Current File : //proc/thread-self/root/usr/lib64/python2.7/HTMLParser.pyo
�
zfc@sdZddlZddlZejd�Zejd�Zejd�Zejd�Zejd�Zejd�Z ejd �Z
ejd
�Zejd�Zejd�Z
ejd
ej�Zejd�Zejd�Zdefd��YZdejfd��YZdS(sA parser for HTML and XHTML.i����Ns[&<]s
&[a-zA-Z#]s%&([a-zA-Z][-.a-zA-Z0-9]*)[^a-zA-Z0-9]s)&#(?:[0-9]+|[xX][0-9a-fA-F]+)[^0-9a-fA-F]s <[a-zA-Z]t>s--\s*>s$([a-zA-Z][^
/>]*)(?:\s|/(?!>))*s[a-zA-Z][^
/>]*s]((?<=[\'"\s/])[^\s/>][^\s/=>]*)(\s*=+\s*(\'[^\']*\'|"[^"]*"|(?![\'"])[^>\s]*))?(?:\s|/(?!>))*s
<[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|/(?!>))*
)*
)?
\s* # trailing whitespace
s#</\s*([a-zA-Z][-.a-zA-Z0-9:_]*)\s*>tHTMLParseErrorcBs#eZdZdd�Zd�ZRS(s&Exception raised for all parse errors.cCs'||_|d|_|d|_dS(Nii(tmsgtlinenotoffset(tselfRtposition((s"/usr/lib64/python2.7/HTMLParser.pyt__init__<s
cCsW|j}|jdk r,|d|j}n|jdk rS|d|jd}n|S(Ns, at line %ds, column %di(RRtNoneR(Rtresult((s"/usr/lib64/python2.7/HTMLParser.pyt__str__Bs N(NN(t__name__t
__module__t__doc__RRR
(((s"/usr/lib64/python2.7/HTMLParser.pyR9st
HTMLParsercBs
eZdZdZd�Zd�Zd�Zd�Zd�ZdZ
d�Zd �Zd
�Z
d�Zd�Zd
d�Zd�Zd�Zd�Zd�Zd�Zd�Zd�Zd�Zd�Zd�Zd�Zd�Zd�Zd�ZdZd�Z RS( s�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.
tscripttstylecCs|j�dS(s#Initialize and reset this instance.N(treset(R((s"/usr/lib64/python2.7/HTMLParser.pyRbscCs8d|_d|_t|_d|_tjj|�dS(s1Reset this instance. Loses all unprocessed data.ts???N( trawdatatlasttagtinteresting_normaltinterestingRt
cdata_elemt
markupbaset
ParserBaseR(R((s"/usr/lib64/python2.7/HTMLParser.pyRfs
cCs!|j||_|jd�dS(s�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').
iN(Rtgoahead(Rtdata((s"/usr/lib64/python2.7/HTMLParser.pytfeednscCs|jd�dS(sHandle any buffered data.iN(R(R((s"/usr/lib64/python2.7/HTMLParser.pytclosewscCst||j���dS(N(Rtgetpos(Rtmessage((s"/usr/lib64/python2.7/HTMLParser.pyterror{scCs|jS(s)Return full source of start tag: '<...>'.(t_HTMLParser__starttag_text(R((s"/usr/lib64/python2.7/HTMLParser.pytget_starttag_text�scCs2|j�|_tjd|jtj�|_dS(Ns</\s*%s\s*>(tlowerRtretcompiletIR(Rtelem((s"/usr/lib64/python2.7/HTMLParser.pytset_cdata_mode�scCst|_d|_dS(N(RRRR(R((s"/usr/lib64/python2.7/HTMLParser.pytclear_cdata_mode�s c
Csj|j}d}t|�}x�||kr|jj||�}|rT|j�}n|jraPn|}||kr�|j|||!�n|j||�}||kr�Pn|j}|d|�r7t j
||�r�|j|�}n�|d|�r |j|�}n�|d|�r*|j
|�}nm|d|�rK|j|�}nL|d|�rl|j|�}n+|d|kr�|jd�|d}nP|dkr"|s�Pn|jd|d�}|dkr|jd|d�}|dkr|d}qn
|d7}|j|||!�n|j||�}q|d |�rtj
||�}|r�|j�d
d!} |j| �|j�}|d|d�s�|d}n|j||�}qqd||kr|j|||d
!�|j||d
�}nPq|d
|�rtj
||�}|r�|jd�} |j| �|j�}|d|d�sv|d}n|j||�}qntj
||�}|r�|r�|j�||kr�|jd�nPq|d|kr|jd
�|j||d�}qPqqW|rY||krY|jrY|j|||!�|j||�}n|||_dS(Nit<s</s<!--s<?s<!iRs&#ii����t;t&s#EOF in middle of entity or char ref(RtlenRtsearchtstartRthandle_datat updatepost
startswithtstarttagopentmatchtparse_starttagtparse_endtagt
parse_commenttparse_pitparse_html_declarationtfindtcharreftgroupthandle_charreftendt entityrefthandle_entityreft
incompleteR (
RR>RtitnR4tjR2tktname((s"/usr/lib64/python2.7/HTMLParser.pyR�s�
