def process_item(self, item, domain):
now = arrow.now()
seen = self.check_seen_before(item)
if len(seen) > 0:
last_seen = max(seen)
time_limit = now.replace(**self.time_scale).timestamp
if last_seen < time_limit:
self.insert_item_price(item, now.timestamp)
raise DropItem("Already seen %s, %s" % (item['url'], arrow.get(last_seen).humanize()))
else:
self.insert_item_price(item, now.timestamp)
self.insert_item_main(item)
self.insert_item_tag_list(item)
self.insert_item_description(item)
self.conn.commit()
return item
评论列表
文章目录