u` d Z ddlmZmZmZmZ ddlT ddlmZ ddl Z ddl
Z
e j d Z e j d Z
e j d Z e j d Z e j d
Z e j d Z e j d Z e j d
Z e j d Z e j d Z e j d Z e j de j Z e j de j Z e j d Z e j d Z G d de Z G d dej Z dS )zLA parser for HTML and XHTML.
Backported for python-future from Python 3.3.
)absolute_importdivisionprint_functionunicode_literals)*) _markupbaseNz[&<]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][-.a-zA-Z0-9:_]*)(?:\s|/(?!>))*z[a-zA-Z][^
/> ]*zJ\s*([a-zA-Z_][-.:a-zA-Z_0-9]*)(\s*=\s*(\'[^\']*\'|"[^"]*"|[^\s"\'=<>`]*))?z]((?<=[\'"\s/])[^\s/>][^\s/=>]*)(\s*=+\s*(\'[^\']*\'|"[^"]*"|(?![\'"])[^>\s]*))?(?:\s|/(?!>))*a
<[a-zA-Z][-.a-zA-Z0-9:_]* # tag name
(?:\s+ # whitespace before attribute name
(?:[a-zA-Z_][-.:a-zA-Z0-9_]* # attribute name
(?:\s*=\s* # value indicator
(?:'[^']*' # LITA-enclosed value
|\"[^\"]*\" # LIT-enclosed value
|[^'\">\s]+ # bare value
)
)?
)
)*
\s* # trailing whitespace
aF
<[a-zA-Z][-.a-zA-Z0-9:_]* # 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 comma
)?(?:\s|/(?!>))*
)*
)?
\s* # trailing whitespace
z#c e Zd ZdZddZd ZdS )HTMLParseErrorz&Exception raised for all parse errors.NNc P |sJ || _ |d | _ |d | _ d S )Nr msglinenooffset)selfr positions n/builddir/build/BUILD/cloudlinux-venv-1.0.10/venv/lib/python3.11/site-packages/future/backports/html/parser.py__init__zHTMLParseError.__init__U s,
s qk qk c j | j }| j
|d| j z z }| j |d| j dz z z }|S )Nz , at line %dz , column %dr r )r results r __str__zHTMLParseError.__str__[ sD ; "nt{ ::F ; "mt{Q ??F
r N)r )__name__
__module__ __qualname____doc__r r r r r r R s= 00" " " " r r c e Zd ZdZdZd dZd Zd Zd Zd Z d Z
d
Z d Z d Z
d
Zd Zd!dZd Zd Zd Zd Zd Zd Zd Zd Zd Zd Zd Zd Z d Z d Z d Z d S )"
HTMLParsera 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). Entity references are
passed by calling self.handle_entityref() with the entity
reference as the argument. Numeric character references are
passed to self.handle_charref() with the string containing the
reference as the argument.
)scriptstyleFc x |r t j dt d || _ | dS )zInitialize and reset this instance.
If strict is set to False (the default) the parser will parse invalid
markup, otherwise it will raise an error. Note that the strict mode
is deprecated.
z The strict mode is deprecated. )
stacklevelN)warningswarnDeprecationWarningstrictreset)r r* s r r zHTMLParser.__init__z sE < M:,
<
<
<
<
r c d| _ d| _ t | _ d| _ t
j | dS )z1Reset this instance. Loses all unprocessed data. z???N)rawdatalasttaginteresting_normal interesting
cdata_elemr
ParserBaser+ r s r r+ zHTMLParser.reset s< - $$T*****r c N | j |z | _ | d dS )zFeed data to the parser.
Call this as often as you want, with as little or as much text
as you want (may include '\n').
r N)r. goaheadr datas r feedzHTMLParser.feed s% |d* Qr c 0 | d dS )zHandle any buffered data.r N)r6 r4 s r closezHTMLParser.close s Qr c F t || N)r getpos)r messages r errorzHTMLParser.error s Wdkkmm444r Nc | j S )z)Return full source of start tag: '<...>'.)_HTMLParser__starttag_textr4 s r get_starttag_textz HTMLParser.get_starttag_text s ##r c | | _ t j d| j z t j | _ d S )Nz )lowerr2 recompileIr1 )r elems r set_cdata_modezHTMLParser.set_cdata_mode s4 **,, :nt&FMMr c , t | _ d | _ d S r= )r0 r1 r2 r4 s r clear_cdata_modezHTMLParser.clear_cdata_mode s - r c | j }d}t | }||k ri| j || }|r| }n | j rn.|}||k r | ||| | || }||k rn|j } |d| rt
|| r| | }n |d| r| | }n |d| r|
| }n |d| r| | }ne |d| r3| j r| | }n<| | }n&|dz |k r| d |dz }nn|dk r|sn| j r| d | d |dz }|dk r%| d|dz }|dk r|dz }n|dz
}| ||| | || }n0 |d
| rt(
|| }|rq| d d } | | | } |d
|dz
s|dz
}| || }d
||d v r3| |dd | |d }n\ |d| r>t0
|| }|rj| d } | | | } |d
|dz
s|dz
}| || }t4
|| }|ra|r^| ||d k r>| j r| d n!||k r|}| ||dz }nJ|dz |k r/| d | ||dz }n nJ d ||k i|r@||k r:| j s3| ||| | || }||d | _ d S )Nr <>D))$/// A%:c1Q3// " E q!,,A "((!44 9u{{}} ; ; ; 9 JJ'L M M M M Avv$% $q!a% 8 8A!eq[[ $$S))) q!a%00AA5555qu !eex %1q55 5 WQqS\ * * *q!$$A qrr{ r c | j }|||dz dk s
J d |||dz dk r| | S |||dz dk r| | S |||dz d k rF| d
|dz }|d k rd S | ||dz | |d z S | | S )
Nr% rR z+unexpected call to parse_html_declaration() rP zV # #%%a(( (
QqsU^u
$
$,,Q// /
QqsU^
!
