a
    CgX                     @   s8  d Z ddlZddlZddlZddlmZ ddlmZmZm	Z	 ddl
mZ ddlmZmZmZmZmZmZ ddlmZ ddlmZ dd	lmZ d
dlmZmZmZ ddlZddlmZ G dd deZG dd deZ G dd deZ!G dd deZ"G dd deZ#G dd deZ$G dd deZ%G dd deZ&dS )z#Tests of Beautiful Soup as a whole.    N)Iterable)BeautifulSoupGuessedAtParserWarningdammit)TreeBuilder)AttributeValueListXMLAttributeDictCommentPYTHON_SPECIFIC_ENCODINGSTagNavigableString)SoupStrainerParserRejectedMarkup)MarkupResemblesLocatorWarning   )default_builderLXML_PRESENTSoupTest)Typec                   @   sl   e Z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
ejddddd gdd ZdS )TestConstructorc                 C   s"   d}|  |}d|jjksJ d S )Nu   <h1>éé</h1>u   éésouph1stringselfdatar    r   R/var/www/lab.imftr.de/x/nb_venv/lib/python3.9/site-packages/bs4/tests/test_soup.pytest_short_unicode_input,   s    
z(TestConstructor.test_short_unicode_inputc                 C   s"   d}|  |}d|jjksJ d S )Nz<h1>foo bar</h1>zfoo barr   r   r   r   r   test_embedded_null1   s    
z"TestConstructor.test_embedded_nullc                 C   s,   d d}| j|dgd}d|jks(J d S )Nu   Räksmörgåsutf-8)Zexclude_encodingszwindows-1252)encoder   original_encoding)r   	utf8_datar   r   r   r   test_exclude_encodings6   s    
z&TestConstructor.test_exclude_encodingsc                 C   s  G dd dt }tddd}tjdd" tdd|i|}W d    n1 sP0    Y  t|j|sjJ tdd	|jjksJ d
|jjksJ |f i |}tjdd}td|dd}W d    n1 s0    Y  t	|d j
}|dsJ ||jksJ ||jksJ d S )Nc                   @   sD   e Zd Zdd Zdd Zdd Zdd Zd	d
 Ze ZZ	dd Z
dS )z7TestConstructor.test_custom_builder_class.<locals>.Mockc                 [   s4   || _ d| _d| _g | _g | _i | _t| _t| _	d S )NTF)
called_withis_xmlZstore_line_numbersZcdata_list_attributesZpreserve_whitespace_tagsstring_containersr   Zattribute_dict_classr   Zattribute_value_list_class)r   kwargsr   r   r   __init__?   s    z@TestConstructor.test_custom_builder_class.<locals>.Mock.__init__c                 S   s   d S Nr   r   r   r   r   r   initialize_soupI   s    zGTestConstructor.test_custom_builder_class.<locals>.Mock.initialize_soupc                 S   s
   || _ d S r,   )fed)r   markupr   r   r   feedL   s    z<TestConstructor.test_custom_builder_class.<locals>.Mock.feedc                 S   s   d S r,   r   r   r   r   r   resetO   s    z=TestConstructor.test_custom_builder_class.<locals>.Mock.resetc                 S   s   d S r,   r   )r   ignorer   r   r   r4   R   s    z>TestConstructor.test_custom_builder_class.<locals>.Mock.ignorec                    s
   dV  d S )N)prepared markupzoriginal encodingzdeclared encodingzcontains replacement charactersr   r   argsr*   r   r   r   prepare_markupW   s    zFTestConstructor.test_custom_builder_class.<locals>.Mock.prepare_markupN)__name__
__module____qualname__r+   r.   r1   r3   r4   Zset_up_substitutionsZcan_be_empty_elementr8   r   r   r   r   Mock>   s   
r<   valueT)varZconvertEntitiesrecord builder)r>   r5   )rB   Zignored_valuer   zCKeyword arguments to the BeautifulSoup constructor will be ignored.)rA   )objectdictwarningscatch_warningsr   
isinstancerB   r'   r/   strmessage
