ó
V]c @ s‚ d Z d d l m Z d d l Z d d l Z d d l m Z m Z m Z d d d d g Z d Z
d
Z d Z d Z
d
Z d Z d e f d „ ƒ YZ e ƒ Z d e f d „ ƒ YZ d e f d „ ƒ YZ d e f d „ ƒ YZ d „ Z d e f d „ ƒ YZ d d l m Z e e ƒ y d d l m Z e e ƒ Wn e k
rKn Xy d d l m Z e e ƒ Wn e k
r}n Xd S( t MITiÿÿÿÿ( t defaultdictN( t CharsetMetaAttributeValuet ContentMetaAttributeValuet nonwhitespace_ret HTMLTreeBuildert SAXTreeBuildert TreeBuildert TreeBuilderRegistryt fastt
permissivet strictt xmlt htmlt html5c B s# e Z d „ Z d „ Z d „ Z RS( c C s t t ƒ | _ g | _ d S( N( R t listt builders_for_featuret builders( t self( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt __init__ s c C sB x( | j D] } | j | j d | ƒ q
W| j j d | ƒ d S( s8 Register a treebuilder based on its advertised features.i N( t featuresR t insertR ( R t treebuilder_classt feature( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt register# s c G s t | j ƒ d k r d St | ƒ d k r6 | j d St | ƒ } | j ƒ d } d } x‚ t | ƒ d k rÜ | j ƒ } | j j | g ƒ } t | ƒ d k r[ | d k rÁ | } t | ƒ } qÙ | j t | ƒ ƒ } q[ q[ W| d k rí d Sx | D] } | | k rô | Sqô Wd S( Ni (
t lenR t NoneR t reverset popR t gett sett intersection( R R t
candidatest
candidate_setR t we_have_the_featuret candidate( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt lookup) s.
( t __name__t
__module__R R R$ ( ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyR s c B s§ e Z d Z d Z g Z g Z e Z e Z d Z
i Z e ƒ Z
e ƒ Z e e d „ Z d „ Z d „ Z d „ Z d „ Z d d d „ Z d „ Z d „ Z d
„ Z RS( s2 Turn a document into a Beautiful Soup object tree.s [Unknown tree builder]c C sU d | _ | | j k r$ | j } n | | _ | | j k rH | j } n | | _ d S( s Constructor.
:param multi_valued_attributes: If this is set to None, the
TreeBuilder will not turn any values for attributes like
'class' into lists. Setting this do a dictionary will
customize this behavior; look at DEFAULT_CDATA_LIST_ATTRIBUTES
for an example.
Internally, these are called "CDATA list attributes", but that
probably doesn't make sense to an end-user, so the argument name
is `multi_valued_attributes`.
:param preserve_whitespace_tags:
N( R t soupt USE_DEFAULTt DEFAULT_CDATA_LIST_ATTRIBUTESt cdata_list_attributest DEFAULT_PRESERVE_WHITESPACE_TAGSt preserve_whitespace_tags( R t multi_valued_attributesR, ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyR g s c C s
| | _ d S( sp The BeautifulSoup object has been initialized and is now
being associated with the TreeBuilder.
N( R' ( R R' ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt initialize_soup~ s c C s d S( N( ( R ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt reset„ s c C s | j d k r t S| | j k S( s¤ 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 tag to be
an empty-element tag (it's not in
HTMLBuilder.empty_element_tags). This means an empty
tag
will be presented as "
", not "".
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.
"" will become "", and "bar" will
be left alone.
N( t empty_element_tagsR t True( R t tag_name( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt can_be_empty_element‡ s c C s
t ƒ ‚ d S( N( t NotImplementedError( R t markup( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt feedœ s c C s | d d t f S( N( R t False( R R5 t user_specified_encodingt document_declared_encoding( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt prepare_markupŸ s c C s | S( s¥ Wrap an HTML fragment to make it look like a document.
