Day4-Hive直播行业基础笔试题

Hive笔试题实战

短视频

题目一：计算各个视频的平均完播率

有用户-视频互动表tb_user_video_log：

uid-用户ID，video_id-视频ID，start_time-开始观看时间，end_time-结束观看时间，if_follow-是否关注，if_like-是否点赞，if_retweet-是否转发，comment_id-评论ID。

有短视频信息表tb_video_info：

video_id-视频ID，author-创作者ID，tag-类别标签，duration-视频时长(秒)，release_time-发布时间。

问题：计算2021年里有播放记录的每个视频的完播率(结果保留三位小数)，并按完播率降序排序。输出结果如下：

注：视频完播率是指完成播放次数占总播放次数的比例。简单起见，结束观看时间与开始播放时间的差≥视频时长时，视为完成播放。

-- 建立用户-视频互动表

drop table if exists tb_user_video_log;

create table tb_user_video_log (

id         int comment '自增ID', 

uid        int comment '用户ID',

video_id   int comment '视频ID',

start_time string COMMENT '开始观看时间',

end_time   string COMMENT '结束观看时间',

    if_follow  int comment '是否关注',

    if_like    int comment '是否点赞',

    if_retweet int comment '是否转发',

    comment_id int comment '评论ID'

) comment '用户-视频互动表'

row format delimited fields terminated by ',';

-- 建立短视频信息表

drop table if exists tb_video_info;

create table tb_video_info (

    id           int comment '自增ID',

    video_id     int comment '视频ID',

    author       int comment '创作者ID',

    tag          string comment '类别标签',

    duration     int comment '视频时长(秒数)',

    release_time string comment '发布时间'

) comment '短视频信息表'

    row format delimited fields terminated by ',';

-- 插入数据

insert into tb_user_video_log

values (1, 101, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:30', 0, 1, 1, null),

       (2, 102, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:24', 0, 0, 1, null),

       (3, 103, 2001, '2021-10-01 11:00:00', '2021-10-01 11:00:34', 0, 1, 0, 1732526),

       (4, 101, 2002, '2021-09-01 10:00:00', '2021-09-01 10:00:42', 1, 0, 1, null),

       (5, 102, 2002, '2021-10-01 11:00:00', '2021-10-01 11:00:30', 1, 0, 1, null);

insert into tb_video_info

values (1, 2001, 901, '影视', 30, '2021-01-01 7:00:00'),

       (2, 2002, 901, '美食', 60, '2021-01-01 7:00:00'),

       (3, 2003, 902, '旅游', 90, '2021-01-01 7:00:00');

参考答案：

-- 第一步：找出2021年有过播放的视频

select * from tb_user_video_log where year(start_time) = 2021;

-- 第二步：计算(每一个视频的)完播次数。完播：结束时间-起始时间>=视频时长

select a.video_id as video_id,

       sum(if(unix_timestamp(a.end_time) - unix_timestamp(a.start_time) >= b.duration, 1, 0))

from (

      select * from tb_user_video_log where year(start_time) = 2021

) a left join tb_video_info b on a.video_id = b.video_id

group by a.video_id;

-- 第三步：计算完播率。完播次数/总的播放次数

select a.video_id as video_id,

       sum(if(unix_timestamp(a.end_time) - unix_timestamp(a.start_time) >= b.duration, 1, 0)) / count(*)

from (

     select * from tb_user_video_log where year(start_time) = 2021

) a left join tb_video_info b on a.video_id = b.video_id

group by a.video_id;

-- 第四步：保留三位小数，还需要降序排序

select a.video_id  as video_id,

       round(sum(if(unix_timestamp(a.end_time) - unix_timestamp(a.start_time) >= b.duration, 1, 0)) / count(*), 3) as avg_comp_play_rate

from (

      select * from tb_user_video_log where year(start_time) = 2021

) a left join tb_video_info b on a.video_id = b.video_id

group by a.video_id

order by avg_comp_play_rate desc;

