U
    lufM                     @   s,  d Z ddlmZ ddlZddlZddlZddlZddlZddlZddl	m
Z
mZmZmZmZ ddlmZmZmZ ddlmZmZmZmZmZ ddlmZmZmZ ddl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.    )	set_traceN)BeautifulSoupBeautifulStoneSoupGuessedAtParserWarningMarkupResemblesLocatorWarningdammit)builder_registryTreeBuilderParserRejectedMarkup)CommentSoupStrainerPYTHON_SPECIFIC_ENCODINGSTagNavigableString   )default_builderLXML_PRESENTSoupTestc                   @   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
dS )TestConstructorc                 C   s"   d}|  |}d|jjkstd S )Nu   <h1>éé</h1>u   éésouph1stringAssertionErrorselfdatar    r   7/tmp/pip-unpacked-wheel-pg4vfqbd/bs4/tests/test_soup.pytest_short_unicode_input)   s    
z(TestConstructor.test_short_unicode_inputc                 C   s"   d}|  |}d|jjkstd S )Nz<h1>foo bar</h1>zfoo barr   r   r   r   r   test_embedded_null.   s    
z"TestConstructor.test_embedded_nullc                 C   s,   d d}| j|dgd}d|jks(td S )Nu   Räksmörgåsutf-8)Zexclude_encodingszwindows-1252)encoder   original_encodingr   )r   	utf8_datar   r   r   r   test_exclude_encodings3   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 5 Q R X t|j|sVttdd	|jjksltd
|jj	ks|t|f |}tjdd}td|dd}W 5 Q R X t
|d j}|dst||jkst||jkst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                 [   s(   || _ d| _d| _g | _g | _i | _d S )NTF)called_withis_xmlZstore_line_numbersZcdata_list_attributesZpreserve_whitespace_tagsstring_containers)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_soupC   s    zGTestConstructor.test_custom_builder_class.<locals>.Mock.initialize_soupc                 S   s
   || _ d S r+   )fed)r   markupr   r   r   feedE   s    z<TestConstructor.test_custom_builder_class.<locals>.Mock.feedc                 S   s   d S r+   r   r   r   r   r   resetG   s    z=TestConstructor.test_custom_builder_class.<locals>.Mock.resetc                 S   s   d S r+   r   )r   ignorer   r   r   r3   I   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_markupL   s    zFTestConstructor.test_custom_builder_class.<locals>.Mock.prepare_markupN)__name__
__module____qualname__r*   r-   r0   r2   r3   Zset_up_substitutionsZcan_be_empty_elementr7   r   r   r   r   Mock;   s   r;   valueT)varZconvertEntitiesrecord builder)r=   r4   )rA   Zignored_valuer   zCKeyword arguments to the BeautifulSoup constructor will be ignored.)r@   )objectdictwarningscatch_warningsr   
isinstancerA   r   r&   r.   strmessage
startswith)r   r;   r)   r   rA   wmsgr   r   r   test_custom_builder_class8   s*    
  z)TestConstructor.test_custom_builder_classc              	   C   sX   G dd dt }dd }dd l}tt}td|d W 5 Q R X dt|jksTt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
