def process_item(self, item, spider): item_url = item['url'] with db_session: estateEntity = EstateEntity.get(url = item_url) if estateEntity: print('already have this url item') return estateEntity = EstateEntity( url = item_url, published_at = item['published_at'], website = item['website'], location = item['location'], html = item['html'] ) print('url: ', item_url) print('save post')