:   r0   r   r   r   test_invalid_flavorU   s
    rD   lxmlc                 C   s<   | dddd}t |ddgd}t |ddgd}t|| d S )	Nr   r   r7   valid_markup.htmlr   rE   )	index_colr:   r2   r   r1   )r   filenameZdfs_lxmlZdfs_bs4r   r   r   test_same_ordering^   s    rJ   r:   )Zmarksc                   @   s$  e Zd Zejdd Zejdd Zejdddd Zd	d
 Zdd Z	ej
jejddddd Zej
jejddddd Zej
jdd Zdd Zdd Zdd Zdd Zdd Zd d! Zd"d# Zd$d% Zd&d' Zd(d) Zd*d+ Zd,d- Zd.d/ Zd0d1 Zd2d3 Zd4d5 Z d6d7 Z!d8d9 Z"d:d; Z#d<d= Z$ej
jejd>d? Z%ej
jej
jejd@dA Z&ej
jdBdC Z'ej
jdDdE Z(dFdG Z)ej
jdHdI Z*ej
jdJdK Z+ej
jdLdM Z,ej
jdNdO Z-ej
jdPdQ Z.ej
jdRdS Z/ej
jdTdU Z0dVdW Z1ej
jejdXdddYdZ Z2ej
jejdXddd[d\ Z3d]d^ Z4d_d` Z5dadb Z6dcdd Z7dedf Z8dgdh Z9ej
jdidj Z:ej
jdkdl Z;dmdn Z<dodp Z=dqdr Z>dsdt Z?dudv Z@dwdx ZAdydz ZBd{d| ZCd}d~ ZDdd ZEdd ZFdd ZGdd ZHej
Idddgdd ZJdd ZKdd ZLdd ZMdd ZNdd ZOdd ZPdd ZQej
jdd ZRdd ZSdd ZTej
IddeUdgdfdeUdgeUdgfgdd ZVej
Wddd ZXdd ZYdd ZZej
jdd Z[dd Z\dd Z]ej
Idddddgdd Z^dd Z_dd Z`dd ZadS )TestReadHtmlc                 C   s   |ddddS )Nr   r   r7   r8   r   selfr   r   r   r   	spam_datap   s    zTestReadHtml.spam_datac                 C   s   |ddddS )Nr   r   r7   banklist.htmlr   rL   r   r   r   banklist_datat   s    zTestReadHtml.banklist_dataT)Zautousec                 c   s   t t|d| _d V  d S )Nr9   )r   r   )rM   r:   r   r   r   set_defaultsx   s    zTestReadHtml.set_defaultsc                 C   sV   t jdddd ddddjt}| }| j|dd	id
dd
 }t || d S )N      c                  W   s
   t j S r    )nprandomZrand)r.   r   r   r   r$      r%   z2TestReadHtml.test_to_html_compat.<locals>.<lambda>F)Z
data_gen_fZc_idx_namesZr_idx_namesz{:.3f}classZ	dataframer   )attrsrG   )	r+   ZmakeCustomDataframeapplymapformatZastypefloatto_htmlr   r,   )rM   dfoutresr   r   r   test_to_html_compat}   s    z TestReadHtml.test_to_html_compatc                    s  t tdtjdgddtdddgddtdtjdgddtdd	dgddd
dd gd
dd
gdddgddd gd}|dkrttjdddgtjd}ttjddtgtjd}n2t	dt
dddg}t
ddd g}|jdd}td| | j||dd }W 5 Q R X t tdtjdgddtdddgddtdtjdgddtdd	dgddtd
dtgddtd
dd
gdd||d|dkrdd lddlm  t  fddjD t| d S )N   rS   ZInt64)dtype   g      ?g      @ZFloat64g       @TFabc)rc   rd   re   defghpythonpyarrowindexzmode.string_storagedtype_backendr   boolean)ArrowExtensionArrayc                    s$   i | ]}| j | d dqS )T)Zfrom_pandas)array).0colrr   expectedpar   r   
<dictcomp>   s    z3TestReadHtml.test_dtype_backend.<locals>.<dictcomp>)r   r   rT   nanr   rs   Zobject_r
   r<   Zimportorskipr   r[   pdZoption_contextr   rl   Zpandas.arraysrr   columnsr+   r,   )rM   Zstring_storagerp   r\   Zstring_arrayZstring_array_nar]   resultr   rv   r   test_dtype_backend   sP    

zTestReadHtml.test_dtype_backendThttps://www.fdic.gov/resources/resolutions/bank-failures/failed-bank-list/index.html)rC   Zcheck_before_testc                 C   s.   d}| j |dd}| j |dd}t|| d S )Nr   First Federal Bank of Floridar5   Metcalf BankrH   rM   rC   df1df2r   r   r   test_banklist_url   s    	zTestReadHtml.test_banklist_url\https://raw.githubusercontent.com/pandas-dev/pandas/main/pandas/tests/io/data/html/spam.htmlc                 C   s.   d}| j |dd}| j |dd}t|| d S )Nr   	.*Water.*r5   UnitrH   r   r   r   r   test_spam_url   s
    
zTestReadHtml.test_spam_urlc                 C   s6   | j |dddid}| j |dddid}t|| d S )Nz.*Florida.*idtabler6   rW   r   rH   )rM   rP   r   r   r   r   r   test_banklist   s    zTestReadHtml.test_banklistc                 C   sV   | j |dd}| j |dd}t|| |d jd dks<t|d jd dksRtd S )Nr   r5   r   r   r   r   
ProximatesZNutrient)r   r1   ilocr'   r|   rM   rN   r   r   r   r   r   	test_spam   s
    
zTestReadHtml.test_spamc                 C   s&   |  |}|D ]}t|tstqd S r    r   r!   r   r'   )rM   rN   dfsr\   r   r   r   test_spam_no_match   s    
