a
    OfJ                  
   @   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 n4 ey Z zdd
lmZ dZW Y dZ[n
dZ[0 0 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                   @   sF   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   s8   || _ |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   Zwarn_if_markup_looks_like_xml)selfmarkupr   Zdocument_declared_encodingZexclude_encodings r   N/var/www/ai-form-bot/venv/lib/python3.9/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sZtrP| j|d< n
| j|d< |j|fi |}t
|tr~d |_n$|jjjd }t
|ts|j}||_d | j_d S )NzYou provided a value for parse_only, but the html5lib tree builder doesn't support parse_only. The entire document will be parsed.   r   )treeZoverride_encodingencodingr   )soupZ
parse_onlyr   r   r   
HTMLParsercreate_treebuilderunderlying_builderparserdict
isinstancestrnew_html5libr   parseoriginal_encoding	tokenizerstreamZcharEncodingname)r   r   r&   extra_kwargsdocr,   r   r   r   feedW   s(    



zHTML5TreeBuilder.feedc                 C   s   t || j| jd| _| jS )N)store_line_numbers)TreeBuilderForHtml5libr"   r3   r%   )r   namespaceHTMLElementsr   r   r   r$   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   featuresZTRACKS_LINE_NUMBERSr   r2   r$   r7   r   r   r   r   r   *   s    
c                       sf   e Zd Zd fdd	Zdd Zdd Zd	d
 Zdd Zdd Zdd Z	dd Z
dd Zdd Z  ZS )r4   NTc                    sL   |r|| _ n ddlm} |dd|i|| _ tt| | d | _|| _d S )Nr   BeautifulSoup html.parserr3   )r@   rA   )r"   bs4r?   superr4   __init__r&   r3   )r   r5   r"   r3   kwargsr?   	__class__r   r   rD      s     zTreeBuilderForHtml5lib.__init__c                 C   s   | j   t| j | j d S N)r"   resetElementr   r   r   r   documentClass   s    
z$TreeBuilderForHtml5lib.documentClassc                 C   s6   |d }|d }|d }t |||}| j| d S )Nr/   publicIdsystemId)r   Zfor_name_and_idsr"   object_was_parsed)r   tokenr/   rM   rN   doctyper   r   r   insertDoctype   s
    z$TreeBuilderForHtml5lib.insertDoctypec                 C   sZ   i }| j r6| jr6| j jj \}}||d< |d |d< | jj||fi |}t|| j|S )N
sourceline   	sourcepos)r&   r3   r-   r.   positionr"   new_tagrJ   )r   r/   	namespacerE   rS   rU   tagr   r   r   elementClass   s    z#TreeBuilderForHtml5lib.elementClassc                 C   s   t t|| jS rH   )TextNoder   r"   )r   datar   r   r   commentClass   s    z#TreeBuilderForHtml5lib.commentClassc                 C   s0   ddl m} |dd| _d| j_t| j| jd S )Nr   r>   r@   rA   z[document_fragment])rB   r?   r"   r/   rJ   )r   r?   r   r   r   fragmentClass   s    z$TreeBuilderForHtml5lib.fragmentClassc                 C   s   | j |j d S rH   )r"   appendelementr   noder   r   r   appendChild   s    z"TreeBuilderForHtml5lib.appendChildc                 C   s   | j S rH   )r"   rK   r   r   r   getDocument   s    z"TreeBuilderForHtml5lib.getDocumentc                 C   s   t j| jS rH   )treebuilder_baseTreeBuildergetFragmentr`   rK   r   r   r   rg      s    z"TreeBuilderForHtml5lib.getFragmentc                    sB   ddl m  g tdd fdd	|d dS )Nr   r>   z8^(.*?)(?: PUBLIC "(.*?)"(?: "(.*?)")?| SYSTEM "(.*?)")?$c           	         s  t |  r
t | tr| }|r|d}|jdkrx|dpBd}|dpZ|dpZd}dd| |||f  qdd| |f  nd	d| f  n:t | tr̈d
d| | f  nt | trdd| | f  n| jrdt	| j | j
f }n| j
}dd| |f  | jrg }t| j D ]N\}}t |trjdt	|j |j
f }t |trd|}|||f qBt|D ]&\}}dd|d  ||f  q|d7 }| jD ]}|| qd S )NrT      r@   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	lastindexr_   r   r   rX   r   r/   attrslistitemsr	   joinsortedchildren)	r`   indentmr/   rM   rN   
attributesvaluechildr?   Z
doctype_rervserializeElementr   r   rz      sH    








 
z?TreeBuilderForHtml5lib.testSerializer.<locals>.serializeElement
)r   )rB   r?   recompilerp   r   r`   r   rx   r   testSerializer   s    
)
z%TreeBuilderForHtml5lib.testSerializer)NT)r8   r9   r:   rD   rL   rR   rZ   r]   r^   rc   rd   rg   r   __classcell__r   r   rF   r   r4      s     r4   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 rH   )r`   r'   rm   r~   r   r   r   rD      s    zAttrList.__init__c                 C   s   t | j  S rH   )rn   rm   ro   __iter__rK   r   r   r   r      s    zAttrList.__iter__c                 C   s^   | j jp
i }||dg v s<| j j|v rP||| j jg v rPt|tsPt|}|| j |< d S )N*)r`   Zcdata_list_attributesgetr/   r(   rn   r
   findall)r   r/   rv   Z	list_attrr   r   r   __setitem__   s    