startswith)r   r<   r*   r   rB   wmsgr   r   r   test_custom_builder_class;   s.    !0$z)TestConstructor.test_custom_builder_classc                 C   sd   G dd dt }dd }tt}td|d W d    n1 sD0    Y  dt|jv s`J d S )Nc                   @   s   e Zd Zdd ZdS )z:TestConstructor.test_parser_markup_rejection.<locals>.Mockc                 _   s   t dd S )NzNope.r   r6   r   r   r   r1      s    z?TestConstructor.test_parser_markup_rejection.<locals>.Mock.feedN)r9   r:   r;   r1   r   r   r   r   r<      s   r<   c                    s    |d d dfV  |d d dfV  d S NFr   )r   r0   r7   r*   r   r   r   r8      s    zDTestConstructor.test_parser_markup_rejection.<locals>.prepare_markuprA   )rB   zoThe markup you provided was rejected by the parser. Trying a different parser or a different encoding may help.)r   pytestraisesr   r   rH   r=   )r   r<   r8   exc_infor   r   r   test_parser_markup_rejection|   s    *z,TestConstructor.test_parser_markup_rejectionc              	   C   s   d}|  |}|j}d|d ks$J ddg|d ks8J | j |td d}d|jd ksZJ ddiddifD ]j}tjd	d
  | j |d |d}W d    n1 s0    Y  |j}ddg|d ksJ d|d ksjJ qjd S )Nz&<a id=" an id " class=" a class "></a>z an id idaclass)rB   Zmulti_valued_attributesz	 a class *Tr?   Zan)r   rT   r   rE   rF   )r   r0   r   rT   Z
switcheroor   r   r   test_cdata_list_attributes   s    
$z*TestConstructor.test_cdata_list_attributesc                    sn   G dd dt G dd dtG dd dt | jdt tt id}t fd	d
|jD sjJ d S )Nc                   @   s   e Zd ZdS )z9TestConstructor.test_replacement_classes.<locals>.TagPlusNr9   r:   r;   r   r   r   r   TagPlus   s   rY   c                   @   s   e Zd ZdS )z<TestConstructor.test_replacement_classes.<locals>.StringPlusNrX   r   r   r   r   
StringPlus   s   rZ   c                   @   s   e Zd ZdS )z=TestConstructor.test_replacement_classes.<locals>.CommentPlusNrX   r   r   r   r   CommentPlus   s   r[   z<a><b>foo</b>bar</a><!--whee-->)Zelement_classesc                 3   s   | ]}t | fV  qd S r,   )rG   .0xr[   rZ   rY   r   r   	<genexpr>   s   z;TestConstructor.test_replacement_classes.<locals>.<genexpr>)r   r   r	   r   allZdescendantsr-   r   r_   r   test_replacement_classes   s    z(TestConstructor.test_replacement_classesc                 C   s   G dd dt }G dd dt }| jd||dd}t|jjd t sJJ t|jjd |s`J |jjD ]}t||shJ qhg |jksJ d S )	Nc                   @   s   e Zd ZdS )zATestConstructor.test_alternate_string_containers.<locals>.PStringNrX   r   r   r   r   PString   s   rc   c                   @   s   e Zd ZdS )zATestConstructor.test_alternate_string_containers.<locals>.BStringNrX   r   r   r   r   BString   s   rd   z4<div>Hello.<p>Here is <b>some <i>bolded</i></b> text)bp)r)   r   )	r   r   rG   divcontentsrf   re   stringsZstring_container_stack)r   rc   rd   r   sr   r   r    test_alternate_string_containers   s    	z0TestConstructor.test_alternate_string_containers
bad_markupr   Fc                 C   s   dS rN   r   )r^   r   r   r   <lambda>       zTestConstructor.<lambda>c                 C   sR   t t}t|d W d    n1 s*0    Y  d|dt|jv sNJ d S )Nhtml.parserz'Incoming markup is of an invalid type: z?. Markup must be a string, a bytestring, or an open filehandle.)rO   rP   	TypeErrorr   rH   r=   )r   rl   rQ   r   r   r   test_invalid_markup_type   s    (
z(TestConstructor.test_invalid_markup_typeN)r9   r:   r;   r    r!   r&   rM   rR   rW   rb   rk   rO   markparametrizerq   r   r   r   r   r   +   s   A!r   c                   @   sT   e Zd Zejdddgdd Zejddd eD d	g d
d Zdd Z	d	S )
TestOutputz!eventual_encoding,actual_encoding)r"   r"   )utf-16ru   c                 C   s0   |  d}d|_d| d|j|dks,J d S )N<tag></tag>Tz<?xml version="1.0" encoding="z"?>
<tag></tag>eventual_encoding)r   r(   decode)r   rx   Zactual_encodingr   r   r   r   test_decode_xml_declaration   s    


