t_ @ s d d l m Z m Z m Z d d l m Z m Z d d l Z d d l m Z d d l m
Z
d d l m Z d d l m
Z
d d l m Z d d
l m Z m Z m Z m Z m Z m Z m Z m Z m Z m Z m Z m Z m Z m Z m Z m Z d d d
d Z! d d d d d Z" d d Z# Gd d d e$ Z% e j& d d Z' d d Z( d d d d d Z) Gd d d e* Z+ d S) )absolute_importdivisionunicode_literals)with_metaclassviewkeysN ) _inputstream)
_tokenizer) treebuilders)Marker)_utils)spaceCharactersasciiUpper2LowerspecialElementsheadingElements
cdataElementsrcdataElements
tokenTypes
tagTokenTypes
namespaces htmlIntegrationPointElements"mathmlTextIntegrationPointElementsadjustForeignAttributesadjustMathMLAttributesadjustSVGAttributesE_ReparseExceptionetreeTc K s1 t j | } t | d | } | j | | S)a Parse an HTML document as a string or file-like object into a tree
:arg doc: the document to parse as a string or file-like object
:arg treebuilder: the treebuilder to use when parsing
:arg namespaceHTMLElements: whether or not to namespace HTML elements
:returns: parsed tree
Example:
>>> from html5lib.html5parser import parse
>>> parse('This is a doc
')
namespaceHTMLElements)r
getTreeBuilder
HTMLParserparse)doc treebuilderr kwargstbp r' /builddir/build/BUILDROOT/alt-python35-pip-20.2.4-1.el7.x86_64/opt/alt/python35/lib/python3.5/site-packages/pip/_vendor/html5lib/html5parser.pyr! s r! divc K s7 t j | } t | d | } | j | d | | S)a# Parse an HTML fragment as a string or file-like object into a tree
:arg doc: the fragment to parse as a string or file-like object
:arg container: the container context to parse the fragment in
:arg treebuilder: the treebuilder to use when parsing
:arg namespaceHTMLElements: whether or not to namespace HTML elements
:returns: parsed tree
Example:
>>> from html5lib.html5libparser import parseFragment
>>> parseFragment('this is a fragment')
r container)r
r r
parseFragment)r" r* r# r r$ r% r& r' r' r( r+ 1 s r+ c s G f d d d t } | S)Nc s" e Z d Z f d d Z d S)z-method_decorator_metaclass..Decoratedc s[ xB | j D]4 \ } } t | t j r7 | } | | | .Decorated.__new__N)__name__
__module__ __qualname__r1 r' )r8 r' r( DecoratedK s r< )r0 )r8 r< r' )r8 r( method_decorator_metaclassJ s r= c @ s e Z d Z d Z d d d d d d Z d d d d d Z d
d Z e d d
Z d d Z d d Z
d d Z d d Z d d Z
d d d d Z d d Z d d Z d d Z d! d" Z d# d$ Z d% d& Z d S)'r z]HTML parser
Generates a tree structure from a stream of (possibly malformed) HTML.
NFTc sh | _ | d k r$ t j d } | | _ g _ f d d t | j D _ d S)a
:arg tree: a treebuilder class controlling the type of tree that will be
returned. Built in treebuilders can be accessed through
html5lib.treebuilders.getTreeBuilder(treeType)
:arg strict: raise an exception when a parse error is encountered
:arg namespaceHTMLElements: whether or not to namespace HTML elements
:arg debug: whether or not to enable debug mode which logs things
Example:
>>> from html5lib.html5parser import HTMLParser
>>> parser = HTMLParser() # generates parser with etree builder
>>> parser = HTMLParser('lxml', strict=True) # generates parser with lxml builder which is strict
Nr c s+ i | ]! \ } } | j | q Sr' )tree).0namecls)selfr' r(
y s z'HTMLParser.__init__..)strictr
r r> errors getPhasesr, phases)rB r> rD r debugr' )rB r( __init__] s zHTMLParser.__init__r) c K s{ | | _ | | _ | | _ t j | d | | | _ | j y | j Wn& t k
rv | j | j Yn Xd S)Nparser)
innerHTMLModer* scriptingr
HTMLTokenizer tokenizerresetmainLoopr )rB stream innerHTMLr* rL r$ r' r' r( _parse| s
zHTMLParser._parsec C s | j j d | _ g | _ g | _ d | _ | j r | j j | _ | j t
k rp | j j | j _
nH | j t k r | j j | j _
n$ | j d k r | j j | j _
n | j d | _ | j j | j n d | _ | j d | _ d | _ d | _ d | _ d S)NFz no quirks plaintext
beforeHtmlinitialT)r> rO
firstStartTagrE log
compatModerK r* lowerrR r rN rcdataStatestater rawtextStateplaintextStaterG phaseinsertHtmlElementresetInsertionMode lastPhasebeforeRCDataPhase
framesetOK)rB r' r' r( rO s*
zHTMLParser.resetc C s' t | d s d S| j j j d j S)zName of the character encoding that was used to decode the input stream, or
:obj:`None` if that is not determined yet
rN Nr )hasattrrN rQ charEncodingr@ )rB r' r' r( documentEncoding s zHTMLParser.documentEncodingc C se | j d k rK | j t d k rK d | j k oJ | j d j t d k S| j | j f t k Sd S)Nzannotation-xmlmathmlencoding text/htmlapplication/xhtml+xml)rj rk )r@ namespacer
attributes translater r )rB elementr' r' r( isHTMLIntegrationPoint s
z!HTMLParser.isHTMLIntegrationPointc C s | j | j f t k S)N)rl r@ r )rB ro r' r' r( isMathMLTextIntegrationPoint s z'HTMLParser.isMathMLTextIntegrationPointc C s5 t d } t d } t d } t d } t d } t d } t d } x| j D]} d } | }
x:|
d k r|
} | j j r | j j d n d } | r | j n d } | r | j n d }
|
d } | | k r| j |
d
|
j d i d }
qe t | j j d k s| | j j k s| j
| ru| | k rc| d
t d d g k s| | | f k s| t d k r|
d k r| | k r| d
d k s| j
| r| | | | f k r| j } n
| j d } | | k r| j |
}
qe | | k r&| j |
}
qe | | k rD| j |
}
qe | | k rb| j |
}
qe | | k r| j |
}
qe | | k re | j |
}
qe W| | k rP | d rP | d rP | j d d
| d
i qP Wd } g } xD | r0| j | j | j j } | r| j | k st qWd S)N
CharactersSpaceCharactersStartTagEndTagCommentDoctype
ParseErrorr r0 datadatavarsr r@ mglyph
malignmarkrh zannotation-xmlsvginForeignContent selfClosingselfClosingAcknowledgedz&non-void-element-with-trailing-solidusT)r rN r> openElementsrl r@
parseErrorgetlendefaultNamespacerq frozensetr rp r_ rG processCharactersprocessSpaceCharactersprocessStartTag
processEndTagprocessCommentprocessDoctypeappend
processEOFAssertionError)rB CharactersTokenSpaceCharactersToken
StartTagToken EndTagToken CommentToken DoctypeTokenParseErrorTokentoken
prev_token new_token currentNodecurrentNodeNamespacecurrentNodeNamer0 r_ reprocessrG r' r' r( rP sp
"
zHTMLParser.mainLoopc O s& | j | d d | | | j j S)a Parse a HTML document into a well-formed tree
:arg stream: a file-like object or string containing the HTML to be parsed
The optional encoding parameter must be a string that indicates
the encoding. If specified, that encoding will be used,
regardless of any BOM or later declaration (such as in a meta
element).
