LongArticleTaskServer/applications/tasks/analysis_task/crawler_detail.py

from applications.api import feishu_robot


class CrawlerDetailAnalysisConst:
    CATEGORY_LIST = [
        "知识科普",
        "国家大事",
        "历史人物",
        "奇闻趣事",
        "名人八卦",
        "怀旧时光",
        "情感故事",
        "社会法治",
        "现代人物",
        "社会现象",
        "健康养生",
        "家长里短",
        "军事历史",
        "财经科技",
        "政治新闻",
    ]

    TRANSFORMED_STATUS = 1
    NOT_TRANSFORMED_STATUS = 0

    CRAWLER_DETAIL_TASK_PLATFORM = "crawler_detail_by_platform"
    CRAWLER_DETAIL_TASK_CATEGORY = "crawler_detail_by_category"

    TRANSFORM_DETAIL_TASK_PLATFORM = "transform_detail_by_platform"
    TRANSFORM_DETAIL_TASK_CATEGORY = "transform_detail_by_category"


class CrawlerDetailBase(CrawlerDetailAnalysisConst):
    def __init__(self):
        super().__init__()

    @staticmethod
    def create_feishu_column_map() -> dict:
        date_column = feishu_robot.create_feishu_columns_sheet(
            sheet_type="plain_text", sheet_name="dt", display_name="日期"
        )
        category_column = feishu_robot.create_feishu_columns_sheet(
            sheet_type="plain_text", sheet_name="category", display_name="文章品类"
        )
        platform_column = feishu_robot.create_feishu_columns_sheet(
            sheet_type="plain_text", sheet_name="platform", display_name="抓取渠道"
        )
        video_cnt_column = feishu_robot.create_feishu_columns_sheet(
            sheet_type="number", sheet_name="video_count", display_name="视频数量"
        )
        avg_score_column = feishu_robot.create_feishu_columns_sheet(
            sheet_type="number",
            sheet_name="average_similarity_score",
            display_name="相关性分均值",
        )
        return {
            "dt": date_column,
            "category": category_column,
            "platform": platform_column,
            "video_count": video_cnt_column,
            "average_similarity_score": avg_score_column,
        }


class CrawlerArticleDetailAnalysis(CrawlerDetailBase):
    pass
    # raise NotImplementedError


class CrawlerVideoDetailAnalysis(CrawlerDetailBase):
    def __init__(self, pool, trace_id):
        super().__init__()
        self.pool = pool
        self.trace_id = trace_id

    async def get_crawler_videos_by_platform(self, start_date, end_date):
        """
        获取 start_dt && end_dt 之间每个渠道抓取的视频数量
        """
        query = """
            SELECT CAST(DATE(FROM_UNIXTIME(crawler_timestamp)) AS CHAR) AS dt, platform, count(1) AS video_count
            FROM publish_single_video_source
            WHERE crawler_timestamp BETWEEN UNIX_TIMESTAMP(%s) AND UNIX_TIMESTAMP(%s)
            GROUP BY dt, platform;
        """
        return await self.pool.async_fetch(query=query, params=(start_date, end_date))

    async def get_crawler_videos_by_category(self, start_date, end_date):
        """
        获取 start_dt && end_dt 之间每个品类抓取的视频数量
        """
        category_place_holders = ", ".join(["%s"] * len(self.CATEGORY_LIST))
        query = f"""
            SELECT CAST(DATE(FROM_UNIXTIME(crawler_timestamp)) AS CHAR) AS dt, category, count(1) AS video_count
            FROM publish_single_video_source
            WHERE crawler_timestamp BETWEEN UNIX_TIMESTAMP(%s) AND UNIX_TIMESTAMP(%s)
              AND category IN ({category_place_holders})
            GROUP BY dt, category;
        """
        return await self.pool.async_fetch(
            query=query, params=tuple([start_date, end_date] + self.CATEGORY_LIST)
        )

    async def get_transform_videos_by_platform(self, start_date, end_date):
        query = """
            SELECT CAST(DATE(create_timestamp) AS CHAR) AS dt, platform, 
                   count(*) AS video_count, avg(score) AS average_similarity_score
            FROM single_video_transform_queue
            WHERE create_timestamp BETWEEN %s AND %s AND status = %s
            GROUP BY dt, platform;
        """
        return await self.pool.async_fetch(
            query=query, params=(start_date, end_date, self.TRANSFORMED_STATUS)
        )

    async def get_transform_videos_by_category(self, start_date, end_date):
        raise NotImplementedError()


class CrawlerDetailDeal(CrawlerVideoDetailAnalysis, CrawlerArticleDetailAnalysis):
    def __init__(self, pool, trace_id):
        super().__init__(pool, trace_id)

    async def analysis_video_pool(self, task, start_date, end_date):
        match task:
            case self.CRAWLER_DETAIL_TASK_PLATFORM:
                return await self.get_crawler_videos_by_platform(start_date, end_date)

            case self.CRAWLER_DETAIL_TASK_CATEGORY:
                return await self.get_crawler_videos_by_category(start_date, end_date)

            case self.TRANSFORM_DETAIL_TASK_PLATFORM:
                return await self.get_transform_videos_by_platform(start_date, end_date)

            case self.TRANSFORM_DETAIL_TASK_CATEGORY:
                return await self.get_transform_videos_by_category(start_date, end_date)

            case _:
                return None

    async def analysis_article_pool(self, task, start_date, end_date):
        raise NotImplementedError()

    async def deal(self, params):
        start_date = params.get("start_date")
        end_date = params.get("end_date")
        media_type = params.get("media_type", "video")
        sub_task = params.get("sub_task_name", self.CRAWLER_DETAIL_TASK_PLATFORM)

        column_dict = self.create_feishu_column_map()

        match media_type:
            case "video":
                crawler_detail = await self.analysis_video_pool(sub_task, start_date, end_date)

            case "article":
                crawler_detail = await self.analysis_article_pool(sub_task, start_date, end_date)
            case _:
                return None

        column_list = list(crawler_detail[0].keys())
        columns = [column_dict[key] for key in column_list]
        await feishu_robot.bot(
            title=f"[{start_date}, {end_date}) 抓取 {media_type} 统计",
            detail={
                "columns": columns,
                "rows": crawler_detail,
            },
            table=True,
            mention=False,
        )
        return None