1997-11-18 11:09:54 -04:00
|
|
|
# A parser for XML, using the derived class as static DTD.
|
1997-11-18 11:27:20 -04:00
|
|
|
# Author: Sjoerd Mullender.
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
import re
|
|
|
|
import string
|
|
|
|
|
|
|
|
|
|
|
|
# Regular expressions used for parsing
|
|
|
|
|
|
|
|
_S = '[ \t\r\n]+'
|
|
|
|
_opS = '[ \t\r\n]*'
|
|
|
|
_Name = '[a-zA-Z_:][-a-zA-Z0-9._:]*'
|
|
|
|
interesting = re.compile('[&<]')
|
|
|
|
incomplete = re.compile('&(' + _Name + '|#[0-9]*|#x[0-9a-fA-F]*)?|'
|
1998-03-26 17:13:24 -04:00
|
|
|
'<([a-zA-Z_:][^<>]*|'
|
|
|
|
'/([a-zA-Z_:][^<>]*)?|'
|
|
|
|
'![^<>]*|'
|
|
|
|
r'\?[^<>]*)?')
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
ref = re.compile('&(' + _Name + '|#[0-9]+|#x[0-9a-fA-F]+);?')
|
|
|
|
entityref = re.compile('&(?P<name>' + _Name + ')[^-a-zA-Z0-9._:]')
|
|
|
|
charref = re.compile('&#(?P<char>[0-9]+[^0-9]|x[0-9a-fA-F]+[^0-9a-fA-F])')
|
|
|
|
space = re.compile(_S)
|
|
|
|
newline = re.compile('\n')
|
|
|
|
|
|
|
|
starttagopen = re.compile('<' + _Name)
|
|
|
|
endtagopen = re.compile('</')
|
|
|
|
starttagend = re.compile(_opS + '(?P<slash>/?)>')
|
|
|
|
endbracket = re.compile('>')
|
|
|
|
tagfind = re.compile(_Name)
|
1998-01-29 10:55:24 -04:00
|
|
|
cdataopen = re.compile(r'<!\[CDATA\[')
|
|
|
|
cdataclose = re.compile(r'\]\]>')
|
|
|
|
doctype = re.compile('<!DOCTYPE' + _S + '(?P<name>' + _Name + ')' + _S)
|
1997-11-18 11:09:54 -04:00
|
|
|
special = re.compile('<!(?P<special>[^<>]*)>')
|
1998-01-29 10:55:24 -04:00
|
|
|
procopen = re.compile(r'<\?(?P<proc>' + _Name + ')' + _S)
|
|
|
|
procclose = re.compile(_opS + r'\?>')
|
1997-11-18 11:09:54 -04:00
|
|
|
commentopen = re.compile('<!--')
|
|
|
|
commentclose = re.compile('-->')
|
|
|
|
doubledash = re.compile('--')
|
|
|
|
attrfind = re.compile(
|
|
|
|
_S + '(?P<name>' + _Name + ')'
|
|
|
|
'(' + _opS + '=' + _opS +
|
|
|
|
'(?P<value>\'[^\']*\'|"[^"]*"|[-a-zA-Z0-9.:+*%?!()_#=~]+))')