   r5   r   r   r   r0   l   s    z?TestConstructor.test_parser_markup_rejection.<locals>.Mock.feedN)r8   r9   r:   r0   r   r   r   r   r;   k   s   r;   c                    s    t d d dfV  t d d dfV  d S )NF)r/   r5   r   r   r   r7   o   s    zDTestConstructor.test_parser_markup_rejection.<locals>.prepare_markupr   r@   )rA   zoThe markup you provided was rejected by the parser. Trying a different parser or a different encoding may help.)	r	   repytestraisesr
   r   rG   r<   r   )r   r;   r7   rM   exc_infor   r   r   test_parser_markup_rejectionh   s    z,TestConstructor.test_parser_markup_rejectionc              
   C   s   d}|  |}|j}d|d ks$tddg|d ks8t| j |td d}d|jd ksZtddiddifD ]V}tjd	d
}| j |d |d}W 5 Q R X |j}ddg|d kstd|d ksjtqjd S )Nz&<a id=" an id " class=" a class "></a>z an id idaclass)rA   Zmulti_valued_attributesz	 a class *Tr>   Zan)r   rS   r   r   rD   rE   )r   r/   r   rS   Z
switcheroorJ   r   r   r   test_cdata_list_attributes{   s    
z*TestConstructor.test_cdata_list_attributesc                    sp   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
| D sltd S )Nc                   @   s   e Zd ZdS )z9TestConstructor.test_replacement_classes.<locals>.TagPlusNr8   r9   r:   r   r   r   r   TagPlus   s   rX   c                   @   s   e Zd ZdS )z<TestConstructor.test_replacement_classes.<locals>.StringPlusNrW   r   r   r   r   
StringPlus   s   rY   c                   @   s   e Zd ZdS )z=TestConstructor.test_replacement_classes.<locals>.CommentPlusNrW   r   r   r   r   CommentPlus   s   rZ   z<a><b>foo</b>bar</a><!--whee-->)Zelement_classesc                 3   s   | ]}t | fV  qd S r+   )rF   .0xrZ   rY   rX   r   r   	<genexpr>   s   z;TestConstructor.test_replacement_classes.<locals>.<genexpr>)r   r   r   r   allZrecursiveChildGeneratorr   r,   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 sJtt|jjd |s`t|jjD ]}t||shtqhg |j	kstd S )	Nc                   @   s   e Zd ZdS )zATestConstructor.test_alternate_string_containers.<locals>.PStringNrW   r   r   r   r   PString   s   rb   c                   @   s   e Zd ZdS )zATestConstructor.test_alternate_string_containers.<locals>.BStringNrW   r   r   r   r   BString   s   rc   z4<div>Hello.<p>Here is <b>some <i>bolded</i></b> text)bp)r(   r   )
r   r   rF   divcontentsr   re   rd   stringsZstring_container_stack)r   rb   rc   r   sr   r   r    test_alternate_string_containers   s    	z0TestConstructor.test_alternate_string_containersN)r8   r9   r:   r   r    r%   rL   rQ   rV   ra   rj   r   r   r   r   r   '   s   0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-16rl   c                 C   s0   |  d}d|_d| d|j|dks,td S )N<tag></tag>Tz<?xml version="1.0" encoding="z"?>
<tag></tag>eventual_encoding)r   r'   decoder   )r   ro   Zactual_encodingr   r   r   r   test_decode_xml_declaration   s
    


