def fromXml (self, s, *args):
import libxml2
input_source = libxml2.inputBuffer(Stream (s))
reader = input_source.newTextReader("")
nodetypes = ["none", "element", "attribute", "text", "cdata", "entref", "entity",
"procinst", "comment", "document", "doctype", "docfragment", "notation",
"whitespace", "preservewhitespace", "endelement", "endentity", "xmldecl"]
def _read():
if not reader.Read():
return self.eos
return [[reader.Name(), nodetypes[reader.NodeType()]], Lazy (_read)]
return _read()
def tst_reader(s):
f = str_io(s)
input = libxml2.inputBuffer(f)
reader = input.newTextReader("tst")
res = ""
while reader.Read():
res=res + "%s (%s) [%s] %d\n" % (reader.NodeType(),reader.Name(),
reader.Value(), reader.IsEmptyElement())
if reader.NodeType() == 1: # Element
while reader.MoveToNextAttribute():
res = res + "-- %s (%s) [%s]\n" % (reader.NodeType(),
reader.Name(),reader.Value())
return res
def process_introspection_data(data):
method_map = {}
XMLREADER_START_ELEMENT_NODE_TYPE = 1
XMLREADER_END_ELEMENT_NODE_TYPE = 15
stream = cStringIO.StringIO(data.encode('utf-8'))
input_source = libxml2.inputBuffer(stream)
reader = input_source.newTextReader("urn:introspect")
ret = reader.Read()
current_iface = None
current_method = None
current_sigstr = ''
while ret == 1:
name = reader.LocalName()
if reader.NodeType() == XMLREADER_START_ELEMENT_NODE_TYPE:
if (not current_iface and not current_method and name == 'interface'):
current_iface = reader.GetAttribute('name')
elif (current_iface and not current_method and name == 'method'):
current_method = reader.GetAttribute('name')
if reader.IsEmptyElement():
method_map[current_iface + '.' + current_method] = ''
current_method = None
current_sigstr = ''
elif (current_iface and current_method and name == 'arg'):
direction = reader.GetAttribute('direction')
if not direction or direction == 'in':
current_sigstr = current_sigstr + reader.GetAttribute('type')
elif reader.NodeType() == XMLREADER_END_ELEMENT_NODE_TYPE:
if (current_iface and not current_method and name == 'interface'):
current_iface = None
if (current_iface and current_method and name == 'method'):
method_map[current_iface + '.' + current_method] = current_sigstr
current_method = None
current_sigstr = ''
ret = reader.Read()
if ret != 0:
raise exceptions.IntrospectionParserException(data)
return method_map
def xmlvalidate(log):
import libxml2
from StringIO import StringIO
from random import random
prefix="...%s..." % str(random()).replace('0.','')
msg=[]
libxml2.registerErrorHandler(lambda msg,str: msg.append(str), msg)
input = libxml2.inputBuffer(StringIO(xmlEncoding.asUTF8(aString)))
reader = input.newTextReader(prefix)
reader.SetParserProp(libxml2.PARSER_VALIDATE, 1)
ret = reader.Read()
while ret == 1: ret = reader.Read()
msg=''.join(msg)
for line in msg.splitlines():
if line.startswith(prefix): log(line.split(':',4)[-1].strip())
str_io = io.StringIO
docstr="""<?xml version='1.0'?>
<!DOCTYPE doc [
<!ENTITY tst "<p>test</p>">
]>
<doc>&tst;</doc>"""
# Memory debug specific
libxml2.debugMemory(1)