题目二：平均播放进度大于60%的视频类别

有用户-视频互动表tb_user_video_log：

uid-用户ID，video_id-视频ID，start_time-开始观看时间，end_time-结束观看时间，if_follow-是否关注，if_like-是否点赞，if_retweet-是否转发，comment_id-评论ID。

有短视频信息表tb_video_info：

video_id-视频ID，author-创作者ID，tag-类别标签，duration-视频时长(秒)，release_time-发布时间。

问题：计算各类视频的平均播放进度，将进度大于60%的类别输出(结果保留两位小数，并按播放进度倒序排序)。示例数据的输出结果如下：

注：播放进度=播放时长÷视频时长*100%，当播放时长大于视频时长时，播放进度均记为100%。

例如：影视类视频2001被用户101、102、103看过，播放进度分别为：30秒（100%）、21秒（70%）、30秒（100%），平均播放进度为(100%+70%+100%)/3=90.00%（保留两位小数）。

-- 建立用户-视频互动表

drop table if exists tb_user_video_log;

create table tb_user_video_log (

    id         int comment '自增ID',

    uid        int comment '用户ID',

    video_id   int comment '视频ID',

    start_time string COMMENT '开始观看时间',

    end_time   string COMMENT '结束观看时间',

    if_follow  int comment '是否关注',

    if_like    int comment '是否点赞',

    if_retweet int comment '是否转发',

    comment_id int comment '评论ID'

) comment '用户-视频互动表'

    row format delimited fields terminated by ',';

-- 建立短视频信息表

drop table if exists tb_video_info;

create table tb_video_info (

    id           int comment '自增ID',

    video_id     int comment '视频ID',

    author       int comment '创作者ID',

    tag          string comment '类别标签',

    duration     int comment '视频时长(秒数)',

    release_time string comment '发布时间'

) comment '短视频信息表'

    row format delimited fields terminated by ',';

-- 插入数据

insert into tb_user_video_log

values (1, 101, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:30', 0, 1, 1, null),

       (2, 102, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:21', 0, 0, 1, null),

       (3, 103, 2001, '2021-10-01 11:00:50', '2021-10-01 11:01:20', 0, 1, 0, 1732526),

       (4, 102, 2002, '2021-10-01 11:00:00', '2021-10-01 11:00:30', 1, 0, 1, null),

       (5, 103, 2002, '2021-10-01 10:59:05', '2021-10-01 11:00:05', 1, 0, 1, null);

insert into tb_video_info

values (1, 2001, 901, '影视', 30, '2021-01-01 7:00:00'),

       (2, 2002, 901, '美食', 60, '2021-01-01 7:00:00'),

       (3, 2003, 902, '旅游', 90, '2021-01-01 7:00:00');

参考答案：

-- 第一步：计算每次播放的播放时长

select video_id, unix_timestamp(end_time) - unix_timestamp(start_time) as total_time from tb_user_video_log;

-- 第二步：计算每一次的播放进度

select a.video_id as video_id,

       if(a.total_time / b.duration > 1, 1, a.total_time / b.duration) as play_progress

from (

    select video_id, unix_timestamp(end_time) - unix_timestamp(start_time) as total_time from tb_user_video_log

) a left join tb_video_info b on a.video_id = b.video_id;

-- 第三步：计算各类视频的平均播放进度

select b.tag,

       avg(if(a.total_time / b.duration > 1, 1, a.total_time / b.duration)) as avg_play_progress

from (

    select video_id, unix_timestamp(end_time) - unix_timestamp(start_time) as total_time from tb_user_video_log

) a left join tb_video_info b on a.video_id = b.video_id

group by b.tag;

-- 第四步：过滤，排序

select b.tag,

       avg(if(a.total_time / b.duration > 1, 1, a.total_time / b.duration)) as avg_play_progress

from (

    select video_id, unix_timestamp(end_time) - unix_timestamp(start_time) as total_time from tb_user_video_log

) a left join tb_video_info b on a.video_id = b.video_id

group by b.tag having avg_play_progress > 0.6 order by avg_play_progress desc;