z&TestOutput.test_decode_xml_declarationrx   c                 C   s   g | ]}|qS r   r   r\   r   r   r   
<listcomp>  rn   zTestOutput.<listcomp>Nc                 C   s(   t dd}d|_d|j|dks$J d S )Nrv   ro   Tz!<?xml version="1.0"?>
<tag></tag>rw   )r   r(   ry   )r   rx   r   r   r   r   Mtest_decode_xml_declaration_with_missing_or_python_internal_eventual_encoding
  s
    	
zXTestOutput.test_decode_xml_declaration_with_missing_or_python_internal_eventual_encodingc                 C   sV   |  d}d|jddksJ d|jddks2J d| ksBJ d| ksRJ d S )Nrv   s   <tag></tag>r"   )encodingz<tag>
</tag>
)r   r#   Zencode_contentsZdecode_contentsZprettifyr-   r   r   r   test  s
    
zTestOutput.test)
r9   r:   r;   rO   rr   rs   rz   r
   r|   r~   r   r   r   r   rt      s   

rt   c                   @   s   e Zd Zeej ee ejdddZeej ddddZ	dd	 Z
d
d Zdd Zdd Zdd Zdd Zdd Zejdg ddd Zejdg ddd Zdd Zdd  Zd!d" Zd#d$ ZdS )%TestWarnings)rE   clsreturnc                 C   s@   |D ]&}t |j|r|jtks"J |  S qtd||f d S )Nz%s warning not found in %r)rG   rI   filename__file__	Exception)r   rE   r   rK   r   r   r   _assert_warning,  s
    
zTestWarnings._assert_warningN)rK   r   c                 C   s2   |  |t}t|j}|tjd d s.J d S )N<   )r   r   rH   rI   rJ   ZMESSAGE)r   rK   warningrI   r   r   r   _assert_no_parser_specified5  s    
z(TestWarnings._assert_no_parser_specifiedc                 C   sB   t jdd}td W d    n1 s*0    Y  | | d S NTr?   <a><b></b></a>rE   rF   r   r   r   rK   r   r   r   #test_warning_if_no_parser_specified:  s    &z0TestWarnings.test_warning_if_no_parser_specifiedc                 C   sD   t jdd}tdd W d    n1 s,0    Y  | | d S )NTr?   r   htmlr   r   r   r   r   *test_warning_if_parser_specified_too_vague?  s    (z7TestWarnings.test_warning_if_parser_specified_too_vaguec                 C   sF   t jdd}| d W d    n1 s,0    Y  g |ksBJ d S r   rE   rF   r   r   r   r   r   ,test_no_warning_if_explicit_parser_specifiedD  s    (z9TestWarnings.test_no_warning_if_explicit_parser_specifiedc                 C   sn   t ddd}tjdd}| jd|d W d    n1 s<0    Y  | |t}t|j}|dsjJ d S )	NrT   re   )namer   Tr?   r   
parse_onlyz7The given value for parse_only will exclude everything:)	r   rE   rF   r   r   UserWarningrH   rI   rJ   )r   strainerrK   r   rL   r   r   r   +test_warning_if_strainer_filters_everythingI  s    ,
z8TestWarnings.test_warning_if_strainer_filters_everythingc                 C   s   t jdd"}tddtdd}W d    n1 s40    Y  | |t}t|j}d|v s`J d|v slJ d	| ks|J d S )
NTr?   r   ro   re   )parseOnlyTheser   r   s   <b></b>)	rE   rF   r   r   r   DeprecationWarningrH   rI   r#   )r   rK   r   r   rL   r   r   r   )test_parseOnlyThese_renamed_to_parse_onlyQ  s    $
z6TestWarnings.test_parseOnlyThese_renamed_to_parse_onlyc                 C   s~   t jdd"}d}t|ddd}W d    n1 s40    Y  | |t}t|j}d|v s`J d|v slJ d|jkszJ d S )	NTr?   s   éro   utf8)fromEncodingr   Zfrom_encoding)rE   rF   r   r   r   rH   rI   r$   )r   rK   r   r   r   rL   r   r   r   *test_fromEncoding_renamed_to_from_encoding^  s    ,
z7TestWarnings.test_fromEncoding_renamed_to_from_encodingc                 C   s<   t t | jddd W d    n1 s.0    Y  d S )Nz<a>T)Zno_such_argument)rO   rP   rp   r   r2   r   r   r   "test_unrecognized_keyword_argumenth  s    z/TestWarnings.test_unrecognized_keyword_argumentr0   )zmarkup.htmlz
markup.htmzmarkup.HTMLz
markup.txtzmarkup.xhtmlz
markup.xmlz/home/user/file.txtz/c:\user\file.html\\server\share\path\file.XhTmlc                 C   sX   t jdd8}t|d | |t}dt|jv s6J W d    n1 sJ0    Y  d S )NTr?   ro   zlooks more like a filenamerE   rF   r   r   r   rH   rI   )r   r0   rK   r   r   r   r   test_resembles_filename_warningl  s    
z,TestWarnings.test_resembles_filename_warning)r   Z
markuphtmlz
markup.comrA   z	markup.jsz
markup.jpgzmarkup.markupz/home/user/filez)c:\user\file.html\\server\share\path\filezAlog message containing a url http://www.url.com/ right there.htmlztwo  consecutive  spaces.htmlztwo//consecutive//slashes.htmlz4looks/like/a/filename/but/oops/theres/a#comment.htmlztwo
lines.htmlzcontains?.htmlzcontains*.htmlzcontains#.htmlzcontains&.htmlzcontains;.htmlzcontains>.htmlzcontains<.htmlzcontains$.htmlzcontains|.htmlzcontains:.htmlz:-at-the-front.htmlc                 C   sF   t jdd}| | W d    n1 s,0    Y  g |ksBJ d S )NTr?   r   )r   r0   rK   r   r   r   "test_resembles_filename_no_warning  s    ((z/TestWarnings.test_resembles_filename_no_warningc                 C   st   d}t jdd}t|d W d    n1 s00    Y  | |t}dt|jv sXJ |t|jdvspJ d S )Ns   http://www.crummybytes.com/Tr?   ro   looks more like a URLr   )rE   rF   r   r   r   rH   rI   r#   r   urlwarning_listr   r   r   r   test_url_warning_with_bytes_url  s    (z,TestWarnings.test_url_warning_with_bytes_urlc                 C   sn   d}t jdd}t|d W d    n1 s00    Y  | |t}dt|jv sXJ |t|jvsjJ d S )Nzhttp://www.crummyunicode.com/Tr?   ro   r   r   r   r   r   r   !test_url_warning_with_unicode_url  s    (z.TestWarnings.test_url_warning_with_unicode_urlc                 C   sP   t jdd}| d W d    n1 s,0    Y  tdd |D rLJ d S )NTr?   s$   http://www.crummybytes.com/ is greatc                 s   s   | ]}d t |jv V  qdS r   NrH   rI   r]   rK   r   r   r   r`     rn   zETestWarnings.test_url_warning_with_bytes_and_space.<locals>.<genexpr>rE   rF   r   anyr   r   r   r   r   %test_url_warning_with_bytes_and_space  s    (z2TestWarnings.test_url_warning_with_bytes_and_spacec                 C   sP   t jdd}| d W d    n1 s,0    Y  tdd |D rLJ d S )NTr?   z&http://www.crummyunicode.com/ is greatc                 s   s   | ]}d t |jv V  qdS r   r   r   r   r   r   r`     rn   zGTestWarnings.test_url_warning_with_unicode_and_space.<locals>.<genexpr>r   r   r   r   r   'test_url_warning_with_unicode_and_space  s    (z4TestWarnings.test_url_warning_with_unicode_and_space)r9   r:   r;   r   rE   WarningMessager   Warningr   r   r   r   r   r   r   r   r   rO   rr   rs   r   r   r   r   r   r   r   r   r   r   r   $  s2   		

