File "parser.cpython-39.opt-1.pyc"
Full Path: /home/attunedd/public_html/byp/izo/con7ext_sym404/rintoar.txt/opt/alt/python39/lib64/python3.9/html/__pycache__/parser.cpython-39.opt-1.pyc
File size: 10.23 KB
MIME-type: text/x-bytecode.python
Charset: 8 bit
a
XC?hC @ s d Z ddlZddlZddlmZ dgZedZedZedZ edZ
ed Z ed
Z ed Z
ed Zed
ZedejZed
ZedZG dd dejZdS )z A parser for HTML and XHTML. N)unescape
HTMLParserz[&<]z
&[a-zA-Z#]z%&([a-zA-Z][-.a-zA-Z0-9]*)[^a-zA-Z0-9]z)&#(?:[0-9]+|[xX][0-9a-fA-F]+)[^0-9a-fA-F]z <[a-zA-Z]>z--\s*>z+([a-zA-Z][^\t\n\r\f />\x00]*)(?:\s|/(?!>))*z]((?<=[\'"\s/])[^\s/>][^\s/=>]*)(\s*=+\s*(\'[^\']*\'|"[^"]*"|(?![\'"])[^>\s]*))?(?:\s|/(?!>))*aF
<[a-zA-Z][^\t\n\r\f />\x00]* # tag name
(?:[\s/]* # optional whitespace before attribute name
(?:(?<=['"\s/])[^\s/>][^\s/=>]* # attribute name
(?:\s*=+\s* # value indicator
(?:'[^']*' # LITA-enclosed value
|"[^"]*" # LIT-enclosed value
|(?!['"])[^>\s]* # bare value
)
\s* # possibly followed by a space
)?(?:\s|/(?!>))*
)*
)?
\s* # trailing whitespace
z#</\s*([a-zA-Z][-.a-zA-Z0-9:_]*)\s*>c @ s e Zd ZdZdZddddZdd Zd d
Zd d Zd
Z dd Z
dd Z dd Z dd Z
dd Zd7ddZdd Zd d Zd d Zd!d" Zd#d$ Zd%d& Zd'd( Zd)d* Zd+d, Zd-d. Zd/d0 Zd1d2 Zd3d4 Z d5d6 Z d
S )8r aE Find tags and other markup and call handler functions.
Usage:
p = HTMLParser()
p.feed(data)
...
p.close()
Start tags are handled by calling self.handle_starttag() or
self.handle_startendtag(); end tags by self.handle_endtag(). The
data between tags is passed from the parser to the derived class
by calling self.handle_data() with the data as argument (the data
may be split up in arbitrary chunks). If convert_charrefs is
True the character references are converted automatically to the
corresponding Unicode character (and self.handle_data() is no
longer split in chunks), otherwise they are passed by calling
self.handle_entityref() or self.handle_charref() with the string
containing respectively the named or numeric reference as the
argument.
)ZscriptZstyleT)convert_charrefsc C s || _ | dS )zInitialize and reset this instance.
If convert_charrefs is True (the default), all character references
are automatically converted to the corresponding Unicode characters.
N)r reset)selfr r 0/opt/alt/python39/lib64/python3.9/html/parser.py__init__V s zHTMLParser.__init__c C s( d| _ d| _t| _d| _tj| dS )z1Reset this instance. Loses all unprocessed data. z???N)rawdatalasttaginteresting_normal interesting
cdata_elem _markupbase
ParserBaser r r r r r _ s
zHTMLParser.resetc C s | j | | _ | d dS )zFeed data to the parser.
Call this as often as you want, with as little or as much text
as you want (may include '\n').