-- 第五步：百分比

select tag,

       concat(round(avg_play_progress * 100, 2), '%') as avg_play_progress

from (

  select b.tag as tag,

         avg(if(a.total_time / b.duration > 1, 1, a.total_time / b.duration)) as avg_play_progress

  from (

      select video_id, unix_timestamp(end_time) - unix_timestamp(start_time) as total_time from tb_user_video_log

  ) a left join tb_video_info b on a.video_id = b.video_id

  group by b.tag

  having avg_play_progress > 0.6

  order by avg_play_progress desc

)t;

题目三：每类视频近一个月的转发量/率

有用户-视频互动表tb_user_video_log：

uid-用户ID，video_id-视频ID，start_time-开始观看时间，end_time-结束观看时间，if_follow-是否关注，if_like-是否点赞，if_retweet-是否转发，comment_id-评论ID。

有短视频信息表tb_video_info：

video_id-视频ID，author-创作者ID，tag-类别标签，duration-视频时长(秒)，release_time-发布时间。

问题：统计在有用户互动的最近一个月（按包含当天在内的近30天算，比如10月31日的近30天为10.2~10.31之间的数据）中，每类视频的转发量和转发率（保留3位小数）。输出结果如下：

注：转发率＝转发量÷播放量。结果按转发率降序排序。

解释：由表tb_user_video_log的数据可得，数据转储当天为2021年10月1日。近30天内，影视类视频2001共有3次播放记录，被转发2次，转发率为0.667；美食类视频2002共有2次播放记录，1次被转发，转发率为0.500。

-- 建立用户-视频互动表

drop table if exists tb_user_video_log;

create table tb_user_video_log (

    id         int comment '自增ID',

    uid        int comment '用户ID',

    video_id   int comment '视频ID',

    start_time string COMMENT '开始观看时间',

    end_time   string COMMENT '结束观看时间',

    if_follow  int comment '是否关注',

    if_like    int comment '是否点赞',

    if_retweet int comment '是否转发',

    comment_id int comment '评论ID'

) comment '用户-视频互动表'

    row format delimited fields terminated by ',';

-- 建立短视频信息表

drop table if exists tb_video_info;

create table tb_video_info (

    id           int comment '自增ID',

    video_id     int comment '视频ID',

    author       int comment '创作者ID',

    tag          string comment '类别标签',

    duration     int comment '视频时长(秒数)',

    release_time string comment '发布时间'

) comment '短视频信息表'

    row format delimited fields terminated by ',';

-- 插入数据

insert into tb_user_video_log

values (1, 101, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:20', 0, 1, 1, null),

       (2, 102, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:15', 0, 0, 1, null),

       (3, 103, 2001, '2021-10-01 11:00:50', '2021-10-01 11:01:15', 0, 1, 0, 1732526),

       (4, 102, 2002, '2021-09-10 11:00:00', '2021-09-10 11:00:30', 1, 0, 1, null),

       (5, 103, 2002, '2021-10-01 10:59:05', '2021-10-01 11:00:05', 1, 0, 0, null);

insert into tb_video_info

values (1, 2001, 901, '影视', 30, '2021-01-01 7:00:00'),

       (2, 2002, 901, '美食', 60, '2021-01-01 7:00:00'),

       (3, 2003, 902, '旅游', 90, '2021-01-01 7:00:00');

参考答案：

-- 1. 找出最后一次的播放时间

select max(start_time) from tb_user_video_log;

-- 2. 基于最后一次的播放时间，向前推29天(包含当天在内的近30天算)，获取到近30天内的所有播放记录

select *

from tb_user_video_log a,

     (select max(start_time) as last_date from tb_user_video_log) b

where datediff(b.last_date, a.start_time) <= 29;

-- 3. 计算每一类视频的转发量和转发率

select t2.tag                               as tag,

       sum(if_retweet)                      as retweet_cut,

       round(sum(if_retweet) / count(*), 3) as retweet_rate

from (

    select *

    from tb_user_video_log a,

           (select max(start_time) as last_date from tb_user_video_log) b

    where datediff(b.last_date, a.start_time) <= 29

) t1 left join tb_video_info t2 on t1.video_id = t2.video_id

group by t2.tag order by retweet_rate desc;