zTestReadHtml.test_spam_no_matchc                 C   s.   | j |ddid}|D ]}t|tstqd S )Nr   r   )rW   r   )rM   rP   r   r\   r   r   r   test_banklist_no_match  s    z#TestReadHtml.test_banklist_no_matchc                 C   s4   | j |dddd }|jd dks&t|jr0td S )Nr   rb   r6   headerr   r   )r   r|   r'   r-   )rM   rN   r\   r   r   r   test_spam_header  s    zTestReadHtml.test_spam_headerc                 C   s.   | j |ddd}| j |ddd}t|| d S Nr   r`   r6   skiprowsr   rH   r   r   r   r   test_skiprows_int  s    zTestReadHtml.test_skiprows_intc                 C   s6   | j |dtdd}| j |dtdd}t|| d S Nr   rb   r   r   )r   ranger1   r   r   r   r   test_skiprows_range  s    z TestReadHtml.test_skiprows_rangec                 C   s6   | j |dddgd}| j |dddgd}t|| d S Nr   r`   rb   r   r   rH   r   r   r   r   test_skiprows_list  s    zTestReadHtml.test_skiprows_listc                 C   s6   | j |dddhd}| j |dddhd}t|| d S r   rH   r   r   r   r   test_skiprows_set  s    zTestReadHtml.test_skiprows_setc                 C   s.   | j |ddd}| j |ddd}t|| d S r   rH   r   r   r   r   test_skiprows_slice#  s    z TestReadHtml.test_skiprows_slicec                 C   s6   | j |dtdd}| j |dtdd}t|| d S r   r   slicer1   r   r   r   r   test_skiprows_slice_short)  s    z&TestReadHtml.test_skiprows_slice_shortc                 C   s<   | j |dtddd}| j |dtdddd}t|| d S )	Nr   rb      r   r   rR   r`   r   r   r   r   r   test_skiprows_slice_long/  s    z%TestReadHtml.test_skiprows_slice_longc                 C   s:   | j |dtdd}| j |dtdd}t|| d S r   )r   rT   Zaranger1   r   r   r   r   test_skiprows_ndarray5  s    z"TestReadHtml.test_skiprows_ndarrayc              	   C   s.   t jtdd | j|ddd W 5 Q R X d S )Nz%is not a valid type for skipping rowsr5   r   Zasdfr   )r<   r=   	TypeErrorr   )rM   rN   r   r   r   test_skiprows_invalid;  s    z"TestReadHtml.test_skiprows_invalidc                 C   s.   | j |ddd}| j |ddd}t|| d S Nr   r   r6   rG   r   rH   r   r   r   r   
test_index?  s    zTestReadHtml.test_indexc                 C   s2   | j |dddd}| j |dddd}t|| d S Nr   r`   r   )r6   r   rG   r   rH   r   r   r   r   test_header_and_index_no_typesD  s    z+TestReadHtml.test_header_and_index_no_typesc                 C   s2   | j |dddd}| j |dddd}t|| d S r   rH   r   r   r   r    test_header_and_index_with_typesI  s    z-TestReadHtml.test_header_and_index_with_typesc                 C   s.   | j |ddd}| j |ddd}t|| d S r   rH   r   r   r   r   test_infer_typesN  s    zTestReadHtml.test_infer_typesc              	   C   sr   t |dd}t| }W 5 Q R X t |dd}t| }W 5 Q R X | j|dd}| j|dd}t|| d S NzUTF-8)encodingr   r5   r   )openr   readr   r1   )rM   rN   rh   data1data2r   r   r   r   r   test_string_ioT  s    zTestReadHtml.test_string_ioc              	   C   sJ   t |dd}| }W 5 Q R X | j|dd}| j|dd}t|| d S r   )r   r   r   r1   )rM   rN   rh   r   r   r   r   r   r   test_string_  s
    zTestReadHtml.test_stringc              	   C   sZ   t |dd}| j|dd}W 5 Q R X t |dd}| j|dd}W 5 Q R X t|| d S r   )r   r   r1   )rM   rN   rh   r   r   r   r   r   test_file_likeh  s
    zTestReadHtml.test_file_likec              	   C   s,   t jtdd | jddd W 5 Q R X d S )Nz#urlopen error unknown url type: gitr5   zgit://github.comr   )r<   r=   r   r   rM   r   r   r   test_bad_url_protocolq  s    z"TestReadHtml.test_bad_url_protocolc              	   C   s4   d}t jttf|d | jddd W 5 Q R X d S )NzNName or service not known|Temporary failure in name resolution|No tables foundr5   zhttp://www.a23950sdfa908sd.comr   )r<   r=   r   rB   r   rM   r0   r   r   r   test_invalid_urlw  s    zTestReadHtml.test_invalid_urlc                 C   sN   |}| j ttj|dddid}t|ts2t|D ]}t|ts6tq6d S )NZFirstr   r   r   )	r   r   ospathabspathr!   listr'   r   rM   rP   rC   r   r\   r   r   r   test_file_url  s      zTestReadHtml.test_file_urlc              	   C   s6   |}t jtdd | j|dddid W 5 Q R X d S )NzNo tables foundr5   r   r   Z	tasdfabler   rA   )rM   rP   rC   r   r   r   test_invalid_table_attrs  s      z%TestReadHtml.test_invalid_table_attrsc                 K   s   | j |fdddid|S )NMetcalfr   r   r   r   )rM   r   r/   r   r   r   
_bank_data  s    zTestReadHtml._bank_datac                 C   s*   | j |ddgdd }t|jts&td S )Nr   r`   r   r   r!   r|   r   r'   rM   rP   r\   r   r   r   test_multiindex_header  s    z#TestReadHtml.test_multiindex_headerc                 C   s*   | j |ddgdd }t|jts&td S )Nr   r`   rG   )r   r!   rn   r   r'   r   r   r   r   test_multiindex_index  s    z"TestReadHtml.test_multiindex_indexc                 C   s@   | j |ddgddgdd }t|jts,tt|jts<td S )Nr   r`   )r   rG   )r   r!   r|   r   r'   rn   r   r   r   r   test_multiindex_header_index  s    z)TestReadHtml.test_multiindex_header_indexc                 C   s,   | j |ddgddd }t|jts(td S Nr   r`   )r   r   r   r   r   r   r   &test_multiindex_header_skiprows_tuples  s    z3TestReadHtml.test_multiindex_header_skiprows_tuplesc                 C   s,   | j |ddgddd }t|jts(td S r   r   r   r   r   r   test_multiindex_header_skiprows  s    z,TestReadHtml.test_multiindex_header_skiprowsc                 C   sB   | j |ddgddgddd }t|jts.tt|jts>td S )Nr   r`   )r   rG   r   )r   r!   rn   r   r'   r|   r   r   r   r   %test_multiindex_header_index_skiprows  s       z2TestReadHtml.test_multiindex_header_index_skiprowsc                 C   sZ   |}| j ttj|ttdddid}t|ts>t	|D ]}t|t