:arg scripting: treat noscript elements as if JavaScript was turned on
:returns: parsed tree
Example:
>>> from html5lib.html5parser import HTMLParser
>>> parser = HTMLParser()
>>> parser.parse('This is a doc
')
FN)rS r> getDocument)rB rQ argsr$ r' r' r( r! s zHTMLParser.parsec O s# | j | d | | | j j S)aZ Parse a HTML fragment into a well-formed tree fragment
:arg container: name of the element we're setting the innerHTML
property if set to None, default to 'div'
:arg stream: a file-like object or string containing the HTML to be parsed
The optional encoding parameter must be a string that indicates
the encoding. If specified, that encoding will be used,
regardless of any BOM or later declaration (such as in a meta
element)
:arg scripting: treat noscript elements as if JavaScript was turned on
:returns: parsed tree
Example:
>>> from html5lib.html5libparser import HTMLParser
>>> parser = HTMLParser()
>>> parser.parseFragment('this is a fragment')
T)rS r> getFragment)rB rQ r r$ r' r' r( r+ s zHTMLParser.parseFragmentzXXX-undefined-errorc C sX | d k r i } | j j | j j j | | f | j rT t t | | d S)N)rE r rN rQ positionrD rx r )rB errorcoderz r' r' r( r ; s
% zHTMLParser.parseErrorc C s t | t d S)N)adjust_attributesr )rB r r' r' r( r C s z!HTMLParser.adjustMathMLAttributesc C s t | t d S)N)r r )rB r r' r' r( r F s z HTMLParser.adjustSVGAttributesc C s t | t d S)N)r adjustForeignAttributesMap)rB r r' r' r( r I s z"HTMLParser.adjustForeignAttributesc C s | j j d S)N)rJ r_ )rB r r' r' r( reparseTokenNormalL s z HTMLParser.reparseTokenNormalc C sC d } d d d d d d d d d d
d d
d d
d
d d d d d d d d d d d d d i } x | j j d d d D] } | j } d } | | j j d k r | j s t d } | j } | d k r | j s t | r | j | j j k r qz | | k r | j | | } Pqz | rz | j d } Pqz W| | _ d S) NFselectinSelecttdinCellthtrinRowtbody inTableBodytheadtfootcaption inCaptioncolgroup
inColumnGrouptableinTableheadinBodybodyframeset
inFramesethtml
beforeHeadr r Tr )r r r r ) r> r r@ rR r rl r rG r_ )rB lastnewModesnodenodeName new_phaser' r' r( ra P sB
z HTMLParser.resetInsertionModec C su | d k s t | j j | | d k rC | j j | j _ n | j j | j _ | j | _ | j d | _ d S)NRAWTEXTRCDATAtext)r r )
r r>
insertElementrN r] r\ r[ r_
originalPhaserG )rB r contentTyper' r' r( parseRCDataRawtext| s z HTMLParser.parseRCDataRawtext)r9 r: r; __doc__rI rS rO propertyrg rp rq rP r! r+ r r r r r ra r r' r' r' r( r V s" "
C ,r c . s d d } d d } Gd d d t | | | Gd d d } Gd d
d
} Gd d d } Gd
d d } Gd d d } Gd d d } G f d d d Gd d d } G f d d d }
G f d d d G f d d d } G f d d d } G f d d d }
G f d! d" d" } G f d# d$ d$ } G f d% d& d& } Gd' d( d( } G f d) d* d* } Gd+ d, d, } G f d- d. d. } G f d/ d0 d0 } Gd1 d2 d2 } Gd3 d4 d4 } d5 | d6 | d7 | d8 | d9 | d: | d; d< | d= |
d> d? | d@ | dA |
dB | dC | dD | dE | dF | dG | dH | dI | dJ | dK | i S)LNc s2 d d t j D f d d } | S)z4Logger that records which phase processes each tokenc S s i | ] \ } } | | q Sr' r' )r? keyvaluer' r' r( rC s z*getPhases..log..c s j j d r t | d k r | d } d | d i } | d t k r` | d | d <| j j j | j j j j | j j j
j | j
j j | f | | | S | | | Sd S)Nprocessr r0 r@ ) r9
startswithr r rJ rX r rN r\ r_ __class__)rB r r$ r info)r8
type_namesr' r( wrapped s $
z'getPhases..log..wrapped)r r, )r8 r r' )r8 r r( rX s zgetPhases..logc S s | r t | St Sd S)N)r= r0 )
use_metaclassmetaclass_funcr' r' r( getMetaclass s
z getPhases..getMetaclassc @ s e Z d Z d Z d Z d d Z d d Z d
d Z d d
Z d d Z d d Z
d d Z d d Z d d Z
d S)zgetPhases..PhasezNBase class for helper object that implements each phase of processing
rJ r> __startTagCache
__endTagCachec S s( | | _ | | _ i | _ i | _ d S)N)rJ r> _Phase__startTagCache_Phase__endTagCache)rB rJ r> r' r' r( rI s z!getPhases..Phase.__init__c S s
t d S)N)NotImplementedError)rB r' r' r( r s z#getPhases..Phase.processEOFc S s! | j j | | j j d d S)Nr r )r>
insertCommentr )rB r r' r' r( r s z'getPhases..Phase.processCommentc S s | j j d d S)Nzunexpected-doctype)rJ r )rB r r' r' r( r s z'getPhases..Phase.processDoctypec S s | j j | d d S)Nry )r>
insertText)rB r r' r' r( r s z*getPhases..Phase.processCharactersc S s | j j | d d S)Nry )r> r )rB r r' r' r( r s z/getPhases..Phase.processSpaceCharactersc S s | d } | | j k r) | j | } n` | j | } | j | .Phase.processStartTagc S s | j j r- | d d k r- | j j d xN | d j D]< \ } } | | j j d j k r> | | j j d j | Wd | j _ d S)Nr@ r z
non-html-rootry r F)rJ rW r r, r> r rm )rB r attrr r' r' r( startTagHtml s z%getPhases..Phase.startTagHtmlc S s | d } | | j k r) | j | } n` | j | } | j | .Phase.processEndTagN)rJ r> r r )r9 r: r; r __slots__rI r r r r r r r r r' r' r' r( Phase s
r c @ sy e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z d S)z getPhases..InitialPhasec S s d S)Nr' )rB r r' r' r( r s z6getPhases..InitialPhase.processSpaceCharactersc S s | j j | | j j d S)N)r> r document)rB r r' r' r( r s z.getPhases..InitialPhase.processCommentc 8 S si | d } | d } | d } | d } | d k sX | d k sX | d k rh | d k rh | j j d | d k rz d } | j j | | d k r | j t } | s
| d d k s
| j dJ s
| dK k s
| j dL r | d k s
| r| j dD k rdE | j _ n6 | j dM sC| j dN rO| d k rOdH | j _ | j j dI | j _
d S)ONr@ publicIdsystemIdcorrectr zabout:legacy-compatzunknown-doctype *+//silmaril//dtd html pro v0r11 19970101//4-//advasoft ltd//dtd html 3.0 aswedit + extensions//*-//as//dtd html 3.0 aswedit + extensions// -//ietf//dtd html 2.0 level 1// -//ietf//dtd html 2.0 level 2//&-//ietf//dtd html 2.0 strict level 1//&-//ietf//dtd html 2.0 strict level 2// -//ietf//dtd html 2.0 strict//-//ietf//dtd html 2.0//-//ietf//dtd html 2.1e//-//ietf//dtd html 3.0// -//ietf//dtd html 3.2 final//-//ietf//dtd html 3.2//-//ietf//dtd html 3//-//ietf//dtd html level 0//-//ietf//dtd html level 1//-//ietf//dtd html level 2//-//ietf//dtd html level 3//"-//ietf//dtd html strict level 0//"-//ietf//dtd html strict level 1//"-//ietf//dtd html strict level 2//"-//ietf//dtd html strict level 3//-//ietf//dtd html strict//-//ietf//dtd html//(-//metrius//dtd metrius presentational//5-//microsoft//dtd internet explorer 2.0 html strict//.-//microsoft//dtd internet explorer 2.0 html//0-//microsoft//dtd internet explorer 2.0 tables//5-//microsoft//dtd internet explorer 3.0 html strict//.-//microsoft//dtd internet explorer 3.0 html//0-//microsoft//dtd internet explorer 3.0 tables//#-//netscape comm. corp.