#
# First test, normal don't substitute entities.
#
f = str_io(docstr)
input = libxml2.inputBuffer(f)
reader = input.newTextReader("test_noent")
ret = reader.Read()
if ret != 1:
print("Error reading to root")
sys.exit(1)
if reader.Name() == "doc" or reader.NodeType() == 10:
ret = reader.Read()
if ret != 1:
print("Error reading to root")
sys.exit(1)
if reader.Name() != "doc" or reader.NodeType() != 1:
print("test_normal: Error reading the root element")
sys.exit(1)
ret = reader.Read()
if ret != 1:
# Parse the Relax NG Schemas
#
rngp = libxml2.relaxNGNewMemParserCtxt(schema, len(schema))
rngs = rngp.relaxNGParse()
del rngp
#
# Parse and validate the correct document
#
docstr="""<foo>
<label>some text</label>
<item>100</item>
</foo>"""
f = StringIO.StringIO(docstr)
input = libxml2.inputBuffer(f)
reader = input.newTextReader("correct")
reader.RelaxNGSetSchema(rngs)
ret = reader.Read()
while ret == 1:
ret = reader.Read()
if ret != 0:
print "Error parsing the document"
sys.exit(1)
if reader.IsValid() != 1:
print "Document failed to validate"
sys.exit(1)
#
def parse(self, source):
self.__parsing = 1
try:
# prepare source and create reader
source = saxutils.prepare_input_source(source)
input = libxml2.inputBuffer(source.getByteStream())
reader = input.newTextReader(source.getSystemId())
reader.SetErrorHandler(self._errorHandler, None)
# configure reader
reader.SetParserProp(libxml2.PARSER_LOADDTD, 1)
reader.SetParserProp(libxml2.PARSER_DEFAULTATTRS, 1)
reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES, 0)
reader.SetParserProp(libxml2.PARSER_VALIDATE, 0)
# we reuse attribute maps (for a slight performance gain)
attributesImpl = xmlreader.AttributesImpl({})
# start loop
self._cont_handler.startDocument()
while 1:
r = reader.Read()
# check for errors
if r == 1:
pass
if self.__errors is not None:
self._reportErrors(0)
elif r == 0:
if self.__errors is not None:
self._reportErrors(0)
break # end of parse
else:
if self.__errors is not None:
self._reportErrors(1)
else:
self._err_handler.fatalError(
SAXException("Read failed (no details available)"))
break # fatal parse error
# get node type
nodeType = reader.NodeType()
# Element
if nodeType == 1:
eltName = reader.Name()
attributesImpl._attrs = attrs = {}
while reader.MoveToNextAttribute():
attName = reader.Name()
attrs[attName] = reader.Value()
reader.MoveToElement()
self._cont_handler.startElement(eltName, attributesImpl)
if reader.IsEmptyElement():
self._cont_handler.endElement(eltName)
# EndElement
elif nodeType == 15:
self._cont_handler.endElement(reader.Name())
# Text
elif nodeType == 3:
self._cont_handler.characters(reader.Value())
# SignificantWhitespace
elif nodeType == 14:
self._cont_handler.characters(reader.Value())
# EntityReference
elif nodeType == 5:
# Treating entity as such
self._cont_handler.entity(reader.Name())
elif nodeType == 10:
# We parse the doctype with a SAX parser
nodeText = str(reader.CurrentNode())
entityDeclParser = libxml2.createPushParser(
self._cont_handler, nodeText, len(nodeText), "doctype")
entityDeclParser.parseChunk("", 0, 1)
pass
# Ignore all other node types
if r == 0:
self._cont_handler.endDocument()
reader.Close()
finally:
self.__parsing = 0
# this tests the basic APIs of the XmlTextReader interface
#
import libxml2
import sys
try:
import StringIO
str_io = StringIO.StringIO
except:
import io
str_io = io.StringIO
# Memory debug specific
libxml2.debugMemory(1)
f = str_io("""<a><b b1="b1"/><c>content of c</c></a>""")
input = libxml2.inputBuffer(f)
reader = input.newTextReader("test1")
ret = reader.Read()
if ret != 1:
print("test1: Error reading to first element")
sys.exit(1)
if reader.Name() != "a" or reader.IsEmptyElement() != 0 or \
reader.NodeType() != 1 or reader.HasAttributes() != 0:
print("test1: Error reading the first element")
sys.exit(1)
ret = reader.Read()
if ret != 1:
print("test1: Error reading to second element")
sys.exit(1)
if reader.Name() != "b" or reader.IsEmptyElement() != 1 or \
reader.NodeType() != 1 or reader.HasAttributes() != 1:
#!/usr/bin/python -u
import sys
import libxml2
try:
import StringIO
str_io = StringIO.StringIO
except:
import io
str_io = io.StringIO
# Memory debug specific
libxml2.debugMemory(1)
i = 0
while i < 5000:
