B
    `<I              
   @   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 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 yddlmZ d	ZW n2 ek
r Z zdd
lmZ dZW ddZ[X Y nX 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)
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   s&   || _ |rt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.F)user_specified_encodingwarningswarn)selfmarkupr   Zdocument_declared_encodingZexclude_encodings r   ]/home/kop/projects/devel/pgwui/test_venv/lib/python3.7/site-packages/bs4/builder/_html5lib.pyprepare_markup?   s    
zHTML5TreeBuilder.prepare_markupc             C   s   | j jd k	rtd tj| jd}|| j_t	 }t
|tsVtrL| j|d< n
| j|d< |j|f|}t
|trvd |_n$|jjjd }t
|ts|j}||_d | j_d S )NzYou provided a value for parse_only, but the html5lib tree builder doesn't support parse_only. The entire document will be parsed.)treeoverride_encodingencodingr   )soupZ
parse_onlyr   r   r   
HTMLParsercreate_treebuilderunderlying_builderparserdict
isinstancestrnew_html5libr   parseoriginal_encoding	tokenizerstreamcharEncodingname)r   r   r"   extra_kwargsdocr(   r   r   r   feedL   s"    




zHTML5TreeBuilder.feedc             C   s   t || j| jd| _| jS )N)store_line_numbers)TreeBuilderForHtml5libr   r0   r!   )r   namespaceHTMLElementsr   r   r   r    h   s    z#HTML5TreeBuilder.create_treebuilderc             C   s   d| S )zSee `TreeBuilder`.z)<html><head></head><body>%s</body></html>r   )r   fragmentr   r   r   test_fragment_to_documento   s    z*HTML5TreeBuilder.test_fragment_to_document)NN)__name__
__module____qualname____doc__NAMEr   r   r   featuresZTRACKS_LINE_NUMBERSr   r/   r    r4   r   r   r   r   r   )   s   
c                   sf   e Zd Zd fdd	Zdd Zdd Zd	d
 Zdd Zdd Zdd Z	dd Z
dd Zdd Z  ZS )r1   NTc                sL   |r|| _ n ddlm} |dd|i|| _ tt| | d | _|| _d S )Nr   )BeautifulSoup html.parserr0   )r<   r=   )r   bs4r;   superr1   __init__r"   r0   )r   r2   r   r0   kwargsr;   )	__class__r   r   r@   v   s    
zTreeBuilderForHtml5lib.__init__c             C   s   | j   t| j | j d S )N)r   resetElement)r   r   r   r   documentClass   s    
z$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,   rF   rG   doctyper   r   r   insertDoctype   s
    z$TreeBuilderForHtml5lib.insertDoctypec             C   sV   i }| j r6| jr6| j jj \}}||d< |d |d< | jj||f|}t|| j|S )N
sourceline   	sourcepos)r"   r0   r)   r*   positionr   new_tagrD   )r   r,   	namespacerA   rL   rN   tagr   r   r   elementClass   s    z#TreeBuilderForHtml5lib.elementClassc             C   s   t t|| jS )N)TextNoder   r   )r   datar   r   r   commentClass   s    z#TreeBuilderForHtml5lib.commentClassc             C   s0   ddl m} |dd| _d| j_t| j| jd S )Nr   )r;   r<   zhtml.parserz[document_fragment])r>   r;   r   r,   rD   )r   r;   r   r   r   fragmentClass   s    z$TreeBuilderForHtml5lib.fragmentClassc             C   s   | j |j d S )N)r   appendelement)r   noder   r   r   appendChild   s    z"TreeBuilderForHtml5lib.appendChildc             C   s   | j S )N)r   )r   r   r   r   getDocument   s    z"TreeBuilderForHtml5lib.getDocumentc             C   s   t j| jS )N)treebuilder_baseTreeBuildergetFragmentrY   )r   r   r   r   r_      s    z"TreeBuilderForHtml5lib.getFragmentc                sB   ddl m  g tdd fdd	|d dS )Nr   )r;   z8^(.*?)(?: PUBLIC "(.*?)"(?: "(.*?)")?| SYSTEM "(.*?)")?$c       	         s  t |  r
t | tr| }|r|d}|jdkrx|dpBd}|dpZ|dpZd}dd| |||f  qdd| |f  nd	d| f  nHt | tr̈d
d| | f  n$t | trdd| | f  n | jrdt	| j | j
f }n| j
}dd| |f  | jrg }x`t| j D ]N\}}t |trndt	|j |j
f }t |trd|}|||f qFW x2t|D ]&\}}dd|d  ||f  qW |d7 }x| jD ]}|| qW d S )NrM      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$   r   matchgroup	lastindexrX   r   r   rQ   r   r,   attrslistitemsr   joinsortedchildren)	rY   indentmr,   rF   rG   
attributesvaluechild)r;   
doctype_rervserializeElementr   r   rt      sD    