//dtd html//*-//netscape comm. corp.//dtd strict html//*-//o'reilly and associates//dtd html 2.0//3-//o'reilly and associates//dtd html extended 1.0//;-//o'reilly and associates//dtd html extended relaxed 1.0//N-//softquad software//dtd hotmetal pro 6.0::19990601::extensions to html 4.0//E-//softquad//dtd hotmetal pro 4.0::19971010::extensions to html 4.0//$-//spyglass//dtd html 2.0 extended//+-//sq//dtd html 2.0 hotmetal + extensions//--//sun microsystems corp.//dtd hotjava html//4-//sun microsystems corp.//dtd hotjava strict html// -//w3c//dtd html 3 1995-03-24// -//w3c//dtd html 3.2 draft// -//w3c//dtd html 3.2 final//-//w3c//dtd html 3.2// -//w3c//dtd html 3.2s draft// -//w3c//dtd html 4.0 frameset//#-//w3c//dtd html 4.0 transitional//(-//w3c//dtd html experimental 19960712//&-//w3c//dtd html experimental 970421//-//w3c//dtd w3 html//-//w3o//dtd w3 html 3.0//#-//webtechs//dtd mozilla html 2.0// -//webtechs//dtd mozilla html//$-//w3o//dtd w3 html strict 3.0//en//"-/w3c/dtd html 4.0 transitional/en -//w3c//dtd html 4.01 frameset//$-//w3c//dtd html 4.01 transitional//z:http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtdquirks -//w3c//dtd xhtml 1.0 frameset//$-//w3c//dtd xhtml 1.0 transitional//zlimited quirksrU )7r r r r r r r r r r r r r r r r r r r r r r r r r r r r r
r r r
r r r r r r r r r r r r r r r r r r r r! r" r# r$ )r% r& r )r' r( )r* r+ )r' r( ) rJ r r>
insertDoctypern r r rZ rY rG r_ )rB r r@ r r r r' r' r( r s
z.getPhases..InitialPhase.processDoctypec S s& d | j _ | j j d | j _ d S)Nr) rU )rJ rY rG r_ )rB r' r' r( anythingElseZ s z,getPhases..InitialPhase.anythingElsec S s | j j d | j | S)Nz expected-doctype-but-got-chars)rJ r r- )rB r r' r' r( r ^ s
z1getPhases..InitialPhase.processCharactersc S s+ | j j d d | d i | j | S)Nz"expected-doctype-but-got-start-tagr@ )rJ r r- )rB r r' r' r( r c s
z/getPhases..InitialPhase.processStartTagc S s+ | j j d d | d i | j | S)Nz expected-doctype-but-got-end-tagr@ )rJ r r- )rB r r' r' r( r i s
z-getPhases..InitialPhase.processEndTagc S s | j j d | j d S)Nz expected-doctype-but-got-eofT)rJ r r- )rB r' r' r( r o s
z*getPhases..InitialPhase.processEOFN)
r9 r: r; tupler r r r r- r r r r r' r' r' r( InitialPhase s _ r/ c @ sm e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d S)z"getPhases..BeforeHtmlPhasec S s3 | j j t d d | j j d | j _ d S)Nr rt r )r>
insertRootimpliedTagTokenrJ rG r_ )rB r' r' r( r` x s z4getPhases..BeforeHtmlPhase.insertHtmlElementc S s | j d S)NT)r` )rB r' r' r( r } s
z-getPhases..BeforeHtmlPhase.processEOFc S s | j j | | j j d S)N)r> r r )rB r r' r' r( r s z1getPhases..BeforeHtmlPhase.processCommentc S s d S)Nr' )rB r r' r' r( r s z9getPhases..BeforeHtmlPhase.processSpaceCharactersc S s | j | S)N)r` )rB r r' r' r( r s
z4getPhases..BeforeHtmlPhase.processCharactersc S s* | d d k r d | j _ | j | S)Nr@ r T)rJ rW r` )rB r r' r' r( r s
z2getPhases..BeforeHtmlPhase.processStartTagc S sB | d d k r0 | j j d d | d i n | j | Sd S)Nr@ r r r brz unexpected-end-tag-before-html)r r r r2 )rJ r r` )rB r r' r' r( r s
z0getPhases..BeforeHtmlPhase.processEndTagN) r9 r: r; r. r r` r r r r r r r' r' r' r( BeforeHtmlPhaset s r3 c @ s e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z e
j d e f d e f g Z e
e _ e
j d e f g Z e e _ d S)z"getPhases..BeforeHeadPhasec S s | j t d d d S)Nr rt T) startTagHeadr1 )rB r' r' r( r s z-getPhases..BeforeHeadPhase.processEOFc S s d S)Nr' )rB r r' r' r( r s z9getPhases..BeforeHeadPhase.processSpaceCharactersc S s | j t d d | S)Nr rt )r4 r1 )rB r r' r' r( r s z4getPhases..BeforeHeadPhase.processCharactersc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r s z/getPhases..BeforeHeadPhase.startTagHtmlc S s@ | j j | | j j d | j _ | j j d | j _ d S)Nr inHeadr )r> r r headPointerrJ rG r_ )rB r r' r' r( r4 s z/getPhases..BeforeHeadPhase.startTagHeadc S s | j t d d | S)Nr rt )r4 r1 )rB r r' r' r(
startTagOther s z0getPhases..BeforeHeadPhase.startTagOtherc S s | j t d d | S)Nr rt )r4 r1 )rB r r' r' r( endTagImplyHead s z2getPhases..BeforeHeadPhase.endTagImplyHeadc S s! | j j d d | d i d S)Nzend-tag-after-implied-rootr@ )rJ r )rB r r' r' r( endTagOther s z.getPhases..BeforeHeadPhase.endTagOtherr r r r2 N)r r r r2 )r9 r: r; r. r r r r r r4 r7 r8 r9 r MethodDispatcherr defaultr r' r' r' r( BeforeHeadPhase s r< c @ sW e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z d d Z
d d Z d d Z d d Z d d Z d d Z d d Z e j d e f d e f d/ e f d# e
f d$ e f d0 e f d* e
f d+ e f g Z e e _ e j d+ e f d1 e f g Z e e _ d. S)2z getPhases..InHeadPhasec S s | j d S)NT)r- )rB r' r' r( r s
z)getPhases..InHeadPhase.processEOFc S s | j | S)N)r- )rB r r' r' r( r s
z0getPhases..InHeadPhase.processCharactersc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r s z+getPhases..InHeadPhase.startTagHtmlc S s | j j d d S)Nz!two-heads-are-not-better-than-one)rJ r )rB r r' r' r( r4 s z+getPhases..InHeadPhase.startTagHeadc S s. | j j | | j j j d | d r r r )rB r r' r' r( startTagBaseLinkCommand s z6getPhases..InHeadPhase.startTagBaseLinkCommandc S s | j j | | j j j d | d <| d } | j j j j d d k r d | k ry | j j j j | d n{ d | k r d | k r | d j d k r t
j | d j d
} t
j
| } | j } | j j j j | d S) NTr ry r tentativecharsetcontentz
http-equivz content-typezutf-8)r> r r r rJ rN rQ rf changeEncodingrZ r
EncodingBytesencodeContentAttrParserr! )rB r rm ry rJ codecr' r' r( startTagMeta s
z+getPhases..InHeadPhase.startTagMetac S s | j j | d d S)Nr )rJ r )rB r r' r' r(
startTagTitle s z,getPhases..InHeadPhase.startTagTitlec S s | j j | d d S)Nr )rJ r )rB r r' r' r( startTagNoFramesStyle s z4getPhases..InHeadPhase.startTagNoFramesStylec S sL | j j r" | j j | d n& | j j | | j j d | j _ d S)Nr inHeadNoscript)rJ rL r r> r rG r_ )rB r r' r' r( startTagNoscript s z/getPhases..InHeadPhase.startTagNoscriptc S sT | j j | | j j j | j j _ | j j | j _ | j j d | j _ d S)Nr ) r> r rJ rN scriptDataStater\ r_ r rG )rB r r' r' r( startTagScript s z-getPhases..InHeadPhase.startTagScriptc S s | j | S)N)r- )rB r r' r' r( r7 s
z,getPhases..InHeadPhase.startTagOtherc S sQ | j j j j } | j d k s7 t d | j | j j d | j _ d S)Nr zExpected head got %s afterHead)rJ r> r r r@ r rG r_ )rB r r r' r' r(
endTagHead
s "z)getPhases..InHeadPhase.endTagHeadc S s | j | S)N)r- )rB r r' r' r( endTagHtmlBodyBr s
