def pre_load(self, data, *args, **kwargs):
data = data["exchange-document"]
bib = data["bibliographic-data"]
del data["bibliographic-data"]
data["publications"] = resolve_list(
bib, "publication-reference.document-id")
ipc_class = resolve_list(bib,
"classifications-ipcr.classification-ipcr")
data["ipc_classes"] = [c["text"] for c in ipc_class]
classifications = resolve_list(
bib, "patent-classifications.patent-classification")
if classifications:
data["cpc_classes"] = [
c for c in classifications
if resolve(c, "classification-scheme.@scheme") == "CPCI"
]
data["us_classes"] = [
c for c in classifications
if resolve(c, "classification-scheme.@scheme") == "UC"
]
data["applications"] = resolve_list(
bib, "application-reference.document-id")
data["priority_claims"] = self.pre_load_priority_claims(bib)
data["applicants"] = resolve_list(bib, "parties.applicants.applicant")
data["inventors"] = resolve_list(bib, "parties.inventors.inventor")
titles = resolve(bib, "invention-title")
if isinstance(titles, list):
data["title"] = next(t["#text"] for t in titles
if t["@lang"] == "en")
elif isinstance(titles, str):
data["title"] = titles
else:
data["title"] = titles["#text"]
return data
def __len__(self):
page = self.get_page(0)
max_length = int(resolve(page, "@total-result-count"))
limit = self.config["limit"]
if limit:
return limit if limit < max_length else max_length
else:
return max_length
def get_page(self, page_number):
if page_number not in self.pages:
query_params = self.query_params(page_number)
response = session.get(self.search_url,
params=query_params,
timeout=10)
data = xmltodict.parse(response.text,
process_namespaces=True,
namespaces=NS)
self.pages[page_number] = resolve(
data, "world-patent-data.biblio-search")
return self.pages[page_number]
def pre_load_priority_claims(self, bib):
pcs = resolve(bib, "priority-claims.priority-claim")
out = list()
pcs = (pcs if isinstance(pcs, list) else [
pcs,
])
for pc in pcs:
if isinstance(pc["document-id"], list):
doc_id = pc["document-id"][0]
else:
doc_id = pc["document-id"]
pc = {**pc, **doc_id}
del pc["document-id"]
out.append(pc)
return out
def result_gen(offset, limit):
num_pages = math.ceil(len(self) / self.page_size)
page_num = int(offset / self.page_size)
counter = page_num * self.page_size
while page_num < num_pages:
page_data = self.get_page(page_num)
data = resolve(page_data, self.item_path)
if not isinstance(data, list):
yield counter, data
counter += 1
else:
for item in data:
if not self.config[
"limit"] or counter < self.config["limit"]:
yield counter, item
counter += 1
page_num += 1