convert_xml.py

#encoding: utf-8

import xml.etree.ElementTree as ET
tree = ET.parse('angkorwat.xml')
root = tree.getroot()

list_title = []
list_image = []
list_image_author = []
list_image_copyright_sign = []
list_desc = []

for strings in root.findall('string-array'):
    print strings.attrib
    for item in strings.findall('item'):
        if strings.attrib['name'].endswith('_images'):
            list_image.append(item.text)
        elif strings.attrib['name'].endswith('_image_copyrights'):
            list_image_author.append(item.text)
        elif strings.attrib['name'].endswith('_image_copyright_signs'):
            list_image_copyright_sign.append(item.text)
        elif strings.attrib['name'].endswith('_picture_titles'):
            list_title.append(item.text)
        elif strings.attrib['name'].endswith('_descriptions'):
            list_desc.append(item.text)

print list_title

resources = ET.Element('resources')
for idx, title in enumerate(list_title):
    slide = ET.SubElement(resources, 'slide')
    title = ET.SubElement(slide, 'title')
    title.text = list_title[idx]
    image = ET.SubElement(slide, 'image')
    image.text = list_image[idx]
    image_author = ET.SubElement(slide, 'image_author')
    image_author.text = list_image_author[idx]
    image_copyright = ET.SubElement(slide, 'image_copyright_sign')
    image_copyright.text = list_image_copyright_sign[idx]
    desc = ET.SubElement(slide, 'desc')
    desc.text = list_desc[idx]
    empty = ET.SubElement(slide, 'empty')

import codecs
out = codecs.open("angkorwat_xml.xml", "w", "utf-8")

from xml.dom import minidom
def newwritexml(self, writer, indent= '', addindent= '', newl= ''):
    if len(self.childNodes)==1 and self.firstChild.nodeType==3:
        writer.write(indent)
        self.oldwritexml(writer) # cancel extra whitespace
        writer.write(newl)
    else:
        self.oldwritexml(writer, indent, addindent, newl)

minidom.Element.oldwritexml= minidom.Element.writexml
minidom.Element.writexml = newwritexml

r = minidom.parseString(ET.tostring(resources))
r.encoding="UTF-8"

r.writexml(out, indent="    ", newl="\n", encoding="UTF-8")