
­­­­­­­­­­­­­­­­­­
<!DOCTYPE html>
<html>
3
6]/                 @   s,  d dl mZ d dlZd dlZd dlmZmZmZmZ ddddgZ	dZ
d	Zd
ZdZdZdZG dd deZe ZG dd deZG dd deZG dd deZdd ZG dd deZddlmZ ee yddlmZ ee W n ek
r   Y nX yddlmZ ee W n ek
r&   Y nX dS )    )defaultdictN)CharsetMetaAttributeValueContentMetaAttributeValueHTMLAwareEntitySubstitutionwhitespace_reHTMLTreeBuilderSAXTreeBuilderTreeBuilderTreeBuilderRegistryZfastZ
permissivestrictZxmlZhtmlZhtml5c               @   s$   e Zd Zdd Zdd Zdd ZdS )r
   c             C   s   t t| _g | _d S )N)r   listbuilders_for_featurebuilders)self r   /usr/lib/python3.6/__init__.py__init__    s    
zTreeBuilderRegistry.__init__c             C   s4   x |j D ]}| j| jd| qW | jjd| dS )z8Register a treebuilder based on its advertised features.r   N)featuresr   insertr   )r   Ztreebuilder_classfeaturer   r   r   register$   s    zTreeBuilderRegistry.registerc             G   s   t | jdkrd S t |dkr(| jd S t|}|j  d }d }xVt |dkr|j }| jj|g }t |dkrB|d kr|}t|}qB|jt|}qBW |d krd S x|D ]}||kr|S qW d S )Nr   )	lenr   r   reversepopr   getsetintersection)r   r   Z
candidatesZcandidate_setr   Zwe_have_the_feature	candidater   r   r   lookup*   s.    


zTreeBuilderRegistry.lookupN)__name__
__module____qualname__r   r   r   r   r   r   r   r
      s   c               @   st   e Zd ZdZdZg Zg ZdZdZe	 Z
dZi Zdd Zdd Zd	d
 Zdd ZdddZdd Zdd Zdd ZdS )r	   z2Turn a document into a Beautiful Soup object tree.z[Unknown tree builder]FNc             C   s
   d | _ d S )N)soup)r   r   r   r   r   f   s    zTreeBuilder.__init__c             C   s   d S )Nr   )r   r   r   r   reseti   s    zTreeBuilder.resetc             C   s   | j dkrdS || j kS )a  Might a tag with this name be an empty-element tag?

        The final markup may or may not actually present this tag as
        self-closing.

        For instance: an HTMLBuilder does not consider a <p> tag to be
        an empty-element tag (it's not in
        HTMLBuilder.empty_element_tags). This means an empty <p> tag
        will be presented as "<p></p>", not "<p />".

        The default implementation has no opinion about which tags are
        empty-element tags, so a tag will be presented as an
        empty-element tag if and only if it has no contents.
        "<foo></foo>" will become "<foo />", and "<foo>bar</foo>" will
        be left alone.
        NT)empty_element_tags)r   tag_namer   r   r   can_be_empty_elementl   s    
z TreeBuilder.can_be_empty_elementc             C   s
   t  d S )N)NotImplementedError)r   markupr   r   r   feed   s    zTreeBuilder.feedc             C   s   |d d dfS )NFr   )r   r(   Zuser_specified_encodingZdocument_declared_encodingr   r   r   prepare_markup   s    zTreeBuilder.prepare_markupc             C   s   |S )a  Wrap an HTML fragment to make it look like a document.

        Different parsers do this differently. For instance, lxml
        introduces an empty <head> tag, and html5lib
        doesn't. Abstracting this away lets us write simple tests
        which run HTML fragments through the parser and compare the
        results against other HTML fragments.

        This method should not be used outside of tests.
        r   )r   Zfragmentr   r   r   test_fragment_to_document   s    z%TreeBuilder.test_fragment_to_documentc             C   s   dS )NFr   )r   tagr   r   r   set_up_substitutions   s    z TreeBuilder.set_up_substitutionsc             C   s   |s|S | j r| j jdg }| j j|j d}xRt|j D ]B}||ksT|r<||kr<|| }t|trrtj|}n|}|||< q<W |S )zaReplaces class="foo bar" with class=["foo", "bar"]