z/getPhases..InHeadPhase.endTagHtmlBodyBrc S s! | j j d d | d i d S)Nzunexpected-end-tagr@ )rJ r )rB r r' r' r( r9 s z*getPhases..InHeadPhase.endTagOtherc S s | j t d d S)Nr )rN r1 )rB r' r' r( r- s z+getPhases..InHeadPhase.anythingElser titlenoframesstylenoscriptscriptbasebasefontbgsoundcommandlinkr2 r r2 r N)rQ rR )rU rV rW rX rY )r2 r r )r9 r: r; r. r r r r r4 r= rF rG rH rJ rL r7 rN rO r9 r- r r: r r; r r' r' r' r( InHeadPhase s> rZ c @ s e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z d d Z
d d Z d d Z d d Z e j d e f d$ e
f d% e f g Z e e _ e j d! e
f d" e f g Z e e _ d# S)&z&getPhases..InHeadNoscriptPhasec S s | j j d | j d S)Nzeof-in-head-noscriptT)rJ r r- )rB r' r' r( r / s
z1getPhases..InHeadNoscriptPhase.processEOFc S s | j j d j | S)Nr5 )rJ rG r )rB r r' r' r( r 4 s z5getPhases..InHeadNoscriptPhase.processCommentc S s | j j d | j | S)Nzchar-in-head-noscript)rJ r r- )rB r r' r' r( r 7 s
z8getPhases..InHeadNoscriptPhase.processCharactersc S s | j j d j | S)Nr5 )rJ rG r )rB r r' r' r( r < s z=getPhases..InHeadNoscriptPhase.processSpaceCharactersc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r ? s z3getPhases..InHeadNoscriptPhase.startTagHtmlc S s | j j d j | S)Nr5 )rJ rG r )rB r r' r' r( r= B s z>getPhases..InHeadNoscriptPhase.startTagBaseLinkCommandc S s! | j j d d | d i d S)Nzunexpected-start-tagr@ )rJ r )rB r r' r' r( startTagHeadNoscriptE s z;getPhases..InHeadNoscriptPhase.startTagHeadNoscriptc S s+ | j j d d | d i | j | S)Nz unexpected-inhead-noscript-tagr@ )rJ r r- )rB r r' r' r( r7 H s
z4getPhases..InHeadNoscriptPhase.startTagOtherc S sQ | j j j j } | j d k s7 t d | j | j j d | j _ d S)NrS zExpected noscript got %sr5 )rJ r> r r r@ r rG r_ )rB r r r' r' r( endTagNoscriptM s "z5getPhases..InHeadNoscriptPhase.endTagNoscriptc S s+ | j j d d | d i | j | S)Nz unexpected-inhead-noscript-tagr@ )rJ r r- )rB r r' r' r( endTagBrR s
z/getPhases..InHeadNoscriptPhase.endTagBrc S s! | j j d d | d i d S)Nzunexpected-end-tagr@ )rJ r )rB r r' r' r( r9 W s z2getPhases..InHeadNoscriptPhase.endTagOtherc S s | j t d d S)NrS )r\ r1 )rB r' r' r( r- Z s z3getPhases..InHeadNoscriptPhase.anythingElser rV rW rY r2 rQ rR r rS r2 N)rV rW rY r2 rQ rR )r rS )r9 r: r; r. r r r r r r r= r[ r7 r\ r] r9 r- r r: r r; r r' r' r' r( InHeadNoscriptPhase, s, r^ c
@ s e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z d d Z
d d Z d d Z e j d e f d e f d e f d& e
f d# e f g Z e e _ e j d' e
f g Z e e _ d% S)(z!getPhases..AfterHeadPhasec S s | j d S)NT)r- )rB r' r' r( r n s
z,getPhases..AfterHeadPhase.processEOFc S s | j | S)N)r- )rB r r' r' r( r r s
z3getPhases..AfterHeadPhase.processCharactersc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r v s z.getPhases..AfterHeadPhase.startTagHtmlc S s6 d | j _ | j j | | j j d | j _ d S)NFr )rJ rd r> r rG r_ )rB r r' r' r( startTagBodyy s z.getPhases..AfterHeadPhase.startTagBodyc S s* | j j | | j j d | j _ d S)Nr )r> r rJ rG r_ )rB r r' r' r( startTagFrameset~ s z2getPhases..AfterHeadPhase.startTagFramesetc S s | j j d d | d i | j j j | j j | j j d j | xD | j j d d d D]) } | j d k rg | j j j | Pqg Wd S)Nz#unexpected-start-tag-out-of-my-headr@ r5 r r r )
rJ r r> r r r6 rG r r@ remove)rB r r r' r' r( startTagFromHead s z2getPhases..AfterHeadPhase.startTagFromHeadc S s! | j j d d | d i d S)Nzunexpected-start-tagr@ )rJ r )rB r r' r' r( r4 s z.getPhases..AfterHeadPhase.startTagHeadc S s | j | S)N)r- )rB r r' r' r( r7 s
z/getPhases..AfterHeadPhase.startTagOtherc S s | j | S)N)r- )rB r r' r' r( rO s
z2getPhases..AfterHeadPhase.endTagHtmlBodyBrc S s! | j j d d | d i d S)Nzunexpected-end-tagr@ )rJ r )rB r r' r' r( r9 s z-getPhases..AfterHeadPhase.endTagOtherc S s? | j j t d d | j j d | j _ d | j _ d S)Nr rt r T)r> r r1 rJ rG r_ rd )rB r' r' r( r- s z.getPhases..AfterHeadPhase.anythingElser r r rU rV rW rY r2 rQ rT rR rP r r2 N) rU rV rW rY r2 rQ rT rR rP )r r r2 )r9 r: r; r. r r r r r_ r` rb r4 r7 rO r9 r- r r: r r; r r' r' r' r( AfterHeadPhasek s0
rc c , s9 e Z d Z d Z f d d Z d d Z d d Z d d Z d
d Z d d
Z d d Z
d d Z d d Z d d Z
d d Z d d Z d d Z d d Z d d Z d d! Z d" d# Z d$ d% Z d& d' Z d( d) Z d* d+ Z d, d- Z d. d/ Z d0 d1 Z d2 d3 Z d4 d5 Z d6 d7 Z d8 d9 Z d: d; Z d< d= Z! d> d? Z" d@ dA Z# dB dC Z$ dD dE Z% dF dG Z& dH dI Z' dJ dK Z( dL dM Z) dN dO Z* dP dQ Z+ dR dS Z, dT dU Z- dV dW Z. dX dY Z/ dZ d[ Z0 d\ d] Z1 d^ d_ Z2 d` da Z3 db dc Z4 dd de Z5 df dg Z6 e7 j8 dh j9 f d e f dr e f ds e
f d e f e: e f d e f d e f d e f d e f d e f d e f d e f d e f d e f d e f d e f d e f d e f d e f d e f d e f d e f d e! f d e" f d e# f d e$ f d e& f d e' f d e% f d e( f d e) f d e* f g! Z; e+ e; _< e7 j8 dr e- f dh e. f d e/ f d e0 f d e, f d e1 f e: e2 f d e3 f d e4 f d e5 f g
Z= e6 e= _< S)z getPhases..InBodyPhaser c s) t | j | | | j | _ d S)N)superrI processSpaceCharactersNonPrer )rB r r$ ) InBodyPhaser r' r( rI s z'getPhases..InBodyPhase.__init__c S s4 | j | j k o3 | j | j k o3 | j | j k S)N)r@ rl rm )rB node1node2r' r' r( isMatchingFormattingElement s z:getPhases..InBodyPhase.isMatchingFormattingElementc S s | j j | | j j d } g } xP | j j d d d D]5 } | t k rV Pq@ | j | | r@ | j | q@ Wt | d k s t t | d k r | j j j | d | j j j | d S)Nr r r r )
r> r r activeFormattingElementsr ri r r r ra )rB r ro matchingElementsr r' r' r( addFormattingElement s z3getPhases..InBodyPhase.addFormattingElementc
S sT t d } xA | j j d d d D]& } | j | k r& | j j d Pq& Wd S)Ndddtlir& r r r r r r r r r z expected-closing-tag-but-got-eof) rn ro rp r& r r r r r r r r r )r r> r r@ rJ r )rB allowed_elementsr r' r' r( r s z)getPhases..InBodyPhase.processEOFc S s | d } | j | _ | j d re | j j d j d k re | j j d j re | d d } | r | j j | j j | d S)
Nry
r prelistingtextarear )rs rt ru r ) re r r r> r r@
hasContent#reconstructActiveFormattingElementsr )rB r ry r' r' r( !processSpaceCharactersDropNewline s
z@getPhases..InBodyPhase.processSpaceCharactersDropNewlinec S sn | d d k r d S| j j | j j | d | j j rj t d d | d D rj d | j _ d S)Nry c S s g | ] } | t k q Sr' )r
)r? charr' r' r(
s zDgetPhases..InBodyPhase.processCharacters..F)r> rw r rJ rd any)rB r r' r' r( r s
z0getPhases..InBodyPhase.processCharactersc S s% | j j | j j | d d S)Nry )r> rw r )rB r r' r' r( re s
