def update(item):
keyword = item.keyword
papers_existing_in_item = [x.id for x in item.papers]
p = PubMedFetcher(keyword, num_of_documents=10, sort="pub+date")
for paper in p.papers.values():
paper_mongo = store_paper(paper)
if paper_mongo.id not in papers_existing_in_item:
item.update(push__papers=paper_mongo)
paper_mongo.update(push__subscriptions=item)
def process_one(self, item):
if len(item) < 30:
return None
m = re.search(
'pmid (\d+).+?year (\d+).+?month (\d+).+?day (\d+).+?title.+?name "(.+?)".+?authors \{(.+?)\},\s*from journal.+?name "(.+?)".+?abstract "(.+?)"',
item,
re.DOTALL,
)
error_count = 1
if m and m.group(6):
id = m.group(1)
title = m.group(5).replace("\n", "").strip()
author = m.group(6)
m_author = re.findall('name ml "(.+?)"', author)
journal = m.group(7)
year = int(m.group(2))
month = int(m.group(3))
day = int(m.group(4))
abstract = m.group(8).replace("\n", "").strip()
if not m_author:
return None
h = {
"Source": "PubMed",
"PMID": id,
"Title": title,
"Author": m_author,
"Journal": journal,
"Year": year,
"Date": datetime(year, month, day),
"Abstract": abstract,
}
h["URL"] = "http://www.ncbi.nlm.nih.gov/pubmed/" + h["PMID"]
try:
paper_mongo = store_paper(h)
h["DBID"] = str(paper_mongo.id)
except:
logging.error("Store paper failed: {}".format(h["PMID"]))
return h
else:
logging.warning("Parse error. #%d", error_count)
error_count += 1
return None