z&TestOutput.test_decode_xml_declarationro   c                 C   s   g | ]}|qS r   r   r[   r   r   r   
<listcomp>   s     zTestOutput.<listcomp>Nc                 C   s(   t dd}d|_d|j|dks$td S )Nrm   html.parserTz!<?xml version="1.0"?>
<tag></tag>rn   )r   r'   rp   r   )r   ro   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td|jddks2td| ksBtd| ksRtd S )Nrm   s   <tag></tag>r!   )encodingz<tag>
</tag>
)r   r"   r   Zencode_contentsZdecode_contentsZprettifyr,   r   r   r   test   s
    
zTestOutput.test)
r8   r9   r:   rN   markparametrizerq   r   rt   rv   r   r   r   r   rk      s   

 
	rk   c                   @   s   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ddddgdd Zejdddddgd d! Zd"d# Zd$d% Zd&d' Zd(d) Zd*S )+TestWarningsc                 C   s@   |D ]&}t |j|r|jtks"t|  S qtd||f d S )Nz%s warning not found in %r)rF   rH   filename__file__r   	Exception)r   rD   clsrJ   r   r   r   _assert_warning  s
    
zTestWarnings._assert_warningc                 C   s2   |  |t}t|j}|tjd d s.td S )N<   )r~   r   rG   rH   rI   r   ZNO_PARSER_SPECIFIED_WARNINGr   )r   rJ   warningrH   r   r   r   _assert_no_parser_specified  s    
z(TestWarnings._assert_no_parser_specifiedc              	   C   s.   t jdd}td}W 5 Q R X | | d S NTr>   <a><b></b></a>rD   rE   r   r   r   rJ   r   r   r   r   #test_warning_if_no_parser_specified  s    z0TestWarnings.test_warning_if_no_parser_specifiedc              	   C   s0   t jdd}tdd}W 5 Q R X | | 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   s2   t jdd}| d}W 5 Q R X g |ks.td S r   rD   rE   r   r   r   r   r   r   ,test_no_warning_if_explicit_parser_specified  s    z9TestWarnings.test_no_warning_if_explicit_parser_specifiedc              	   C   sl   t jdd}tddtdd}W 5 Q R X | |t}t|j}d|ksLtd|ksXtd	|	 kshtd S )
NTr>   r   rs   rd   )parseOnlyTheser   
parse_onlys   <b></b>)
rD   rE   r   r   r~   DeprecationWarningrG   rH   r   r"   )r   rJ   r   r   rK   r   r   r   )test_parseOnlyThese_renamed_to_parse_only"  s     
z6TestWarnings.test_parseOnlyThese_renamed_to_parse_onlyc              	   C   sj   t jdd}d}t|ddd}W 5 Q R X | |t}t|j}d|ksLtd|ksXtd|jksftd S )	NTr>   s   érs   utf8)fromEncodingr   Zfrom_encoding)	rD   rE   r   r~   r   rG   rH   r   r#   )r   rJ   r   r   r   rK   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 5 Q R X d S )Nz<a>T)Zno_such_argument)rN   rO   	TypeErrorr   r1   r   r   r   "test_unrecognized_keyword_argument:  s    z/TestWarnings.test_unrecognized_keyword_argument	extensionzmarkup.htmlz
markup.htmzmarkup.HTMLz
markup.txtzmarkup.xhtmlz
markup.xmlz/home/user/filezc:\userilec              	   C   sH   t jdd2}td| d}| |t}dt|jks:tW 5 Q R X d S )NTr>   r/   rs   zlooks more like a filenamerD   rE   r   r~   r   rG   rH   r   )r   r   rJ   r   r   r   r   r   test_resembles_filename_warning>  s    z,TestWarnings.test_resembles_filename_warningZ
markuphtmlz
markup.comr@   z	markup.jsc              	   C   s6   t jdd}| d| }W 5 Q R X g |ks2td S )NTr>   r/   r   )r   r   rJ   r   r   r   r   "test_resembles_filename_no_warningK  s    z/TestWarnings.test_resembles_filename_no_warningc              	   C   s`   d}t jdd}t|d}W 5 Q R X | |t}dt|jksDt|t|jdks\td S )Ns   http://www.crummybytes.com/Tr>   rs   looks more like a URLr   )	rD   rE   r   r~   r   rG   rH   r   r"   r   urlwarning_listr   r   r   r   r   test_url_warning_with_bytes_urlW  s     z,TestWarnings.test_url_warning_with_bytes_urlc              	   C   sZ   d}t jdd}t|d}W 5 Q R X | |t}dt|jksDt|t|jksVtd S )Nzhttp://www.crummyunicode.com/Tr>   rs   r   r   r   r   r   r   !test_url_warning_with_unicode_urla  s     z.TestWarnings.test_url_warning_with_unicode_urlc              	   C   s<   t jdd}| d}W 5 Q R X tdd |D r8td S )NTr>   s$   http://www.crummybytes.com/ is greatc                 s   s   | ]}d t |jkV  qdS r   NrG   rH   r\   rJ   r   r   r   r_   r  s   zETestWarnings.test_url_warning_with_bytes_and_space.<locals>.<genexpr>rD   rE   r   anyr   r   r   r   r   r   r   %test_url_warning_with_bytes_and_spacem  s
    z2TestWarnings.test_url_warning_with_bytes_and_spacec              	   C   s<   t jdd}| d}W 5 Q R X tdd |D r8td S )NTr>   z&http://www.crummyunicode.com/ is greatc                 s   s   | ]}d t |jkV  qdS r   r   r   r   r   r   r_   x  s   zGTestWarnings.test_url_warning_with_unicode_and_space.<locals>.<genexpr>r   r   r   r   r   'test_url_warning_with_unicode_and_spaceu  s
    z4TestWarnings.test_url_warning_with_unicode_and_spaceN)r8   r9   r:   r~   r   r   r   r   r   r   r   rN   rw   rx   r   r   r   r   r   r   r   r   r   r   ry      s6      



