MfK: d Z dZdgZddlmZ ddlZddlZddlmZm Z m
Z
m Z m Z ddl
mZmZ ddlmZmZmZmZmZ d Z G d
d ee Z G d de ZdS )
zCUse the HTMLParser library to parse HTML files that aren't too bad.MITHTMLParserTreeBuilder )
HTMLParserN)CDataComment DeclarationDoctypeProcessingInstruction)EntitySubstitution
UnicodeDammit)DetectsXMLParsedAsHTMLParserRejectedMarkupHTMLHTMLTreeBuilderSTRICTz html.parserc f e Zd ZdZdZdZd Zd Zd ZddZ dd Z
d
Z d Z d Z
d
Zd Zd Zd ZdS )BeautifulSoupHTMLParserzA subclass of the Python standard library's HTMLParser class, which
listens for HTMLParser events and translates them into calls
to Beautiful Soup's tree construction API.
ignorereplacec | d| j | _ t j | g|R i | g | _ | dS )a Constructor.
:param on_duplicate_attribute: A strategy for what to do if a
tag includes the same attribute more than once. Accepted
values are: REPLACE (replace earlier values with later
ones, the default), IGNORE (keep the earliest value
encountered), or a callable. A callable must take three
arguments: the dictionary of attributes already processed,
the name of the duplicate attribute, and the most recent value
encountered.
on_duplicate_attributeN)popREPLACEr r __init__ already_closed_empty_element_initialize_xml_detector)selfargskwargss V/var/www/surfInsights/venv3-11/lib/python3.11/site-packages/bs4/builder/_htmlparser.pyr z BeautifulSoupHTMLParser.__init__. sf '-jj $dl'
'
# D242226222 -/) %%''''' c t | )N)r )r messages r errorz BeautifulSoupHTMLParser.errorJ s #7+++r! c ` | ||d }| | dS )zHandle an incoming empty-element tag.
This is only called when the markup looks like .
:param name: Name of the tag.
:param attrs: Dictionary of the tag's attributes.
F)handle_empty_elementN)handle_starttag
handle_endtag)r nameattrstags r handle_startendtagz*BeautifulSoupHTMLParser.handle_startendtagZ s8 ""4U"KK 4 r! Tc i }|D ]F\ }}|d}||v r2| j }|| j k rn$|d| j fv r|||< n |||| n|||< d}G| \ } }
| j |dd|| |
} | r:| j r3|r1| |d | j | | j
| | dS dS )a3 Handle an opening tag, e.g. ''
:param name: Name of the tag.
:param attrs: Dictionary of the tag's attributes.
:param handle_empty_element: True if this tag is known to be
an empty-element tag (i.e. there is not expected to be any
closing tag).
N z"")
sourceline sourceposF)check_already_closed) r IGNOREr getpossoupr' is_empty_elementr( r append _root_tag_root_tag_encountered) r r) r* r& attr_dictkeyvalueon_dupe attrvaluer/ r0 r+ s r r' z'BeautifulSoupHTMLParser.handle_starttagi sG JC }i 5dk)) t| 444%*IcNNGIsE2222!& # II $
Ii'' $iJ (
;3' ;,@ ;
t% @ @ @
- 4 4T : : : > ! & &t , , , , , " !r! c |r%|| j v r | j | dS | j | dS )zHandle a closing tag, e.g. ''
:param name: A tag name.
:param check_already_closed: True if this tag is expected to
be the closing portion of an empty-element tag,
e.g. ''.
N)r remover4 r( )r r) r1 s r r( z%BeautifulSoupHTMLParser.handle_endtag sT *DD,M$M$M
- 4 4T : : : : : I # #D ) ) ) ) )r! c : | j | dS )z4Handle some textual data that shows up between tags.N)r4 handle_datar datas r rA z#BeautifulSoupHTMLParser.handle_data s d#####r! c J | d r$t | d d }nH| d r$t | d d }nt | }d}|dk rO| j j dfD ]@}|s t |g | }*# t $ r
}Y d}~9d}~ww xY w|s/ t | }n # t t f$ r
}Y d}~nd}~ww xY w|pd}| | dS )zHandle a numeric character reference by converting it to the
corresponding Unicode character and treating it as textual
data.
:param name: Character number, possibly in hexadecimal.
x XN z windows-1252u �)
startswithintlstripr4 original_encoding bytearraydecodeUnicodeDecodeErrorchr
ValueError
OverflowErrorrA )r r) real_namerC encodinges r handle_charrefz&BeautifulSoupHTMLParser.handle_charref sa ??3 "DKK , ,b11II
__S
!
! "DKK , ,b11IID I s?? "Y 8. I
$i[1188BBDD) DDDD
9~~
.
22 s$ #C
CC C, ,DDc t j | }||}nd|z }| | dS )zHandle a named entity reference by converting it to the
corresponding Unicode character(s) and treating it as textual
data.
:param name: Name of the entity reference.
Nz&%s)r HTML_ENTITY_TO_CHARACTERgetrA )r r) characterrC s r handle_entityrefz(BeautifulSoupHTMLParser.handle_entityref sL '?CCDII DD 4BJN/7 /7 /7 /7b1 1 1 1 1 1 1r! )rl __license____all__ html.parserr syswarnings bs4.elementr r r r r
bs4.dammitr r bs4.builderr
r r r r r r r rm r! r r se I I # " " " " "
9 8 8 8 8 8 8 8
v1 v1 v1 v1 v1j*@ v1 v1 v1rf1 f1 f1 f1 f1O f1 f1 f1 f1 f1r!