sBt	qBd S )NZFloridar   r   r   )r   r   r   r   r   recompiler!   r   r'   r   r   r   r   r   test_regex_idempotency  s    z#TestReadHtml.test_regex_idempotencyc              	   C   s2   d}t jt|d | j|ddd W 5 Q R X d S )Nz\(you passed a negative value\)r5   ZWaterr   r   rA   rM   rN   r0   r   r   r   test_negative_skiprows  s    z#TestReadHtml.test_negative_skiprowshttps://docs.python.org/2/c                 C   s&   d}| j |dd}t|dks"td S )Nr   Pythonr5   r`   r   r&   r'   )rM   rC   r   r   r   r   test_multiple_matches  s    z"TestReadHtml.test_multiple_matchesc                 C   s<   d}| j |dd}dd |D }t|tddgks8td S )Nr   r   r5   c                 S   s   g | ]}|j d  dd qS )r   r   rR   )r   )rt   r\   r   r   r   
<listcomp>  s     z7TestReadHtml.test_python_docs_table.<locals>.<listcomp>ZRepoZWhat)r   sortedr'   )rM   rC   r   zzr   r   r   test_python_docs_table  s    z#TestReadHtml.test_python_docs_tablec                 C   s"   d}|  |}t|dkstdS )z@
        Make sure that read_html ignores empty tables.
        a  
            <table>
                <thead>
                    <tr>
                        <th>A</th>
                        <th>B</th>
                    </tr>
                </thead>
                <tbody>
                    <tr>
                        <td>1</td>
                        <td>2</td>
                    </tr>
                </tbody>
            </table>
            <table>
                <tbody>
                </tbody>
            </table>
        r`   Nr   )rM   r7   r}   r   r   r   test_empty_tables  s    
zTestReadHtml.test_empty_tablesc                 C   s:   |  dd }tddgddggddgd	}t|| d S )
Na  <table>
            <thead>
                <tr>
                    <th>A</th>
                    <th>B</th>
                </tr>
            </thead>
            <tbody>
                <tr>
                    <td>1</td>
                    <td>2</td>
                </tr>
            </tbody>
            <tbody>
                <tr>
                    <td>3</td>
                    <td>4</td>
                </tr>
            </tbody>
        </table>r   r`   rb   rS   rR   ABr   r|   r   r   r+   r,   rM   r}   rw   r   r   r   test_multiple_tbody  s    z TestReadHtml.test_multiple_tbodyc                 C   s0   |  dd }tddidgd}t|| dS )zt
        Don't fail with bs4 when there is a header and only one column
        as described in issue #9178
        a3  <table>
                <thead>
                    <tr>
                        <th>Header</th>
                    </tr>
                </thead>
                <tbody>
                    <tr>
                        <td>first</td>
                    </tr>
                </tbody>
            </table>r   Headerfirstr   rn   Nr   r   r   r   r   test_header_and_one_column  s    z'TestReadHtml.test_header_and_one_columnc                 C   s8   |  dd }tdddggdddgd	}t|| d
S )zK
        Ensure parser adds <tr> within <thead> on malformed HTML.
        a  <table>
            <thead>
                <tr>
                    <th>Country</th>
                    <th>Municipality</th>
                    <th>Year</th>
                </tr>
            </thead>
            <tbody>
                <tr>
                    <td>Ukraine</td>
                    <th>Odessa</th>
                    <td>1944</td>
                </tr>
            </tbody>
        </table>r   ZUkraineZOdessa  ZCountryZMunicipalityZYearr   Nr   r   r   r   r   test_thead_without_tr*  s    
z"TestReadHtml.test_thead_without_trc                 C   s   d}t ddggddgd}t ddgddggddgd}|jd	d
}|jdd
}| |d }| |d }t|| t|| dS )zh
        Make sure that read_html reads tfoot, containing td or th.