ry   c                   @   s   e Zd Zdd ZdS )TestSelectiveParsingc                 C   s.   d}t d}| j||d}| dks*td S )Nz&No<b>Yes</b><a>No<b>Yes <c>Yes</c></b>rd   )r   s   <b>Yes</b><b>Yes <c>Yes</c></b>)r   r   r"   r   )r   r/   Zstrainerr   r   r   r   test_parse_with_soupstrainer~  s    z1TestSelectiveParsing.test_parse_with_soupstrainerN)r8   r9   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   sb   |  d}|jddddid}t|ts,td|jks:ttddd|jksPtd |jks^td S )Nr@   fooZbaznameza name)barattrs)r   r   )	r   new_tagrF   r   r   r   rC   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.td| ks>td S )Nr@   xmlbrre      <br/>s   <p/>r   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.td| ks>td S )Nr@   rs   r   re   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)r8   r9   r:   __doc__r   rN   rw   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 tt|ts.td S Nr@   r   )r   
new_stringr   rF   r   r   r   ri   r   r   r   'test_new_string_creates_navigablestring  s    

z5TestNewString.test_new_string_creates_navigablestringc                 C   s4   |  d}|dt}d|ks"tt|ts0td S r   )r   r   r   r   rF   r   r   r   r   3test_new_string_can_create_navigablestring_subclass  s    
zATestNewString.test_new_string_can_create_navigablestring_subclassN)r8   r9   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.td S )Nz<a>some markup</a>some markup)r   pickledumpsloadsrS   r   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s2td S )Nr   )r   rA   r   r   r   r   r   r   r   r   r   test_pickle_with_no_builder  s
    


z&TestPickle.test_pickle_with_no_builderN)r8   r9   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"t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$   r   r1   r   r   r   setup_method  s    z#TestEncodingConversion.setup_methodc              	   C   s   t j}ttj z^dd }|t _d}| |}| }t|t	sFt
|| | ks\t
|j dksnt
W 5 ttj |t _X d S )Nc                 S   s   d S r+   r   )rG   r   r   r   noop  s    z>TestEncodingConversion.test_ascii_in_unicode_out.<locals>.noops   <foo>a</foo>r!   )r   Zchardet_dammitloggingdisableWARNINGNOTSETr   rp   rF   rG   r   Zdocument_forr#   lower)r   chardetr   asciiZsoup_from_asciiZunicode_outputr   r   r   test_ascii_in_unicode_out  s    
z0TestEncodingConversion.test_ascii_in_unicode_outc                 C   s@   |  | j}| | jkst|jjdks.t|jd ks<td S Nu   Sacré bleu!)r   r   rp   r   r   r   r#   r   Zsoup_from_unicoder   r   r   test_unicode_in_unicode_out  s    z2TestEncodingConversion.test_unicode_in_unicode_outc                 C   s2   |  | j}| | jkst|jjdks.td S r   )r   r$   rp   r   r   r   r   )r   Zsoup_from_utf8r   r   r   test_utf8_in_unicode_out  s    z/TestEncodingConversion.test_utf8_in_unicode_outc                 C   s$   |  | j}|d| jks td S )Nr!   )r   r   r"   r$   r   r   r   r   r   test_utf8_out  s    z$TestEncodingConversion.test_utf8_outN)r8   r9   r:   r   r   r   r   r   r   r   r   r   r     s
   r   )&r   pdbr   r   osr   rN   systempfileZbs4r   r   r   r   r   Zbs4.builderr   r	   r
   Zbs4.elementr   r   r   r   r   r@   r   r   r   rD   r   rk   ry   r   r   r   r   r   r   r   r   r   <module>   s*    0)}	"