def get_relationships(self):
if self.relationships:
return self.relationships
doc_path = os.path.join(self.file_path, "word/_rels/document.xml.rels")
with open(doc_path, encoding="UTF-8") as f:
doc = f.read()
doc = BeautifulSoup(doc, "xml")
self.relationships = Relationships(doc)
return self.relationships
def _write_chartsheet_rels_files(self):
# Write the chartsheet .rels files for links to drawing files.
index = 0
for worksheet in self.workbook.worksheets():
if not worksheet.is_chartsheet:
continue
index += 1
external_links = worksheet.external_drawing_links
if not external_links:
continue
# Create the chartsheet .rels xlsx_dir.
rels = Relationships()
for link_data in external_links:
rels._add_worksheet_relationship(*link_data)
# Create .rels file such as /xl/chartsheets/_rels/sheet1.xml.rels.
rels._set_xml_writer(self._filename('xl/chartsheets/_rels/sheet'
+ str(index) + '.xml.rels'))
rels._assemble_xml_file()
def _write_worksheet_rels_files(self):
# Write data such as hyperlinks or drawings.
index = 0
for worksheet in self.workbook.worksheets():
if worksheet.is_chartsheet:
continue
index += 1
external_links = (worksheet.external_hyper_links +
worksheet.external_drawing_links +
worksheet.external_vml_links +
worksheet.external_table_links +
worksheet.external_comment_links)
if not external_links:
continue
# Create the worksheet .rels dirs.
rels = Relationships()
for link_data in external_links:
rels._add_worksheet_relationship(*link_data)
# Create .rels file such as /xl/worksheets/_rels/sheet1.xml.rels.
rels._set_xml_writer(self._filename('xl/worksheets/_rels/sheet'
+ str(index) + '.xml.rels'))
rels._assemble_xml_file()
def _write_vml_drawing_rels_file(self, worksheet, index):
# Write the vmlDdrawing .rels files for worksheets with images in
# headers or footers.
# Create the drawing .rels dir.
rels = Relationships()
for drawing_data in worksheet.vml_drawing_links:
rels._add_document_relationship(*drawing_data)
# Create .rels file such as /xl/drawings/_rels/vmlDrawing1.vml.rels.
rels._set_xml_writer(self._filename('xl/drawings/_rels/vmlDrawing'
+ str(index)
+ '.vml.rels'))
rels._assemble_xml_file()
def _find_relationships(self, list_tagged, global_entities):
relationships = Relationships()
relation_stops_type = ['CONJ', 'WPRO', ',', '(', ')']
relationship_stop_words = ['ex']
for tagged in list_tagged:
for index_sentence, sentence in enumerate(tagged):
last_entity = None
last_entity_index = 0
last_relation = None
for index, item in enumerate(sentence):
# In order to avoid stop words
if( len(item[0]) == 1 or item[0].lower() in relationship_stop_words):
continue
# to get the entity already identified
elif( item[1] == 'NE'):
# In order to build the relationship
if(last_entity is not None and self._contain_main_entity(last_entity[0], item[0])):
# to build a relationship with anything between entities
# just if there is only one token between entities
if(index-last_entity_index == 2 and len(sentence[index-1][0])>1 ):
id1 = self._search_parent_entity(last_entity[2], global_entities).id()
id2 = self._search_parent_entity(item[2], global_entities).id()
relation = (sentence[index-1][0], id1, last_entity[0], id2, item[0])
relationships.add(relation)
# In order to build a relationship from relation already identified
elif(last_relation is not None):
id1 = self._search_parent_entity(last_entity[2], global_entities).id()
id2 = self._search_parent_entity(item[2], global_entities).id()
relation = (last_relation, id1, last_entity[0], id2, item[0])
relationships.add(relation)
last_entity = item
last_entity_index = index
last_relation = None
# In order to get just relationships between entities
if(last_entity is None):
continue
# In order to get relationship composed by verb and noun
elif('N' in item[1]):
last_relation = self._compose_verb_noun(sentence, index, last_entity_index, relation_stops_type)
# In order to get relationship composed by one or more verbs
elif('VB' in item[1]):
last_relation = self._get_composed_verbs(sentence, index, last_entity_index, relation_stops_type)
# In order to break relationships
elif(item[1] in relation_stops_type):
last_relation = None
last_entity = None
last_entity_index = 0
# In order to remove relationships if a conjuction is found
if last_relation is not None and last_relation[0].isupper():
last_relation = None
return relationships
def _write_drawing_rels_files(self):
