def __init__(self, tree):
if hasattr(tree, "getroot"):
tree = Root(tree)
elif isinstance(tree, list):
tree = FragmentRoot(tree)
_base.NonRecursiveTreeWalker.__init__(self, tree)
self.filter = ihatexml.InfosetFilter()
def testSerializer(element):
rv = []
finalText = None
filter = ihatexml.InfosetFilter()
def serializeElement(element, indent=0):
if not hasattr(element, "tag"):
if hasattr(element, "getroot"):
#Full tree case
rv.append("#document")
if element.docinfo.internalDTD:
if not (element.docinfo.public_id
or element.docinfo.system_url):
dtd_str = "<!DOCTYPE %s>" % element.docinfo.root_name
else:
dtd_str = """<!DOCTYPE %s "%s" "%s">""" % (
element.docinfo.root_name,
element.docinfo.public_id,
element.docinfo.system_url)
rv.append("|%s%s" % (' ' * (indent + 2), dtd_str))
next_element = element.getroot()
while next_element.getprevious() is not None:
next_element = next_element.getprevious()
while next_element is not None:
serializeElement(next_element, indent + 2)
next_element = next_element.getnext()
elif isinstance(element, basestring):
#Text in a fragment
rv.append("|%s\"%s\"" % (' ' * indent, element))
else:
#Fragment case
rv.append("#document-fragment")
for next_element in element:
serializeElement(next_element, indent + 2)
elif type(element.tag) == type(etree.Comment):
rv.append("|%s<!-- %s -->" % (' ' * indent, element.text))
else:
rv.append("|%s<%s>" %
(' ' * indent, filter.fromXmlName(element.tag)))
if hasattr(element, "attrib"):
for name, value in element.attrib.iteritems():
rv.append('|%s%s="%s"' %
(' ' *
(indent + 2), filter.fromXmlName(name), value))
if element.text:
rv.append("|%s\"%s\"" % (' ' * (indent + 2), element.text))
indent += 2
for child in element.getchildren():
serializeElement(child, indent)
if hasattr(element, "tail") and element.tail:
rv.append("|%s\"%s\"" % (' ' * (indent - 2), element.tail))
serializeElement(element, 0)
if finalText is not None:
rv.append("|%s\"%s\"" % (' ' * 2, finalText))
return "\n".join(rv)
def tostring(element):
"""Serialize an element and its child nodes to a string"""
rv = []
finalText = None
filter = ihatexml.InfosetFilter()
def serializeElement(element):
if type(element) == type(ElementTree.ElementTree):
element = element.getroot()
if element.tag == "<!DOCTYPE>":
if element.get("publicId") or element.get("systemId"):
publicId = element.get("publicId") or ""
systemId = element.get("systemId") or ""
rv.append("""<!DOCTYPE %s PUBLIC "%s" "%s">""" %
(element.text, publicId, systemId))
else:
rv.append("<!DOCTYPE %s>" % (element.text, ))
elif element.tag == "<DOCUMENT_ROOT>":
if element.text:
rv.append(element.text)
if element.tail:
finalText = element.tail
for child in element.getchildren():
serializeElement(child)
elif type(element.tag) == type(ElementTree.Comment):
rv.append("<!--%s-->" % (element.text, ))
else:
#This is assumed to be an ordinary element
if not element.attrib:
rv.append("<%s>" % (filter.fromXmlName(element.tag), ))
else:
attr = " ".join([
"%s=\"%s\"" % (filter.fromXmlName(name), value)
for name, value in element.attrib.iteritems()
])
rv.append("<%s %s>" % (element.tag, attr))
if element.text:
rv.append(element.text)
for child in element.getchildren():
serializeElement(child)
rv.append("</%s>" % (element.tag, ))
if element.tail:
rv.append(element.tail)
serializeElement(element)
if finalText is not None:
rv.append("%s\"" % (' ' * 2, finalText))
return "".join(rv)
def testSerializer(element):
rv = []
finalText = None
infosetFilter = ihatexml.InfosetFilter()
def serializeElement(element, indent=0):
