def _build_data(self, path, *args, **kwargs):
with ZipFile(path) as f:
# These must be in the ZIP in order for the docx to be valid.
self.document_text = f.read('word/document.xml')
self.relationship_text = f.read('word/_rels/document.xml.rels')
# These are all optional.
self.styles_text = self._extract_xml(f, 'word/styles.xml')
self.fonts = self._extract_xml(f, 'word/fontTable.xml')
self.numbering_text = self._extract_xml(f, 'word/numbering.xml')
self.comment_text = self._extract_xml(f, 'word/comments.xml')
zipped_image_files = [
e for e in f.infolist() if e.filename.startswith('word/media/')
]
for e in zipped_image_files:
self._image_data[e.filename] = f.read(e.filename)
self.root = parse_xml_from_string(self.document_text)
self.numbering_root = None
if self.numbering_text:
self.numbering_root = parse_xml_from_string(self.numbering_text)
self.comment_root = None
if self.comment_text:
self.comment_root = parse_xml_from_string(self.comment_text)
def _build_data(
self,
path,
document_xml=None,
rels_dict=None,
numbering_dict=None,
styles_dict=None,
*args, **kwargs):
self._test_rels_dict = rels_dict
if rels_dict:
for value in rels_dict.values():
self._image_data['word/%s' % value] = 'word/%s' % value
self.numbering_root = None
if numbering_dict is not None:
self.numbering_root = parse_xml_from_string(
DXB.numbering(numbering_dict),
)
self.numbering_dict = numbering_dict
# Intentionally not calling super
if document_xml is not None:
self.root = parse_xml_from_string(document_xml)
self.zip_path = ''
# This is the standard page width for a word document, Also the page
# width that we are looking for in the test.
self.page_width = 612
self.styles_dict = styles_dict
def _build_data(self, path, *args, **kwargs):
with ZipFile(path) as f:
# These must be in the ZIP in order for the docx to be valid.
self.document_text = f.read('word/document.xml')
self.relationship_text = f.read('word/_rels/document.xml.rels')
# These are all optional.
self.styles_text = self._extract_xml(f, 'word/styles.xml')
self.fonts = self._extract_xml(f, 'word/fontTable.xml')
self.numbering_text = self._extract_xml(f, 'word/numbering.xml')
self.comment_text = self._extract_xml(f, 'word/comments.xml')
zipped_image_files = [
e for e in f.infolist()
if e.filename.startswith('word/media/')
]
for e in zipped_image_files:
self._image_data[e.filename] = f.read(e.filename)
self.root = parse_xml_from_string(self.document_text)
self.numbering_root = None
if self.numbering_text:
self.numbering_root = parse_xml_from_string(self.numbering_text)
self.comment_root = None
if self.comment_text:
self.comment_root = parse_xml_from_string(self.comment_text)
def _parse_rels_root(self):
tree = parse_xml_from_string(self.relationship_text)
rels_dict = {}
for el in tree:
rId = el.get("Id")
target = el.get("Target")
rels_dict[rId] = target
return rels_dict
def _parse_rels_root(self):
tree = parse_xml_from_string(self.relationship_text)
rels_dict = {}
for el in tree:
rId = el.get('Id')
target = el.get('Target')
rels_dict[rId] = target
return rels_dict
def _parse_styles(self):
if self.styles_text is None:
return {}
tree = parse_xml_from_string(self.styles_text)
styles_dict = {}
for style in find_all(tree, "style"):
style_val = find_first(style, "name").attrib["val"]
run_properties = find_first(style, "rPr")
styles_dict[style.attrib["styleId"]] = {
"style_name": style_val,
"default_run_properties": self._parse_run_properties(run_properties),
}
return styles_dict
def _parse_styles(self):
if self.styles_text is None:
return {}
tree = parse_xml_from_string(self.styles_text)
styles_dict = {}
for style in find_all(tree, 'style'):
style_val = find_first(style, 'name').attrib['val']
run_properties = find_first(style, 'rPr')
styles_dict[style.attrib['styleId']] = {