z;getPhases..InBodyPhase.processSpaceCharactersNonPrec S s | j j d j | S)Nr5 )rJ rG r )rB r r' r' r( startTagProcessInHead s z4getPhases..InBodyPhase.startTagProcessInHeadc S s | j j d d d i t | j j d k sJ | j j d j d k r_ | j j s t n] d | j _ xN | d j D]< \ } } | | j j d j
k r| | | j j d j
| r r@ rR r rd r, rm )rB r r r r' r' r( r_ s z+getPhases..InBodyPhase.startTagBodyc S s | j j d d d i t | j j d k sJ | j j d j d k r_ | j j s t n | j j sn n | j j d j r | j j d j j
| j j d x- | j j d j d k r | j j j q W| j j | | j j
d | j _ d S) Nzunexpected-start-tagr@ r r r r r r )rJ r r r> r r@ rR r rd parent removeChildr r rG r_ )rB r r' r' r( r` s 1 $ z/getPhases..InBodyPhase.startTagFramesetc S s? | j j d d d r+ | j t d | j j | d S)Nr& variantbutton)r> elementInScopeendTagPr1 r )rB r r' r' r( startTagCloseP s z-getPhases..InBodyPhase.startTagClosePc S sW | j j d d d r+ | j t d | j j | d | j _ | j | _ d S)Nr& r r F) r> r r r1 r rJ rd rx r )rB r r' r' r( startTagPreListing s
z1getPhases..InBodyPhase.startTagPreListingc S s} | j j r( | j j d d d i nQ | j j d d d rS | j t d | j j | | j j d | j _ d S) Nzunexpected-start-tagr@ formr& r r r r ) r> formPointerrJ r r r r1 r r )rB r r' r' r( startTagForm s z+getPhases..InBodyPhase.startTagFormc S s d | j _ d d g d d d g d d d g i } | | d } xh t | j j D]T } | j | k r | j j j t | j d P| j t
k rT | j d k rT PqT W| j j d d
d r | j j j t d d | j j | d S)
NFrp ro rn r@ ru addressr) r& r r )r r) r& )
rJ rd reversedr> r r@ r_ r r1 nameTupler r r )rB r stopNamesMap stopNamesr r' r' r( startTagListItem& s" z/getPhases..InBodyPhase.startTagListItemc S sW | j j d d d r+ | j t d | j j | | j j j | j j _ d S)Nr& r r ) r> r r r1 r rJ rN r^ r\ )rB r r' r' r( startTagPlaintext< s z0getPhases..InBodyPhase.startTagPlaintextc S s | j j d d d r+ | j t d | j j d j t k rq | j j d d | d i | j j j | j j
| d S)Nr& r r r zunexpected-start-tagr@ r ) r> r r r1 r r@ r rJ r r r )rB r r' r' r( startTagHeadingB s z.getPhases..InBodyPhase.startTagHeadingc S s | j j d } | r | j j d d d d d i | j t d | | j j k ro | j j j | | | j j k r | j j j | | j j | j
| d S)Naz$unexpected-start-tag-implies-end-tag startNameendName) r> !elementInActiveFormattingElementsrJ r endTagFormattingr1 r ra rk rw rm )rB r afeAElementr' r' r( startTagAJ s
z(getPhases..InBodyPhase.startTagAc S s | j j | j | d S)N)r> rw rm )rB r r' r' r( startTagFormattingW s
z1getPhases..InBodyPhase.startTagFormattingc S so | j j | j j d r^ | j j d d d d d i | j t d | j j | j | d S)Nnobrz$unexpected-start-tag-implies-end-tagr r )r> rw r rJ r r r1 rm )rB r r' r' r( startTagNobr[ s
z+getPhases..InBodyPhase.startTagNobrc S su | j j d rH | j j d d d d d i | j t d | S| j j | j j | d | j _ d S)Nr z$unexpected-start-tag-implies-end-tagr r F) r> r rJ r r r1 rw r rd )rB r r' r' r( startTagButtone s
z-getPhases..InBodyPhase.startTagButtonc S s@ | j j | j j | | j j j t d | j _ d S)NF)r> rw r rk r r rJ rd )rB r r' r' r( startTagAppletMarqueeObjectp s
z:getPhases..InBodyPhase.startTagAppletMarqueeObjectc S s[ | j j d d d r+ | j t d | j j d | j _ | j j | d d S)Nr& r r Fr )r> r r r1 rw rJ rd r )rB r r' r' r( startTagXmpv s
z*getPhases..InBodyPhase.startTagXmpc S ss | j j d k r= | j j d d d r= | j t d | j j | d | j _ | j j d | j _ d S)Nr) r& r r Fr )
rJ rY r> r r r1 r rd rG r_ )rB r r' r' r(
startTagTable} s z,getPhases..InBodyPhase.startTagTablec S sG | j j | j j | | j j j d | d rw r r r rJ rd )rB r r' r' r( startTagVoidFormatting s
z5getPhases..InBodyPhase.startTagVoidFormattingc S sV | j j } | j | d | d k rR | d d j t d k rR | | j _ d S)Nr0 ry hidden)rJ rd r rn r )rB r rd r' r' r(
startTagInput s
z,getPhases..InBodyPhase.startTagInputc S s. | j j | | j j j d | d r r r )rB r r' r' r( startTagParamSource s z2getPhases..InBodyPhase.startTagParamSourcec S se | j j d d d r+ | j t d | j j | | j j j d | d r r r1 r r r rJ rd )rB r r' r' r(
startTagHr s
z)getPhases..InBodyPhase.startTagHrc S sM | j j d d d d d i | j t d d d | d d | d d S)
Nz unexpected-start-tag-treated-as originalNameimagenewNameimgrt rm ry r )rJ r r r1 )rB r r' r' r(
startTagImage s
z,getPhases..InBodyPhase.startTagImagec S s | j j d d d i | j j r) d Si } d | d k rQ | d d | d <| j t d d d | | j t d d | j t d
d d | d k r | d d } n d } | j d
t d d | i | d j } d | k r | d =d | k r| d =d | d <| j t d d d | d | d | j t d
| j t d d | j t d d S)Nzdeprecated-tagr@ isindexactionry r rt rm hrlabelpromptz3This is a searchable index. Enter search keywords: r0 rr inputr )
rJ r r> r r r1 r r copyr )rB r
form_attrsr rm r' r' r( startTagIsIndex s6
z.getPhases..InBodyPhase.startTagIsIndexc S sD | j j | | j j j | j j _ | j | _ d | j _ d S)NF) r> r rJ rN r[ r\ rx r rd )rB r r' r' r( startTagTextarea s z/getPhases..InBodyPhase.startTagTextareac S s d | j _ | j | d S)NF)rJ rd startTagRawtext)rB r r' r' r( startTagIFrame s z-getPhases..InBodyPhase.startTagIFramec S s- | j j r | j | n
| j | d S)N)rJ rL r r7 )rB r r' r' r( rJ s z/getPhases..InBodyPhase.startTagNoscriptc S s | j j | d d S)z8iframe, noembed noframes, noscript(if scripting enabled)r N)rJ r )rB r r' r' r( r s z.getPhases..InBodyPhase.startTagRawtextc S sV | j j d j d k r2 | j j j t d | j j | j j j | d S)Nr optionr ) r> r r@ rJ r_ r r1 rw r )rB r r' r' r( startTagOpt s
z*getPhases..InBodyPhase.startTagOptc S s | j j | j j | d | j _ | j j | j j d | j j d | j j d | j j d | j j d | j j d f k r | j j d | j _ n | j j d | j _ d S)
NFr r r r r r inSelectInTabler )r> rw r rJ rd r_ rG )rB r r' r' r( startTagSelect s
z-getPhases..InBodyPhase.startTagSelectc S sY | j j d rE | j j | j j d j d k rE | j j | j j | d S)Nrubyr r )r> r generateImpliedEndTagsr r@ rJ r r )rB r r' r' r( startTagRpRt s
z+getPhases..InBodyPhase.startTagRpRtc S ss | j j | j j | | j j | t d | d <| j j | | d ro | j j j d | d rw rJ r r r r r r )rB r r' r' r( startTagMath s
z+getPhases..InBodyPhase.startTagMathc S ss | j j | j j | | j j | t d | d <| j j | | d ro | j j j d | d rw rJ r r r r r r )rB r r' r' r( startTagSvg s
z*getPhases..InBodyPhase.startTagSvgc S s! | j j d d | d i d S)a5 Elements that should be children of other elements that have a
different insertion mode; here they are ignored
"caption", "col", "colgroup", "frame", "frameset", "head",
"option", "optgroup", "tbody", "td", "tfoot", "th", "thead",
"tr", "noscript"