        Ignores empty tfoot
        a  <table>
            <thead>
                <tr>
                    <th>A</th>
                    <th>B</th>
                </tr>
            </thead>
            <tbody>
                <tr>
                    <td>bodyA</td>
                    <td>bodyB</td>
                </tr>
            </tbody>
            <tfoot>
                {footer}
            </tfoot>
        </table>ZbodyAZbodyBr   r   r   ZfootAZfootB )footerz%<tr><td>footA</td><th>footB</th></tr>r   N)r   rY   r   r+   r,   )rM   Zdata_templateZ	expected1Z	expected2r   r   Zresult1Zresult2r   r   r   test_tfoot_readH  s     zTestReadHtml.test_tfoot_readc                 C   s4   | j dddd }tddggdd}t|| d S )Na
  
            <table>
                <tr>
                    <td>S</td>
                    <td>I</td>
                </tr>
                <tr>
                    <td>text</td>
                    <td>1944</td>
                </tr>
            </table>
        r   r   textr   )SIr|   r   r   r   r   r   &test_parse_header_of_non_string_columnn  s    z3TestReadHtml.test_parse_header_of_non_string_columnc              
      s   ddl m   fdd}| j|dddidd }t|d	d
ddttdd}|j|jks\tddddddddddg
}dddddddd d!d"g
}||||}||}	|}
d#d$g}|
| 	t
|
|< t|
|	 d S )%Nr   _remove_whitespacec                    s(   z
 | W S  t k
r"   |  Y S X d S r    )AttributeErrorr"   r   r   r   try_remove_ws  s    
z8TestReadHtml.test_banklist_header.<locals>.try_remove_wsr   r   r   r   r   r   csvzbanklist.csv)Updated DateClosing Date
convertersz,First Vietnamese American Bank In Vietnamesez"Westernbank Puerto Rico En Espanolz*R-G Premier Bank of Puerto Rico En EspanolzEurobank En EspanolzSanderson State Bank En EspanolzLWashington Mutual Bank (Including its subsidiary Washington Mutual Bank FSB)zSilver State Bank En Espanolz%AmTrade International Bank En EspanolzHamilton Bank, NA En Espanolz6The Citizens Savings Bank Pioneer Community Bank, Inc.zFirst Vietnamese American BankzWesternbank Puerto RicozR-G Premier Bank of Puerto RicoZEurobankzSanderson State BankzWashington Mutual BankzSilver State BankzAmTrade International BankzHamilton Bank, NAzThe Citizens Savings Bankr   r   )pandas.io.htmlr   r   r   r   shaper'   rX   replaceapplyr   r+   r,   )rM   rP   r   r   r\   Zground_trutholdnewZdfnewZgtnewZ	convertedZ	date_colsr   r   r   test_banklist_header  sH    
z!TestReadHtml.test_banklist_headerc              	   C   sX   d}t |}| }W 5 Q R X ||ks,t| j|dddidd }|| ksTtd S )NzGold Canyonr   r   r   r   )r   r   r'   r   Z	to_string)rM   rP   gcrh   Zraw_textr\   r   r   r   test_gold_canyon  s    
zTestReadHtml.test_gold_canyonc                 C   s4   | j dddd }| j dddd }t|| d S )Na  <table>
                        <thead>
                            <tr style="text-align: right;">
                            <th></th>
                            <th>C_l0_g0</th>
                            <th>C_l0_g1</th>
                            <th>C_l0_g2</th>
                            <th>C_l0_g3</th>
                            <th>C_l0_g4</th>
                            </tr>
                        </thead>
                        <tbody>
                            <tr>
                            <th>R_l0_g0</th>
                            <td> 0.763</td>
                            <td> 0.233</td>
                            <td> nan</td>
                            <td> nan</td>
                            <td> nan</td>
                            </tr>
                            <tr>
                            <th>R_l0_g1</th>
                            <td> 0.244</td>
                            <td> 0.285</td>
                            <td> 0.392</td>
                            <td> 0.137</td>
                            <td> 0.222</td>
                            </tr>
                        </tbody>
                    </table>r   r   a  <table>
                    <thead>
                        <tr style="text-align: right;">
                        <th></th>
                        <th>C_l0_g0</th>
                        <th>C_l0_g1</th>
                        <th>C_l0_g2</th>
                        <th>C_l0_g3</th>
                        <th>C_l0_g4</th>
                        </tr>
                    </thead>
                    <tbody>
                        <tr>
                        <th>R_l0_g0</th>
                        <td> 0.763</td>
                        <td> 0.233</td>
                        </tr>
                        <tr>
                        <th>R_l0_g1</th>
                        <td> 0.244</td>
                        <td> 0.285</td>