$

r   c                   @   s   e Zd Zdd ZdS )TestSelectiveParsingc                 C   s.   d}t d}| j||d}| dks*J d S )Nz&No<b>Yes</b><a>No<b>Yes <c>Yes</c></b>re   r   s   <b>Yes</b><b>Yes <c>Yes</c></b>)r   r   r#   )r   r0   r   r   r   r   r   test_parse_with_soupstrainer  s    z1TestSelectiveParsing.test_parse_with_soupstrainerN)r9   r:   r;   r   r   r   r   r   r     s   r   c                   @   s:   e Zd ZdZdd Zejje dddd Z	dd	 Z
d
S )
TestNewTagz(Test the BeautifulSoup.new_tag() method.c                 C   s   |  d}|jdddddid}t|ts.J d|jks<J |jdksJJ tddd|jks`J d |ju snJ |d}d |ju sJ |jdd d	}d |ju sJ |jddd	}d|jksJ d S )
NrA   footxtZbazr   za name)r   barattrs)r   r   )r   )	r   new_tagrG   r   r   r   rD   r   parent)r   r   r   r   r   r   test_new_tag  s    

zTestNewTag.test_new_tagz-lxml not installed, cannot parse XML document)reasonc                 C   sB   t dd}|d}|d}d| ks.J d| ks>J d S )NrA   xmlbrrf      <br/>s   <p/>r   r   r#   )r   Zxml_soupZxml_brZxml_pr   r   r   5test_xml_tag_inherits_self_closing_rules_from_builder  s
    


