LongArticlesJob-fork/tasks/task2.py

"""
@author: luojunhui
"""

from applications import AIDTApi, DeNetMysql, PQMySQL, Functions
from config import poolLevelConfig, cateMap, coldPoolArticlesNum


class ColdStartTask(object):
    """
    冷启分配任务
    """
    AidApi = AIDTApi()
    DeMysql = DeNetMysql()
    PqMysql = PQMySQL()
    Fun = Functions()
    pool4 = poolLevelConfig['1']

    @classmethod
    def getTopArticles(cls, category, limit_count):
        """
        获取高分享的文章list
        :return:
        """
        sql = f"""
            select content_id, content_link, title 
            from cold_start_article_pool
            where category = '{category}'
            order by view_count DESC, publish_time_stamp DESC
            limit {limit_count};
        """
        result = cls.PqMysql.select(sql)
        return result

    @classmethod
    def computeScore(cls):
        """
        和每个账号计算相关性分数
        :return:
        """
        category_list = ["军事政法", "健康养生", "宗教历史"]
        L = []
        for category in category_list:
            limit_count = coldPoolArticlesNum * cateMap.get(category, 0.1)
            article_tuple = cls.getTopArticles(category, int(limit_count))
            title_list = [article[2] for article in article_tuple]
            score_list = cls.Fun.getTitleScore(title_list, "指尖奇文")['指尖奇文']['score_list']
            for index, score in enumerate(score_list):
                obj = {
                    "id": article_tuple[index][0],
                    "url": article_tuple[index][1],
                    "title": article_tuple[index][2],
                    "cate": category,
                    "score": score
                }
                L.append(obj)
        result = [i for i in L if i['score'] >= 0.35]
        return result

    @classmethod
    def sendToColdPool(cls, plan_id, plan_name, plan_tag):
        """
        把文章send至第四层
        :return:
        """
        result = cls.computeScore()
        army = [i for i in result if i['cate'] == '军事政法']
        healthy = [i for i in result if i['cate'] == '健康养生']
        history = [i for i in result if i['cate'] == '宗教历史']
        url_list = [i['url'] for i in result]
        # 再加一次配比，每个品类的数量占比
        cls.AidApi.updateArticleIntoCrawlerPlan(
            plan_id=plan_id,
            plan_name=plan_name,
            plan_tag=plan_tag,
            url_list=[i['url'] for i in result]
        )


if __name__ == '__main__':
    CST = ColdStartTask()
    CST.sendToColdPool(
        plan_id=None,
        plan_name="冷启池子--0729--Monday--分品类抓取--6个品类",
        plan_tag="autoArticlePoolLevel1",
    )