                        <td> 0.392</td>
                        <td> 0.137</td>
                        <td> 0.222</td>
                        </tr>
                    </tbody>
                 </table>)r   r+   r,   )rM   rw   r}   r   r   r   test_different_number_of_cols  s     "z*TestReadHtml.test_different_number_of_colsc                 C   s8   |  dd }tdddggdddgd	}t|| d S )
NaZ  
            <table>
                <tr>
                    <th>A</th>
                    <th colspan="1">B</th>
                    <th rowspan="1">C</th>
                </tr>
                <tr>
                    <td>a</td>
                    <td>b</td>
                    <td>c</td>
                </tr>
            </table>
        r   rc   rd   re   r   r   Cr   r   r   r   r   r   test_colspan_rowspan_1  s    z#TestReadHtml.test_colspan_rowspan_1c                 C   sD   | j dddd }tdddddggdd	d
ddgd}t|| d S )Na  
            <table>
                <tr>
                    <td colspan="2">X</td>
                    <td>Y</td>
                    <td rowspan="2">Z</td>
                    <td>W</td>
                </tr>
                <tr>
                    <td>A</td>
                    <td colspan="2">B</td>
                    <td>C</td>
                </tr>
            </table>
        r   r   r   r   Zr  XzX.1YWr   r   r   r   r   r    test_colspan_rowspan_copy_values  s     z-TestReadHtml.test_colspan_rowspan_copy_valuesc                 C   sD   | j dddd }tdddddggddddd	gd
}t|| d S )Na(  
            <table>
                <tr>
                    <td rowspan="2">A</td>
                    <td rowspan="2" colspan="3">B</td>
                    <td>C</td>
                </tr>
                <tr>
                    <td>D</td>
                </tr>
            </table>
        r   r   r   r   DzB.1zB.2r  r   r   r   r   r   r   test_colspan_rowspan_both_not_1=  s     z,TestReadHtml.test_colspan_rowspan_both_not_1c                 C   s8   | j dddd }tddggddgd}t|| d S )Nz
            <table>
                <tr>
                    <td>A</td>
                    <td rowspan="2">B</td>
                </tr>
                <tr>
                    <td>C</td>
                </tr>
            </table>
        r   r   r  r   r   r   r   r   r   r   r   test_rowspan_at_end_of_row[  s    z'TestReadHtml.test_rowspan_at_end_of_rowc                 C   s>   | j dddd }tddgddggddgd}t|| d S )Nz
            <table>
                <tr>
                    <td rowspan="3">A</td>
                    <td rowspan="3">B</td>
                </tr>
            </table>
        r   r   r   r   r   r   r   r   r   r   test_rowspan_only_rowsv  s    
z#TestReadHtml.test_rowspan_only_rowsc                 C   sT   |  dd }tddgddggddgddggd}tdd	gg|d
}t|| d S )Nam  
            <table>
                <tr>
                    <th>A</th>
                    <th>B</th>
                </tr>
                <tr>
                    <th>a</th>
                    <th>b</th>
                </tr>
                <tr>
                    <td>1</td>
                    <td>2</td>
                </tr>
            </table>
        r   r   r   rc   rd   r`   Zlevelscodesrb   r   r   r   r   r+   r,   rM   r}   r|   rw   r   r   r   +test_header_inferred_from_rows_with_only_th  s    $z8TestReadHtml.test_header_inferred_from_rows_with_only_thc                 C   sd   t dtdddi}| }| j|dgdd}t||d  | j|dgdd}t||d  d S )Ndate1/1/2001
   Zperiodsr`   r   Zparse_datesrG   )r   r   r[   r   r+   r,   )rM   r\   rw   r^   r   r   r   test_parse_dates_list  s    z"TestReadHtml.test_parse_dates_listc                 C   sn   t tddd}t|dd |dd d}| j| dd	d
gid	d}td|i}t||d  d S )Nr  r  r  c                 S   s   t |  S r    )strr  r   r   r   r   r$     r%   z7TestReadHtml.test_parse_dates_combine.<locals>.<lambda>c                 S   s   t |  S r    )r"  timer   r   r   r   r$     r%   )r  r#  datetimer`   rb   r   r   )r   r   r   r)   r   r[   r+   r,   )rM   Z	raw_datesr\   r^   Znewdfr   r   r   test_parse_dates_combine  s     
 z%TestReadHtml.test_parse_dates_combinec                 C   s   |dddd}t j|s,tt| dt j|sJtt| d| j|ddd	d
 }|jdksltd|jd ks~t|d j	t
	dkstt
|jd dstd S )Nr   r   r7   wikipedia_states.htmlz is not a filez is an empty fileArizonar`   r   r   )<      Unnamedr   sq mifloat64)r   r+  HzPN$A)r   r   isfiler'   reprgetsizer   r  r|   ra   rT   allcloselocrM   r   r   r}   r   r   r   test_wikipedia_states_table  s    z(TestReadHtml.test_wikipedia_states_tablec                 C   sp   |dddd}| j |dddd }|jdks0td	|jd
