o
    RhJ                  
   @   s  d Z dg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 ddlZddlmZmZ ddl
mZmZmZmZ z
ddlmZ d	ZW n ey_ Z zdd
lmZ dZW Y dZ[ndZ[ww G dd de	ZG dd dejZG dd deZ G dd dej!Z"G dd de"Z#dS )MITHTML5TreeBuilder    N)DetectsXMLParsedAsHTML
PERMISSIVEHTMLHTML_5HTMLTreeBuilder)NamespacedAttributenonwhitespace_re)
namespacesprefixes)CommentDoctypeNavigableStringTag)_baseF)baseTc                   @   sH   e Zd ZdZdZeeeegZdZ		dddZ
dd Zd	d
 Zdd ZdS )r   a  Use html5lib to build a tree.

    Note that this TreeBuilder does not support some features common
    to HTML TreeBuilders. Some of these features could theoretically
    be implemented, but at the very least it's quite difficult,
    because html5lib moves the parse tree around as it's being built.

    * This TreeBuilder doesn't use different subclasses of NavigableString
      based on the name of the tag in which the string was found.

    * You can't use a SoupStrainer to parse only part of a document.
    html5libTNc                 c   s:    || _ |rtjddd tj|dd |d d dfV  d S )NzjYou provided a value for exclude_encoding, but the html5lib tree builder doesn't support exclude_encoding.   
stacklevelF)user_specified_encodingwarningswarnr   warn_if_markup_looks_like_xml)selfmarkupr   document_declared_encodingexclude_encodings r   W/home/air/sanwanet/backup_V2/venv/lib/python3.10/site-packages/bs4/builder/_html5lib.pyprepare_markup@   s   zHTML5TreeBuilder.prepare_markupc                 C   s   | j jd urtjddd tj| jd}|| j_t	 }t
|ts-tr(| j|d< n| j|d< |j|fi |}t
|tr?d |_n|jjjd }t
|tsN|j}||_d | j_d S )NzYou provided a value for parse_only, but the html5lib tree builder doesn't support parse_only. The entire document will be parsed.   r   )treeoverride_encodingencodingr   )soup
parse_onlyr   r   r   
HTMLParsercreate_treebuilderunderlying_builderparserdict
isinstancestrnew_html5libr   parseoriginal_encoding	tokenizerstreamcharEncodingname)r   r   r+   extra_kwargsdocr1   r   r   r    feedW   s(   