|
|
|
|
|
|
|
|
|
|
|
|
# XML parser base class -- find tags and call handler functions.
|
|
|
|
# Usage: p = XMLParser(); p.feed(data); ...; p.close().
|
1997-11-18 11:27:20 -04:00
|
|
|
# The dtd is defined by deriving a class which defines methods with
|
|
|
|
# special names to handle tags: start_foo and end_foo to handle <foo>
|
|
|
|
# and </foo>, respectively. The data between tags is passed to the
|
|
|
|
# parser by calling self.handle_data() with some data as argument (the
|
|
|
|
# data may be split up in arbutrary chunks). Entity references are
|
|
|
|
# passed by calling self.handle_entityref() with the entity reference
|
|
|
|
# as argument.
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
class XMLParser:
|
|
|
|
|
|
|
|
# Interface -- initialize and reset this instance
|
|
|
|
def __init__(self, verbose=0):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.verbose = verbose
|
|
|
|
self.reset()
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Interface -- reset this instance. Loses all unprocessed data
|
|
|
|
def reset(self):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.rawdata = ''
|
|
|
|
self.stack = []
|
|
|
|
self.nomoretags = 0
|
|
|
|
self.literal = 0
|
|
|
|
self.lineno = 1
|
|
|
|
self.__at_start = 1
|
|
|
|
self.__seen_doctype = None
|
|
|
|
self.__seen_starttag = 0
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# For derived classes only -- enter literal mode (CDATA) till EOF
|
|
|
|
def setnomoretags(self):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.nomoretags = self.literal = 1
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# For derived classes only -- enter literal mode (CDATA)
|
|
|
|
def setliteral(self, *args):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.literal = 1
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Interface -- feed some data to the parser. Call this as
|
|
|
|
# often as you want, with as little or as much text as you
|
|
|
|
# want (may include '\n'). (This just saves the text, all the
|
|
|
|
# processing is done by goahead().)
|
|
|
|
def feed(self, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.rawdata = self.rawdata + data
|
|
|
|
self.goahead(0)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Interface -- handle the remaining data
|
|
|
|
def close(self):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.goahead(1)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Interface -- translate references
|
|
|
|
def translate_references(self, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
newdata = []
|
|
|
|
i = 0
|
|
|
|
while 1:
|
|
|
|
res = ref.search(data, i)
|
|
|
|
if res is None:
|
|
|
|
newdata.append(data[i:])
|
|
|
|
return string.join(newdata, '')
|
|
|
|
if data[res.end(0) - 1] != ';':
|
|
|
|
self.syntax_error("`;' missing after entity/char reference")
|
|
|
|
newdata.append(data[i:res.start(0)])
|
|
|
|
str = res.group(1)
|
|
|
|
if str[0] == '#':
|
|
|
|
if str[1] == 'x':
|
|
|
|
newdata.append(chr(string.atoi(str[2:], 16)))
|
|
|
|
else:
|
|
|
|
newdata.append(chr(string.atoi(str[1:])))
|
|
|
|
else:
|
|
|
|
try:
|
|
|
|
newdata.append(self.entitydefs[str])
|
|
|
|
except KeyError:
|
|
|
|
# can't do it, so keep the entity ref in
|
|
|
|
newdata.append('&' + str + ';')
|
|
|
|
i = res.end(0)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Internal -- handle data as far as reasonable. May leave state
|
|
|
|
# and data to be processed by a subsequent call. If 'end' is
|
|
|
|
# true, force handling all data as if followed by EOF marker.
|
|
|
|
def goahead(self, end):
|
1998-03-26 17:13:24 -04:00
|
|
|
rawdata = self.rawdata
|
|
|
|
i = 0
|
|
|
|
n = len(rawdata)
|
|
|
|
while i < n:
|
|
|
|
if i > 0:
|
|
|
|