# Write the drawing .rels files for worksheets with charts or drawings.
index = 0
for worksheet in self.workbook.worksheets():
if not worksheet.drawing_links:
continue
index += 1
# Create the drawing .rels xlsx_dir.
rels = Relationships()
for drawing_data in worksheet.drawing_links:
rels._add_document_relationship(*drawing_data)
# Create .rels file such as /xl/drawings/_rels/sheet1.xml.rels.
rels._set_xml_writer(self._filename('xl/drawings/_rels/drawing'
+ str(index) + '.xml.rels'))
rels._assemble_xml_file()
def _write_workbook_rels_file(self):
# Write the _rels/.rels xml file.
rels = Relationships()
worksheet_index = 1
chartsheet_index = 1
for worksheet in self.workbook.worksheets():
if worksheet.is_chartsheet:
rels._add_document_relationship('/chartsheet',
'chartsheets/sheet'
+ str(chartsheet_index)
+ '.xml')
chartsheet_index += 1
else:
rels._add_document_relationship('/worksheet',
'worksheets/sheet'
+ str(worksheet_index)
+ '.xml')
worksheet_index += 1
rels._add_document_relationship('/theme', 'theme/theme1.xml')
rels._add_document_relationship('/styles', 'styles.xml')
# Add the sharedString rel if there is string data in the workbook.
if self.workbook.str_table.count:
rels._add_document_relationship('/sharedStrings',
'sharedStrings.xml')
# Add vbaProject if present.
if self.workbook.vba_project:
rels._add_ms_package_relationship('/vbaProject', 'vbaProject.bin')
rels._set_xml_writer(self._filename('xl/_rels/workbook.xml.rels'))
rels._assemble_xml_file()
def _write_root_rels_file(self):
# Write the _rels/.rels xml file.
rels = Relationships()
rels._add_document_relationship('/officeDocument', 'xl/workbook.xml')
rels._add_package_relationship('/metadata/core-properties',
'docProps/core.xml')
rels._add_document_relationship('/extended-properties',
'docProps/app.xml')
rels._set_xml_writer(self._filename('_rels/.rels'))
rels._assemble_xml_file()
class Docx(IdAble):
def __init__(self, path):
super(Docx, self).__init__()
if path is None or not isinstance(path, str):
raise Exception("Path is not allowed None")
if not os.path.exists(TEMP_BASE_DIR):
try:
os.mkdir(TEMP_BASE_DIR)
except FileExistsError as e:
pass
self.document = None
self.content_types = None
self.relationships = None
self.numbering = None
self.styles = None
self.base_dir = uuid1().hex
file = ZipFile(path)
self.file_path = os.path.join(TEMP_BASE_DIR, self.base_dir)
os.mkdir(self.file_path)
file.extractall(self.file_path)
file.close()
self.get_document()
self.get_content_types()
self.get_numbering()
self.get_relationships()
self.get_styles()
def get_numbering(self):
if self.numbering:
return self.numbering
numbering_path = os.path.join(self.file_path, "word/numbering.xml")
if not os.path.exists(numbering_path):
self.numbering = Numbering()
return self.numbering
with open(numbering_path, encoding="UTF-8") as f:
numbering = f.read()
numbering = BeautifulSoup(numbering, "xml")
self.numbering = Numbering(numbering)
return self.numbering
def get_document(self):
if self.document:
return self.document
doc_path = os.path.join(self.file_path, "word/document.xml")
with open(doc_path, encoding="UTF-8") as f:
document = f.read()
document = BeautifulSoup(document, "xml")
self.document = Document(document)
return self.document
def get_relationships(self):
if self.relationships:
return self.relationships
doc_path = os.path.join(self.file_path, "word/_rels/document.xml.rels")
with open(doc_path, encoding="UTF-8") as f:
doc = f.read()
doc = BeautifulSoup(doc, "xml")
self.relationships = Relationships(doc)
return self.relationships
def get_content_types(self):
if self.content_types:
return self.content_types
content_path = os.path.join(self.file_path, "[Content_Types].xml")
with open(content_path, encoding="UTF-8") as f:
content_types = f.read()
content_types = BeautifulSoup(content_types, "xml")
self.content_types = ContentTypes(content_types)
return self.content_types
def get_styles(self):
if self.styles:
return self.styles
