title_with_video/applications/match_alg/rank.py

"""
@author: luojunhui
"""
from applications.match_alg.recall import recall_videos
from applications.log import logging


def jac_score(d1, d2):
    """
    通过交并集来判断
    :param d1:
    :param d2:
    :return:
    """
    f1_keys = set(d1["key_words"])
    f2_keys = set(d2["key_words"])
    keys_union = f1_keys | f2_keys
    keys_intersection = f1_keys & f2_keys
    f1_search_keys = set(d1["search_keys"])
    f2_search_keys = set(d2["search_keys"])
    search_keys_union = f1_search_keys | f2_search_keys
    search_keys_intersection = f1_search_keys & f2_search_keys
    f1_extra_keys = set(d1["extra_keys"])
    f2_extra_keys = set(d2["extra_keys"])
    extra_keys_union = f1_extra_keys | f2_extra_keys
    extra_keys_intersection = f1_extra_keys & f2_extra_keys
    score_1 = len(keys_intersection) / len(keys_union)
    score_2 = len(search_keys_intersection) / len(search_keys_union)
    score_3 = len(extra_keys_intersection) / len(extra_keys_union)
    return score_1 * 0.4 + score_2 * 0.4 + score_3 * 0.2, d2['video_id']


async def best_choice(params_obj, trace_id, search_videos):
    """
    计算，返回出最合适的 video_id
    :return: video_id
    """
    pq_list, search_list = await recall_videos(trace_id=trace_id, s_videos=search_videos)

    def best_video_id(target_list):
        """
        :param target_list:
        :return:
        """
        score_list = []
        for video_obj in target_list:
            try:
                score, video_id = jac_score(d1=params_obj, d2=video_obj)
                score_list.append((video_id, score))
            except Exception as e:
                print(e)
        sorted_list = sorted(score_list, key=lambda x: x[1], reverse=True)
        return sorted_list[0] if sorted_list else (0, 0)

    if search_list:
        logging(
            code="1003",
            info="Return Best Search Video",
            data=search_list,
            trace_id=trace_id
        )
        return search_list[0]
        # return best_video_id(search_list)[0]
        # best_search_tuple = best_video_id(search_list)
        # if best_search_tuple[1] > 0:
        #     logging(
        #         code="1003",
        #         info="search_score---{}".format(best_search_tuple[1]),
        #         trace_id=trace_id
        #     )
        #     return best_search_tuple[0]
        # else:
        #     best_pq_tuple = best_video_id(pq_list)
        #     if best_pq_tuple[1] > 0:
        #         logging(
        #             code="1003",
        #             info="pq_score---{}".format(best_pq_tuple[1]),
        #             trace_id=trace_id
        #         )
        #         return best_pq_tuple[0]
        #     else:
        #         return None
    else:
        best_pq_tuple = best_video_id(pq_list)
        if best_pq_tuple[1] > 0:
            logging(
                code="1003",
                info="pq_score---{}".format(best_pq_tuple[1]),
                trace_id=trace_id
            )
            return best_pq_tuple[0]
        else:
            return None