self.__at_start = 0
|
|
|
|
if self.nomoretags:
|
|
|
|
data = rawdata[i:n]
|
|
|
|
self.handle_data(data)
|
|
|
|
self.lineno = self.lineno + string.count(data, '\n')
|
|
|
|
i = n
|
|
|
|
break
|
|
|
|
res = interesting.search(rawdata, i)
|
|
|
|
if res:
|
|
|
|
j = res.start(0)
|
|
|
|
else:
|
|
|
|
j = n
|
|
|
|
if i < j:
|
|
|
|
self.__at_start = 0
|
|
|
|
data = rawdata[i:j]
|
|
|
|
self.handle_data(data)
|
|
|
|
self.lineno = self.lineno + string.count(data, '\n')
|
|
|
|
i = j
|
|
|
|
if i == n: break
|
|
|
|
if rawdata[i] == '<':
|
|
|
|
if starttagopen.match(rawdata, i):
|
|
|
|
if self.literal:
|
|
|
|
data = rawdata[i]
|
|
|
|
self.handle_data(data)
|
|
|
|
self.lineno = self.lineno + string.count(data, '\n')
|
|
|
|
i = i+1
|
|
|
|
continue
|
|
|
|
k = self.parse_starttag(i)
|
|
|
|
if k < 0: break
|
|
|
|
self.__seen_starttag = 1
|
|
|
|
self.lineno = self.lineno + string.count(rawdata[i:k], '\n')
|
|
|
|
i = k
|
|
|
|
continue
|
|
|
|
if endtagopen.match(rawdata, i):
|
|
|
|
k = self.parse_endtag(i)
|
|
|
|
if k < 0: break
|
|
|
|
self.lineno = self.lineno + string.count(rawdata[i:k], '\n')
|
|
|
|
i = k
|
|
|
|
self.literal = 0
|
|
|
|
continue
|
|
|
|
if commentopen.match(rawdata, i):
|
|
|
|
if self.literal:
|
|
|
|
data = rawdata[i]
|
|
|
|
self.handle_data(data)
|
|
|
|
self.lineno = self.lineno + string.count(data, '\n')
|
|
|
|
i = i+1
|
|
|
|
continue
|
|
|
|
k = self.parse_comment(i)
|
|
|
|
if k < 0: break
|
|
|
|
self.lineno = self.lineno + string.count(rawdata[i:k], '\n')
|
|
|
|
i = k
|
|
|
|
continue
|
|
|
|
if cdataopen.match(rawdata, i):
|
|
|
|
k = self.parse_cdata(i)
|
|
|
|
if k < 0: break
|
|
|
|
self.lineno = self.lineno + string.count(rawdata[i:i], '\n')
|
|
|
|
i = k
|
|
|
|
continue
|
|
|
|
res = procopen.match(rawdata, i)
|
|
|
|
if res:
|
|
|
|
k = self.parse_proc(i)
|
|
|
|
if k < 0: break
|
|
|
|
self.lineno = self.lineno + string.count(rawdata[i:k], '\n')
|
|
|
|
i = k
|
|
|
|
continue
|
|
|
|
res = doctype.match(rawdata, i)
|
|
|
|
if res:
|
|
|
|
if self.literal:
|
|
|
|
data = rawdata[i]
|
|
|
|
self.handle_data(data)
|
|
|
|
self.lineno = self.lineno + string.count(data, '\n')
|
|
|
|
i = i+1
|
|
|
|
continue
|
|
|
|
if self.__seen_doctype:
|
|
|
|
self.syntax_error('multiple DOCTYPE elements')
|
|
|
|
if self.__seen_starttag:
|
|
|
|
self.syntax_error('DOCTYPE not at beginning of document')
|
|
|
|
k = self.parse_doctype(res)
|
|
|
|
if k < 0: break
|
|
|
|
self.__seen_doctype = res.group('name')
|
|
|
|
self.lineno = self.lineno + string.count(rawdata[i:k], '\n')
|
|
|
|
i = k
|
|
|
|
continue
|
|
|
|
res = special.match(rawdata, i)
|
|
|
|
if res:
|
|
|
|
if self.literal:
|
|
|
|
data = rawdata[i]
|
|
|
|
self.handle_data(data)
|
|
|
|
self.lineno = self.lineno + string.count(data, '\n')
|
|
|
|
i = i+1
|
|
|
|
continue
|
|
|
|
self.handle_special(res.group('special'))
|
|
|
|
self.lineno = self.lineno + string.count(res.group(0), '\n')
|
|
|
|
i = res.end(0)
|
|
|
|
continue
|
|
|
|
elif rawdata[i] == '&':
|
|
|
|
res = charref.match(rawdata, i)
|
|
|
|
if res is not None:
|
|
|
|
i = res.end(0)
|
|
|
|
if rawdata[i-1] != ';':
|
|
|
|
self.syntax_error("`;' missing in charref")
|
|
|
|
i = i-1
|
|
|
|
self.handle_charref(res.group('char')[:-1])
|
|
|
|
self.lineno = self.lineno + string.count(res.group(0), '\n')
|
|
|
|
continue
|
|
|
|
res = entityref.match(rawdata, i)
|
|
|
|
if res is not None:
|
|
|
|
i = res.end(0)
|
|
|
|
if rawdata[i-1] != ';':
|
|
|
|
self.syntax_error("`;' missing in entityref")
|
|
|
|
i = i-1
|
|
|
|
self.handle_entityref(res.group('name'))
|
|
|
|
self.lineno = self.lineno + string.count(res.group(0), '\n')
|
|
|
|
continue
|
|
|
|
else:
|
|
|
|
raise RuntimeError, 'neither < nor & ??'
