# Import ElementTree according to Fredrik Lundh's own recipe
# see: http://mail.python.org/pipermail/python-dev/2006-August/068504.html
try:
from xml.etree import cElementTree as ET
except ImportError:
try:
import cElementTree as ET
except ImportError:
import elementtree.ElementTree as ET
# Converter
def DOM2ET(domelem):
"""Converts a DOM node object of type element to an ElementTree Element.
domelem: DOM node object of type element (domelem.nodeType == domelem.ELEMENT_NODE)
returns an 'equivalent' ElementTree Element
"""
# make some local variables for fast processing
tyCDATA = domelem.CDATA_SECTION_NODE
tyTEXT = domelem.TEXT_NODE
tyPI = domelem.PROCESSING_INSTRUCTION_NODE
tyCOMMENT = domelem.COMMENT_NODE
tyELEMENT = domelem.ELEMENT_NODE
# calculate the attributes of the domelem
attribs = domelem.attributes
attrs = dict((x.name, x.value) for x in (attribs.item(i) for i in range(attribs.length)))
# build the ET Element
etelem = ET.Element(domelem.tagName, attrs)
last = None # to differentiate between 'text' and 'tail'
for node in domelem.childNodes:
nodeType = node.nodeType
if (tyTEXT == nodeType or tyCDATA == nodeType) and node.data:
data = node.data
if last is None: etelem.text = etelem.text + data if etelem.text else data
else: last.tail = last.tail + data if last.tail else data
elif tyELEMENT == nodeType:
last = DOM2ET(node)
etelem.append(last)
elif tyCOMMENT == nodeType:
last = ET.Comment(node.data)
etelem.append(last)
elif tyPI == nodeType:
last = ET.ProcessingInstruction(node.target, node.data)
etelem.append(last)
return etelem
if __name__ == "__main__":
import xml.dom.minidom as minidom
xmltext = '<ELEM key="value">text<SUBELEM/>tail</ELEM>'
doc = minidom.parseString(xmltext)
topelem = doc.documentElement
etelem = DOM2ET(topelem)
print ET.tostring(etelem) # prints: <ELEM key="value">text<SUBELEM />tail</ELEM>