z@TestNewTag.test_xml_tag_inherits_self_closing_rules_from_builderc                 C   sB   t dd}|d}|d}d| ks.J d| ks>J d S )NrA   ro   r   rf   r   s   <p></p>r   )r   Z	html_soupZhtml_brZhtml_pr   r   r   1test_tag_inherits_self_closing_rules_from_builder  s
    


z<TestNewTag.test_tag_inherits_self_closing_rules_from_builderN)r9   r:   r;   __doc__r   rO   rr   Zskipifr   r   r   r   r   r   r   r     s   

r   c                   @   s    e Zd ZdZdd Zdd ZdS )TestNewStringz+Test the BeautifulSoup.new_string() method.c                 C   s2   |  d}|d}d|ks J t|ts.J d S NrA   r   )r   
new_stringrG   r   r   r   rj   r   r   r   'test_new_string_creates_navigablestring  s    

z5TestNewString.test_new_string_creates_navigablestringc                 C   s4   |  d}|dt}d|ks"J t|ts0J d S r   )r   r   r	   rG   r   r   r   r   3test_new_string_can_create_navigablestring_subclass  s    
zATestNewString.test_new_string_can_create_navigablestring_subclassN)r9   r:   r;   r   r   r   r   r   r   r   r     s   r   c                   @   s   e Zd Zdd Zdd ZdS )
TestPicklec                 C   s2   |  d}t|}t|}d|jjks.J d S )Nz<a>some markup</a>some markup)r   pickledumpsloadsrT   r   r   r   ZpickledZ	unpickledr   r   r   test_normal_pickle  s    


zTestPickle.test_normal_picklec                 C   s6   |  d}d |_t|}t|}d|jks2J d S )Nr   )r   rB   r   r   r   r   r   r   r   r   test_pickle_with_no_builder  s
    


z&TestPickle.test_pickle_with_no_builderN)r9   r:   r;   r   r   r   r   r   r   r     s   r   c                   @   s4   e Zd Zdd Zdd Zdd Zdd Zd	d
 ZdS )TestEncodingConversionc                 C   s&   d| _ | j d| _| jdks"J d S )NuU   <html><head><meta charset="utf-8"/></head><body><foo>Sacré bleu!</foo></body></html>r"   sU   <html><head><meta charset="utf-8"/></head><body><foo>Sacré bleu!</foo></body></html>)unicode_datar#   r%   r2   r   r   r   setup_method)  s    z#TestEncodingConversion.setup_methodc              	   C   s   t j}ttj zpdd }|t _d}| |}| }t|tsFJ || 	| ks\J |j
 dksnJ W ttj |t _nttj |t _0 d S )Nc                 S   s   d S r,   r   )rH   r   r   r   noop9  s    z>TestEncodingConversion.test_ascii_in_unicode_out.<locals>.noops   <foo>a</foo>r"   )r   Z_chardet_dammitloggingdisableWARNINGr   ry   rG   rH   Zdocument_forr$   lowerNOTSET)r   chardetr   asciiZsoup_from_asciiZunicode_outputr   r   r   test_ascii_in_unicode_out2  s    
z0TestEncodingConversion.test_ascii_in_unicode_outc                 C   s@   |  | j}| | jksJ |jjdks.J |jd u s<J d S Nu   Sacré bleu!)r   r   ry   r   r   r$   r   Zsoup_from_unicoder   r   r   test_unicode_in_unicode_outH  s    z2TestEncodingConversion.test_unicode_in_unicode_outc                 C   s2   |  | j}| | jksJ |jjdks.J d S r   )r   r%   ry   r   r   r   )r   Zsoup_from_utf8r   r   r   test_utf8_in_unicode_outP  s    z/TestEncodingConversion.test_utf8_in_unicode_outc                 C   s$   |  | j}|d| jks J d S )Nr"   )r   r   r#   r%   r   r   r   r   test_utf8_outW  s    z$TestEncodingConversion.test_utf8_outN)r9   r:   r;   r   r   r   r   r   r   r   r   r   r   %  s
   	r   )'r   r   r   rO   typingr   Zbs4r   r   r   Zbs4.builderr   Zbs4.elementr   r   r	   r
   r   r   Z
bs4.filterr   Zbs4.exceptionsr   Zbs4._warningsr   rA   r   r   r   rE   r   r   rt   r   r   r   r   r   r   r   r   r   r   <module>   s.     L. ).