|
|
|
|
# We get here only if incomplete matches but
|
|
|
|
# nothing else
|
|
|
|
res = incomplete.match(rawdata, i)
|
|
|
|
if not res:
|
|
|
|
data = rawdata[i]
|
|
|
|
self.handle_data(data)
|
|
|
|
self.lineno = self.lineno + string.count(data, '\n')
|
|
|
|
i = i+1
|
|
|
|
continue
|
|
|
|
j = res.end(0)
|
|
|
|
if j == n:
|
|
|
|
break # Really incomplete
|
|
|
|
self.syntax_error("bogus `<' or `&'")
|
|
|
|
data = res.group(0)
|
|
|
|
self.handle_data(data)
|
|
|
|
self.lineno = self.lineno + string.count(data, '\n')
|
|
|
|
i = j
|
|
|
|
# end while
|
|
|
|
if end and i < n:
|
|
|
|
data = rawdata[i:n]
|
|
|
|
self.handle_data(data)
|
|
|
|
self.lineno = self.lineno + string.count(data, '\n')
|
|
|
|
i = n
|
|
|
|
self.rawdata = rawdata[i:]
|
|
|
|
if end:
|
|
|
|
if self.stack:
|
|
|
|
self.syntax_error('missing end tags')
|
|
|
|
while self.stack:
|
|
|
|
self.finish_endtag(self.stack[-1])
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Internal -- parse comment, return length or -1 if not terminated
|
|
|
|
def parse_comment(self, i):
|
1998-03-26 17:13:24 -04:00
|
|
|
rawdata = self.rawdata
|
|
|
|
if rawdata[i:i+4] <> '<!--':
|
|
|
|
raise RuntimeError, 'unexpected call to handle_comment'
|
|
|
|
res = commentclose.search(rawdata, i+4)
|
|
|
|
if not res:
|
|
|
|
return -1
|
|
|
|
# doubledash search will succeed because it's a subset of commentclose
|
|
|
|
if doubledash.search(rawdata, i+4).start(0) < res.start(0):
|
|
|
|
self.syntax_error("`--' inside comment")
|
|
|
|
self.handle_comment(rawdata[i+4: res.start(0)])
|
|
|
|
return res.end(0)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
1998-01-29 10:55:24 -04:00
|
|
|
# Internal -- handle DOCTYPE tag, return length or -1 if not terminated
|
|
|
|
def parse_doctype(self, res):
|
1998-03-26 17:13:24 -04:00
|
|
|
rawdata = self.rawdata
|
|
|
|
n = len(rawdata)
|
|
|
|
name = res.group('name')
|
|
|
|
j = k = res.end(0)
|
|
|
|
level = 0
|
|
|
|
while k < n:
|
|
|
|
c = rawdata[k]
|
|
|
|
if c == '<':
|
|
|
|
level = level + 1
|
|
|
|
elif c == '>':
|
|
|
|
if level == 0:
|
|
|
|
self.handle_doctype(name, rawdata[j:k])
|
|
|
|
return k+1
|
|
|
|
level = level - 1
|
|
|
|
k = k+1
|
|
|
|
return -1
|
1998-01-29 10:55:24 -04:00
|
|
|
|
|
|
|
# Internal -- handle CDATA tag, return length or -1 if not terminated
|
1997-11-18 11:09:54 -04:00
|
|
|
def parse_cdata(self, i):
|
1998-03-26 17:13:24 -04:00
|
|
|
rawdata = self.rawdata
|
|
|
|
if rawdata[i:i+9] <> '<![CDATA[':
|
|
|
|
raise RuntimeError, 'unexpected call to handle_cdata'
|
|
|
|
res = cdataclose.search(rawdata, i+9)
|
|
|
|
if not res:
|
|
|
|
return -1
|
|
|
|
self.handle_cdata(rawdata[i+9:res.start(0)])
|
|
|
|
return res.end(0)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
1998-01-29 10:55:24 -04:00
|
|
|
__xml_attributes = {'version': '1.0', 'standalone': 'no', 'encoding': None}
|
|
|
|
# Internal -- handle a processing instruction tag
|
|
|
|
def parse_proc(self, i):
|
1998-03-26 17:13:24 -04:00
|
|
|
rawdata = self.rawdata
|
|
|
|
end = procclose.search(rawdata, i)
|
|
|
|
if not end:
|
|
|
|
return -1
|
|
|
|
j = end.start(0)
|
|
|
|
res = tagfind.match(rawdata, i+2)
|
|
|
|
if not res:
|
|
|
|
raise RuntimeError, 'unexpected call to parse_proc'
|
|
|
|
k = res.end(0)
|
|
|
|
name = res.group(0)
|
|
|
|
if name == 'xml':
|
|
|
|
if self.__at_start:
|
|
|