 d ksFt|jjdksVtt|jd dsltd S )Nr   r   r7   r&  r'  r   r   )r(     r*  r   r`   rb   )ZAlaska)zTotal area[2]r+  r-  )r   r  r'   r|   ZnlevelsrT   r1  r2  r3  r   r   r    test_wikipedia_states_multiindex  s    z-TestReadHtml.test_wikipedia_states_multiindexc                 C   sB   | j dddgd}tddggtddgd	}t|d | d S )
NaK  
                <table>
                    <thead>
                        <tr><th></th><th></tr>
                        <tr><th>A</th><th>B</th></tr>
                    </thead>
                    <tbody>
                        <tr><td>a</td><td>b</td></tr>
                    </tbody>
                </table>
            r   r`   r   rc   rd   )Unnamed: 0_level_0r   )zUnnamed: 1_level_0r   r   )r   r   r   from_tuplesr+   r,   r   r   r   r   %test_parser_error_on_empty_header_row  s    z2TestReadHtml.test_parser_error_on_empty_header_rowc                 C   sL   | j dddd }tddidgd}|d jtdks<tt|| d S )	Na  <html>
            <body>
             <table>
                <thead>
                    <tr>
                        <th>Header</th>
                    </tr>
                </thead>
                <tbody>
                    <tr>
                        <td>1100#101</td>
                    </tr>
                </tbody>
            </table>
            </body>
        </html>#)decimalr   r   gClg0@r   r,  )r   r   ra   rT   r'   r+   r,   r   r   r   r   test_decimal_rows  s    zTestReadHtml.test_decimal_rowsargFc              	   C   s6   t d}tjt|d | j||d W 5 Q R X d S )NzPassing a bool to header is invalid. Use header=None for no header or header=int or list-like of ints to specify the row(s) making up the column namesr5   r   )r   escaper<   r=   r   r   )rM   rN   r=  r0   r   r   r   test_bool_header_arg  s
    z!TestReadHtml.test_bool_header_argc                 C   s6   | j ddtidd }tdddgi}t|| d S )Na  <table>
                 <thead>
                   <tr>
                     <th>a</th>
                    </tr>
                 </thead>
                 <tbody>
                   <tr>
                     <td> 0.763</td>
                   </tr>
                   <tr>
                     <td> 0.244</td>
                   </tr>
                 </tbody>
               </table>rc   r   r   z0.763z0.244)r   r"  r   r+   r,   r   r   r   r   test_converters  s    zTestReadHtml.test_convertersc                 C   s6   | j ddgdd }tddtjgi}t|| d S )Na  <table>
                 <thead>
                   <tr>
                     <th>a</th>
                   </tr>
                 </thead>
                 <tbody>
                   <tr>
                     <td> 0.763</td>
                   </tr>
                   <tr>
                     <td> 0.244</td>
                   </tr>
                 </tbody>
               </table>gZd;?)Z	na_valuesr   rc   g"~j?r   r   rT   rz   r+   r,   r   r   r   r   test_na_values!  s    zTestReadHtml.test_na_valuesc                 C   sh   d}t dddgi}| j|ddd }t|| t dtjtjgi}| j|ddd }t|| d S )	Na  <table>
                        <thead>
                            <tr>
                            <th>a</th>
                            </tr>
                        </thead>
                        <tbody>
                            <tr>
                            <td> N/A</td>
                            </tr>
                            <tr>
                            <td> NA</td>
                            </tr>
                        </tbody>
                    </table>rc   zN/Ar
   F)Zkeep_default_nar   T)r   r   r+   r,   rT   rz   )rM   Z	html_dataexpected_dfhtml_dfr   r   r   test_keep_default_na:  s    z!TestReadHtml.test_keep_default_nac                 C   s>   |  dd }tddgtjtjggddgd}t|| d S )Nak  
            <table>
                <tr>
                    <th>A</th>
                    <th>B</th>
                </tr>
                <tr>
                    <td>a</td>
                    <td>b</td>
                </tr>
                <tr>
                    <td></td>
                    <td></td>
                </tr>
            </table>
        r   rc   rd   r   r   r   rA  r   r   r   r   test_preserve_empty_rowsS  s     z%TestReadHtml.test_preserve_empty_rowsc                 C   sT   |  dd }tddgddggddgddggd}tdd	gg|d
}t|| d S )NaU  
            <table>
                <thead>
                    <tr><th></th><th></tr>
                    <tr><th>A</th><th>B</th></tr>
                    <tr><th>a</th><th>b</th></tr>
                </thead>
                <tbody>
                    <tr><td>1</td><td>2</td></tr>
                </tbody>
            </table>