"z?TreeBuilderForHtml5lib.testSerializer.<locals>.serializeElement
)r   )r>   r;   recompilerj   )r   rY   r   )r;   rr   rs   rt   r   testSerializer   s    
)
z%TreeBuilderForHtml5lib.testSerializer)NT)r5   r6   r7   r@   rE   rK   rS   rV   rW   r[   r\   r_   rx   __classcell__r   r   )rB   r   r1   t   s    r1   c               @   sL   e Zd Zdd Zdd Zdd Zdd Zd	d
 Zdd Zdd Z	dd Z
dS )AttrListc             C   s   || _ t| j j| _d S )N)rY   r#   rg   )r   rY   r   r   r   r@      s    zAttrList.__init__c             C   s   t | j  S )N)rh   rg   ri   __iter__)r   r   r   r   r{      s    zAttrList.__iter__c             C   sR   | j j}||d ks0| j j|krD||| j j krDt|tsDt|}|| j |< d S )N*)rY   Zcdata_list_attributesr,   r$   rh   r	   findall)r   r,   rp   Z	list_attrr   r   r   __setitem__   s    

zAttrList.__setitem__c             C   s   t | j S )N)rh   rg   ri   )r   r   r   r   ri      s    zAttrList.itemsc             C   s   t | j S )N)rh   rg   keys)r   r   r   r   r      s    zAttrList.keysc             C   s
   t | jS )N)lenrg   )r   r   r   r   __len__  s    zAttrList.__len__c             C   s
   | j | S )N)rg   )r   r,   r   r   r   __getitem__  s    zAttrList.__getitem__c             C   s   |t | j kS )N)rh   rg   r   )r   r,   r   r   r   __contains__  s    zAttrList.__contains__N)r5   r6   r7   r@   r{   r~   ri   r   r   r   r   r   r   r   r   rz      s   rz   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 )rD   c             C   s&   t j| |j || _|| _|| _d S )N)r]   Noder@   r,   rY   r   rQ   )r   rY   r   rQ   r   r   r   r@     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 k	rt|j  |d k	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 k	r| j	 }n| j}| j	j|| j|d d S )NF)parentmost_recent_element)r$   r%   r   rY   rB   r   r   extractcontentsr   
new_stringreplace_withZ_most_recent_element_last_descendantnext_elementrH   )r   rZ   Zstring_childrq   Zold_elementZnew_elementr   r   r   r   r[     s8    







zElement.appendChildc             C   s   t | jtri S t| jS )N)r$   rY   r   rz   )r   r   r   r   getAttributesG  s    zElement.getAttributesc             C   s   |d k	rt |dkrg }x8t| D ](\}}t|tr&t| }||= |||< q&W | jj| j	| x"t| D ]\}}|| j
|< qrW | jj| j
 d S )Nr   )r   rh   ri   r$   tupler   r   ZbuilderZ$_replace_cdata_list_attribute_valuesr,   rY   Zset_up_substitutions)r   ro   Zconverted_attributesr,   rp   new_namer   r   r   setAttributesL  s    

zElement.setAttributesNc             C   s4   t | j|| j}|r&| || n
| | d S )N)rT   r   r   insertBeforer[   )r   rU   r   textr   r   r   
insertTextb  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 )NrM   )
rY   indexrB   r   r   r   r   r   insertr   )r   rZ   refNoder   Zold_nodeZnew_strr   r   r   r   i  s    zElement.insertBeforec             C   s   |j   d S )N)rY   r   )r   rZ   r   r   r   removeChildu  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k	rr||	_n||	_||	_|dk	r|	|_n|	|_|dk	r|	|_|d dd}
||
_|dk	r|
|_d|
_x|D ]}||_|j	| qW g |_||_dS )z1Move all of this tag's children into another tag.Fr   r   NT)
rY   Znext_siblingr   r   r   r   Zprevious_elementZprevious_siblingr   rX   )r   Z
new_parentrY   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_descendantrq   r   r   r   reparentChildrenx  s>    

zElement.reparentChildrenc             C   sF   | j | jj| j}t|| j | j}x| jD ]\}}||j|< q,W |S )N)r   rP   rY   r,   rQ   rD   ro   )r   rR   rZ   keyrp   r   r   r   	cloneNode  s
    zElement.cloneNodec             C   s   | j jS )N)rY   r   )r   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)rQ   r
   r,   )r   r   r   r   getNameTuple  s    
zElement.getNameTuple)N)r5   r6   r7   r@   r[   r   r   propertyro   r   r   r   r   r   r   r   	nameTupler   r   r   r   rD   
  s   6

BrD   c               @   s   e Zd Zdd Zdd ZdS )rT   c             C   s   t j| d  || _|| _d S )N)r]   r   r@   rY   r   )r   rY   r   r   r   r   r@     s    zTextNode.__init__c             C   s   t d S )N)NotImplementedError)r   r   r   r   r     s    zTextNode.cloneNodeN)r5   r6   r7   r@   r   r   r   r   r   rT     s   rT   )#__license____all__r   rv   Zbs4.builderr   r   r   r   Zbs4.elementr   r	   r   Zhtml5lib.constantsr
   r   r   r   r   r   Zhtml5lib.treebuildersr   r]   r&   ImportErrorer   r   r^   r1   objectrz   r   rD   rT   r   r   r   r   <module>   s(   Kx C