|
attrdict, k = self.parse_attributes('xml', k, j,
|
|
|
|
self.__xml_attributes)
|
|
|
|
if k != j:
|
|
|
|
self.syntax_error('garbage at end of <?xml?>')
|
|
|
|
if attrdict['version'] != '1.0':
|
|
|
|
self.syntax_error('only XML version 1.0 supported')
|
|
|
|
self.handle_xml(attrdict.get('encoding', None),
|
|
|
|
attrdict['standalone'])
|
|
|
|
return end.end(0)
|
|
|
|
else:
|
|
|
|
self.syntax_error("<?xml?> tag not at start of document")
|
|
|
|
self.handle_proc(name, rawdata[k:j])
|
|
|
|
return end.end(0)
|
1998-01-29 10:55:24 -04:00
|
|
|
|
|
|
|
# Internal -- parse attributes between i and j
|
|
|
|
def parse_attributes(self, tag, k, j, attributes = None):
|
1998-03-26 17:13:24 -04:00
|
|
|
rawdata = self.rawdata
|
|
|
|
# Now parse the data between k and j into a tag and attrs
|
|
|
|
attrdict = {}
|
|
|
|
try:
|
|
|
|
# convert attributes list to dictionary
|
|
|
|
d = {}
|
|
|
|
for a in attributes:
|
|
|
|
d[a] = None
|
|
|
|
attributes = d
|
|
|
|
except TypeError:
|
|
|
|
pass
|
|
|
|
while k < j:
|
|
|
|
res = attrfind.match(rawdata, k)
|
|
|
|
if not res: break
|
|
|
|
attrname, attrvalue = res.group('name', 'value')
|
|
|
|
if attrvalue is None:
|
|
|
|
self.syntax_error('no attribute value specified')
|
|
|
|
attrvalue = attrname
|
|
|
|
elif attrvalue[:1] == "'" == attrvalue[-1:] or \
|
|
|
|
attrvalue[:1] == '"' == attrvalue[-1:]:
|
|
|
|
attrvalue = attrvalue[1:-1]
|
|
|
|
else:
|
|
|
|
self.syntax_error('attribute value not quoted')
|
|
|
|
if attributes is not None and not attributes.has_key(attrname):
|
|
|
|
self.syntax_error('unknown attribute %s of element %s' %
|
|
|
|
(attrname, tag))
|
|
|
|
if attrdict.has_key(attrname):
|
|
|
|
self.syntax_error('attribute specified twice')
|
|
|
|
attrdict[attrname] = self.translate_references(attrvalue)
|
|
|
|
k = res.end(0)
|
|
|
|
if attributes is not None:
|
|
|
|
# fill in with default attributes
|
|
|
|
for key, val in attributes.items():
|
|
|
|
if val is not None and not attrdict.has_key(key):
|
|
|
|
attrdict[key] = val
|
|
|
|
return attrdict, k
|
1998-01-29 10:55:24 -04:00
|
|
|
|
|
|
|
# Internal -- handle starttag, return length or -1 if not terminated
|
|
|
|
def parse_starttag(self, i):
|
1998-03-26 17:13:24 -04:00
|
|
|
rawdata = self.rawdata
|
|
|
|
# i points to start of tag
|
|
|
|
end = endbracket.search(rawdata, i+1)
|
|
|
|
if not end:
|
|
|
|
return -1
|
|
|
|
j = end.start(0)
|
|
|
|
res = tagfind.match(rawdata, i+1)
|
|
|
|
if not res:
|
|
|
|
raise RuntimeError, 'unexpected call to parse_starttag'
|
|
|
|
k = res.end(0)
|
|
|
|
tag = res.group(0)
|
|
|
|
if not self.__seen_starttag and self.__seen_doctype:
|
|
|
|
if tag != self.__seen_doctype:
|
|
|
|
self.syntax_error('starttag does not match DOCTYPE')
|
|
|
|
if hasattr(self, tag + '_attributes'):
|
|
|
|
attributes = getattr(self, tag + '_attributes')
|
|
|
|
else:
|
|
|
|
attributes = None
|
|
|
|
attrdict, k = self.parse_attributes(tag, k, j, attributes)
|
|
|
|
res = starttagend.match(rawdata, k)
|
|
|
|
if not res:
|
|
|
|
self.syntax_error('garbage in start tag')
|
|
|
|
self.finish_starttag(tag, attrdict)
|
|
|
|
if res and res.group('slash') == '/':
|
|
|
|
self.finish_endtag(tag)
|
|
|
|
return end.end(0)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Internal -- parse endtag
|
|
|
|
def parse_endtag(self, i):
|
1998-03-26 17:13:24 -04:00
