
    nhtF                        d Z ddlZddlZddlmZ dg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ej                  Z ej        d          Z ej        d          Z G d dej                  ZdS )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
</[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                       e Zd ZdZdZdddZd Zd Zd Zd	Z	d
 Z
d Zd Zd Zd Zdd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d Zd Zd	S ) r   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.
    )scriptstyleT)convert_charrefsc                <    || _         |                                  dS )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	   s     "/usr/lib/python3.11/html/parser.py__init__zHTMLParser.__init__W   s     !1

    c                     d| _         d| _        t          | _        d| _        t
          j                            |            dS )z1Reset this instance.  Loses all unprocessed data. z???N)rawdatalasttaginteresting_normalinteresting
cdata_elem_markupbase
ParserBaser   r   s    r   r   zHTMLParser.reset`   s<    -$$T*****r   c                 N    | j         |z   | _         |                     d           dS )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   goaheadr   datas     r   feedzHTMLParser.feedh   s%     |d*Qr   c                 0    |                      d           dS )zHandle any buffered data.   N)r   r   s    r   closezHTMLParser.closeq   s    Qr   Nc                     | j         S )z)Return full source of start tag: '<...>'.)_HTMLParser__starttag_textr   s    r   get_starttag_textzHTMLParser.get_starttag_textw   s    ##r   c                     |                                 | _        t          j        d| j        z  t          j                  | _        d S )Nz</\s*%s\s*>)lowerr   recompileIr   )r   elems     r   set_cdata_modezHTMLParser.set_cdata_mode{   s4    **,,:nt&FMMr   c                 ,    t           | _        d | _        d S N)r   r   r   r   s    r   clear_cdata_modezHTMLParser.clear_cdata_mode   s    -r   c                    | j         }d}t          |          }||k     rL| j        r}| j        sv|                    d|          }|dk     rY|                    dt          ||dz
                      }|dk    r*t          j        d          	                    ||          sn|}n=| j
        	                    ||          }|r|                                }n| j        rn|}||k     rV| j        r2| j        s+|                     t          |||                              n|                     |||                    |                     ||          }||k    rn|j        } |d|          rt                               ||          r|                     |          }	n |d|          r|                     |          }	n |d|          r|                     |          }	nl |d|          r|                     |          }	nJ |d	|          r|                     |          }	n(|d
z   |k     s|r|                     d           |d
z   }	nn|	dk     r|snt                               ||          rn |d|          r_|dz   |k    r|                     d           nzt.                              ||          rn]|                     ||dz   d                     n; |d|          rU|}dD ]/}
|                    |
|dz             r|t          |
          z  } n0|                     ||dz   |                    n |d|          r!|                     ||dz   d                     n|||dz                                            dk    r!|                     ||dz   d                     ni |d	|          r!|                     ||dz   d                     n< |d|          r!|                     ||dz   d                     nt=          d          |}	|                     ||	          }n- |d|          rt>                              ||          }|rq|                                 dd         }| !                    |           |"                                }	 |d|	d
z
            s|	d
z
  }	|                     ||	          }d||d          v r9|                     |||dz                       |                     ||dz             }nS |d|          r5tF                              ||          }|rj|                     d