if not hasattr(element, "tag"):
if hasattr(element, "getroot"):
#Full tree case
rv.append("#document")
if element.docinfo.internalDTD:
if not (element.docinfo.public_id
or element.docinfo.system_url):
dtd_str = "<!DOCTYPE %s>" % element.docinfo.root_name
else:
dtd_str = """<!DOCTYPE %s "%s" "%s">""" % (
element.docinfo.root_name,
element.docinfo.public_id,
element.docinfo.system_url)
rv.append("|%s%s" % (' ' * (indent + 2), dtd_str))
next_element = element.getroot()
while next_element.getprevious() is not None:
next_element = next_element.getprevious()
while next_element is not None:
serializeElement(next_element, indent + 2)
next_element = next_element.getnext()
elif isinstance(element, str) or isinstance(element, bytes):
#Text in a fragment
assert isinstance(element, str) or sys.version_info.major == 2
rv.append("|%s\"%s\"" % (' ' * indent, element))
else:
#Fragment case
rv.append("#document-fragment")
for next_element in element:
serializeElement(next_element, indent + 2)
elif type(element.tag) == type(etree.Comment):
rv.append("|%s<!-- %s -->" % (' ' * indent, element.text))
if hasattr(element, "tail") and element.tail:
rv.append("|%s\"%s\"" % (' ' * indent, element.tail))
else:
assert isinstance(element, etree._Element)
nsmatch = etree_builders.tag_regexp.match(element.tag)
if nsmatch is not None:
ns = nsmatch.group(1)
tag = nsmatch.group(2)
prefix = constants.prefixes[ns]
rv.append(
"|%s<%s %s>" %
(' ' * indent, prefix, infosetFilter.fromXmlName(tag)))
else:
rv.append(
"|%s<%s>" %
(' ' * indent, infosetFilter.fromXmlName(element.tag)))
if hasattr(element, "attrib"):
attributes = []
for name, value in element.attrib.items():
nsmatch = tag_regexp.match(name)
if nsmatch is not None:
ns, name = nsmatch.groups()
name = infosetFilter.fromXmlName(name)
prefix = constants.prefixes[ns]
attr_string = "%s %s" % (prefix, name)
else:
attr_string = infosetFilter.fromXmlName(name)
attributes.append((attr_string, value))
for name, value in sorted(attributes):
rv.append('|%s%s="%s"' % (' ' * (indent + 2), name, value))
if element.text:
rv.append("|%s\"%s\"" % (' ' * (indent + 2), element.text))
indent += 2
for child in element.getchildren():
serializeElement(child, indent)
if hasattr(element, "tail") and element.tail:
rv.append("|%s\"%s\"" % (' ' * (indent - 2), element.tail))
serializeElement(element, 0)
if finalText is not None:
rv.append("|%s\"%s\"" % (' ' * 2, finalText))
return "\n".join(rv)
def __init__(self, namespaceHTMLElements, fullTree=False):
builder = etree_builders.getETreeModule(etree, fullTree=fullTree)
infosetFilter = self.infosetFilter = ihatexml.InfosetFilter()
self.namespaceHTMLElements = namespaceHTMLElements
class Attributes(dict):
def __init__(self, element, value={}):
self._element = element
dict.__init__(self, value)
for key, value in self.items():
if isinstance(key, tuple):
name = "{%s}%s" % (
key[2], infosetFilter.coerceAttribute(key[1]))
else:
name = infosetFilter.coerceAttribute(key)
self._element._element.attrib[name] = value
def __setitem__(self, key, value):
dict.__setitem__(self, key, value)
if isinstance(key, tuple):
name = "{%s}%s" % (key[2],
infosetFilter.coerceAttribute(key[1]))
else:
name = infosetFilter.coerceAttribute(key)
self._element._element.attrib[name] = value
class Element(builder.Element):
def __init__(self, name, namespace):
name = infosetFilter.coerceElement(name)
builder.Element.__init__(self, name, namespace=namespace)
self._attributes = Attributes(self)
def _setName(self, name):
