def read_section(self, node: Any) -> None:
attrs = XmlPreprocessor.get_clear_attributes(node)
num_str = attrs.get('abstractNumId')
if not num_str:
return
sect_index = int(num_str)
sect = NumberingSetsSection()
restart_num_str = attrs.get('restartNumberingAfterBreak') or ''
sect.restart_after_break = restart_num_str != '0'
self.collections[sect_index] = sect
go_deeper = True
for child in node:
tag = XmlPreprocessor.get_clear_tag(child)
# find "<w:lvl w:ilvl="0">"
if tag == 'lvl':
self.read_numbering_sets(child, sect)
go_deeper = False
try:
if not go_deeper:
return
for child in node:
self.read_section(child)
finally:
sect.initialize()
def explore_paragraph_numpr(self, pr: DocParagraph, node: Any):
for elt in node:
tag = XmlPreprocessor.get_clear_tag(elt)
if tag == 'ilvl':
style_val = XmlPreprocessor.get_clear_attributes(elt).get('val')
if style_val:
pr.list_level = int(style_val)
elif tag == 'numId':
style_val = XmlPreprocessor.get_clear_attributes(elt).get('val')
if style_val:
pr.list_number = int(style_val)
def explore_paragraph_properties(self, pr: DocParagraph, node: Any):
for elt in node:
tag = XmlPreprocessor.get_clear_tag(elt)
if tag == 'pStyle':
style_val = XmlPreprocessor.get_clear_attributes(elt).get('val')
if style_val:
self.apply_paragraph_style(pr, style_val)
elif tag == 'numPr':
self.explore_paragraph_numpr(pr, elt)
else:
self.explore_paragraph_properties(pr, elt)
def add_hyperlink(self, node: Any, container: List[DocItem]) -> None:
text_child = XmlPreprocessor.find_children_by_tag(node,
't',
first_only=True)
if not text_child or not text_child[0].text:
return
link = self.make_hyperlink_instance()
link.text = text_child[0].text
# get link's reference
# r:id="rId4" -> self.relationship
rel_id = XmlPreprocessor.get_clear_attribute_val(node, 'id')
if rel_id:
rel_link = self.relationship.rel_by_id.get(rel_id) or ''
link.link = rel_link
container.append(link)
def explore_section(self, node: Any, style_set: StyleSets):
for child in node:
tag = XmlPreprocessor.get_clear_tag(child)
if tag == 'aliases':
val = XmlPreprocessor.get_clear_attribute_val(child, 'val')
aliases = set((val or '').split(','))
style_set.aliases = aliases
continue
if tag == 'ilvl':
val = XmlPreprocessor.get_clear_attribute_val(child, 'val')
style_set.ilvl = int(val)
continue
if tag == 'numId':
val = XmlPreprocessor.get_clear_attribute_val(child, 'val')
style_set.numId = int(val)
continue
self.explore_section(child, style_set)
def traverse_doc_tree(self, node: Any, items_container: List[DocItem]):
# fill "Items"
for elt in node:
tag = XmlPreprocessor.get_clear_tag(elt)
if tag == 'p':
self.add_paragraph(elt, items_container)
elif tag == 'tbl':
self.add_table(elt, items_container)
else:
self.traverse_doc_tree(elt, items_container)
def read_num_map(self, node: Any) -> None:
"""
<w:num w:numId="1">
<w:abstractNumId w:val="1" />
</w:num>
"""
num_id_str = XmlPreprocessor.get_clear_attribute_val(node, 'numId')
if not num_id_str:
return
num_id = int(num_id_str) # list number
for child in node:
tag = XmlPreprocessor.get_clear_tag(child)
if tag != 'abstractNumId':
continue
val_str = XmlPreprocessor.get_clear_attribute_val(child, 'val')
if not val_str:
break
sect_id = int(val_str)
self.list_num_to_section_id[num_id] = sect_id
return
def traverse_tree(self, node) -> None:
go_deeper = True
for child in node:
tag = XmlPreprocessor.get_clear_tag(child)
if tag == 'Relationship':