zAttrList.__setitem__c                 C   s   t | j S rH   )rn   rm   ro   rK   r   r   r   ro     s    zAttrList.itemsc                 C   s   t | j S rH   rn   rm   keysrK   r   r   r   r     s    zAttrList.keysc                 C   s
   t | jS rH   )lenrm   rK   r   r   r   __len__  s    zAttrList.__len__c                 C   s
   | j | S rH   )rm   r   r/   r   r   r   __getitem__  s    zAttrList.__getitem__c                 C   s   |t | j v S rH   r   r   r   r   r   __contains__  s    zAttrList.__contains__N)r8   r9   r:   rD   r   r   ro   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 )rJ   c                 C   s&   t j| |j || _|| _|| _d S rH   )re   NoderD   r/   r`   r"   rX   )r   r`   r"   rX   r   r   r   rD     s    zElement.__init__c                 C   s*  d  }}t |tr| }}n:t |tr,|}n*|jjtkrJ|j }}| |_n|j}| |_t |tst|jd urt|j  |d ur| jjr| jjd jtkr| jjd }| j	
|| }|| || j	_n`t |tr| j	
|}| jjr| 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   r`   rG   r   r   extractcontentsr"   
new_stringreplace_withZ_most_recent_element_last_descendantnext_elementrO   )r   rb   Zstring_childrw   Zold_elementZnew_elementr   r   r   r   rc     s<    







zElement.appendChildc                 C   s   t | jtri S t| jS rH   )r(   r`   r   r   rK   r   r   r   getAttributesU  s    zElement.getAttributesc                 C   s   |d urt |dkrg }t| D ](\}}t|tr$t| }||= |||< q$| jj| j	| t| D ]\}}|| j
|< ql| jj| j
 d S )Nr   )r   rn   ro   r(   tupler	   r"   builderZ$_replace_cdata_list_attribute_valuesr/   r`   Zset_up_substitutions)r   ru   Zconverted_attributesr/   rv   new_namer   r   r   setAttributesZ  s    

zElement.setAttributesNc                 C   s4   t | j|| j}|r&| || n
| | d S rH   )r[   r"   r   insertBeforerc   )r   r\   r   textr   r   r   
insertTextp  s    zElement.insertTextc                 C   s   | j |j }|j jtkrf| j jrf| j j|d  jtkrf| j j|d  }| j||j  }|| n| j ||j  | |_	d S )NrT   )
r`   indexrG   r   r   r"   r   r   insertr   )r   rb   ZrefNoder   old_nodeZnew_strr   r   r   r   w  s    zElement.insertBeforec                 C   s   |j   d S rH   )r`   r   ra   r   r   r   removeChild  s    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r|d }	|durr||	_n||	_||	_|dur|	|_n|	|_|dur|	|_|d dd}
||
_|dur|
|_d|
_|D ]}||_|j	| qg |_||_dS )z1Move all of this tag's children into another tag.Fr   r   NT)
r`   Znext_siblingr   r   r   r   Zprevious_elementZprevious_siblingr   r_   )r   Z
new_parentr`   Znew_parent_elementZfinal_next_elementZnew_parents_last_descendantZnew_parents_last_childZ(new_parents_last_descendant_next_elementZ	to_appendZfirst_childZlast_childs_last_descendantrw   r   r   r   reparentChildren  s>    
zElement.reparentChildrenc                 C   sB   | j | jj| j}t|| j | j}| jD ]\}}||j|< q*|S rH   )r"   rW   r`   r/   rX   rJ   ru   )r   rY   rb   keyrv   r   r   r   	cloneNode  s
    zElement.cloneNodec                 C   s   | j jS rH   )r`   r   rK   r   r   r   
hasContent  s    zElement.hasContentc                 C   s(   | j d krtd | jfS | j | jfS d S )Nhtml)rX   r   r/   rK   r   r   r   getNameTuple  s    
zElement.getNameTuple)N)r8   r9   r:   rD   rc   r   r   propertyru   r   r   r   r   r   r   r   Z	nameTupler   r   r   r   rJ     s   6

BrJ   c                   @   s   e Zd Zdd Zdd ZdS )r[   c                 C   s   t j| d  || _|| _d S rH   )re   r   rD   r`   r"   )r   r`   r"   r   r   r   rD     s    zTextNode.__init__c                 C   s   t d S rH   )NotImplementedErrorrK   r   r   r   r     s    zTextNode.cloneNodeN)r8   r9   r:   rD   r   r   r   r   r   r[     s   r[   )$__license____all__r   r|   Zbs4.builderr   r   r   r   r   Zbs4.elementr	   r
   r   Zhtml5lib.constantsr   r   r   r   r   r   Zhtml5lib.treebuildersr   re   r*   ImportErrorer   r   rf   r4   objectr   r   rJ   r[   r   r   r   r   <module>   s*   Xx C