|
|
|
rawdata = self.rawdata
|
|
|
|
end = endbracket.search(rawdata, i+1)
|
|
|
|
if not end:
|
|
|
|
return -1
|
|
|
|
res = tagfind.match(rawdata, i+2)
|
|
|
|
if not res:
|
|
|
|
self.syntax_error('no name specified in end tag')
|
|
|
|
tag = ''
|
|
|
|
k = i+2
|
|
|
|
else:
|
|
|
|
tag = res.group(0)
|
|
|
|
k = res.end(0)
|
|
|
|
if k != end.start(0):
|
|
|
|
# check that there is only white space at end of tag
|
|
|
|
res = space.match(rawdata, k)
|
|
|
|
if res is None or res.end(0) != end.start(0):
|
|
|
|
self.syntax_error('garbage in end tag')
|
|
|
|
self.finish_endtag(tag)
|
|
|
|
return end.end(0)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Internal -- finish processing of start tag
|
|
|
|
# Return -1 for unknown tag, 1 for balanced tag
|
|
|
|
def finish_starttag(self, tag, attrs):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.stack.append(tag)
|
|
|
|
try:
|
|
|
|
method = getattr(self, 'start_' + tag)
|
|
|
|
except AttributeError:
|
|
|
|
self.unknown_starttag(tag, attrs)
|
|
|
|
return -1
|
|
|
|
else:
|
|
|
|
self.handle_starttag(tag, method, attrs)
|
|
|
|
return 1
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Internal -- finish processing of end tag
|
|
|
|
def finish_endtag(self, tag):
|
1998-03-26 17:13:24 -04:00
|
|
|
if not tag:
|
|
|
|
self.syntax_error('name-less end tag')
|
|
|
|
found = len(self.stack) - 1
|
|
|
|
if found < 0:
|
|
|
|
self.unknown_endtag(tag)
|
|
|
|
return
|
|
|
|
else:
|
|
|
|
if tag not in self.stack:
|
|
|
|
self.syntax_error('unopened end tag')
|
|
|
|
try:
|
|
|
|
method = getattr(self, 'end_' + tag)
|
|
|
|
except AttributeError:
|
|
|
|
self.unknown_endtag(tag)
|
|
|
|
return
|
|
|
|
found = len(self.stack)
|
|
|
|
for i in range(found):
|
|
|
|
if self.stack[i] == tag:
|
|
|
|
found = i
|
|
|
|
while len(self.stack) > found:
|
|
|
|
if found < len(self.stack) - 1:
|
|
|
|
self.syntax_error('missing close tag for %s' % self.stack[-1])
|
|
|
|
tag = self.stack[-1]
|
|
|
|
try:
|
|
|
|
method = getattr(self, 'end_' + tag)
|
|
|
|
except AttributeError:
|
|
|
|
method = None
|
|
|
|
if method:
|
|
|
|
self.handle_endtag(tag, method)
|
|
|
|
else:
|
|
|
|
self.unknown_endtag(tag)
|
|
|
|
del self.stack[-1]
|
1997-11-18 11:09:54 -04:00
|
|
|
|
1998-01-29 10:55:24 -04:00
|
|
|
# Overridable -- handle xml processing instruction
|
|
|
|
def handle_xml(self, encoding, standalone):
|
1998-03-26 17:13:24 -04:00
|
|
|
pass
|
1998-01-29 10:55:24 -04:00
|
|
|
|
|
|
|
# Overridable -- handle DOCTYPE
|
|
|
|
def handle_doctype(self, tag, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
pass
|
1998-01-29 10:55:24 -04:00
|
|
|
|
1997-11-18 11:09:54 -04:00
|
|
|
# Overridable -- handle start tag
|
|
|
|
def handle_starttag(self, tag, method, attrs):
|
1998-03-26 17:13:24 -04:00
|
|
|
method(attrs)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Overridable -- handle end tag
|
|
|
|
def handle_endtag(self, tag, method):
|
1998-03-26 17:13:24 -04:00
|
|
|
method()
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Example -- handle character reference, no need to override
|
|
|
|
def handle_charref(self, name):
|
1998-03-26 17:13:24 -04:00
|
|
|
try:
|
|
|
|
if name[0] == 'x':
|
|
|
|
n = string.atoi(name[1:], 16)
|
|
|
|
else:
|
|
|
|
n = string.atoi(name)
|
|
|
|
except string.atoi_error:
|
|
|
|
self.unknown_charref(name)
|
|
|
|