Different parsers do this differently. For instance, lxml
introduces an empty 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 t fragment( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt test_fragment_to_document£ s c C s t S( N( R7 ( R t tag( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt set_up_substitutions° s c C s½ | s
| S| j r¹ | j j d g ƒ } | j j | j ƒ d ƒ } xs | j ƒ D]b } | | k st | rP | | k rP | | } t | t ƒ rŸ t j | ƒ } n | } | | | RK ( ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyR S s(
c B sq e Z d Z d „ Z d „ Z d „ Z d „ Z d „ Z d „ Z d „ Z d „ Z
d „ Z d
„ Z d „ Z
RS( s9 A Beautiful Soup treebuilder that listens for SAX events.c C s
t ƒ ‚ d S( N( R4 ( R R5 ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyR6 Ô s c C s d S( N( ( R ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt close× s c C s9 t d „ t | j ƒ ƒ Dƒ ƒ } | j j | | ƒ d S( Nc s s% | ] \ } } | d | f Vq d S( i N( ( t .0t keyRI ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pys Û s ( t dictR t itemsR' t handle_starttag( R t nameRE ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt startElementÚ s "c C s | j j | ƒ d S( N( R' t
handle_endtag( R RX ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt
endElementß s c C s | j | | ƒ d S( N( RY ( R t nsTuplet nodeNameRE ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt startElementNSã s c C s | j | ƒ d S( N( R[ ( R R\ R] ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt endElementNSç s c C s d S( N( ( R t prefixt nodeValue( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt startPrefixMappingì s c C s d S( N( ( R R` ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt endPrefixMappingð s c C s | j j | ƒ d S( N( R' t handle_data( R t content( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt
charactersõ s c C s d S( N( ( R ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt
startDocumentø s c C s d S( N( ( R ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt endDocumentû s ( R% R& RL R6 RR RY R[ R^ R_ Rb Rc Rf Rg Rh ( ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyR Ñ s c $ B s| e Z d 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 ƒ 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/ d0 d1 d2 d3 d4 d5 d6 d7 d8 d9 d: g# ƒ Z i d; d< d= g d> 6d? d@ g dA 6d? d@ g d
6dB g dC 6dB g dD 6dB g dC 6dE g d& 6dF g dG 6d? g d 6dH g dI 6dJ g dK 6dL g d3 6Z e d5 dM g ƒ Z dN „ Z RS(O s] This TreeBuilder knows facts about HTML.
Such as which tags are empty-element tags.
t areat baset brt colt embedt hrt imgt inputt keygent linkt menuitemt metat paramt sourcet trackt wbrt basefontt bgsoundt commandt framet imaget isindext nextidt spacert addresst articlet asidet
blockquotet canvast ddt divt dlt dtt fieldsett
figcaptiont figuret footert formt h1t h2t h3t h4t h5t h6t headert lit maint navt noscriptt olt outputt pt pret sectiont tablet tfoott ult videot classt accesskeyt dropzoneR? t relt revt at headerst tdt ths accept-charsett archiveRQ t sizest icont sandboxt iframet fort textareac C s² | j d k r t S| j d ƒ } | j d ƒ } | j d ƒ } d } | d k rk | } t | ƒ | d - s ( R% R& RL R R0 t block_elementsR) R+ R> ( ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyR ÿ s( 0!u
c C sv t j d } xb | j D]W } t | | ƒ } t | t ƒ r t | | | ƒ | j j | ƒ | j j | ƒ q q Wd S( s9 Copy TreeBuilders from the given module into this module.s bs4.builderN(
t syst modulest __all__t getattrt
issubclassR t setattrt appendt builder_registryR ( t modulet this_moduleRX t obj( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt register_treebuilders_fromN s
t ParserRejectedMarkupc B s e Z RS( ( R% R& ( ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyRÃ [ s i ( t _htmlparser( t _html5lib( t _lxml( t __license__t collectionsR t itertoolsR· t bs4.elementR R R R¹ t FASTt
PERMISSIVEt STRICTt XMLt HTMLt HTML_5RQ R R¾ R R R RÂ t ExceptionRÃ t RÄ RÅ t ImportErrorRÆ ( ( ( s3 lib/python2.7/site-packages/bs4/builder/__init__.pyt sB 4 ~.O