r N)r goaheadr datar r r feedg s zHTMLParser.feedc C s | d dS )zHandle any buffered data. N)r r r r r closep s zHTMLParser.closeNc C s | j S )z)Return full source of start tag: '<...>'.)_HTMLParser__starttag_textr r r r get_starttag_textv s z HTMLParser.get_starttag_textc C s$ | | _td| j tj| _d S )Nz </\s*%s\s*>)lowerr recompileIr )r elemr r r set_cdata_modez s
zHTMLParser.set_cdata_modec C s t | _d | _d S N)r r r r r r r clear_cdata_mode~ s zHTMLParser.clear_cdata_modec C sJ | j }d}t|}||k r| jrv| jsv|d|}|dk r|dt||d }|dkrptd ||spq|}n*| j
||}|r| }n| jrq|}||k r| jr| js| t
||| n| ||| | ||}||krq|j}|d|rJt||r"| |} n|d|r:| |} nn|d|rR| |} nV|d|rj| |} n>|d |r| |} n&|d
|k r| d |d
} nq| dk r<|sq|d |d
} | dk r|d|d
} | dk r|d
} n| d
7 } | jr*| js*| t
||| n| ||| | || }q|d |rt||}|r| d
d }
| |
| } |d| d
s| d
} | || }qn<d||d v r| |||d
| ||d
}qq|d|rt||}|rN|d
}
| |
| } |d| d
s@| d
} | || }qt ||}|r|r| ||d kr| } | |kr|} | ||d
}qn.|d
|k r| d | ||d
}nqqq|r8||k r8| js8| jr| js| t
||| n| ||| | ||}||d | _ d S )Nr <&" z[\s;]z</<!--z<?z<!r r z&# ;) r lenr r findrfindmaxr r searchr start handle_datar Z updatepos
startswith starttagopenmatchparse_starttag parse_endtag
parse_commentparse_piparse_html_declarationcharrefgrouphandle_charrefend entityrefhandle_entityref
incomplete) r r= r injZampposr4 r2 knamer r r r s
zHTMLParser.goaheadc C s | j }|||d dkr$| |S |||d dkrB| |S |||d dkr|d|d }|dkrvdS | ||d | |d
S | |S d S ) N r' z<![ z <!doctyper r) r( r )r r7 Zparse_marked_sectionr r, handle_declparse_bogus_comment)r rA r gtposr r r r9 s
z!HTMLParser.parse_html_declarationr c C sD | j }|d|d }|dkr"dS |r<| ||d | |d S )Nr r( r) r )r r, handle_comment)r rA Zreportr posr r r rJ s z HTMLParser.parse_bogus_commentc C sH | j }t||d }|s dS | }| ||d | | }|S )Nr( r) )r picloser/ r0 handle_pir= )r rA r r4 rC r r r r8 s zHTMLParser.parse_pic C s d | _ | |}|dk r |S | j}||| | _ g }t||d }| }|d | _}||k r t ||}|s~q |ddd\} }
} |
sd } nZ| d d d kr| dd ksn | d d d kr| dd krn n | dd } | rt
| } | | | f | }q`||| } | dvr|
\}
}d | j v rz|
| j d }
t| j | j d }n|t| j }| ||| |S | d
r| || n"| || || jv r| | |S ) Nr r r( rG 'r) ")r />
rR )r check_for_whole_start_tagr tagfind_tolerantr4 r= r; r r
attrfind_tolerantr appendstripZgetposcountr+ r- r1 endswithhandle_startendtaghandle_starttagCDATA_CONTENT_ELEMENTSr! )r rA endposr attrsr4 rD tagmattrnamerestZ attrvaluer= linenooffsetr r r r5 , sX
&
zHTMLParser.parse_starttagc C s | j }t||}|r| }|||d }|dkr>|d S |dkr~|d|rZ|d S |d|rjdS ||krv|S |d S |dkrdS |dv rdS ||kr|S |d S td d S )
Nr r /rR r( r) r z6abcdefghijklmnopqrstuvwxyz=/ABCDEFGHIJKLMNOPQRSTUVWXYZzwe should not get here!)r locatestarttagend_tolerantr4 r= r2 AssertionError)r rA r ra rC nextr r r rT _ s. z$HTMLParser.check_for_whole_start_tagc C s | j }t||d }|s dS | }t||}|s| jd urV| ||| |S t||d }|s|||d dkr|d S | |S |
d }| d| }|
| |d S |
d }| jd ur|| jkr| ||| |S |
| | |S )Nr r) r( rG z</>r )r endendtagr/ r=
endtagfindr4 r r1 rU rJ r; r r,
handle_endtagr# )r rA r r4 rK Z namematchZtagnamer r r r r6 s6
zHTMLParser.parse_endtagc C s | || | | d S r" )r\ rl r r` r_ r r r r[ s z HTMLParser.handle_startendtagc C s d S r" r rm r r r r\ s zHTMLParser.handle_starttagc C s d S r" r )r r` r r r rl s zHTMLParser.handle_endtagc C s d S r" r r rE r r r r< s zHTMLParser.handle_charrefc C s d S r" r rn r r r r? s zHTMLParser.handle_entityrefc C s d S r" r r r r r r1 s zHTMLParser.handle_datac C s d S r" r r r r r rL s zHTMLParser.handle_commentc C s d S r" r )r Zdeclr r r rI s zHTMLParser.handle_declc C s d S r" r r r r r rO s zHTMLParser.handle_pic C s d S r" r r r r r unknown_decl s zHTMLParser.unknown_decl)r ) __name__
__module__ __qualname____doc__r] r
r r r r r r! r# r r9 rJ r8 r5 rT r6 r[ r\ rl r< r? r1 rL rI rO ro r r r r r > s6 z
3"()rs r r Zhtmlr __all__r r r@ r> r: r3 rN Z commentcloserU rV VERBOSErg rj rk r r r r r r <module> s*