f = str_io("foobar")
buf = libxml2.inputBuffer(f)
i = i + 1
del f
del buf
# Memory debug specific
libxml2.cleanupParser()
if libxml2.debugMemory(1) == 0:
print("OK")
else:
print("Memory leak %d bytes" % (libxml2.debugMemory(1)))
libxml2.dumpMemory()
def parse(self, source):
self.__parsing = 1
try:
# prepare source and create reader
if type(source) in StringTypes:
reader = libxml2.newTextReaderFilename(source)
else:
source = saxutils.prepare_input_source(source)
input = libxml2.inputBuffer(source.getByteStream())
reader = input.newTextReader(source.getSystemId())
reader.SetErrorHandler(self._errorHandler, None)
# configure reader
reader.SetParserProp(libxml2.PARSER_LOADDTD, 1)
reader.SetParserProp(libxml2.PARSER_DEFAULTATTRS, 1)
reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES, 1)
reader.SetParserProp(libxml2.PARSER_VALIDATE, self.__validate)
# we reuse attribute maps (for a slight performance gain)
if self.__ns:
attributesNSImpl = xmlreader.AttributesNSImpl({}, {})
else:
attributesImpl = xmlreader.AttributesImpl({})
# prefixes to pop (for endPrefixMapping)
prefixes = []
# start loop
self._cont_handler.startDocument()
while 1:
r = reader.Read()
# check for errors
if r == 1:
if not self.__errors is None:
self._reportErrors(0)
elif r == 0:
if not self.__errors is None:
self._reportErrors(0)
break # end of parse
else:
if not self.__errors is None:
self._reportErrors(1)
else:
self._err_handler.fatalError(\
SAXException("Read failed (no details available)"))
break # fatal parse error
# get node type
nodeType = reader.NodeType()
# Element
if nodeType == 1:
if self.__ns:
eltName = (_d(reader.NamespaceUri()),\
_d(reader.LocalName()))
eltQName = _d(reader.Name())
attributesNSImpl._attrs = attrs = {}
attributesNSImpl._qnames = qnames = {}
newPrefixes = []
while reader.MoveToNextAttribute():
qname = _d(reader.Name())
value = _d(reader.Value())
if qname.startswith("xmlns"):
if len(qname) > 5:
newPrefix = qname[6:]
else:
newPrefix = None
newPrefixes.append(newPrefix)
self._cont_handler.startPrefixMapping(\
newPrefix,value)
if not self.__nspfx:
continue # don't report xmlns attribute
attName = (_d(reader.NamespaceUri()),
_d(reader.LocalName()))
qnames[attName] = qname
attrs[attName] = value
reader.MoveToElement()
self._cont_handler.startElementNS( \
eltName,eltQName,attributesNSImpl)
if reader.IsEmptyElement():
self._cont_handler.endElementNS(eltName, eltQName)
for newPrefix in newPrefixes:
self._cont_handler.endPrefixMapping(newPrefix)
else:
prefixes.append(newPrefixes)
else:
eltName = _d(reader.Name())
attributesImpl._attrs = attrs = {}
while reader.MoveToNextAttribute():
attName = _d(reader.Name())
attrs[attName] = _d(reader.Value())
reader.MoveToElement()
self._cont_handler.startElement( \
eltName,attributesImpl)
if reader.IsEmptyElement():
self._cont_handler.endElement(eltName)
# EndElement
elif nodeType == 15:
if self.__ns:
self._cont_handler.endElementNS( \
(_d(reader.NamespaceUri()),_d(reader.LocalName())),
_d(reader.Name()))
for prefix in prefixes.pop():
self._cont_handler.endPrefixMapping(prefix)
else:
self._cont_handler.endElement(_d(reader.Name()))
# Text
elif nodeType == 3:
self._cont_handler.characters(_d(reader.Value()))
# Whitespace
elif nodeType == 13:
self._cont_handler.ignorableWhitespace(_d(reader.Value()))
# SignificantWhitespace
elif nodeType == 14:
self._cont_handler.characters(_d(reader.Value()))
# CDATA
elif nodeType == 4:
if not self.__lex_handler is None:
self.__lex_handler.startCDATA()
self._cont_handler.characters(_d(reader.Value()))
if not self.__lex_handler is None:
self.__lex_handler.endCDATA()
# EntityReference
elif nodeType == 5:
if not self.__lex_handler is None:
self.startEntity(_d(reader.Name()))
reader.ResolveEntity()
# EndEntity
elif nodeType == 16:
if not self.__lex_handler is None:
self.endEntity(_d(reader.Name()))
# ProcessingInstruction
elif nodeType == 7:
self._cont_handler.processingInstruction( \
_d(reader.Name()),_d(reader.Value()))
# Comment
elif nodeType == 8:
if not self.__lex_handler is None:
self.__lex_handler.comment(_d(reader.Value()))
# DocumentType
elif nodeType == 10:
#if not self.__lex_handler is None:
# self.__lex_handler.startDTD()
pass # TODO (how to detect endDTD? on first non-dtd event?)