style_path = os.path.join(self.file_path, "word/styles.xml")
with open(style_path, encoding="UTF-8") as f:
styles = f.read()
styles = BeautifulSoup(styles, "xml")
self.styles = Styles(styles)
return self.styles
def extract_media_files(self, path):
relationships = self.get_relationships()
file_mapping = relationships.get_file_mapping()
template = "cp {} {}"
base_dir = os.path.join(self.file_path, "word")
#print(file_mapping)
for file in file_mapping.keys():
from_file = os.path.join(base_dir, file)
to_file = os.path.join(path, file_mapping[file])
dir_name = os.path.dirname(to_file)
if not os.path.exists(dir_name):
os.makedirs(dir_name)
extract = template.format(from_file, to_file)
os.system(extract)
def merge(self, doc, page=False):
if not isinstance(doc, Docx):
raise Exception("merge parameter is not docx")
source_content_types = doc.get_content_types()
self.get_content_types().merge_content_types(source_content_types)
source_relationships = doc.get_relationships()
#print(source_relationships.get_file_mapping())
source_relationships.generate_id(doc.id)
doc.extract_media_files(os.path.join(self.file_path, "word"))
self.get_relationships().merge_relationships(source_relationships)
source_styles = doc.get_styles()
source_styles.generate_id(doc.id)
self.styles.merge(source_styles)
source_numberings = doc.get_numbering()
source_numberings.generate_id(doc.num)
self.numbering.merge(source_numberings)
source_document = doc.get_document()
source_document.generate_id(doc.id, doc.num)
self.get_document().merge(source_document, page)
def save(self, name):
import zipfile
self._save_document()
self._save_content_types()
self._save_relationships()
self._save_numbering()
self._save_styles()
file = ZipFile(name, "w", compression=zipfile.ZIP_DEFLATED)
for base, children, files in os.walk(self.file_path):
base_name = base.split(self.base_dir)[-1]
for f in files:
zip_path = os.path.join(base_name, f)
real_path = os.path.join(base, f)
file.write(real_path, zip_path)
file.close()
def _save_document(self):
with open(os.path.join(self.file_path, "word/document.xml"),
mode="w",
encoding="UTF-8") as f:
f.write(str(self.document.get_dom()))
def _save_content_types(self):
with open(os.path.join(self.file_path, "[Content_Types].xml"),
mode="w",
encoding="UTF-8") as f:
f.write(str(self.content_types.get_dom()))
def _save_relationships(self):
with open(os.path.join(self.file_path, "word/_rels/document.xml.rels"),
mode="w",
encoding="UTF-8") as f:
f.write(str(self.relationships.get_dom()))
def _save_numbering(self):
numbering = self.numbering.get_dom()
if not numbering:
return
numbering_path = os.path.join(self.file_path, "word/numbering.xml")
with open(numbering_path, "w+", encoding="UTF-8") as f:
f.write(str(numbering))
def _save_styles(self):
with open(os.path.join(self.file_path, "word/styles.xml"),
"w+",
encoding="UTF-8") as f:
f.write(str(self.styles.get_dom()))
def append_paragraph(self, text, align="left"):
self.document.append_paragraph(text, align)
def append_picture(self, filepath, align="left"):
if not os.path.exists(filepath):
return
media_dir = os.path.join(self.file_path, "word/media")
if not os.path.exists(media_dir):
os.mkdir(media_dir)
suffix = filepath.split(".")[-1]
self.content_types.append_extension(suffix)
id_file = self.relationships.append_relationship(suffix)
#print(id_file)
file_path = os.path.join(
self.file_path,
"word/media/{filename}".format(filename=id_file["filename"]))
os.system("cp {f_file} {t_file}".format(f_file=filepath,
t_file=file_path))
img = Image.open(file_path)
width, height = img.size
img.close()
self.document.append_picture(id_file["rid"], width * 6350,
height * 6350, align)
def close(self):
os.system("rm -rf {0}".format(self.file_path))
from relationships import Relationships from relationshipstats import RelationshipStats from history import History Relationships.create_table(fail_silently=True) RelationshipStats.create_table(fail_silently=True) History.create_table(fail_silently=True)