zHTML5TreeBuilder.feedc                 C   s   t || j| jd| _| jS )N)store_line_numbers)TreeBuilderForHtml5libr&   r9   r*   )r   namespaceHTMLElementsr   r   r    r)   v   s
   z#HTML5TreeBuilder.create_treebuilderc                 C   s   d| S )zSee `TreeBuilder`.z)<html><head></head><body>%s</body></html>r   )r   fragmentr   r   r    test_fragment_to_document}   s   z*HTML5TreeBuilder.test_fragment_to_document)NN)__name__
__module____qualname____doc__NAMEr   r   r   featuresTRACKS_LINE_NUMBERSr!   r8   r)   r=   r   r   r   r    r   *   s    
c                       sj   e Zd Z		d f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  ZS )r:   NTc                    sN   |r|| _ nddlm} |	dd|i|| _ tt| | d | _|| _d S )Nr   BeautifulSoup html.parserr9   )rG   rH   )r&   bs4rF   superr:   __init__r+   r9   )r   r;   r&   r9   kwargsrF   	__class__r   r    rK      s   
zTreeBuilderForHtml5lib.__init__c                 C   s   | j   t| j | j d S N)r&   resetElementr   r   r   r    documentClass   s   
z$TreeBuilderForHtml5lib.documentClassc                 C   s6   |d }|d }|d }t |||}| j| d S )Nr5   publicIdsystemId)r   for_name_and_idsr&   object_was_parsed)r   tokenr5   rT   rU   doctyper   r   r    insertDoctype   s
   z$TreeBuilderForHtml5lib.insertDoctypec                 C   sZ   i }| j r| jr| j jj \}}||d< |d |d< | jj||fi |}t|| j|S )N
sourceline   	sourcepos)r+   r9   r2   r3   positionr&   new_tagrQ   )r   r5   	namespacerL   r[   r]   tagr   r   r    elementClass   s   z#TreeBuilderForHtml5lib.elementClassc                 C   s   t t|| jS rO   )TextNoder   r&   )r   datar   r   r    commentClass   s   z#TreeBuilderForHtml5lib.commentClassc                 C   s0   ddl m} |dd| _d| j_t| j| jd S )Nr   rE   rG   rH   z[document_fragment])rI   rF   r&   r5   rQ   )r   rF   r   r   r    fragmentClass   s   z$TreeBuilderForHtml5lib.fragmentClassc                 C   s   | j |j d S rO   )r&   appendelementr   noder   r   r    appendChild   s   z"TreeBuilderForHtml5lib.appendChildc                 C   s   | j S rO   )r&   rR   r   r   r    getDocument   s   z"TreeBuilderForHtml5lib.getDocumentc                 C   s   t j| jS rO   )treebuilder_baseTreeBuildergetFragmentrh   rR   r   r   r    ro         z"TreeBuilderForHtml5lib.getFragmentc                    sB   ddl m  g tdd fdd	|d dS )Nr   rE   z8^(.*?)(?: PUBLIC "(.*?)"(?: "(.*?)")?| SYSTEM "(.*?)")?$c           	         s  t |  r	 t | trW| }|rK|d}|jdkr>|dp"d}|dp.|dp.d}dd| |||f  d S dd| |f  d S d	d| f  d S t | trid
d| | f  d S t | tr{dd| | f  d S | jrdt	| j | j
f }n| j
}dd| |f  | jrg }t| j D ]$\}}t |trdt	|j |j
f }t |trd|}|||f qt|D ]\}}dd|d  ||f  q|d7 }| jD ]}|| qd S )Nr\      rG   r   r"   z|%s<!DOCTYPE %s "%s" "%s"> z|%s<!DOCTYPE %s>z|%s<!DOCTYPE >z|%s<!-- %s -->z|%s"%s"z%s %sz|%s<%s>z
|%s%s="%s")r-   r   matchgroup	lastindexrg   r   r   r`   r   r5   attrslistitemsr	   joinsortedchildren)	rh   indentmr5   rT   rU   
attributesvaluechildrF   
doctype_rervserializeElementr   r    r      sL   












z?TreeBuilderForHtml5lib.testSerializer.<locals>.serializeElement
)r   )rI   rF   recompilery   r   rh   r   r   r    testSerializer   s   

)
z%TreeBuilderForHtml5lib.testSerializer)NT)r>   r?   r@   rK   rS   rZ   rb   re   rf   rk   rl   ro   r   __classcell__r   r   rM   r    r:      s    r:   c                   @   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 )AttrListc                 C   s   || _ t| j j| _d S rO   )rh   r,   rv   r   r   r   r    rK      s   zAttrList.__init__c                 C   s   t | j  S rO   )rw   rv   rx   __iter__rR   r   r   r    r         zAttrList.__iter__c                 C   s^   | j jpi }||dg v s| j j|v r(||| j jg v r(t|ts(t|}|| j |< d S )N*)rh   cdata_list_attributesgetr5   r-   rw   r
   findall)r   r5   r   	list_attrr   r   r    __setitem__   s   

zAttrList.__setitem__c                 C      t | j S rO   )rw   rv   rx   rR   r   r   r    rx     rp   zAttrList.itemsc                 C   r   rO   rw   rv   keysrR   r   r   r    r     rp   zAttrList.keysc                 C   s
   t | jS rO   )lenrv   rR   r   r   r    __len__     
zAttrList.__len__c                 C   s
   | j | S rO   )rv   r   r5   r   r   r    __getitem__  r   zAttrList.__getitem__c                 C   s   |t | j v S rO   r   r   r   r   r    __contains__  r   zAttrList.__contains__N)r>   r?   r@   rK   r   r   rx   r   r   r   r   r   r   r   r    r      s    r   c                   @   sx   e Zd Zdd Zdd Zdd Zdd ZeeeZdd
dZ	dd Z
dd Zdd Zdd Zdd Zdd ZeeZd	S )rQ   c                 C   s&   t j| |j || _|| _|| _d S rO   )rm   NoderK   r5   rh   r&   r`   )r   rh   r&   r`   r   r   r    rK     s   
zElement.__init__c                 C   s*  d  }}t |tr| }}nt |tr|}n|jjtkr%|j }}| |_n|j}| |_t |ts:|jd ur:|j  |d urd| jjrd| jjd jtkrd| jjd }| j	
|| }|| || j	_d S t |tro| j	
|}| jjrz| jd}n| jjd ur| j	 }n| j}| j	j|| j|d d S )NF)parentmost_recent_element)r-   r.   r   rh   rN   r   r   extractcontentsr&   
new_stringreplace_with_most_recent_element_last_descendantnext_elementrW   )r   rj   string_childr   old_elementnew_elementr   r   r   r    rk     s:   