题目四：每个创作者每月的涨粉率及截止当前的总粉丝量

有用户-视频互动表tb_user_video_log：

uid-用户ID，video_id-视频ID，start_time-开始观看时间，end_time-结束观看时间，if_follow-是否关注，if_like-是否点赞，if_retweet-是否转发，comment_id-评论ID。

有短视频信息表tb_video_info：

video_id-视频ID，author-创作者ID，tag-类别标签，duration-视频时长(秒)，release_time-发布时间。

问题：计算2021年里每个创作者每月的涨粉率及截止当月的总粉丝量。输出结果如下：

注：涨粉率=(加粉量 - 掉粉量) / 播放量。结果按创作者ID、总粉丝量升序排序。if_follow-是否关注，为1表示用户观看视频中关注了视频创作者，为0表示此次互动前后关注状态未发生变化，为2表示本次观看过程中取消了关注。

解释：示例数据中表tb_user_video_log里只有视频2001和2002的播放记录，都来自创作者901，播放时间在2021年9月和10月；其中9月里加粉量为1，掉粉量为0，播放量为2，因此涨粉率为0.500（保留3位小数）；其中10月里加粉量为2，掉份量为1，播放量为4，因此涨粉率为0.250，截止当前总粉丝数为2。

-- 建立用户-视频互动表

drop table if exists tb_user_video_log;

create table tb_user_video_log (

    id         int comment '自增ID',

    uid        int comment '用户ID',

    video_id   int comment '视频ID',

    start_time string COMMENT '开始观看时间',

    end_time   string COMMENT '结束观看时间',

    if_follow  int comment '是否关注',

    if_like    int comment '是否点赞',

    if_retweet int comment '是否转发',

    comment_id int comment '评论ID'

) comment '用户-视频互动表'

    row format delimited fields terminated by ',';

-- 建立短视频信息表

drop table if exists tb_video_info;

create table tb_video_info (

    id           int comment '自增ID',

    video_id     int comment '视频ID',

    author       int comment '创作者ID',

    tag          string comment '类别标签',

    duration     int comment '视频时长(秒数)',

    release_time string comment '发布时间'

) comment '短视频信息表'

    row format delimited fields terminated by ',';

-- 插入数据

insert into tb_user_video_log

values (1, 101, 2001, '2021-09-01 10:00:00', '2021-09-01 10:00:20', 0, 1, 1, null),

       (2, 105, 2002, '2021-09-10 11:00:00', '2021-09-10 11:00:30', 1, 0, 1, null),

       (3, 101, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:20', 1, 1, 1, null),

       (4, 102, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:15', 0, 0, 1, null),

       (5, 103, 2001, '2021-10-01 11:00:50', '2021-10-01 11:01:15', 1, 1, 0, 1732526),

       (6, 106, 2002, '2021-10-01 10:59:05', '2021-10-01 11:00:05', 2, 0, 0, null);

 

insert into tb_video_info

VALUES (1, 2001, 901, '影视', 30, '2021-01-01 7:00:00'),

       (2, 2002, 901, '影视', 60, '2021-01-01 7:00:00'),

       (3, 2003, 902, '旅游', 90, '2020-01-01 7:00:00'),

       (4, 2004, 902, '美女', 90, '2020-01-01 8:00:00');

参考答案：

-- 1. 获取2021年的数据，日期整理成月的形式

select video_id, date_format(start_time, 'yyyy-MM') as m, if_follow

from tb_user_video_log

where year(start_time) = 2021;

-- 2. 计算每一个作者每一个月的粉丝变化数量以及视频的播放次数

select b.author                                  as author,

       a.m                                       as m,

       sum(if(a.if_follow = 2, -1, a.if_follow)) as total_fans_m,

       count(*)                                  as total_play_m

from (

    select video_id, date_format(start_time, 'yyyy-MM') as m, if_follow

    from tb_user_video_log

    where year(start_time) = 2021

) a left join tb_video_info b on a.video_id = b.video_id

group by b.author, a.m;