self._name = infosetFilter.coerceElement(name)
self._element.tag = self._getETreeTag(self._name,
self._namespace)
def _getName(self):
return infosetFilter.fromXmlName(self._name)
name = property(_getName, _setName)
def _getAttributes(self):
return self._attributes
def _setAttributes(self, attributes):
self._attributes = Attributes(self, attributes)
attributes = property(_getAttributes, _setAttributes)
def insertText(self, data, insertBefore=None):
data = infosetFilter.coerceCharacters(data)
builder.Element.insertText(self, data, insertBefore)
def appendChild(self, child):
builder.Element.appendChild(self, child)
class Comment(builder.Comment):
def __init__(self, data):
data = infosetFilter.coerceComment(data)
builder.Comment.__init__(self, data)
def _setData(self, data):
data = infosetFilter.coerceComment(data)
self._element.text = data
def _getData(self):
return self._element.text
data = property(_getData, _setData)
self.elementClass = Element
self.commentClass = builder.Comment
#self.fragmentClass = builder.DocumentFragment
_base.TreeBuilder.__init__(self, namespaceHTMLElements)
def getDomBuilder(DomImplementation):
Dom = DomImplementation
infoset_filter = ihatexml.InfosetFilter()
class AttrList:
def __init__(self, element):
self.element = element
def __iter__(self):
return self.element.attributes.items().__iter__()
def __setitem__(self, name, value):
self.element.setAttribute(infoset_filter.coerceAttribute(name),
infoset_filter.coerceCharacters(value))
def items(self):
return [(infoset_filter.fromXmlName(item[0]), item[1])
for item in self.element.attributes.items()]
def keys(self):
return [
infoset_filter.fromXmlName(item)
for item in self.element.attributes.keys()
]
def __getitem__(self, name):
name = infoset_filter.toXmlName(name)
return self.element.getAttribute(name)
def __contains__(self, name):
if isinstance(name, tuple):
raise NotImplementedError
else:
return self.element.hasAttribute(
infoset_filter.toXmlName(name))
class NodeBuilder(_base.Node):
def __init__(self, element):
_base.Node.__init__(self, element.localName)
self.element = element
namespace = property(lambda self: hasattr(self.element, "namespaceURI")
and self.element.namespaceURI or None)
def appendChild(self, node):
node.parent = self
self.element.appendChild(node.element)
def insertText(self, data, insertBefore=None):
data = infoset_filter.coerceCharacters(data)
text = self.element.ownerDocument.createTextNode(data)
if insertBefore:
self.element.insertBefore(text, insertBefore.element)
else:
self.element.appendChild(text)
def insertBefore(self, node, refNode):
self.element.insertBefore(node.element, refNode.element)
node.parent = self
def removeChild(self, node):
if node.element.parentNode == self.element:
self.element.removeChild(node.element)
node.parent = None
def reparentChildren(self, newParent):
while self.element.hasChildNodes():
child = self.element.firstChild
self.element.removeChild(child)
newParent.element.appendChild(child)
self.childNodes = []
def getAttributes(self):
return AttrList(self.element)
def setAttributes(self, attributes):
if attributes:
for name, value in attributes.items():
if isinstance(name, tuple):
if name[0] is not None:
qualifiedName = (
name[0] + ":" +
infoset_filter.coerceAttribute(name[1]))
else:
qualifiedName = infoset_filter.coerceAttribute(
name[1])
self.element.setAttributeNS(name[2], qualifiedName,
value)
else:
self.element.setAttribute(
infoset_filter.coerceAttribute(name), value)
attributes = property(getAttributes, setAttributes)
def cloneNode(self):