!
#
#{
2
2LLac**E{{r WQqSY / 0 0 07N++A.. .r r c | j }|||dz dv s
J d | d|dz }|dk rdS |r | ||dz | |dz S )Nr% )rR rO z"unexpected call to parse_comment()r rT r )r. re handle_comment)r rm reportr. poss r rx z HTMLParser.parse_bogus_comment- s ,q1u~ --- 1B---ll3!$$ "992 2 !C 0 1 1 1Qwr c | j }|||dz dk s
J d t ||dz }|sdS | }| ||dz | | }|S )Nr% rQ z unexpected call to parse_pi()rT )r. picloserX rY handle_piri )r rm r. r^ ro s r rb zHTMLParser.parse_pi9 s ,q1u~%%%'F%%%w!,, 2 KKMM wqsAv''' IIKKr c d | _ | | }|dk r|S | j }||| | _ g }t ||dz }|s
J d | }| d x| _ }||k r| j r t || }nt || }|sn| ddd \ } }
} |
sd } nI| d d dcxk r| dd k s"n | d d dcxk r| dd k r
n n
| dd } | r| | } |
| | f | }||k ||| } | d vr| \ }
}d
| j v rM|
| j d
z }
t# | j | j d
z
}n|t# | j z }| j r(| d ||| d d | ||| |S | d
r| || n4| || || j v r| | |S )Nr r z#unexpected call to parse_starttag()r% rt 'rT ")r />
z junk characters in start tag: r )rB check_for_whole_start_tagr. tagfindr^ ri rg rE r/ r* attrfindattrfind_tolerantunescapeappendstripr> countrW rfindr@ rZ endswithhandle_startendtaghandle_starttagCDATA_CONTENT_ELEMENTSrJ )r rm endposr. attrsr^ rp tagmattrnamerest attrvalueri r r s r r_ zHTMLParser.parse_starttagE sE # //22 A::M, &qx 0
gqs++;;;;;u IIKK "[[^^ 1 1 3 33 s&jj{
8 NN7A..%++GQ77
()1a(8(8 %HdI
, 2A2 $8888)BCC.88882A2 #7777233 77777 %adO
5 MM) 4 4 LL(..**I6 7 7 7A! &jj$ ah %%'' k ! ! ![[]]NFFt+++ $"6"<"Ds{{1u s{{%%dA.. ! q5L%%c1-- 2; <NN1a!e,,,JJ :;;;q55 H q5Lrzzr 5 6 6 r{
2q!$$$
01111uu1u 6777r c | j }|||dz dk s
J d t ||dz }|sdS | }t || }|s | j | ||| |S | j r | d||| t ||dz }|s+|||dz dk r|dz S | | S |
}| d | }| | |dz S | d
}| j *|| j k r | ||| |S | |
| |S )
Nr% rO z unexpected call to parse_endtagr rT z
bad end tag: rt zr )r. endendtagrX ri
endtagfindr^ r2 rZ r* r@ tagfind_tolerantrx rg rE re
handle_endtagrL )r rm r. r^ ry namematchtagnamerI s r r` zHTMLParser.parse_endtag s ,q1u~%%%'H%%% !A#.. 2 !,, * 5!1222 {
D
%0@0@BCCC(..w!<U** Q3J 33A666 oo''--//G
LLimmoo66E w ' ' '7N{{1~~##%% ? &t&& 5!1222 4::<<((( r c \ | || | | d S r= )r r r r r s r r z HTMLParser.handle_startendtag s2 S%((( 3 r c d S r= r r s r r zHTMLParser.handle_starttag r c d S r= r )r r s r r zHTMLParser.handle_endtag r r c d S r= r r rq s r rh zHTMLParser.handle_charref r r c d S r= r r s r rk zHTMLParser.handle_entityref r r c d S r= r r7 s r rZ zHTMLParser.handle_data r r c d S r= r r7 s r r{ zHTMLParser.handle_comment r r c d S r= r )r decls r rw zHTMLParser.handle_decl r r c d S r= r r7 s r r zHTMLParser.handle_pi r r c H | j r| d| d S d S )Nzunknown declaration: )r* r@ r7 s r unknown_declzHTMLParser.unknown_decl s5 ; < JJJDD: ; ; ; ; ; < .replaceEntities s^
1
A
#Q43;;!""Aty(( !"" S 1 12 6 6
. . q66M
A@@@@@:: 8OZZ__ # 7N q#a&&)) # #A!u~~ $QrrU|ae 3333 & 7N
ax
s Ar s% : : : : : : : : : : : : ( ( ( ( ( ( RZ''
RZ
%
%
BJ> ? ?
"*@
A
Arz+&&
"*S//rz)$$
"*?
@
@ 2: ;<< 2:, - - BJ=> > BJ
Z
(RZ ) Z
BJsOO RZ=
>
>
Y $t* t* t* t* t*' t* t* t* t* t*r