return
|
|
|
|
if not 0 <= n <= 255:
|
|
|
|
self.unknown_charref(name)
|
|
|
|
return
|
|
|
|
self.handle_data(chr(n))
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Definition of entities -- derived classes may override
|
1998-01-29 10:55:24 -04:00
|
|
|
entitydefs = {'lt': '<', 'gt': '>', 'amp': '&', 'quot': '"', 'apos': "'"}
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Example -- handle entity reference, no need to override
|
|
|
|
def handle_entityref(self, name):
|
1998-03-26 17:13:24 -04:00
|
|
|
table = self.entitydefs
|
|
|
|
if table.has_key(name):
|
|
|
|
self.handle_data(table[name])
|
|
|
|
else:
|
|
|
|
self.unknown_entityref(name)
|
|
|
|
return
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Example -- handle data, should be overridden
|
|
|
|
def handle_data(self, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
pass
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Example -- handle cdata, could be overridden
|
|
|
|
def handle_cdata(self, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
pass
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Example -- handle comment, could be overridden
|
|
|
|
def handle_comment(self, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
pass
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Example -- handle processing instructions, could be overridden
|
|
|
|
def handle_proc(self, name, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
pass
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Example -- handle special instructions, could be overridden
|
|
|
|
def handle_special(self, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
pass
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# Example -- handle relatively harmless syntax errors, could be overridden
|
1998-01-29 10:55:24 -04:00
|
|
|
def syntax_error(self, message):
|
1998-03-26 17:13:24 -04:00
|
|
|
raise RuntimeError, 'Syntax error at line %d: %s' % (self.lineno, message)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
# To be overridden -- handlers for unknown objects
|
|
|
|
def unknown_starttag(self, tag, attrs): pass
|
|
|
|
def unknown_endtag(self, tag): pass
|
|
|
|
def unknown_charref(self, ref): pass
|
|
|
|
def unknown_entityref(self, ref): pass
|
|
|
|
|
|
|
|
|
|
|
|
class TestXMLParser(XMLParser):
|
|
|
|
|
|
|
|
def __init__(self, verbose=0):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.testdata = ""
|
|
|
|
XMLParser.__init__(self, verbose)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
1998-01-29 10:55:24 -04:00
|
|
|
def handle_xml(self, encoding, standalone):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.flush()
|
|
|
|
print 'xml: encoding =',encoding,'standalone =',standalone
|
1998-01-29 10:55:24 -04:00
|
|
|
|
|
|
|
def handle_doctype(self, tag, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.flush()
|
|
|
|
print 'DOCTYPE:',tag, `data`
|
1998-01-29 10:55:24 -04:00
|
|
|
|
1997-11-18 11:09:54 -04:00
|
|
|
def handle_data(self, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.testdata = self.testdata + data
|
|
|
|
if len(`self.testdata`) >= 70:
|
|
|
|
self.flush()
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def flush(self):
|
1998-03-26 17:13:24 -04:00
|
|
|
data = self.testdata
|
|
|
|
if data:
|
|
|
|
self.testdata = ""
|
|
|
|
print 'data:', `data`