return NodeBuilder(self.element.cloneNode(False))
def hasContent(self):
return self.element.hasChildNodes()
def getNameTuple(self):
if self.namespace == None:
return namespaces["html"], self.name
else:
return self.namespace, self.name
nameTuple = property(getNameTuple)
class TreeBuilder(_base.TreeBuilder):
def documentClass(self):
self.dom = Dom.getDOMImplementation().createDocument(
None, None, None)
return self
def insertDoctype(self, token):
name = token["name"]
publicId = token["publicId"]
systemId = token["systemId"]
domimpl = Dom.getDOMImplementation()
doctype = domimpl.createDocumentType(name, publicId, systemId)
self.document.appendChild(NodeBuilder(doctype))
if Dom == minidom:
doctype.ownerDocument = self.dom
def elementClass(self, name, namespace=None):
if namespace is None and self.defaultNamespace is None:
node = self.dom.createElement(name)
else:
node = self.dom.createElementNS(namespace, name)
return NodeBuilder(node)
def commentClass(self, data):
return NodeBuilder(self.dom.createComment(data))
def fragmentClass(self):
return NodeBuilder(self.dom.createDocumentFragment())
def appendChild(self, node):
self.dom.appendChild(node.element)
def testSerializer(self, element):
return testSerializer(element)
def getDocument(self):
return self.dom
def getFragment(self):
return _base.TreeBuilder.getFragment(self).element
def insertText(self, data, parent=None):
data = infoset_filter.coerceCharacters(data)
if parent <> self:
_base.TreeBuilder.insertText(self, data, parent)
else:
# HACK: allow text nodes as children of the document node
if hasattr(self.dom, '_child_node_types'):
if not Node.TEXT_NODE in self.dom._child_node_types:
self.dom._child_node_types = list(
self.dom._child_node_types)
self.dom._child_node_types.append(Node.TEXT_NODE)
self.dom.appendChild(self.dom.createTextNode(data))
name = None
def testSerializer(element):
element.normalize()
rv = []
def serializeElement(element, indent=0):
if element.nodeType == Node.DOCUMENT_TYPE_NODE:
if element.name:
if element.publicId or element.systemId:
publicId = element.publicId or ""
systemId = element.systemId or ""
rv.append(
"""|%s<!DOCTYPE %s "%s" "%s">""" %
(' ' * indent, element.name, publicId, systemId))
else:
rv.append("|%s<!DOCTYPE %s>" %
(' ' * indent, element.name))
else:
rv.append("|%s<!DOCTYPE >" % (' ' * indent, ))
elif element.nodeType == Node.DOCUMENT_NODE:
rv.append("#document")
elif element.nodeType == Node.DOCUMENT_FRAGMENT_NODE:
rv.append("#document-fragment")
elif element.nodeType == Node.COMMENT_NODE:
rv.append("|%s<!-- %s -->" % (' ' * indent, element.nodeValue))
elif element.nodeType == Node.TEXT_NODE:
rv.append("|%s\"%s\"" % (' ' * indent, element.nodeValue))
else:
if (hasattr(element, "namespaceURI") and element.namespaceURI
not in (None, constants.namespaces["html"])):
name = "%s %s" % (constants.prefixes[element.namespaceURI],
element.nodeName)
else:
name = element.nodeName
rv.append("|%s<%s>" % (' ' * indent, name))
if element.hasAttributes():
i = 0
attr = element.attributes.item(i)
while attr:
name = infoset_filter.fromXmlName(attr.localName)
value = attr.value
ns = attr.namespaceURI
if ns:
name = "%s %s" % (constants.prefixes[ns], name)
i += 1
attr = element.attributes.item(i)
rv.append('|%s%s="%s"' % (' ' *
(indent + 2), name, value))
indent += 2
for child in element.childNodes:
serializeElement(child, indent)
serializeElement(element, 0)
return "\n".join(rv)
def dom2sax(node, handler, nsmap={'xml': XML_NAMESPACE}):
if node.nodeType == Node.ELEMENT_NODE:
if not nsmap:
handler.startElement(node.nodeName, node.attributes)
for child in node.childNodes:
dom2sax(child, handler, nsmap)
handler.endElement(node.nodeName)
else:
attributes = dict(node.attributes.itemsNS())
# gather namespace declarations
prefixes = []
for attrname in node.attributes.keys():
attr = node.getAttributeNode(attrname)
if (attr.namespaceURI == XMLNS_NAMESPACE
or (attr.namespaceURI == None
and attr.nodeName.startswith('xmlns'))):
prefix = (attr.localName != 'xmlns' and attr.localName
or None)
handler.startPrefixMapping(prefix, attr.nodeValue)
prefixes.append(prefix)
nsmap = nsmap.copy()
nsmap[prefix] = attr.nodeValue
del attributes[(attr.namespaceURI, attr.localName)]
# apply namespace declarations
for attrname in node.attributes.keys():
attr = node.getAttributeNode(attrname)
if attr.namespaceURI == None and ':' in attr.nodeName:
prefix = attr.nodeName.split(':')[0]
if nsmap.has_key(prefix):
del attributes[(attr.namespaceURI, attr.localName)]
attributes[(nsmap[prefix],
attr.localName)] = attr.nodeValue
# SAX events
ns = node.namespaceURI or nsmap.get(None, None)
handler.startElementNS((ns, node.nodeName), node.nodeName,
attributes)
for child in node.childNodes:
dom2sax(child, handler, nsmap)
handler.endElementNS((ns, node.nodeName), node.nodeName)
for prefix in prefixes:
handler.endPrefixMapping(prefix)
elif node.nodeType in [Node.TEXT_NODE, Node.CDATA_SECTION_NODE]:
handler.characters(node.nodeValue)
elif node.nodeType == Node.DOCUMENT_NODE:
handler.startDocument()
for child in node.childNodes:
dom2sax(child, handler, nsmap)
handler.endDocument()
elif node.nodeType == Node.DOCUMENT_FRAGMENT_NODE:
for child in node.childNodes:
dom2sax(child, handler, nsmap)
else:
# ATTRIBUTE_NODE
# ENTITY_NODE
# PROCESSING_INSTRUCTION_NODE
# COMMENT_NODE
# DOCUMENT_TYPE_NODE
# NOTATION_NODE
pass
return locals()
def __init__(self, fullTree=False):
builder = etree_builders.getETreeModule(etree, fullTree=fullTree)
filter = self.filter = ihatexml.InfosetFilter()
class Attributes(dict):
def __init__(self, element, value={}):
self._element = element
dict.__init__(self, value)
for k, v in self.iteritems():
self._element._element.attrib[filter.coerceAttribute(
k)] = v
def __setitem__(self, key, value):
dict.__setitem__(self, key, value)
self._element._element.attrib[filter.coerceAttribute(
key)] = value
class Element(builder.Element):
def __init__(self, name):
self._name = name
builder.Element.__init__(self, filter.coerceElement(name))
self._attributes = Attributes(self)
def _setName(self, name):
self._name = name
self._element.tag = filter.coerceElement(name)
def _getName(self):
return self._name
name = property(_getName, _setName)
def _getAttributes(self):
return self._attributes
def _setAttributes(self, attributes):
self._attributes = Attributes(self, attributes)
attributes = property(_getAttributes, _setAttributes)
def insertText(self, data, insertBefore=None):
data = filter.coerceCharacters(data)
builder.Element.insertText(self, data, insertBefore)
def appendChild(self, child):
builder.Element.appendChild(self, child)
class Comment(builder.Comment):
def __init__(self, data):
data = filter.coerceComment(data)
builder.Comment.__init__(self, data)
def _setData(self, data):
data = filter.coerceComment(data)
self._element.text = data
def _getData(self):
return self._element.text
data = property(_getData, _setData)
self.elementClass = Element
self.commentClass = builder.Comment
#self.fragmentClass = builder.DocumentFragment
_base.TreeBuilder.__init__(self)