'style_name': style_val,
'default_run_properties': self._parse_run_properties(
run_properties,
),
}
return styles_dict
def _parse_styles(self):
if self.styles_text is None:
return {}
tree = parse_xml_from_string(self.styles_text)
styles_dict = {}
for style in find_all(tree, 'style'):
style_val = find_first(style, 'name').attrib['val']
run_properties = find_first(style, 'rPr')
styles_dict[style.attrib['styleId']] = {
'style_name':
style_val,
'default_run_properties':
self._parse_run_properties(run_properties, ),
}
return styles_dict
def _load(self):
self.document = WordprocessingDocument(path=None)
package = self.document.package
document_part = package.create_part(
uri='/word/document.xml',
)
if self.relationships:
for relationship in self.relationships:
target_mode = 'Internal'
if relationship['external']:
target_mode = 'External'
target_uri = relationship['target_path']
if 'data' in relationship:
full_target_uri = posixpath.join(
package.uri,
'word',
target_uri,
)
package.streams[full_target_uri] = BytesIO(
relationship['data'],
)
package.create_part(uri=full_target_uri)
document_part.create_relationship(
target_uri=target_uri,
target_mode=target_mode,
relationship_type=relationship['relationship_type'],
relationship_id=relationship['relationship_id'],
)
package.streams[document_part.uri] = BytesIO(self.document_xml)
package.create_relationship(
target_uri=document_part.uri,
target_mode='Internal',
relationship_type=MainDocumentPart.relationship_type,
)
self.numbering_root = None
if self.numbering_dict is not None:
self.numbering_root = parse_xml_from_string(
DXB.numbering(self.numbering_dict),
)
# This is the standard page width for a word document (in points), Also
# the page width that we are looking for in the test.
self.page_width = 612
self.parse_begin(self.document.main_document_part.root_element)
def test_get_image_sizes(self):
parser = XMLDocx2Html(
document_xml=self.get_xml(),
relationships=self.relationships,
)
tree = parse_xml_from_string(self.get_xml())
els = []
els.extend(find_all(tree, 'drawing'))
els.extend(find_all(tree, 'pict'))
image_ids = []
for el in els:
image_ids.append(parser._get_image_size(el))
expected = [
('40px', '20px'),
('41pt', '21pt'),
]
self.assertEqual(
set(image_ids),
set(expected),
)
def test_get_image_sizes(self):
parser = XMLDocx2Html(
document_xml=self.get_xml(),
rels_dict=self.relationship_dict,
)
tree = parse_xml_from_string(self.get_xml())
els = []
els.extend(find_all(tree, 'drawing'))
els.extend(find_all(tree, 'pict'))
image_ids = []
for el in els:
image_ids.append(parser._get_image_size(el))
expected = [
('40px', '20px'),
('41pt', '21pt'),
]
self.assertEqual(
set(image_ids),
set(expected),
)
def test_get_image_id(self):
parser = XMLDocx2Html(
document_xml=self.get_xml(),
rels_dict=self.relationship_dict,
)
tree = parse_xml_from_string(self.get_xml())
els = []
els.extend(find_all(tree, 'drawing'))
els.extend(find_all(tree, 'pict'))
image_ids = []
for el in els:
image_ids.append(parser._get_image_id(el))
expected = [
'rId0',
'rId1',
]
self.assertEqual(
set(image_ids),
set(expected),
)
def test_get_image_id(self):
parser = XMLDocx2Html(
document_xml=self.get_xml(),
relationships=self.relationships,
)
tree = parse_xml_from_string(self.get_xml())
els = []
els.extend(find_all(tree, 'drawing'))
els.extend(find_all(tree, 'pict'))
image_ids = []
for el in els:
image_ids.append(parser._get_image_id(el))
expected = [
'rId0',
'rId1',
]
self.assertEqual(
set(image_ids),
set(expected),
)
def root_element(self):
if self._root_element is None:
self._root_element = parse_xml_from_string(self.stream.read())
return self._root_element
def root_element(self):
if self._root_element is None:
self._root_element = parse_xml_from_string(self.stream.read())
return self._root_element