          }| $                    |           |"                                }	 |d|	d
z
            s|	d
z
  }	|                     ||	          }tJ                              ||          }|rX|rU|                                 ||d          k    r5|"                                }	|	|k    r|}	|                     ||d
z             }nJ|d
z   |k     r/|                     d           |                     ||d
z             }nnJ d            ||k     L|ry||k     rs| j        sl| j        r2| j        s+|                     t          |||                              n|                     |||                    |                     ||          }||d          | _         d S )Nr   <&"   z[\s;]</<!--<?<!r       )z--!z---   z	<![CDATA[   	   	<!doctypewe should not get here!z&#;zinteresting.search() lied)&r   lenr	   r   findrfindmaxr'   r(   searchr   starthandle_datar   	updatepos
startswithstarttagopenmatchparse_starttagparse_endtagparse_commentparse_piparse_html_declaration
endtagopenhandle_commentendswithunknown_declr&   handle_decl	handle_piAssertionErrorcharrefgrouphandle_charrefend	entityrefhandle_entityref
incomplete)r   rZ   r   injampposrJ   rH   ksuffixnames               r   r   zHTMLParser.goahead   s   ,LL!ee$ T_ LLa((q55 %]]3Aqt==F!Jx0077HH $A(//;; AA A1uu( 3 3$$Xgacl%;%;<<<<$$WQqS\222q!$$AAvvu +Jz#q!! \6%%gq11 ++A..AAZa(( ))!,,AAZ** 
**1--AAZa(( a((AAZa(( 33A66AA!eq[[C[$$S)))AAAq55 #))'155 H#D!,, Hq5A:: ,,T2222'--gq99 ?  !//!>>>>#FA.. H&8 & &F&//!<< & !S[[ 0 %& ++GAaCEN;;;;#K33 
H))'!A#$$-8888 1Q3--//;>>((17777#D!,, H++GAaCDDM::::#D!,, Hwqstt}5555,-FGGGANN1a((D!$$ +6gq11  ;;==2.D''---		A%:c1Q3// "Eq!,,Agabbk))((1Q3888 NN1ac22C## 6!33  ;;q>>D))$///		A%:c1Q3// "Eq!,,A"((!44  5u{{}};;!IIKK66 !A NN1a!e44!eq[[ $$S)))q!a%00AA5555qw !eez  	%1q555$ /T_ /  '!A#,!7!78888  1...q!$$Aqrr{r   c                    | j         }|||dz            dk    s
J d            |||dz            dk    r|                     |          S |||dz            dk    r|                     |          S |||dz                                            d	k    rF|                    d
|dz             }|dk    rdS |                     ||dz   |                    |dz   S |                     |          S )Nr7   r6   z+unexpected call to parse_html_declaration()r9   r4   r:   z<![r;   r<   r   r>   r    )r   rM   parse_marked_sectionr&   rA   rT   parse_bogus_comment)r   r^   r   gtposs       r   rO   z!HTMLParser.parse_html_declaration  s   ,q1u~%%% )C%%%1QqS5>V##%%a(((QqsU^u$$,,Q///QqsU^!!##{22LLac**E{{rWQqSY/0007N++A...r   r    c                     | j         }|||dz            dv s
J d            |                    d|dz             }|dk    rdS |r |                     ||dz   |                    |dz   S )Nr7   )r6   r3   z"unexpected call to parse_comment()r   r>   r    )r   rA   rQ   )r   r^   reportr   poss        r   rg   zHTMLParser.parse_bogus_comment'  s    ,q1u~--- 1B---ll3!$$"992 	2!C 0111Qwr   c                    | j         }|||dz            dk    s
J d            t                              ||dz             }|sdS |                                }|                     ||dz   |                    |                                }|S )Nr7   r5   zunexpected call to parse_pi()r>   )r   picloserD   rE   rU   rZ   )r   r^   r   rJ   r`   s        r   rN   zHTMLParser.parse_pi3  s    ,q1u~%%%'F%%%w!,, 	2KKMMwqsAv'''IIKKr   c                 V   d | _         |                     |          }|dk     r|S | j        }|||         | _         g }t                              ||dz             }|s
J d            |                                }|                    d                                          x| _        }||k     rt                              ||          }|sn|                    ddd          \  }	}
}|
sd }nI|d d         dcxk    r|dd          k    s"n |d d         dcxk    r|dd          k    rn n
|dd         }|rt          |          }|                    |	                                |f           |                                }||k     |||                                         }|d	vr|                     |||                    |S |                    d