z unexpected-start-tag-ignoredr@ N)rJ r )rB r r' r' r( startTagMisplaced s z0getPhases..InBodyPhase.startTagMisplacedc S s! | j j | j j | d S)N)r> rw r )rB r r' r' r( r7 s
z,getPhases..InBodyPhase.startTagOtherc S s | j j d d d s` | j t d d | j j d d d i | j t d d n| | j j d | j j d j d k r | j j d d d i | j j j
} x% | j d k r | j j j
} q Wd S)
Nr& r r rt zunexpected-end-tagr@ ru r r ) r> r r r1 rJ r r r r r@ r )rB r r r' r' r( r s z&getPhases..InBodyPhase.endTagPc S s | j j d s# | j j d S| j j d j d k r xV | j j d d D]> } | j t d k rS | j j d d d d | j i PqS W| j j d | j _ d S)Nr r rn ro rp optgroupr r& rprtr r r r r r r z$expected-one-end-tag-but-got-anothergotName expectedName afterBodyr )rn ro rp r r r& r r r r r r r r r r ) r> r rJ r r r@ r rG r_ )rB r r r' r' r(
endTagBody) s
z)getPhases..InBodyPhase.endTagBodyc S s- | j j d r) | j t d | Sd S)Nr )r> r r r1 )rB r r' r' r(
endTagHtml; s z)getPhases..InBodyPhase.endTagHtmlc S s | d d k r | j | _ | j j | d } | rE | j j | j j d j | d k r | j j d d | d i | r | j j j } x) | j | d k r | j j j } q Wd S)Nr@ rs r zend-tag-too-earlyr )
re r r> r r r r@ rJ r r )rB r inScoper r' r' r( endTagBlockA s
z*getPhases..InBodyPhase.endTagBlockc S s | j j } d | j _ | d k s7 | j j | rS | j j d d d i nO | j j | j j d | k r | j j d d d i | j j j | d S)Nzunexpected-end-tagr@ r r zend-tag-too-early-ignoredr )r> r r rJ r r r ra )rB r r r' r' r(
endTagFormO s
z)getPhases..InBodyPhase.endTagFormc S s | d d k r d } n d } | j j | d d | s[ | j j d d | d i n | j j d | d | j j d j | d k r | j j d d | d i | j j j } x) | j | d k r | j j j } q Wd S)
Nr@ rp listr zunexpected-end-tagexcluder zend-tag-too-earlyr )r> r rJ r r r r@ r )rB r r r r' r' r( endTagListItem\ s z-getPhases..InBodyPhase.endTagListItemc S s x. t D]& } | j j | r | j j Pq W| j j d j | d k rk | j j d d | d i x[ t D]S } | j j | rr | j j j } x% | j t k r | j j j } q WPqr Wd S)Nr r@ zend-tag-too-earlyr ) r r> r r r r@ rJ r r )rB r itemr' r' r(
endTagHeadingm s
z,getPhases..InBodyPhase.endTagHeadingc S s! d } x| d k r | d 7} | j j | d } | sd | | j j k ru | j j | j ru | j | d S| | j j k r | j j d d | d i | j j j | d S| j j | j s | j j d d | d i d S| | j j d k r$| j j d d | d i | j j j
| } d } x4 | j j | d D] } | j t k rV| } PqVW| d k r| j j j
} x" | | k r| j j j
} qW| j j j | d S| j j | d } | j j j
| } | } }
d } | j j j
|
} x
| d k r*| d 7} | d 8} | j j | }
|
| j j k ry| j j j |
q!|
| k rP| | k r| j j j
|
d } |
j }
|
| j j | j j j
|
<|
| j j | j j j
|
<|
}
| j r| j j | |
j | |
} q!W| j rD| j j | | j t d k r| j j \ } } | j | | n
| j | | j }
| j |
| j |
| j j j | | j j j | |
| j j j | | j j j | j j j
| d |
q Wd S)z)The much-feared adoption agency algorithmr r r@ Nzadoption-agency-1.2zadoption-agency-4.4zadoption-agency-1.3rj r r r r r r )r r r r r )r> r r r r@ r9 rJ r rk ra indexr r r cloneNoder~ r appendChildr getTableMisnestedNodePosition insertBeforereparentChildreninsert)rB r outerLoopCounterformattingElementafeIndex
furthestBlockro commonAncestorbookmarklastNoder innerLoopCounterr cloner~ r r' r' r( r | s
z/getPhases..InBodyPhase.endTagFormattingc S s | j j | d r# | j j | j j d j | d k r] | j j d d | d i | j j | d r | j j j } x) | j | d k r | j j j } q W| j j d S)Nr@ r zend-tag-too-earlyr ) r> r r r r@ rJ r r clearActiveFormattingElements)rB r ro r' r' r( endTagAppletMarqueeObject s
z8getPhases..InBodyPhase.endTagAppletMarqueeObjectc S sY | j j d d d d d i | j j | j j t d d | j j j d S)Nz unexpected-end-tag-treated-asr r2 r z
br elementrt )rJ r r> rw r r1 r r )rB r r' r' r( r] + s
z'getPhases..InBodyPhase.endTagBrc S s x | j j d d d D] } | j | d k r | j j d | d | j j d j | d k r | j j d d | d i x | j j j | k r q WPq | j t k r | j j d d | d i Pq Wd S)Nr r@ r zunexpected-end-tagr r ) r> r r@ r rJ r r r r )rB r r r' r' r( r9 2 s z*getPhases..InBodyPhase.endTagOtherr rU rV rW rX rY r2 rT rR rP r r r articleaside
blockquotecenterdetailsdirr) dlfieldset
figcaptionfigurefooterheaderhgroupmainmenunavolr& sectionsummaryulrs rt r rp rn ro rT r bbigcodeemfontissmallstrikestrongttur r appletmarqueeobjectxmpr arear2 embedr keygenwbrparamsourcetrackr r r r ru iframerS noembedrQ r r r r r mathr} r colr framer r r r r r r dialog)r ) rU rV rW rX rY r2 rT rR rP )r r r r r r r r) r r r r r r r r r r r r& r r r )rs rt )rp rn ro ) r r r r r r r r r r r r )r r r )r r2 r r r r
)r r r
)r rQ )r r )r r ) r r r r r r r r r r r )r r r r r r r r r r) r r r r r r r rt r r r r rs r r r )rn ro rp )r r r r r r r r r r r r r r )r r r )>r9 r: r; r rI ri rm r rx r re r} r_ r` r r r r r r r r r r r r r r r r r r r r r rJ r r r r r r r r7 r r r r r r r r r r] r9 r r: r r r r; r r' )rf r )r r( rf s
rf c @ s e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z e
j g Z e e _
e
j d e f g Z e e _
d S)
z getPhases..TextPhasec S s | j j | d d S)Nry )r> r )rB r r' r' r( r s z.getPhases..TextPhase.processCharactersc S sL | j j d d | j j d j i | j j j | j j | j _ d S)Nz&expected-named-closing-tag-but-got-eofr@ r Tr )rJ r r> r r@ r r r_ )rB r' r' r( r s
z'getPhases..TextPhase.processEOFc S s d s t d | d d S)NFz4Tried to process start tag %s in RCDATA/RAWTEXT moder@ )r )rB r r' r' r( r7 s z*getPhases..TextPhase.startTagOtherc S s= | j j j } | j d k s' t | j j | j _ d S)NrT )r> r r r@ r rJ r r_ )rB r r r' r' r( endTagScript s z)getPhases..TextPhase.endTagScriptc S s& | j j j | j j | j _ d S)N)r> r r rJ r r_ )rB r r' r' r( r9 s z(getPhases..TextPhase.endTagOtherrT N)r9 r: r; r. r r r r7 r r9 r r: r r; r r' r' r' r( TextPhase s r c
s e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z d d Z
d d Z d d Z d d Z d d Z d d Z d d Z d d Z d! d" Z d# d$ Z e j d% j f d& e
f d' e f d( e f d6 e
f d7 e f d/ e f d8 e f d2 e f d3 e f g
Z e e _ e j d/ e f d9 e f g Z e e _ d5 S):z getPhases..InTablePhasec S s4 x- | j j d j d k r/ | j j j q Wd S)Nr r r r )r r )r> r r@ r )rB r' r' r( clearStackToTableContext s z8getPhases..InTablePhase.clearStackToTableContextc S sB | j j d j d k r, | j j d n | j j s> t d S)Nr r z eof-in-tabler )r> r r@ rJ r rR r )rB r' r' r( r s z*getPhases..InTablePhase.processEOFc S sH | j j } | j j d | j _ | | j j _ | j j j | d S)N inTableText)rJ r_ rG r r )rB r r r' r' r( r s z6getPhases..InTablePhase.processSpaceCharactersc S sH | j j } | j j d | j _ | | j j _ | j j j | d S)Nr )rJ r_ rG r r )rB r r r' r' r( r s z1getPhases..InTablePhase.processCharactersc S s3 d | j _ | j j d j | d | j _ d S)NTr F)r> insertFromTablerJ rG r )rB r r' r' r( r s z*getPhases..InTablePhase.insertTextc S sG | j | j j j t | j j | | j j d | j _ d S)Nr ) r r> rk r r r rJ rG r_ )rB r r' r' r( startTagCaption s
