def search(args):
keyword = args.get('keyword')
# Store history in database
if SearchItem.objects(keyword=keyword).count() == 0:
search_item = SearchItem(keyword=keyword)
else:
search_item = SearchItem.objects(keyword=keyword).get()
search_item.count += 1
search_item.save()
# Load cache
# search_results = g.get('search_results', None)
# if search_results is None:
# g.search_results = {}
# search_results = g.search_results
#
# search_id_to_results = g.get('search_id_to_results', None)
# if search_id_to_results is None:
# g.search_id_to_results = {}
# search_id_to_results = g.search_id_to_results
query_result = PaperProcessor(keyword)
papers = query_result.papers_array
# paper_ids = [x["DBID"] for x in papers]
# search_item.update(add_to_set__papers=paper_ids)
if flask_login.current_user.is_authenticated:
search_history = SearchHistory(item=search_item,
user=User.objects(id=flask_login.current_user.id).get(),
papers=[x for x in query_result.papers_array])
else:
search_history = SearchHistory(item=search_item,
papers=[x for x in query_result.papers_array])
search_history.save()
# # Word bag
# bag = AbstractProcessor().process_list(return_list)
# words = [[y, bag[y]] for y in sorted(list(bag.keys()), key=lambda x: bag[x], reverse=True)[:30]]
# Return result
return jsonify(
response=str(search_history.id),
meta_info={
'page_count': math.ceil(len(papers)/RESULTS_PER_PAGE)
}
)
def jump(args):
try:
search_history_id = args.get('search_history_id')
paper_id = args.get('paper_id')
search_history = SearchHistory.objects(id=search_history_id).get()
search_item = search_history.item
paper = Paper.objects(id=paper_id).get()
click_history = ClickHistory(
search_item=search_item,
search_history=search_history,
paper=paper,
user=User.objects(id=flask_login.current_user.id).get() if flask_login.current_user.is_authenticated else None
)
click_history.save()
if ClickCount.objects(search_item=search_item, paper=paper).count() > 0:
click_count = ClickCount.objects(search_item=search_item, paper=paper).get()
else:
click_count = ClickCount(
search_item=search_item, paper=paper
)
click_count.count = click_count.count + 1
click_count.save()
return redirect(paper.url)
except Exception as e:
logging.warning(e)
abort(401)
def fetch(args):
# Get request arguments
page = int(args.get('page', 1))
order_by = args.get('order_by', '')
filter_by = args.get('filter_by', '')
search_history_id = args.get("search_history_id")
search_history = SearchHistory.objects(id=search_history_id).get()
papers = search_history.papers
if page <= 0 or (page - 1) * RESULTS_PER_PAGE >= len(papers):
logging.warning("No more papers.")
return jsonify(
response=None,
error="No more papers"
)
# Filtering
# TODO : Outdated
# logging.debug("Order by : " + order_by)
# if order_by == "Default":
# # logging.debug("Order by default")
# query_result.papers_array.sort(key=lambda x: x["Score"], reverse=True)
# else:
# for paper in query_result.papers_array:
# try:
# paper["ParsedDate"] = datetime.datetime.strptime(paper["PubDate"], "%Y %b %d").strftime("%Y%m%d")
# except:
# try:
# paper["ParsedDate"] = datetime.datetime.strptime(paper["PubDate"], "%Y %b").strftime("%Y%m%d")
# except:
# try:
# paper["ParsedDate"] = datetime.datetime.strptime(paper["PubDate"], "%Y").strftime("%Y%m%d")
# except:
# paper["ParsedDate"] = "19000000"
#
# if order_by == "Publication Date(Ascending)":
# query_result.papers_array.sort(key=lambda x: x["ParsedDate"])
#
# if order_by == "Publication Date(Descending)":
# query_result.papers_array.sort(key=lambda x: x["ParsedDate"], reverse=True)
#
# logging.debug("Filter by : " + filter_by)
#
# if filter_by != "Default":
# return_list = [paper for paper in query_result.papers_array if
# paper["Abstract"].lower().find(filter_by) != -1]
# else:
# return_list = query_result.papers_array
return jsonify(
response=papers_searilizer(search_history.papers)
)
def recommend_from_user(user):
bag = defaultdict(int)
for subscription in user.subscriptions:
all_tokens = []
papers = subscription.papers
num_paper = len(papers)
# voc = Vocabulary()
for paper in papers:
try:
abstract = paper.abstract
words = list(set([x.strip().strip(",.").lower() for x in abstract.split()]))
for word in words:
if word not in common_words:
bag[word] += 1
except Exception as e:
logging.warning(e)
for paper in papers:
# Map to terms
abstract = paper.abstract
tokens = ''.join(c for c in abstract if c.isalnum() or c.isspace()).split()
tokens.extend([p.plural(token) for token in tokens])
# two_word = [ " ".join(tokens[i:i+2]) for i in range(len(tokens)-1)]
# two_word_plural = [ p.plural(w) for w in two_word ]
# tokens.extend(two_word)
# tokens.extend(two_word_plural)
tokens.extend([t[0].upper()+t[1:] if t[0].islower() else t[0].lower()+t[1:] for t in tokens])
all_tokens.extend(tokens)
# print(Term.objects(name__in=all_tokens).count())
all_tokens = set(all_tokens)
try:
all_tokens.remove('toes')
all_tokens.remove('Toes')
except:
pass
terms = [term.name for term in Term.objects(name__in=all_tokens)]
for term in terms:
bag[term] += num_paper
for history in SearchHistory.objects(user=user.id):
keyword = history.item.keyword
bag[keyword] += 100
return [(x, bag.__getitem__(x)) for x in bag]
# return [(x, bag.__getitem__(x)) for x in sorted(bag, key=bag.__getitem__, reverse=True)]