# XmlDeclaration
elif nodeType == 17:
pass # TODO
# Entity
elif nodeType == 6:
pass # TODO (entity decl)
# Notation (decl)
elif nodeType == 12:
pass # TODO
# Attribute (never in this loop)
#elif nodeType == 2:
# pass
# Document (not exposed)
#elif nodeType == 9:
# pass
# DocumentFragment (never returned by XmlReader)
#elif nodeType == 11:
# pass
# None
#elif nodeType == 0:
# pass
# -
else:
raise SAXException("Unexpected node type %d" % nodeType)
if r == 0:
self._cont_handler.endDocument()
reader.Close()
finally:
self.__parsing = 0
import libxml2
docstr = """<?xml version='1.0'?>
<!DOCTYPE doc [
<!ENTITY tst "<p>test</p>">
]>
<doc>&tst;</doc>"""
# Memory debug specific
libxml2.debugMemory(1)
#
# First test, normal don't substitute entities.
#
f = StringIO.StringIO(docstr)
input = libxml2.inputBuffer(f)
reader = input.newTextReader("test_noent")
ret = reader.Read()
if ret != 1:
print "Error reading to root"
sys.exit(1)
if reader.Name() == "doc" or reader.NodeType() == 10:
ret = reader.Read()
if ret != 1:
print "Error reading to root"
sys.exit(1)
if reader.Name() != "doc" or reader.NodeType() != 1:
print "test_normal: Error reading the root element"
sys.exit(1)
ret = reader.Read()
if ret != 1:
str_io = io.StringIO
docstr = """<?xml version='1.0'?>
<!DOCTYPE doc [
<!ENTITY tst "<p>test</p>">
]>
<doc>&tst;</doc>"""
# Memory debug specific
libxml2.debugMemory(1)
#
# First test, normal don't substitute entities.
#
f = str_io(docstr)
input = libxml2.inputBuffer(f)
reader = input.newTextReader("test_noent")
ret = reader.Read()
if ret != 1:
print("Error reading to root")
sys.exit(1)
if reader.Name() == "doc" or reader.NodeType() == 10:
ret = reader.Read()
if ret != 1:
print("Error reading to root")
sys.exit(1)
if reader.Name() != "doc" or reader.NodeType() != 1:
print("test_normal: Error reading the root element")
sys.exit(1)
ret = reader.Read()
if ret != 1:
1,x
1,c
3,#text
15,c
15,x
14,#text
1,b
3,#text
15,b
14,#text
15,test
"""
res=""
err=""
input = libxml2.inputBuffer(StringIO.StringIO(s))
reader = input.newTextReader("test2")
reader.SetParserProp(libxml2.PARSER_LOADDTD,1)
reader.SetParserProp(libxml2.PARSER_DEFAULTATTRS,1)
reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES,1)
reader.SetParserProp(libxml2.PARSER_VALIDATE,1)
while reader.Read() == 1:
res = res + "%s,%s\n" % (reader.NodeType(),reader.Name())
if res != expect:
print "test2 failed: unexpected output"
print res
sys.exit(1)
if err != "":
print "test2 failed: validation error found"
print err
def parse(self, source):
self.__parsing = 1
try:
# prepare source and create reader
if type(source) in StringTypes:
reader = libxml2.newTextReaderFilename(source)
else:
source = saxutils.prepare_input_source(source)
input = libxml2.inputBuffer(source.getByteStream())
reader = input.newTextReader(source.getSystemId())
reader.SetErrorHandler(self._errorHandler,None)
# configure reader
reader.SetParserProp(libxml2.PARSER_LOADDTD,1)
reader.SetParserProp(libxml2.PARSER_DEFAULTATTRS,1)
reader.SetParserProp(libxml2.PARSER_SUBST_ENTITIES,1)
reader.SetParserProp(libxml2.PARSER_VALIDATE,self.__validate)
# we reuse attribute maps (for a slight performance gain)
if self.__ns:
attributesNSImpl = xmlreader.AttributesNSImpl({},{})
else:
attributesImpl = xmlreader.AttributesImpl({})
# prefixes to pop (for endPrefixMapping)
prefixes = []
# start loop
self._cont_handler.startDocument()
while 1:
r = reader.Read()
# check for errors
if r == 1:
if not self.__errors is None:
self._reportErrors(0)
elif r == 0:
if not self.__errors is None:
self._reportErrors(0)
break # end of parse
else:
if not self.__errors is None:
self._reportErrors(1)
else:
self._err_handler.fatalError(\