          r|                     ||           n4|                     ||           || j        v r|                     |           |S )Nr   r    z#unexpected call to parse_starttag()r7   r:   'r>   ")r   />rq   )r#   check_for_whole_start_tagr   tagfind_tolerantrJ   rZ   rX   r&   r   attrfind_tolerantr   appendstriprF   rR   handle_startendtaghandle_starttagCDATA_CONTENT_ELEMENTSr+   )r   r^   endposr   attrsrJ   rb   tagmattrnamerest	attrvaluerZ   s                r   rK   zHTMLParser.parse_starttag?  si   #//22A::M,&qx0  &&w!44;;;;;uIIKK"[[^^11333s&jj!''33A ()1a(8(8%HdI , 		2A2$8888)BCC.88882A2#777723377777%adO	 0$Y//	LL(..**I6777A &jj ah%%''k!!WQvX.///M<< 	)##C////  e,,,d111##C(((r   c                    | j         }t                              ||          }|r|                                }|||dz            }|dk    r|dz   S |dk    r@|                    d|          r|dz   S |                    d|          rdS ||k    r|S |dz   S |dk    rdS |dv rdS ||k    r|S |dz   S t          d	          )
Nr    r   /rq   r7   r>   r   z6abcdefghijklmnopqrstuvwxyz=/ABCDEFGHIJKLMNOPQRSTUVWXYZr=   )r   locatestarttagend_tolerantrJ   rZ   rH   rV   )r   r^   r   r}   r`   nexts         r   rr   z$HTMLParser.check_for_whole_start_tagk  s   ,&,,Wa88 	A1QqS5>Ds{{1us{{%%dA.. !q5L%%c1-- 2q55Hq5Lrzzr 5 6 6 r1uu1u6777r   c                    | j         }|||dz            dk    s
J d            t                              ||dz             }|sdS |                                }t                              ||          }|s| j        |                     |||                    |S t                              ||dz             }|s+|||dz            dk    r|dz   S | 	                    |          S |
                    d                                          }|                    d|                                          }|                     |           |dz   S |
                    d                                          }| j        *|| j        k    r|                     |||                    |S |                     |           |                                  |S )	Nr7   r3   zunexpected call to parse_endtagr    r>   r:   z</>r   )r   	endendtagrD   rZ   
endtagfindrJ   r   rF   rs   rg   rX   r&   rA   handle_endtagr.   )r   r^   r   rJ   rh   	namematchtagnamer*   s           r   rL   zHTMLParser.parse_endtag  s   ,q1u~%%%'H%%%  !A#.. 	2		  !,, 	*  5!1222(..w!<<I 71QqS5>U**Q3J33A666ooa((..00G
 LLimmoo66Ew'''7N{{1~~##%%?&t&&  5!12224   r   c                 \    |                      ||           |                     |           d S r-   )rx   r   r   r|   r{   s      r   rw   zHTMLParser.handle_startendtag  s2    S%(((3r   c                     d S r-    r   s      r   rx   zHTMLParser.handle_starttag      r   c                     d S r-   r   )r   r|   s     r   r   zHTMLParser.handle_endtag  r   r   c                     d S r-   r   r   rd   s     r   rY   zHTMLParser.handle_charref  r   r   c                     d S r-   r   r   s     r   r\   zHTMLParser.handle_entityref  r   r   c                     d S r-   r   r   s     r   rF   zHTMLParser.handle_data  r   r   c                     d S r-   r   r   s     r   rQ   zHTMLParser.handle_comment  r   r   c                     d S r-   r   )r   decls     r   rT   zHTMLParser.handle_decl  r   r   c                     d S r-   r   r   s     r   rU   zHTMLParser.handle_pi  r   r   c                     d S r-   r   r   s     r   rS   zHTMLParser.unknown_decl  r   r   )r    )__name__
__module____qualname____doc__ry   r   r   r   r!   r#   r$   r+   r.   r   rO   rg   rN   rK   rr   rL   rw   rx   r   rY   r\   rF   rQ   rT   rU   rS   r   r   r   r   r   ?   s        * 1+/     + + +     O$ $ $N N N  G# G# G#X/ / /*	 	 	 		 	 	( ( (X8 8 8D% % %P     
                    r   )r   r'   r   htmlr   __all__r(   r   r]   r[   rW   rI   rP   rm   commentclosers   rt   VERBOSEr   r   r   r   r   r   r   r   <module>r      s~   " " 
			           .  RZ'' RZ%%
BJ>??	
"*@
A
Arz+&&RZ%%

"*S//rz)$$ 2:LMM BJ=> >  (RZ ) Z   BJsOO	 RZ>??
[ [ [ [ [' [ [ [ [ [r   