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def handle_cdata(self, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.flush()
|
|
|
|
print 'cdata:', `data`
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def handle_proc(self, name, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.flush()
|
|
|
|
print 'processing:',name,`data`
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def handle_special(self, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.flush()
|
|
|
|
print 'special:',`data`
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def handle_comment(self, data):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.flush()
|
|
|
|
r = `data`
|
|
|
|
if len(r) > 68:
|
|
|
|
r = r[:32] + '...' + r[-32:]
|
|
|
|
print 'comment:', r
|
1997-11-18 11:09:54 -04:00
|
|
|
|
1998-01-29 10:55:24 -04:00
|
|
|
def syntax_error(self, message):
|
1998-03-26 17:13:24 -04:00
|
|
|
print 'error at line %d:' % self.lineno, message
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def unknown_starttag(self, tag, attrs):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.flush()
|
|
|
|
if not attrs:
|
|
|
|
print 'start tag: <' + tag + '>'
|
|
|
|
else:
|
|
|
|
print 'start tag: <' + tag,
|
|
|
|
for name, value in attrs.items():
|
|
|
|
print name + '=' + '"' + value + '"',
|
|
|
|
print '>'
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def unknown_endtag(self, tag):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.flush()
|
|
|
|
print 'end tag: </' + tag + '>'
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def unknown_entityref(self, ref):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.flush()
|
|
|
|
print '*** unknown entity ref: &' + ref + ';'
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def unknown_charref(self, ref):
|
1998-03-26 17:13:24 -04:00
|
|
|
self.flush()
|
|
|
|
print '*** unknown char ref: &#' + ref + ';'
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def close(self):
|
1998-03-26 17:13:24 -04:00
|
|
|
XMLParser.close(self)
|
|
|
|
self.flush()
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
def test(args = None):
|
|
|
|
import sys
|
|
|
|
|
|
|
|
if not args:
|
1998-03-26 17:13:24 -04:00
|
|
|
args = sys.argv[1:]
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
if args and args[0] == '-s':
|
1998-03-26 17:13:24 -04:00
|
|
|
args = args[1:]
|
|
|
|
klass = XMLParser
|
1997-11-18 11:09:54 -04:00
|
|
|
else:
|
1998-03-26 17:13:24 -04:00
|
|
|
klass = TestXMLParser
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
if args:
|
1998-03-26 17:13:24 -04:00
|
|
|
file = args[0]
|
1997-11-18 11:09:54 -04:00
|
|
|
else:
|
1998-03-26 17:13:24 -04:00
|
|
|
file = 'test.xml'
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
if file == '-':
|
1998-03-26 17:13:24 -04:00
|
|
|
f = sys.stdin
|
1997-11-18 11:09:54 -04:00
|
|
|
else:
|
1998-03-26 17:13:24 -04:00
|
|
|
try:
|
|
|
|
f = open(file, 'r')
|
|
|
|
except IOError, msg:
|
|
|
|
print file, ":", msg
|
|
|
|
sys.exit(1)
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
data = f.read()
|
|
|
|
if f is not sys.stdin:
|
1998-03-26 17:13:24 -04:00
|
|
|
f.close()
|
1997-11-18 11:09:54 -04:00
|
|
|
|
|
|
|
x = klass()
|
|
|
|
for c in data:
|
1998-03-26 17:13:24 -04:00
|
|
|
x.feed(c)
|
1997-11-18 11:09:54 -04:00
|
|
|
x.close()
|
|
|
|
|
|
|
|
|
|
|
|
if __name__ == '__main__':
|
|
|
|
test()
|