SAXException("Read failed (no details available)"))
break # fatal parse error
# get node type
nodeType = reader.NodeType()
# Element
if nodeType == 1:
if self.__ns:
eltName = (_d(reader.NamespaceUri()),\
_d(reader.LocalName()))
eltQName = _d(reader.Name())
attributesNSImpl._attrs = attrs = {}
attributesNSImpl._qnames = qnames = {}
newPrefixes = []
while reader.MoveToNextAttribute():
qname = _d(reader.Name())
value = _d(reader.Value())
if qname.startswith("xmlns"):
if len(qname) > 5:
newPrefix = qname[6:]
else:
newPrefix = None
newPrefixes.append(newPrefix)
self._cont_handler.startPrefixMapping(\
newPrefix,value)
if not self.__nspfx:
continue # don't report xmlns attribute
attName = (_d(reader.NamespaceUri()),
_d(reader.LocalName()))
qnames[attName] = qname
attrs[attName] = value
reader.MoveToElement()
self._cont_handler.startElementNS( \
eltName,eltQName,attributesNSImpl)
if reader.IsEmptyElement():
self._cont_handler.endElementNS(eltName,eltQName)
for newPrefix in newPrefixes:
self._cont_handler.endPrefixMapping(newPrefix)
else:
prefixes.append(newPrefixes)
else:
eltName = _d(reader.Name())
attributesImpl._attrs = attrs = {}
while reader.MoveToNextAttribute():
attName = _d(reader.Name())
attrs[attName] = _d(reader.Value())
reader.MoveToElement()
self._cont_handler.startElement( \
eltName,attributesImpl)
if reader.IsEmptyElement():
self._cont_handler.endElement(eltName)
# EndElement
elif nodeType == 15:
if self.__ns:
self._cont_handler.endElementNS( \
(_d(reader.NamespaceUri()),_d(reader.LocalName())),
_d(reader.Name()))
for prefix in prefixes.pop():
self._cont_handler.endPrefixMapping(prefix)
else:
self._cont_handler.endElement(_d(reader.Name()))
# Text
elif nodeType == 3:
self._cont_handler.characters(_d(reader.Value()))
# Whitespace
elif nodeType == 13:
self._cont_handler.ignorableWhitespace(_d(reader.Value()))
# SignificantWhitespace
elif nodeType == 14:
self._cont_handler.characters(_d(reader.Value()))
# CDATA
elif nodeType == 4:
if not self.__lex_handler is None:
self.__lex_handler.startCDATA()
self._cont_handler.characters(_d(reader.Value()))
if not self.__lex_handler is None:
self.__lex_handler.endCDATA()
# EntityReference
elif nodeType == 5:
if not self.__lex_handler is None:
self.startEntity(_d(reader.Name()))
reader.ResolveEntity()
# EndEntity
elif nodeType == 16:
if not self.__lex_handler is None:
self.endEntity(_d(reader.Name()))
# ProcessingInstruction
elif nodeType == 7:
self._cont_handler.processingInstruction( \
_d(reader.Name()),_d(reader.Value()))
# Comment
elif nodeType == 8:
if not self.__lex_handler is None:
self.__lex_handler.comment(_d(reader.Value()))
# DocumentType
elif nodeType == 10:
#if not self.__lex_handler is None:
# self.__lex_handler.startDTD()
pass # TODO (how to detect endDTD? on first non-dtd event?)
# XmlDeclaration
elif nodeType == 17:
pass # TODO
# Entity
elif nodeType == 6:
pass # TODO (entity decl)
# Notation (decl)
elif nodeType == 12:
pass # TODO
# Attribute (never in this loop)
#elif nodeType == 2:
# pass
# Document (not exposed)
#elif nodeType == 9:
# pass
# DocumentFragment (never returned by XmlReader)
#elif nodeType == 11:
# pass
# None
#elif nodeType == 0:
# pass
# -
else:
raise SAXException("Unexpected node type %d" % nodeType)
if r == 0:
self._cont_handler.endDocument()
reader.Close()
finally:
self.__parsing = 0
#!/usr/bin/python -u
import sys
import libxml2
import StringIO
# Memory debug specific
libxml2.debugMemory(1)
i = 0
while i < 5000:
f = StringIO.StringIO("foobar")
buf = libxml2.inputBuffer(f)
i = i + 1
del f
del buf
# Memory debug specific
libxml2.cleanupParser()
if libxml2.debugMemory(1) == 0:
print "OK"
else:
print "Memory leak %d bytes" % (libxml2.debugMemory(1))
libxml2.dumpMemory()