-- 3. 计算每一个作者到当前月的粉丝变化率以及总粉丝量

select author,

       m as `month`,

       round(total_fans_m / total_play_m, 3) as fans_growth_rate,

       sum(total_fans_m) over (partition by author order by m rows between unbounded preceding and current row ) as total_fans

from (

    select b.author as author,

           a.m as m,

           sum(if(a.if_follow = 2, -1, a.if_follow)) as total_fans_m,

           count(*) as total_play_m

    from (

        select video_id, date_format(start_time, 'yyyy-MM') as m, if_follow

        from tb_user_video_log

        where year(start_time) = 2021) a left join tb_video_info b on a.video_id = b.video_id

      group by b.author, a.m

) t order by author, total_fans;

题目五：国庆期间每类视频点赞量和转发量

有用户-视频互动表tb_user_video_log：

uid-用户ID，video_id-视频ID，start_time-开始观看时间，end_time-结束观看时间，if_follow-是否关注，if_like-是否点赞，if_retweet-是否转发，comment_id-评论ID。

有短视频信息表tb_video_info：

video_id-视频ID，author-创作者ID，tag-类别标签，duration-视频时长(秒)，release_time-发布时间。

问题：统计2021年国庆头3天每类视频每天的近一周总点赞量和一周内最大单天转发量，结果按视频类别降序、日期升序排序。假设数据库中数据足够多，至少每个类别下国庆头3天及之前一周的每天都有播放记录。结果如下：

解释：由表tb_user_video_log里的数据可得只有旅游类视频的播放，2021年9月25到10月3日每天的点赞量和转发量如下：

因此国庆头3天（10.01~10.03）里10.01的近7天（9.25~10.01）总点赞量为5次，单天最大转发量为2次（9月25那天最大）；同理可得10.02和10.03的两个指标。

-- 建立用户-视频互动表

drop table if exists tb_user_video_log;

create table tb_user_video_log (

    id         int comment '自增ID',

    uid        int comment '用户ID',

    video_id   int comment '视频ID',

    start_time string COMMENT '开始观看时间',

    end_time   string COMMENT '结束观看时间',

    if_follow  int comment '是否关注',

    if_like    int comment '是否点赞',

    if_retweet int comment '是否转发',

    comment_id int comment '评论ID'

) comment '用户-视频互动表'

    row format delimited fields terminated by ',';

-- 建立短视频信息表

drop table if exists tb_video_info;

create table tb_video_info (

    id           int comment '自增ID',

    video_id     int comment '视频ID',

    author       int comment '创作者ID',

    tag          string comment '类别标签',

    duration     int comment '视频时长(秒数)',

    release_time string comment '发布时间'

) comment '短视频信息表'

    row format delimited fields terminated by ',';

-- 插入数据

insert into tb_user_video_log

values (1, 101, 2001, '2021-09-24 10:00:00', '2021-09-24 10:00:20', 1, 1, 0, null),

       (2, 105, 2002, '2021-09-25 11:00:00', '2021-09-25 11:00:30', 0, 0, 1, null),

       (3, 102, 2002, '2021-09-25 11:00:00', '2021-09-25 11:00:30', 1, 1, 1, null),

       (4, 101, 2002, '2021-09-26 11:00:00', '2021-09-26 11:00:30', 1, 0, 1, null),

       (5, 101, 2002, '2021-09-27 11:00:00', '2021-09-27 11:00:30', 1, 1, 0, null),

       (6, 102, 2002, '2021-09-28 11:00:00', '2021-09-28 11:00:30', 1, 0, 1, null),

       (7, 103, 2002, '2021-09-29 11:00:00', '2021-09-29 11:00:30', 1, 0, 1, null),

       (8, 102, 2002, '2021-09-30 11:00:00', '2021-09-30 11:00:30', 1, 1, 1, null),

       (9, 101, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:20', 1, 1, 0, null),