z/getPhases..InTablePhase.startTagCaptionc S s4 | j | j j | | j j d | j _ d S)Nr )r r> r rJ rG r_ )rB r r' r' r( startTagColgroup s
z0getPhases..InTablePhase.startTagColgroupc S s | j t d d | S)Nr rt )r r1 )rB r r' r' r( startTagCol s z+getPhases..InTablePhase.startTagColc S s4 | j | j j | | j j d | j _ d S)Nr )r r> r rJ rG r_ )rB r r' r' r( startTagRowGroup s
z0getPhases..InTablePhase.startTagRowGroupc S s | j t d d | S)Nr rt )r r1 )rB r r' r' r( startTagImplyTbody s z2getPhases..InTablePhase.startTagImplyTbodyc S sL | j j d d d d d i | j j j t d | j j sH | Sd S)Nz$unexpected-start-tag-implies-end-tagr r r )rJ r r_ r r1 rR )rB r r' r' r( r s
z-getPhases..InTablePhase.startTagTablec S s | j j d j | S)Nr5 )rJ rG r )rB r r' r' r( startTagStyleScript s z3getPhases..InTablePhase.startTagStyleScriptc S sq d | d k r` | d d j t d k r` | j j d | j j | | j j j n
| j | d S)Nr0 ry r z unexpected-hidden-input-in-table) rn r rJ r r> r r r r7 )rB r r' r' r( r s z-getPhases..InTablePhase.startTagInputc S s\ | j j d | j j d k rX | j j | | j j d | j _ | j j j d S)Nzunexpected-form-in-tabler r )rJ r r> r r r r )rB r r' r' r( r s
z,getPhases..InTablePhase.startTagFormc S sP | j j d d | d i d | j _ | j j d j | d | j _ d S)Nz)unexpected-start-tag-implies-table-voodoor@ Tr F)rJ r r> r rG r )rB r r' r' r( r7 s z-getPhases..InTablePhase.startTagOtherc S s | j j d d d r | j j | j j d j d k rj | j j d d d d | j j d j i x- | j j d j d k r | j j j qm W| j j j | j j n | j j s t
| j j d S)
Nr r r zend-tag-too-early-namedr r r r r ) r> r r r r@ rJ r r ra rR r )rB r r' r' r( endTagTable s
z+getPhases..InTablePhase.endTagTablec S s! | j j d d | d i d S)Nzunexpected-end-tagr@ )rJ r )rB r r' r' r( endTagIgnore s z,getPhases..InTablePhase.endTagIgnorec S sP | j j d d | d i d | j _ | j j d j | d | j _ d S)Nz'unexpected-end-tag-implies-table-voodoor@ Tr F)rJ r r> r rG r )rB r r' r' r( r9 s z+getPhases..InTablePhase.endTagOtherr r r r r r r r r r r rR rT r r r N)r r r )r r r )rR rT ) r r r r r r r r r r r ) r9 r: r; r. r r r r r r r r r r r r r r r r7 r r r9 r r: r r r; r r' )r r' r( InTablePhase sH
r! c s e Z d Z d Z f d d Z d d Z d d Z d d
Z d d Z d
d Z d d Z
d d Z S)z#getPhases..InTableTextPhaser characterTokensc s/ t | j | | d | _ g | _ d S)N)rd rI r r" )rB r r$ )InTableTextPhaser r' r( rI 6 s z,getPhases..InTableTextPhase.__init__c S s d j d d | j D } t d d | D rh d t d d | i } | j j d j | n | r~ | j j | g | _ d S) Nr c S s g | ] } | d q S)ry r' )r? r r' r' r( r{ < s zGgetPhases..InTableTextPhase.flushCharacters..c S s g | ] } | t k q Sr' )r
)r? r r' r' r( r{ = s r0 rr ry r )joinr" r| r rJ rG r r> )rB ry r r' r' r( flushCharacters; s z3getPhases..InTableTextPhase.flushCharactersc S s | j | j | j _ | S)N)r% r rJ r_ )rB r r' r' r( r D s
z2getPhases..InTableTextPhase.processCommentc S s | j | j | j _ d S)NT)r% r rJ r_ )rB r' r' r( r I s
z.getPhases..InTableTextPhase.processEOFc S s( | d d k r d S| j j | d S)Nry ry )r" r )rB r r' r' r( r N s z5getPhases..InTableTextPhase.processCharactersc S s | j j | d S)N)r" r )rB r r' r' r( r S s z:getPhases..InTableTextPhase.processSpaceCharactersc S s | j | j | j _ | S)N)r% r rJ r_ )rB r r' r' r( r X s
z3getPhases..InTableTextPhase.processStartTagc S s | j | j | j _ | S)N)r% r rJ r_ )rB r r' r' r( r ] s
z1getPhases..InTableTextPhase.processEndTag)r r" ) r9 r: r; r rI r% r r r r r r r' )r# )r r( r# 3 s r# c
s e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z d d Z
e j d j f d e f g Z e e _ e j d e
f d e f d! e f g Z e
e _ d S)"z!getPhases..InCaptionPhasec S s | j j d d d S)Nr r r )r> r )rB r' r' r( ignoreEndTagCaptionf s z5getPhases..InCaptionPhase.ignoreEndTagCaptionc S s | j j d j d S)Nr )rJ rG r )rB r' r' r( r i s z,getPhases..InCaptionPhase.processEOFc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r l s z3getPhases..InCaptionPhase.processCharactersc S s@ | j j | j } | j j j t d | s< | Sd S)Nr )rJ r r& r_ r r1 )rB r ignoreEndTagr' r' r( startTagTableElemento s
z6getPhases..InCaptionPhase.startTagTableElementc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r7 w s z/getPhases..InCaptionPhase.startTagOtherc S s | j s | j j | j j d j d k r^ | j j d d d d | j j d j i x- | j j d j d k r | j j j qa W| j j j | j j | j j d | j _
n | j j s t | j j d S)
Nr r z$expected-one-end-tag-but-got-anotherr r r r r r )
r& r> r r r@ rJ r r r rG r_ rR r )rB r r' r' r(
endTagCaptionz s
z/getPhases..InCaptionPhase.endTagCaptionc S s@ | j j | j } | j j j t d | s< | Sd S)Nr )rJ r r& r_ r r1 )rB r r' r' r' r( r s
z-getPhases..InCaptionPhase.endTagTablec S s! | j j d d | d i d S)Nzunexpected-end-tagr@ )rJ r )rB r r' r' r( r s z.getPhases..InCaptionPhase.endTagIgnorec S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r9 s z-getPhases..InCaptionPhase.endTagOtherr r r r r r r r r r r r N) r r r r r r r r r )
r r r r r r r r r r )r9 r: r; r. r r& r r r( r7 r) r r r9 r r: r r r; r r' )r r' r( InCaptionPhaseb s* r* c s e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z e
j d j f d e f g Z e e _ e
j d e
f d e f g Z e e _ d S)z%getPhases..InColumnGroupPhasec S s | j j d j d k S)Nr r r )r> r r@ )rB r' r' r( ignoreEndTagColgroup s z:getPhases..InColumnGroupPhase.ignoreEndTagColgroupc S s\ | j j d j d k r/ | j j s+ t d S| j } | j t d | sX d Sd S)Nr r r Tr ) r> r r@ rJ rR r r+ endTagColgroupr1 )rB r' r' r' r( r s z0getPhases..InColumnGroupPhase.processEOFc S s- | j } | j t d | s) | Sd S)Nr )r+ r, r1 )rB r r' r' r' r( r s z7getPhases..InColumnGroupPhase.processCharactersc S s. | j j | | j j j d | d r r r )rB r r' r' r( r s z1getPhases..InColumnGroupPhase.startTagColc S s- | j } | j t d | s) | Sd S)Nr )r+ r, r1 )rB r r' r' r' r( r7 s z3getPhases..InColumnGroupPhase.startTagOtherc S sX | j r. | j j s t | j j n& | j j j | j j d | j _ d S)Nr )
r+ rJ rR r r r> r r rG r_ )rB r r' r' r( r, s
z4getPhases..InColumnGroupPhase.endTagColgroupc S s | j j d d d i d S)Nz