        r   r   r   rc   rd   r`   r  rb   r   r  r  r   r   r   ,test_ignore_empty_rows_when_inferring_headerk  s    $z9TestReadHtml.test_ignore_empty_rows_when_inferring_headerc                 C   sP   t dddgd}dddgdd	d
gg|_|jdd}| |d }t|| d S )N)ZHillaryD   r  )ZBernieJ   r  )ZDonaldE   R)r   r7  ZAgeZPartyNamezUnnamed: 1_level_1zUnnamed: 2_level_1Frm   r   )r   r|   r[   r   r+   r,   )rM   rC  r7   rD  r   r   r   test_multiple_header_rows  s    z&TestReadHtml.test_multiple_header_rowsc                 C   s@   |dddd}| j |dd}t|ts*tt|d ts<td S )Nr   r   r7   rF   r   r   )r   r!   r   r'   r   )rM   r   rI   r   r   r   r   test_works_on_valid_markup  s    z'TestReadHtml.test_works_on_valid_markupc                 C   s&   |dddd}| j |dddgd d S )	Nr   r   r7   rO   r   rE   r3   r@   r   )rM   r   rP   r   r   r   test_fallback_success  s    z"TestReadHtml.test_fallback_successc                 C   s:   t ddd}ttjdd|d}| }d|ks6td S )Nz
2000-01-01r  r  rR   rm   )r   r   rT   rU   Zrandnr[   r'   )rM   rngr\   r}   r   r   r   test_to_html_timestamp  s    z#TestReadHtml.test_to_html_timestampc                 C   s   t dddg}| }|jdd}|jdd}|jdd}|jdd}|jdd}d|ks`t||kslt||ksxt||kstd	|kstd
|kstd|kst||kstd S )Nr`   rb   )r   r   T)borderr   Fz border="1"z border="2"z border="0"z border)r   r[   r'   )rM   r\   Zout_border_defaultZout_border_trueZout_border_explicit_defaultZout_border_nondefaultZout_border_zeroZout_border_falser   r   r   test_to_html_borderless  s    z$TestReadHtml.test_to_html_borderlesszdisplayed_only,exp0,exp1ZfooNzfoo  bar  baz  quxc                 C   sT   t d}| j||d}t|d | |d k	r@t|d | nt|dksPtd S )Na  <html>
          <body>
            <table>
              <tr>
                <td>
                  foo
                  <span style="display:none;text-align:center">bar</span>
                  <span style="display:none">baz</span>
                  <span style="display: none">qux</span>
                </td>
              </tr>
            </table>
            <table style="display: none">
              <tr>
                <td>foo</td>
              </tr>
            </table>
          </body>
        </html>)displayed_onlyr   r`   )r   r   r+   r,   r&   r'   )rM   rT  Zexp0Zexp1r   r   r   r   r   test_displayed_only  s    	z TestReadHtml.test_displayed_onlyz\ignore:You provided Unicode markup but also provided a value for from_encoding.*:UserWarningc           
   	   C   s   t j|}t j|d }|d\}}zt|d}| j| |dd }W 5 Q R X t|d"}| jt	| |dd }W 5 Q R X | j||dd }	t
|| t
||	 W n4 tk
r   t rd|ksd|krt   Y nX d S )Nr   _rb)r   rG   Z16Z32)r   r   basenamesplitextsplitr   r   r   popr   r+   r,   	Exceptionr	   r<   skip)
rM   r   	base_pathrootrV  r   ZfobjZfrom_stringZfrom_file_likefrom_filenamer   r   r   test_encode  s8      
    
zTestReadHtml.test_encodec              	   C   sj   | j jddkrtd G dd dt}|d}|  |sBttjtdd |  | W 5 Q R X d S )	Nr:   rE   zNot applicable for lxmlc                   @   s   e Zd Zdd ZdS )zFTestReadHtml.test_parse_failure_unseekable.<locals>.UnseekableStringIOc                 S   s   dS NFr   r   r   r   r   seekable  s    zOTestReadHtml.test_parse_failure_unseekable.<locals>.UnseekableStringIO.seekableN)__name__
__module____qualname__rc  r   r   r   r   UnseekableStringIO   s   rg  z?
            <table><tr><td>spam<foobr />eggs</td></tr></table>z#passed a non-rewindable file objectr5   )	r   keywordsgetr<   r]  r   r'   r=   rB   )rM   rg  badr   r   r   test_parse_failure_unseekable  s    
z*TestReadHtml.test_parse_failure_unseekablec                 C   s>   G dd d}|d}|d}|  |s,t|  |s:td S )Nc                   @   sJ   e Zd ZddddZdddZdd Zd	d
 Zdd ZedddZ	dS )z9TestReadHtml.test_parse_failure_rewinds.<locals>.MockFileN)returnc                 S   s   || _ d| _d S rb  )r   at_end)rM   r   r   r   r   __init__  s    zBTestReadHtml.test_parse_failure_rewinds.<locals>.MockFile.__init__c                 S   s   | j r
dn| j}d| _ |S )Nr   T)rm  r   )rM   sizer   r   r   r   r     s    z>TestReadHtml.test_parse_failure_rewinds.<locals>.MockFile.readc                 S   s
   d| _ d S rb  )rm  )rM   offsetr   r   r   seek  s    z>TestReadHtml.test_parse_failure_rewinds.<locals>.MockFile.seekc                 S   s   dS )NTr   r   r   r   r   rc    s    zBTestReadHtml.test_parse_failure_rewinds.<locals>.MockFile.seekablec                 S   s   d S r    r   r   r   r   r   __next__"  s    zBTestReadHtml.test_parse_failure_rewinds.<locals>.MockFile.__next__c                 S   s   | S r    r   r   r   r   r   __iter__%  s    zBTestReadHtml.test_parse_failure_rewinds.<locals>.MockFile.__iter__)N)
rd  re  rf  rn  r   rq  rc  rr  r   rs  r   r   r   r   MockFile  s   