zElement.appendChildc                 C   s   t | jtri S t| jS rO   )r-   rh   r   r   rR   r   r   r    getAttributesU  s   
zElement.getAttributesc                 C   s   |d urJt |dkrLg }t| D ]\}}t|tr&t| }||= |||< q| jj| j	| t| D ]	\}}|| j
|< q6| jj| j
 d S d S d S )Nr   )r   rw   rx   r-   tupler	   r&   builder$_replace_cdata_list_attribute_valuesr5   rh   set_up_substitutions)r   r~   converted_attributesr5   r   new_namer   r   r    setAttributesZ  s   
zElement.setAttributesNc                 C   s6   t | j|| j}|r| || d S | | d S rO   )rc   r&   r   insertBeforerk   )r   rd   r   textr   r   r    
insertTextp  s   zElement.insertTextc                 C   s   | j |j }|j jtkr4| j jr4| j j|d  jtkr4| j j|d  }| j||j  }|| d S | j ||j  | |_	d S )Nr\   )
rh   indexrN   r   r   r&   r   r   insertr   )r   rj   refNoder   old_nodenew_strr   r   r    r   w  s   
zElement.insertBeforec                 C   s   |j   d S rO   )rh   r   ri   r   r   r    removeChild  rp   zElement.removeChildc                 C   s   | j }|j }|j}|dd}t|jdkr|jd }|j}nd}|j}|j}t|dkrf|d }	|dur9||	_n||	_||	_|durG|	|_n|	|_|durQ|	|_|d dd}
||
_|durc|
|_d|
_|D ]}||_|j	| qhg |_||_dS )z1Move all of this tag's children into another tag.Fr   r   NT)
rh   next_siblingr   r   r   r   previous_elementprevious_siblingr   rg   )r   
new_parentrh   new_parent_elementfinal_next_elementnew_parents_last_descendantnew_parents_last_child(new_parents_last_descendant_next_element	to_appendfirst_childlast_childs_last_descendantr   r   r   r    reparentChildren  s>   

zElement.reparentChildrenc                 C   sB   | j | jj| j}t|| j | j}| jD ]	\}}||j|< q|S rO   )r&   r_   rh   r5   r`   rQ   r~   )r   ra   rj   keyr   r   r   r    	cloneNode  s
   zElement.cloneNodec                 C   s   | j jS rO   )rh   r   rR   r   r   r    
hasContent  s   zElement.hasContentc                 C   s$   | j d krtd | jfS | j | jfS )Nhtml)r`   r   r5   rR   r   r   r    getNameTuple  s   
zElement.getNameTuplerO   )r>   r?   r@   rK   rk   r   r   propertyr~   r   r   r   r   r   r   r   	nameTupler   r   r   r    rQ     s    6

BrQ   c                   @   s   e Zd Zdd Zdd ZdS )rc   c                 C   s   t j| d  || _|| _d S rO   )rm   r   rK   rh   r&   )r   rh   r&   r   r   r    rK     s   
zTextNode.__init__c                 C   s   t rO   )NotImplementedErrorrR   r   r   r    r     s   zTextNode.cloneNodeN)r>   r?   r@   rK   r   r   r   r   r    rc     s    rc   )$__license____all__r   r   bs4.builderr   r   r   r   r   bs4.elementr	   r
   r   html5lib.constantsr   r   r   r   r   r   html5lib.treebuildersr   rm   r/   ImportErrorer   r   rn   r:   objectr   r   rQ   rc   r   r   r   r    <module>   s0   Xx C