def from_file(cls, filename):
# unzip vsdx file to temp. folder
directory = './{}'.format(filename.rsplit('.', 1)[0])
with zipfile.ZipFile(filename, "r") as zip_ref:
zip_ref.extractall(directory)
# Read [Content_Types].xml
content_types = ContentTypes.from_xml(directory +
'/[Content_Types].xml')
# Read relationships
package_rels = Relationship.from_xml(
'{}/_rels/.rels'.format(directory))
document_rels = Relationship.from_xml(
'{}/visio/_rels/document.xml.rels'.format(directory))
# Read pages and relationships
page_collection = PageCollection.from_xml(directory, content_types)
# Remove extracted folder again
shutil.rmtree(directory)
return cls(page_collection=page_collection,
package_rels=package_rels,
document_rels=document_rels,
content_types=content_types)
def from_xml(cls, dir, content_types):
"""Generate PageCollection from files
:param dir: The directory of the extracted visio package
:param content_types: Instance of :class:`ContentType`
"""
rel_dir = '{}/visio/pages/_rels/'.format(dir)
page_dir = '{}/visio/pages/'.format(dir)
pages = []
rels = Relationship.from_xml(rel_dir + 'pages.xml.rels')
# Parse pages.xml for page info
tree = ET.parse(page_dir + 'pages.xml')
root = tree.getroot()
for child in root:
name = child.attrib.get('Name', '')
id = child.attrib['ID']
# TODO Parse these namespaces properly
rel_id = child[1].attrib[
'{http://schemas.openxmlformats.org/officeDocument/2006/relationships}id']
pages.append(
Page.from_xml(page_dir + rels.rels[rel_id][0], name, id,
rel_id))
return cls(content_types, rels=rels, pages=pages)