ó
§l§hc @ s) d Z d d l Z d d l Z e j d ƒ Z e j d ƒ Z e j d ƒ Z e j d ƒ Z e j d ƒ Z e j d ƒ Z e j d ƒ Z
e j d
ƒ Z e j d ƒ Z e j d ƒ Z
e j d
ƒ Z e j d e j ƒ Z e j d ƒ Z e j d ƒ Z d e f d „ ƒ YZ d e j f d „ ƒ YZ d S( s A parser for HTML and XHTML.iÿÿÿÿNs [&<]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]s
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# t HTMLParseErrorc B s# e Z d Z d d „ Z d „ Z RS( s& Exception raised for all parse errors.c C s3 | s t ‚ | | _ | d | _ | d | _ d S( Ni i ( t AssertionErrort msgt linenot offset( t selfR t position( ( s /usr/lib/python2.7/HTMLParser.pyt __init__= s
c C sW | j } | j d k r, | d | j } n | j d k rS | d | j d } n | S( Ns , at line %ds , column %di ( R R t NoneR ( R t result( ( s /usr/lib/python2.7/HTMLParser.pyt __str__C s N( NN( t __name__t
__module__t __doc__R R R ( ( ( s /usr/lib/python2.7/HTMLParser.pyR : s t
HTMLParserc B s
e 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
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 Z d „ Z RS( sÇ 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.
t scriptt stylec C s | j ƒ d S( s# Initialize and reset this instance.N( t reset( R ( ( s /usr/lib/python2.7/HTMLParser.pyR c s c C s8 d | _ d | _ t | _ d | _ t j j | ƒ d S( s1 Reset this instance. Loses all unprocessed data.t s ???N( t rawdatat lasttagt interesting_normalt interestingR t
cdata_elemt
markupbaset
ParserBaseR ( R ( ( s /usr/lib/python2.7/HTMLParser.pyR g s
c C s! | j | | _ | j d ƒ d S( 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').
i N( R t goahead( R t data( ( s /usr/lib/python2.7/HTMLParser.pyt feedo s c C s | j d ƒ d S( s Handle any buffered data.i N( R ( R ( ( s /usr/lib/python2.7/HTMLParser.pyt closex s c C s t | | j ƒ ƒ ‚ d S( N( R t getpos( R t message( ( s /usr/lib/python2.7/HTMLParser.pyt error| s c C s | j S( s) Return full source of start tag: '<...>'.( t _HTMLParser__starttag_text( R ( ( s /usr/lib/python2.7/HTMLParser.pyt get_starttag_text s c C s2 | j ƒ | _ t j d | j t j ƒ | _ d S( Ns ( t lowerR t ret compilet IR ( R t elem( ( s /usr/lib/python2.7/HTMLParser.pyt set_cdata_mode… s c C s t | _ d | _ d S( N( R R R R ( R ( ( s /usr/lib/python2.7/HTMLParser.pyt clear_cdata_mode‰ s c C s¨ | j } d } t | ƒ } x4| | k rQ| j j | | ƒ } | rT | j ƒ } n | j ra Pn | } | | k rŠ | j | | | !ƒ n | j | | ƒ } | | k r¬ Pn | j } | d | ƒ rct j
| | ƒ rè | j | ƒ } nµ | d | ƒ r | j | ƒ } n” | d | ƒ r*| j
| ƒ } ns | d | ƒ rK| j | ƒ } nR | d | ƒ rl| j | ƒ } n1 | d | k s‚| rœ| j d ƒ | d } n P| d k rN| s³Pn t j
| | ƒ rÈn}| d | ƒ r$| d | k r÷| j d ƒ qEt j
| | ƒ r qE| j | | d ƒ n!| d | ƒ r| } x8 d D]0 } | j | | d ƒ r@| t | ƒ 8} Pq@q@W| j | | d | !ƒ n¶ | d
| ƒ r¶| j | | d ƒ n | | | d !j ƒ d k rë| j | | d ƒ nZ | d | ƒ r| j | | d ƒ n3 | d | ƒ rE| j | | d ƒ t d ƒ ‚ n | } n | j | | ƒ } q | d | ƒ r3t j
| | ƒ } | rî| j ƒ d d !}
| j |
ƒ | j ƒ } | d | d ƒ sÖ| d } n | j | | ƒ } q qNd | | k r/| j | | | d !ƒ | j | | d ƒ } n Pq | d | ƒ r<t j
| | ƒ } | rº| j d ƒ }
| j |
ƒ | j ƒ } | d | d ƒ s¢| d } n | j | | ƒ } q n t j
| | ƒ } | r| rþ| j ƒ | | k rþ| j d ƒ n PqN| d | k r8| j d ƒ | j | | d ƒ } qNPq d s t d ƒ ‚ q W| r—| | k r—| j r—| j | | | !ƒ | j | | ƒ } n | | | _ d S( Ni t