rt  z/<table><tr><td>spam<br />eggs</td></tr></table>z2<table><tr><td>spam<foobr />eggs</td></tr></table>)r   r'   )rM   rt  Zgoodrj  r   r   r   test_parse_failure_rewinds  s
    z'TestReadHtml.test_parse_failure_rewindsc                 C   s   G dd dt j}ttjj |dddd}|| j|fd}|| j|fd}|  |  | s\| rnq\d |j	  kr|j	ksn t
d S )Nc                       s   e Zd Z fddZ  ZS )z@TestReadHtml.test_importcheck_thread_safety.<locals>.ErrorThreadc              
      sB   zt    W n( tk
r6 } z
|| _W 5 d }~X Y nX d | _d S r    )superrunr\  err)rM   rx  	__class__r   r   rw  5  s
    zDTestReadHtml.test_importcheck_thread_safety.<locals>.ErrorThread.run)rd  re  rf  rw  __classcell__r   r   ry  r   ErrorThread4  s   r|  r   r   r7   rF   )targetr.   )	threadingThreadr   pandasr   r7   r   startis_aliverx  r'   )rM   r   r|  rI   Zhelper_thread1Zhelper_thread2r   r   r   test_importcheck_thread_safety0  s    
z+TestReadHtml.test_importcheck_thread_safetyc                 C   sB   |dddd}t |}| |d }| |d }t|| d S )Nr   r   r7   r8   r   )r   r   r+   r,   )rM   r   Zfile_path_string	file_pathr   r   r   r   r   test_parse_path_objectK  s
    z#TestReadHtml.test_parse_path_objectc                 C   s0   |  dd }tdggdgd}t|| d S )Nz
            <table>
                <tr>
                    <th>A</th>
                </tr>
                <tr>
                    <td>word1<br>word2</td>
                </tr>
            </table>
        r   zword1 word2r   r   r   r   r   r   r   test_parse_br_as_spaceS  s    z#TestReadHtml.test_parse_br_as_spacer(   bodyr   r   c           	      C   s   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d}|d }|d }|d }|dkrt|d }|d }|d }n4|dkr|d }n"|dkr|d }n|dkr|d }| j ||dd }t||g|d}t|| d S )Na  
          <table>
            <tr>
              <th>HTTP</th>
              <th>FTP</th>
              <th><a href="https://en.wiktionary.org/wiki/linkless">Linkless</a></th>
            </tr>
            <tr>
              <td><a href="https://en.wikipedia.org/">Wikipedia</a></td>
              <td>SURROUNDING <a href="ftp://ftp.us.debian.org/">Debian</a> TEXT</td>
              <td>Linkless</td>
            </tr>
            <tfoot>
              <tr>
                <td><a href="https://en.wikipedia.org/wiki/Page_footer">Footer</a></td>
                <td>
                  Multiple <a href="1">links:</a> <a href="2">Only first captured.</a>
                </td>
              </tr>
            </tfoot>
          </table>
          HTTPFTPLinkless)r  N)r  N)r  z'https://en.wiktionary.org/wiki/linkless	WikipediaSURROUNDING Debian TEXT)r  zhttps://en.wikipedia.org/)r  zftp://ftp.us.debian.org/)r  NFooter$Multiple links: Only first captured.)r  z)https://en.wikipedia.org/wiki/Page_footer)r  1)head_ignorehead_extractbody_ignorebody_extractfooter_ignorefooter_extractr  r  r  r(   r  r  r  r  r   r   Zextract_linksr   r   r   )	rM   r=  Zgh_13141_dataZgh_13141_expectedZdata_expZfoot_expZhead_expr}   rw   r   r   r   test_extract_linksf  sH    


zTestReadHtml.test_extract_linksc              	   C   s.   d}t jt|d t|dd W 5 Q R X d S )NzY`extract_links` must be one of {None, "header", "footer", "body", "all"}, got "incorrect"r5   Z	incorrectr  rA   r   r   r   r   test_extract_links_bad  s    z#TestReadHtml.test_extract_links_badc                 C   s2   d}| j |ddd }tdgg}t|| d S )Nz
        <table>
          <tr>
            <td>
              <a href='https://google.com'>Google.com</a>
            </td>
          </tr>
        </table>
        r(   r  r   )z
Google.comzhttps://google.comr   )rM   r   r}   rw   r   r   r    test_extract_links_all_no_header  s    	z-TestReadHtml.test_extract_links_all_no_headerc              	   C   s.   d}t jt|d tddd W 5 Q R X d S )NzPdtype_backend numpy is invalid, only 'numpy_nullable' and 'pyarrow' are allowed.r5   testnumpyro   rA   r   r   r   r   test_invalid_dtype_backend  s    z'TestReadHtml.test_invalid_dtype_backend)brd  re  rf  r<   fixturerN   rP   rQ   r_   r~   marknetworkr+   r   r   Zslowr   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   r   r   r   r   r   r   r   r   r   r   r   r   r  r
  r  r  r  r  r  r  r  r!  r%  r4  r6  r9  r<  parametrizer?  r@  rB  rE  rF  rG  rM  rN  rO  rQ  rS  r   rU  filterwarningsra  rk  ru  r  r  r  r  r  r  r  r   r   r   r   rK   h   s
  



7

		
	







&
/
D!




 
"

CrK   )8	functoolsr   	importlibr   r   r   r   r   pathlibr   r   r~  typingr   urllib.errorr   r  rT   r<   Zpandas.compatr	   Zpandas.util._test_decoratorsutilZ_test_decoratorstdr  r{   r
   r   r   r   r   r   r   r   Zpandas._testingZ_testingr+   Zpandas.core.arraysr   r   Zpandas.io.commonr   r  r   r  r   r1   Z
skip_if_nor?   rD   rJ   r  r  r   rK   r   r   r   r   <module>   sV   (

	