cCs�|j}|||d!dkr0|jd�n|||d!dkrT|j|�S|||d!dkrx|j|�S|||d!j�d kr�|jd
|d�}|dkr�dS|j||d|!�|dS|j|�SdS(
Nis<!s+unexpected call to parse_html_declaration()is<!--is<![i s <!doctypeRi����i(RR R7tparse_marked_sectionR#R:thandle_decltparse_bogus_comment(RRBRtgtpos((s"/usr/lib64/python2.7/HTMLParser.pyR9�s
icCs|j}|||d!dkr0|jd�n|jd|d�}|dkrVdS|rw|j||d|!�n|dS( Nis<!s</s"unexpected call to parse_comment()Ri����i(s<!s</(RR R:thandle_comment(RRBtreportRtpos((s"/usr/lib64/python2.7/HTMLParser.pyRIs cCs]|j}tj||d�}|s)dS|j�}|j||d|!�|j�}|S(Nii����(RtpicloseR.R/t handle_piR>(RRBRR4RD((s"/usr/lib64/python2.7/HTMLParser.pyR8s cCsnd|_|j|�}|dkr(|S|j}|||!|_g}tj||d�}|j�}|jd�j�|_ }x�||krut
j||�}|s�Pn|jddd�\} }
}|
s�d}nX|d dko�|dkns%|d dko |dknr5|dd!}n|rM|j|�}n|j| j�|f�|j�}q�W|||!j
�}|dkr|j�\}
}d
|jkr�|
|jjd
�}
t|j�|jjd
�}n|t|j�}|j|||!�|S|jd �r;|j||�n/|j||�||jkrj|j|�n|S(Niiiis'i����t"Rs/>s
(Rs/>(RR!tcheck_for_whole_start_tagRttagfindR4R>R<R#RtattrfindtunescapetappendtstripRtcountR-trfindR0tendswiththandle_startendtagthandle_starttagtCDATA_CONTENT_ELEMENTSR((RRBtendposRtattrsR4REttagtmtattrnametrestt attrvalueR>RR((s"/usr/lib64/python2.7/HTMLParser.pyR5sP $$cCs�|j}tj||�}|r�|j�}|||d!}|dkrR|dS|dkr�|jd|�rx|dS|jd|�r�dS|j||d�|jd�n|dkr�dS|d kr�dS||kr�|S|dSntd
��dS(NiRt/s/>ii����smalformed empty start tagRs6abcdefghijklmnopqrstuvwxyz=/ABCDEFGHIJKLMNOPQRSTUVWXYZswe should not get here!(RtlocatestarttagendR4R>R2R1R tAssertionError(RRBRR`RDtnext((s"/usr/lib64/python2.7/HTMLParser.pyRQNs, cCsj|j}tj||d�}|s)dS|j�}tj||�}|s|jdk rt|j|||!�|St j||d�}|s�|||d!dkr�|dS|j
|�Sn|jd�j�}|j
d|j��}|j|�|dS|jd�j�}|jdk rO||jkrO|j|||!�|Sn|j|�|j�|S(Nii����iis</>R(Rt endendtagR.R>t
endtagfindR4RRR0RRRIR<R#R:t
handle_endtagR)(RRBRR4RJt namematchttagnameR'((s"/usr/lib64/python2.7/HTMLParser.pyR6ns6
cCs!|j||�|j|�dS(N(R[Rj(RR_R^((s"/usr/lib64/python2.7/HTMLParser.pyRZ�scCsdS(N((RR_R^((s"/usr/lib64/python2.7/HTMLParser.pyR[�scCsdS(N((RR_((s"/usr/lib64/python2.7/HTMLParser.pyRj�scCsdS(N((RRF((s"/usr/lib64/python2.7/HTMLParser.pyR=�scCsdS(N((RRF((s"/usr/lib64/python2.7/HTMLParser.pyR@�scCsdS(N((RR((s"/usr/lib64/python2.7/HTMLParser.pyR0�scCsdS(N((RR((s"/usr/lib64/python2.7/HTMLParser.pyRK�scCsdS(N((Rtdecl((s"/usr/lib64/python2.7/HTMLParser.pyRH�scCsdS(N((RR((s"/usr/lib64/python2.7/HTMLParser.pyRO�scCsdS(N((RR((s"/usr/lib64/python2.7/HTMLParser.pytunknown_decl�scs2d|kr|S�fd�}tjd||�S(NR,cs|j�d}yZ|ddkri|d}|dd
krSt|dd�}nt|�}t|�SWntk
r�d|dSXtjdkr�d dl}id
d6}x-|jj �D]\}}t|�||<q�W|t_ny�j|SWnt
k
rd|dSXdS(Nit#itxtXis&#R+i����u'taposR,(RpRq(tgroupstinttunichrt
ValueErrorRt
entitydefsRthtmlentitydefstname2codepointt iteritemstKeyError(tstcRxRwREtv(R(s"/usr/lib64/python2.7/HTMLParser.pytreplaceEntities�s(
s#&(#?[xX]?(?:[0-9a-fA-F]+|\w{1,8}));(R$tsub(RR|R((Rs"/usr/lib64/python2.7/HTMLParser.pyRT�s(RRN(!RRR
R\RRRRR RR!R"R(R)RR9RIR8R5RQR6RZR[RjR=R@R0RKRHRORnRwRT(((s"/usr/lib64/python2.7/HTMLParser.pyRKs< ^ 4 ( (R
RR$R%RRAR?R;R3RNtcommentcloseRRttagfind_tolerantRStVERBOSEReRhRit ExceptionRRR(((s"/usr/lib64/python2.7/HTMLParser.pyt<module>s&