o ­=?hðCã@s¸dZddlZddlZddlmZdgZe d¡Ze d¡Ze d¡Z e d¡Z e d ¡Z e d ¡Z e d ¡Z e d ¡Ze d ¡Ze dej¡Ze d ¡Ze d¡ZGdd„dejƒZdS)zA parser for HTML and XHTML.éN)ÚunescapeÚ 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]ú>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#c@sàeZdZdZdZddœdd„Zdd„Zd d „Zd d „Zd Z dd„Z dd„Z dd„Z dd„Z dd„Zd7dd„Zdd„Zdd„Zdd „Zd!d"„Zd#d$„Zd%d&„Zd'd(„Zd)d*„Zd+d,„Zd-d.„Zd/d0„Zd1d2„Zd3d4„Zd5d6„Zd S)8raEFind tags and other markup and call handler functions. Usage: p = HTMLParser() p.feed(data) ... p.close() Start tags are handled by calling self.handle_starttag() or self.handle_startendtag(); end tags by self.handle_endtag(). The data between tags is passed from the parser to the derived class by calling self.handle_data() with the data as argument (the data may be split up in arbitrary chunks). If convert_charrefs is True the character references are converted automatically to the corresponding Unicode character (and self.handle_data() is no longer split in chunks), otherwise they are passed by calling self.handle_entityref() or self.handle_charref() with the string containing respectively the named or numeric reference as the argument. )ZscriptZstyleT)Úconvert_charrefscCs||_| ¡dS)zÆInitialize and reset this instance. If convert_charrefs is True (the default), all character references are automatically converted to the corresponding Unicode characters. N)rÚreset)Úselfr©rú2/opt/alt/python310/lib64/python3.10/html/parser.pyÚ__init__Vs zHTMLParser.__init__cCs(d|_d|_t|_d|_tj |¡dS)z1Reset this instance. Loses all unprocessed data.Úz???N)ÚrawdataÚlasttagÚinteresting_normalÚ interestingÚ cdata_elemÚ _markupbaseÚ ParserBaser©rrrr r_s zHTMLParser.resetcCs|j||_| d¡dS)z‘Feed data to the parser. Call this as often as you want, with as little or as much text as you want (may include '\n'). rN)r Úgoahead©rÚdatarrr Úfeedgs zHTMLParser.feedcCs| d¡dS)zHandle any buffered data.éN)rrrrr ÚclosepszHTMLParser.closeNcCs|jS)z)Return full source of start tag: '<...>'.)Ú_HTMLParser__starttag_textrrrr Úget_starttag_textvszHTMLParser.get_starttag_textcCs$| ¡|_t d|jtj¡|_dS)Nz )ÚlowerrÚreÚcompileÚIr)rÚelemrrr Úset_cdata_modezs zHTMLParser.set_cdata_modecCst|_d|_dS©N)rrrrrrr Úclear_cdata_mode~s zHTMLParser.clear_cdata_modec Cs2|j}d}t|ƒ}||krá|jr;|js;| d|¡}|dkr:| dt||dƒ¡}|dkr8t d¡  ||¡s8n©|}n|j   ||¡}|rI|  ¡}n|jrNn“|}||kro|jrf|jsf|  t |||…ƒ¡n |  |||…¡| ||¡}||kr{nf|j}|d|ƒrt ||¡r| |¡} n>|d|ƒr›| |¡} n3|d|ƒr¦| |¡} n(|d|ƒr±| |¡} n|d |ƒr¼| |¡} n|d |krÌ|  d¡|d } nn| dkr|s×n | d |d ¡} | dkrô| d|d ¡} | dkró|d } n| d 7} |jr |js |  t ||| …ƒ¡n |  ||| …¡| || ¡}nÁ|d |ƒrlt ||¡}|rO| ¡d d…} | | ¡| ¡} |d| d ƒsH| d } | || ¡}q d||d…vrk|  |||d …¡| ||d ¡}nu|d|ƒrÝt ||¡}|rœ| d ¡} | | ¡| ¡} |d| d ƒs•| d } | || ¡}q t ||¡}|rÇ|rÆ| ¡||d…krÆ| ¡} | |kr¾|} | ||d ¡}n|d |krÜ|  d¡| ||d ¡}nn||ks|r||kr|js|jr|js|  t |||…ƒ¡n |  |||…¡| ||¡}||d…|_dS)Nrú<ú&é"z[\s;]zÚ rR)rÚcheck_for_whole_start_tagr Útagfind_tolerantr4r=r;rr Úattrfind_tolerantrÚappendÚstripZgetposÚcountr+r-r1ÚendswithÚhandle_startendtagÚhandle_starttagÚCDATA_CONTENT_ELEMENTSr!)rrAÚendposr Úattrsr4rDÚtagÚmÚattrnameÚrestZ attrvaluer=ÚlinenoÚoffsetrrr r5,sV  &( ó   ÿ   ý  zHTMLParser.parse_starttagcCs²|j}t ||¡}|rU| ¡}|||d…}|dkr|dS|dkr?| d|¡r-|dS| d|¡r5dS||kr;|S|dS|dkrEdS|dvrKdS||krQ|S|dStd ƒ‚) Nrrú/rRr(r)r z6abcdefghijklmnopqrstuvwxyz=/ABCDEFGHIJKLMNOPQRSTUVWXYZzwe should not get here!)r Úlocatestarttagend_tolerantr4r=r2ÚAssertionError)rrAr rarCÚnextrrr rT_s.   z$HTMLParser.check_for_whole_start_tagcCs|j}t ||d¡}|sdS| ¡}t ||¡}|s`|jdur+| |||…¡|St ||d¡}|sH|||d…dkrC|dS|  |¡S|  d¡  ¡}|  d| ¡¡}|  |¡|dS|  d¡  ¡}|jdur|||jkr|| |||…¡|S|  |¡| ¡|S)Nrr)r(rGzr)r Ú endendtagr/r=Ú endtagfindr4rr1rUrJr;rr,Ú handle_endtagr#)rrAr r4rKZ namematchZtagnamer rrr r6s6       zHTMLParser.parse_endtagcCs| ||¡| |¡dSr")r\rl©rr`r_rrr r[©s zHTMLParser.handle_startendtagcCódSr"rrmrrr r\®ózHTMLParser.handle_starttagcCrnr"r)rr`rrr rl²rozHTMLParser.handle_endtagcCrnr"r©rrErrr r<¶rozHTMLParser.handle_charrefcCrnr"rrprrr r?ºrozHTMLParser.handle_entityrefcCrnr"rrrrr r1¾rozHTMLParser.handle_datacCrnr"rrrrr rLÂrozHTMLParser.handle_commentcCrnr"r)rZdeclrrr rIÆrozHTMLParser.handle_declcCrnr"rrrrr rOÊrozHTMLParser.handle_picCrnr"rrrrr Ú unknown_declÍrozHTMLParser.unknown_decl)r)Ú__name__Ú __module__Ú __qualname__Ú__doc__r]r rrrrrr!r#rr9rJr8r5rTr6r[r\rlr<r?r1rLrIrOrqrrrr r>s8  z   3"( )rurrZhtmlrÚ__all__rrr@r>r:r3rNZ commentcloserUrVÚVERBOSErgrjrkrrrrrr Ús,          ÿò