def get_classcode(patent_document):
classcode = ""
sdobi = patent_document[0]
for index, sub_node in enumerate(sdobi.iter('classification-ipcr')):
if len(classcode.split()) == 4:
break
for class_ in sub_node:
if class_.tag == 'text':
temp = th.get_node_value(class_)
temp = temp.replace(" ", "")
temp = temp.replace("/", "")
if len(temp) > 3:
if classcode == "":
classcode = temp[:4]
elif temp[:4] not in classcode:
classcode += " " + temp[:4]
if classcode == "":
for sub_node in sdobi.iter('B510'):
for class_ in sub_node:
temp = th.get_node_value(class_)
temp = temp.replace(" ", "")
temp = temp.replace("/", "")
if len(temp) > 3:
if classcode == "":
classcode = temp[1:5]
elif temp[1:5] not in classcode:
classcode += " " + temp[1:5]
return classcode
def get_citations(patent_document): # B500, B550, B560
citations = ""
sdobi = patent_document[0]
for node in sdobi.iter('B560'):
# snm, iid (number of opponent), irf (), adr
for sub_node in node:
if sub_node.tag == 'B561':
for cit in sub_node:
if cit.tag == 'text':
if len(citations) != 0:
temp_citation = th.get_node_value(cit)
temp_citation = temp_citation.replace(" ", "")
temp_citation = temp_citation[temp_citation.
rfind('-') + 1:]
citations += " " + temp_citation
else:
temp_citation = th.get_node_value(cit)
temp_citation = temp_citation.replace(" ", "")
temp_citation = temp_citation[temp_citation.
rfind('-') + 1:]
citations += temp_citation
elif sub_node.tag == 'B565EP':
for cit in sub_node:
if cit.tag == 'date':
if len(citations) != 0:
citations += " " + th.get_node_value(cit)
else:
citations += th.get_node_value(cit)
return citations
def get_country(patent_document):
country = ""
sdobi = patent_document[0]
for sub_node in sdobi.iter('B330'):
for ctry in sub_node:
if ctry.tag == 'ctry' and th.get_node_value(ctry) not in country:
if len(country) != 0:
country += " " + th.get_node_value(ctry)
else:
country += th.get_node_value(ctry)
return country
def get_applicant(patent_document):
applicant = ""
sdobi = patent_document[0]
for sub_node in sdobi.iter('B711'):
# snm, iid (number of opponent), irf (), adr
for appl in sub_node:
if appl.tag == 'snm':
if len(applicant) != 0:
applicant += "," + th.get_node_value(appl)
else:
applicant += th.get_node_value(appl)
return applicant
def get_alternative_title(node):
text = list(
map(
lambda sub_node: th.get_node_value(sub_node.getnext()),
filter(
lambda sub_node: th.get_node_value(sub_node).upper() == 'EN',
node.iter('B541'))))
if text:
if None in text:
text.remove(None)
return th.get_string_from_list(text, " ")
return ""
def get_alternative_text(patent_document, marker):
text = list(map(lambda sub_node : th.get_node_value(sub_node), filter(lambda node: node.tag == marker, patent_document)))
if text:
if None in text:
text.remove(None)
return th.get_string_from_list(text, " ")
return ""
def get_nested_text(node):
text = ""
if not node.tag is etree.Comment:
remove_tags(node)
text = th.get_node_value(node)
if text == None:
return ""
return text
def get_text(patent_document, marker):
for node in patent_document:
if node.tag == marker:
# print("original: ", th.get_node_value(node))
return th.get_node_value(node)
return ""
def get_title(node):
text = ""
for sub_node in node.iter('B541'):
if th.get_node_value(sub_node).upper() == 'EN':
text = th.get_node_value(sub_node.getnext())
return text