no-end-tagr@ r )rJ r )rB r r' r' r( endTagCol s z/getPhases..InColumnGroupPhase.endTagColc S s- | j } | j t d | s) | Sd S)Nr )r+ r, r1 )rB r r' r' r' r( r9 s z1getPhases..InColumnGroupPhase.endTagOtherr r r N)r9 r: r; r. r r+ r r r r7 r, r- r9 r r: r r r; r r' )r r' r( InColumnGroupPhase s"
r. c s e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z d d Z
d d Z d d Z d d Z e j d j f d e f d& e
f d' e f g Z e e _ e j d( e
f d# e f d) e f g Z e e _ d% S)*z#getPhases..InTableBodyPhasec S s_ x- | j j d j d k r/ | j j j q W| j j d j d k r[ | j j s[ t d S) Nr r r r r r )r r r r r )r> r r@ r rJ rR r )rB r' r' r( clearStackToTableBodyContext s
z@getPhases..InTableBodyPhase.clearStackToTableBodyContextc S s | j j d j d S)Nr )rJ rG r )rB r' r' r( r s z.getPhases..InTableBodyPhase.processEOFc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r s z:getPhases..InTableBodyPhase.processSpaceCharactersc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r s z5getPhases..InTableBodyPhase.processCharactersc S s4 | j | j j | | j j d | j _ d S)Nr )r/ r> r rJ rG r_ )rB r r' r' r(
startTagTr s
z.getPhases..InTableBodyPhase.startTagTrc S s7 | j j d d | d i | j t d d | S)Nz unexpected-cell-in-table-bodyr@ r rt )rJ r r0 r1 )rB r r' r' r( startTagTableCell s z5getPhases..InTableBodyPhase.startTagTableCellc S s | j j d d d sH | j j d d d sH | j j d d d rv | j | j t | j j d j | S| j j s t | j j
d S)Nr r r r r r r ) r> r r/ endTagTableRowGroupr1 r r@ rJ rR r r )rB r r' r' r( startTagTableOther s
z6getPhases..InTableBodyPhase.startTagTableOtherc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r7 s z1getPhases..InTableBodyPhase.startTagOtherc S sp | j j | d d d rO | j | j j j | j j d | j _ n | j j d d | d i d S)Nr@ r r r z unexpected-end-tag-in-table-body) r> r r/ r r rJ rG r_ r )rB r r' r' r( r2 s
z7getPhases..InTableBodyPhase.endTagTableRowGroupc S s | j j d d d sH | j j d d d sH | j j d d d rv | j | j t | j j d j | S| j j s t | j j
d S)Nr r r r r r r ) r> r r/ r2 r1 r r@ rJ rR r r )rB r r' r' r( r % s
z/getPhases..InTableBodyPhase.endTagTablec S s! | j j d d | d i d S)Nz unexpected-end-tag-in-table-bodyr@ )rJ r )rB r r' r' r( r 2 s z0getPhases..InTableBodyPhase.endTagIgnorec S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r9 6 s z/getPhases..InTableBodyPhase.endTagOtherr r r r r r r r r r r r N)r r )r r r r r r )r r r )r r r r r r r r )r9 r: r; r. r r/ r r r r0 r1 r3 r7 r2 r r r9 r r: r r r; r r' )r r' r( InTableBodyPhase s4
r4 c s' e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z d d Z
d d Z d d Z d d Z d d Z e j d j f d( e
f d) e f g Z e e _ e j d$ e
f d% e f d* e f d+ e f g Z e e _ d' S),z getPhases..InRowPhasec S sZ xS | j j d j d k rU | j j d d | j j d j i | j j j q Wd S) Nr r r z'unexpected-implied-end-tag-in-table-rowr@ r )r r r )r> r r@ rJ r r )rB r' r' r( clearStackToTableRowContextO s z9getPhases..InRowPhase.clearStackToTableRowContextc S s | j j d d d S)Nr r r )r> r )rB r' r' r( ignoreEndTagTrU s z,getPhases..InRowPhase.ignoreEndTagTrc S s | j j d j d S)Nr )rJ rG r )rB r' r' r( r Y s z(getPhases..InRowPhase.processEOFc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r \ s z4getPhases..InRowPhase.processSpaceCharactersc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r _ s z/getPhases..InRowPhase.processCharactersc S sG | j | j j | | j j d | j _ | j j j t d S)Nr ) r5 r> r rJ rG r_ rk r r )rB r r' r' r( r1 b s
z/getPhases..InRowPhase.startTagTableCellc S s- | j } | j t d | s) | Sd S)Nr )r6 endTagTrr1 )rB r r' r' r' r( r3 h s z0getPhases..InRowPhase.startTagTableOtherc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r7 o s z+getPhases..InRowPhase.startTagOtherc S sb | j s? | j | j j j | j j d | j _ n | j j sQ t | j j
d S)Nr ) r6 r5 r> r r rJ rG r_ rR r r )rB r r' r' r( r7 r s
z&getPhases..InRowPhase.endTagTrc S s- | j } | j t d | s) | Sd S)Nr )r6 r7 r1 )rB r r' r' r' r( r | s z)getPhases..InRowPhase.endTagTablec S sD | j j | d d d r3 | j t d | S| j j d S)Nr@ r r r )r> r r7 r1 rJ r )rB r r' r' r( r2 s z1getPhases..InRowPhase.endTagTableRowGroupc S s! | j j d d | d i d S)Nz unexpected-end-tag-in-table-rowr@ )rJ r )rB r r' r' r( r s z*getPhases..InRowPhase.endTagIgnorec S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r9 s z)getPhases..InRowPhase.endTagOtherr r r r r r r r r r r r N)r r )r r r r r r r )r r r )r r r r r r r )r9 r: r; r. r r5 r6 r r r r1 r3 r7 r7 r r2 r r9 r r: r r r; r r' )r r' r(
InRowPhaseJ s6
r8 c s e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z d d Z
e j d j f d e f g Z e e _ e j d! e
f d" e f d# e f g Z e
e _ d S)$z getPhases..InCellPhasec S s] | j j d d d r. | j t d n+ | j j d d d rY | j t d d S)Nr r r r )r> r endTagTableCellr1 )rB r' r' r( closeCell s z(getPhases..InCellPhase.closeCellc S s | j j d j d S)Nr )rJ rG r )rB r' r' r( r s z)getPhases..InCellPhase.processEOFc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r s z0getPhases..InCellPhase.processCharactersc S sa | j j d d d s0 | j j d d d r> | j | S| j j sP t | j j d S)Nr r r r )r> r r: rJ rR r r )rB r r' r' r( r3 s
z1getPhases..InCellPhase.startTagTableOtherc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r7 s z,getPhases..InCellPhase.startTagOtherc S s | j j | d d d r | j j | d | j j d j | d k r | j j d d | d i x= | j j j } | j | d k rm Pqm Wn | j j j | j j | j j d | j _
n | j j d d | d i d S) Nr@ r r r zunexpected-cell-end-tagr zunexpected-end-tagr ) r> r r r r@ rJ r r r rG r_ )rB r r r' r' r( r9 s
z.getPhases..InCellPhase.endTagTableCellc S s! | j j d d | d i d S)Nzunexpected-end-tagr@ )rJ r )rB r r' r' r( r s z+getPhases..InCellPhase.endTagIgnorec S s; | j j | d d d r* | j | S| j j d S)Nr@ r r )r> r r: rJ r )rB r r' r' r( endTagImply s
z*getPhases..InCellPhase.endTagImplyc S s | j j d j | S)Nr )rJ rG r )rB r r' r' r( r9 s z*getPhases..InCellPhase.endTagOtherr r r r r r r r r r r r N) r r r r r r r r r )r r )r r r r r )r r r r r )r9 r: r; r. r r: r r r3 r7 r9 r r; r9 r r: r r r; r r' )r r' r( InCellPhase s(
r< c s- e Z d Z e Z d d Z d d Z d d Z d d Z d d
Z d d Z
d
d Z d d Z d d Z
d d Z d d Z d d Z e j d j f d e f d e f d e f d" e
f d e f g Z e e _ e j d e
f d e f d e f g Z e e _ d! S)#z getPhases..InSelectPhasec S sB | j j d j d k r, | j j d n | j j s> t d S)Nr r z
eof-in-selectr )r> r r@ rJ r rR r )rB r' r' r( r s z+getPhases..InSelectPhase.processEOFc S s, | d d k r d S| j j | d d S)Nry ry )r> r )rB r r' r' r( r s z2getPhases.