        Modifies its input in place.
        *N)	cdata_list_attributesr   lowerr   keys
isinstancestrr   split)r   r%   attrsZ	universalZtag_specificattrvaluevaluesr   r   r   $_replace_cdata_list_attribute_values   s    
z0TreeBuilder._replace_cdata_list_attribute_values)NN)r   r    r!   __doc__NAMEZALTERNATE_NAMESr   Zis_xmlZ	picklabler   preserve_whitespace_tagsr$   r/   r   r#   r&   r)   r*   r+   r-   r9   r   r   r   r   r	   T   s$    
c               @   sh   e Zd ZdZdd Zdd Zdd Zdd	 Zd
d Zdd Z	dd Z
dd Zdd Zdd Zdd ZdS )r   z9A Beautiful Soup treebuilder that listens for SAX events.c             C   s
   t  d S )N)r'   )r   r(   r   r   r   r)      s    zSAXTreeBuilder.feedc             C   s   d S )Nr   )r   r   r   r   close   s    zSAXTreeBuilder.closec             C   s,   t dd t|j D }| jj|| d S )Nc             s   s   | ]\}}|d  |fV  qdS )   Nr   ).0keyr7   r   r   r   	<genexpr>   s    z.SAXTreeBuilder.startElement.<locals>.<genexpr>)dictr   itemsr"   Zhandle_starttag)r   namer5   r   r   r   startElement   s    zSAXTreeBuilder.startElementc             C   s   | j j| d S )N)r"   Zhandle_endtag)r   rD   r   r   r   
endElement   s    zSAXTreeBuilder.endElementc             C   s   | j || d S )N)rE   )r   nsTuplenodeNamer5   r   r   r   startElementNS   s    zSAXTreeBuilder.startElementNSc             C   s   | j | d S )N)rF   )r   rG   rH   r   r   r   endElementNS   s    zSAXTreeBuilder.endElementNSc             C   s   d S )Nr   )r   prefixZ	nodeValuer   r   r   startPrefixMapping   s    z!SAXTreeBuilder.startPrefixMappingc             C   s   d S )Nr   )r   rK   r   r   r   endPrefixMapping   s    zSAXTreeBuilder.endPrefixMappingc             C   s   | j j| d S )N)r"   Zhandle_data)r   contentr   r   r   
characters   s    zSAXTreeBuilder.charactersc             C   s   d S )Nr   )r   r   r   r   startDocument   s    zSAXTreeBuilder.startDocumentc             C   s   d S )Nr   )r   r   r   r   endDocument   s    zSAXTreeBuilder.endDocumentN)r   r    r!   r:   r)   r=   rE   rF   rI   rJ   rL   rM   rO   rP   rQ   r   r   r   r   r      s   c            $   @   s   e Zd ZdZejZedddddddd	d
dddddddddddddddgZeddddddd d!d"d#d$d%d&d'd(d)d*d+d,d-d.dd/d0d1d2d3d4d5d6d7d8d9d:d;g#Zd<d=d>gd?d@gd?d@gdAgdAgdAgdBgdCgd?gdDgdEgdFgdGZ	dHdI Z
dJS )Kr   z]This TreeBuilder knows facts about HTML.

    Such as which tags are empty-element tags.
    areabasebrcolZembedZhrZimginputZkeygenlinkZmenuitemmetaZparamsourceZtrackZwbrZbasefontZbgsoundZcommandframeZimageZisindexZnextidZspacerZaddressZarticleZasideZ
blockquoteZcanvasZddZdivZdlZdtZfieldsetZ
figcaptionZfigureZfooterformZh1Zh2Zh3Zh4Zh5Zh6headerZlimainZnavZnoscriptZoloutputpZpreZsectiontableZtfootZulZvideoclassZ	accesskeyZdropzoneZrelZrevZheaderszaccept-charsetarchiveZsizesZsandboxfor)r.   arW   tdZthre   r[   objectrR   ZiconZiframer^   c             C   sz   |j dkrdS |jd}|jd}|jd}d }|d k	rJ|}t||d< n(|d k	rr|d k	rr|j dkrrt||d< |d k	S )NrX   Fz
http-equivrN   charsetzcontent-type)rD   r   r   r0   r   )r   r,   Z
http_equivrN   rg   Zmeta_encodingr   r   r   r-     s    



z$HTMLTreeBuilder.set_up_substitutionsN)r   r    r!   r:   r   r<   r   r$   Zblock_elementsr/   r-   r   r   r   r   r      s&    N

c             C   sV   t jd }xF| jD ]<}t| |}t|trt||| |jj| |jj	| qW dS )z9Copy TreeBuilders from the given module into this module.zbs4.builderN)
sysmodules__all__getattr
issubclassr	   setattrappendbuilder_registryr   )moduleZthis_modulerD   objr   r   r   register_treebuilders_from2  s    


rr   c               @   s   e Zd ZdS )ParserRejectedMarkupN)r   r    r!   r   r   r   r   rs   ?  s   rs   r>   )_htmlparser)	_html5lib)_lxml)collectionsr   	itertoolsrh   Zbs4.elementr   r   r   r   rj   ZFASTZ
PERMISSIVEZSTRICTZXMLZHTMLZHTML_5rf   r
   ro   r	   r   r   rr   	Exceptionrs    rt   ru   ImportErrorrv   r   r   r   r   <module>   s@   4b.N