self.parse_relationship(child)
go_deeper = False
if go_deeper:
for child in node:
self.traverse_tree(child)
def read_numbering_sets(self, node: Any, sect: NumberingSetsSection):
nm_set = NumberingSets()
lvl = XmlPreprocessor.get_clear_attribute_val(node, 'ilvl')
nm_set.level = int(lvl)
for child in node:
tag = XmlPreprocessor.get_clear_tag(child)
if tag == 'start':
nm_set.start = int(
XmlPreprocessor.get_clear_attribute_val(child, 'val')
or '0')
continue
if tag == 'numFmt':
nm_set.num_fmt = \
XmlPreprocessor.get_clear_attribute_val(child, 'val') or 'bullet'
continue
if tag == 'lvlText':
nm_set.lvl_text = \
XmlPreprocessor.get_clear_attribute_val(child, 'val') or '*'
continue
if tag == 'lvlJc':
nm_set.lvl_jc = \
XmlPreprocessor.get_clear_attribute_val(child, 'val') or 'left'
continue
sect.sets.append(nm_set)
def fill_table_rows(self, tbl: DocTable, node: Any):
go_deeper = True
for elt in node:
tag = XmlPreprocessor.get_clear_tag(elt)
if tag == 'tr':
go_deeper = False
self.fill_table_row(tbl, elt)
if not go_deeper:
return
for elt in node:
self.fill_table_rows(tbl, elt)
def traverse_tree(self, node: Any) -> None:
# looking for "<w:style w:type="paragraph" w:styleId="Heading2">"
get_section = False
for child in node:
tag = XmlPreprocessor.get_clear_tag(child)
if tag == 'style':
self.read_section(child)
get_section = True
if get_section:
return
for child in node:
self.traverse_tree(child)
def fill_table_row(self, tbl: DocTable, node: Any):
row = DocTableRow()
tbl.rows.append(row)
go_deeper = True
for elt in node:
tag = XmlPreprocessor.get_clear_tag(elt)
if tag == 'tc':
go_deeper = False
self.fill_table_cell(row, elt)
if not go_deeper:
return
for elt in node:
self.fill_table_row(tbl, elt)
def explore_paragraph(self, pr: DocParagraph, node: Any):
for child in node:
tag = XmlPreprocessor.get_clear_tag(child)
if tag == 'pPr':
self.explore_paragraph_properties(pr, child)
elif tag == 'hyperlink':
self.add_hyperlink(child, pr.text_items)
elif tag == 'tab':
pr.text_items.append(DocText('\t'))
elif tag == 't':
if child.text:
pr.text_items.append(DocText(child.text))
elif tag == 'r':
self.explore_paragraph(pr, child)
else:
self.explore_paragraph(pr, child)
def traverse_tree(self, node: Any) -> None:
# looking for "<w:abstractNum w:abstractNumId="1" restartNumberingAfterBreak="0">"
get_section = False
get_num_map = False
for child in node:
tag = XmlPreprocessor.get_clear_tag(child)
if tag == 'abstractNum':
self.read_section(child)
get_section = True
elif tag == 'num':
get_num_map = True
self.read_num_map(child)
if get_section and get_num_map:
return
for child in node:
self.traverse_tree(child)
def read_section(self, node: Any) -> None:
"""
<w:style w:type="paragraph" w:styleId="Heading2">
<w:aliases w:val="1.1,2nd,B Sub/Bold,B Sub/Bold1,B Sub/Bold11,B Sub/Bold12,B Sub/Bold13" />
<w:pPr>
<w:keepNext />
<w:numPr>
<w:ilvl w:val="1" />
<w:numId w:val="19" />
</w:numPr>
</w:pPr>
</w:num>
"""
style_id = XmlPreprocessor.get_clear_attribute_val(node, 'styleId')
if not style_id:
return
style_set = StyleSets()
style_set.styleId = style_id
self.sets[style_id] = style_set
try:
self.explore_section(node, style_set)
except:
del self.sets[style_id]
def test_remove_namespace(self):
text = '{http://schemas.openxmlformats.org/wordprocessingml/2006/main"}val'
self.assertEqual('val', XmlPreprocessor.remove_namespace(text))
self.assertEqual('val', XmlPreprocessor.remove_namespace('val'))