       (10, 102, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:15', 0, 0, 1, null),

       (11, 103, 2001, '2021-10-01 11:00:50', '2021-10-01 11:01:15', 1, 1, 0, 1732526),

       (12, 106, 2002, '2021-10-02 10:59:05', '2021-10-02 11:00:05', 2, 0, 1, null),

       (13, 107, 2002, '2021-10-02 10:59:05', '2021-10-02 11:00:05', 1, 0, 1, null),

       (14, 108, 2002, '2021-10-02 10:59:05', '2021-10-02 11:00:05', 1, 1, 1, null),

       (15, 109, 2002, '2021-10-03 10:59:05', '2021-10-03 11:00:05', 0, 1, 0, null);

 

insert into tb_video_info

VALUES (1, 2001, 901, '影视', 30, '2021-01-01 7:00:00'),

       (2, 2002, 901, '影视', 60, '2021-01-01 7:00:00'),

       (3, 2003, 902, '旅游', 90, '2020-01-01 7:00:00'),

       (4, 2004, 902, '美女', 90, '2020-01-01 8:00:00');

参考答案：

-- 1. 锁定数据范围：2021-09.25~2021-10-03

select video_id, date(start_time), if_like, if_retweet

from tb_user_video_log

where datediff('2021-10-03', start_time) < 9;

-- 2. 统计每一类视频每天的点赞量和转发量

select b.tag             as tag,

       a.dt              as dt,

       sum(a.if_like)    as total_like_d,

       sum(a.if_retweet) as total_retweet_d

from (

     select video_id, date(start_time) as dt, if_like, if_retweet

      from tb_user_video_log

      where datediff('2021-10-03', start_time) < 9

) a left join tb_video_info b on a.video_id = b.video_id

group by b.tag, a.dt;

-- 3. 统计最近7天的点赞总量和最大转发量

select tag,

       dt,

       sum(total_like_d) over (partition by tag order by dt rows between 6 preceding and current row ) as sum_like_cnt_7d,

       max(total_retweet_d) over (partition by tag order by dt rows between 6 preceding and current row)  as max_retweet_cnt_7d

from (

      select b.tag             as tag,

             a.dt              as dt,

             sum(a.if_like)    as total_like_d,

             sum(a.if_retweet) as total_retweet_d

      from (

            select video_id, date(start_time) as dt, if_like, if_retweet

            from tb_user_video_log

            where datediff('2021-10-03', start_time) < 9

      ) a left join tb_video_info b on a.video_id = b.video_id

      group by b.tag, a.dt

) t1;

-- 4. 过滤出10-01~10-03

select *

from (

      select tag,

             dt,

             sum(total_like_d) over (partition by tag order by dt rows between 6 preceding and current row ) as sum_like_cnt_7d,

             max(total_retweet_d) over (partition by tag order by dt rows between 6 preceding and current row)  as max_retweet_cnt_7d

      from (

            select b.tag             as tag,

                   a.dt              as dt,

                   sum(a.if_like)    as total_like_d,

                   sum(a.if_retweet) as total_retweet_d

            from (

                  select video_id, date(start_time) as dt, if_like, if_retweet

                  from tb_user_video_log

                  where datediff('2021-10-03', start_time) < 9

            ) a left join tb_video_info b on a.video_id = b.video_id

            group by b.tag, a.dt

      ) t1

) t2 where month(dt) = 10

order by tag desc, dt asc;

题目六：近一个月发布的视频中热度最高的top3视频

有用户-视频互动表tb_user_video_log：

uid-用户ID，video_id-视频ID，start_time-开始观看时间，end_time-结束观看时间，if_follow-是否关注，if_like-是否点赞，if_retweet-是否转发，comment_id-评论ID。

有短视频信息表tb_video_info：

video_id-视频ID，author-创作者ID，tag-类别标签，duration-视频时长(秒)，release_time-发布时间。

问题：找出近一个月发布的视频中热度最高的top3视频。结果如下：

注意：

1)热度=(a*视频完播率+b*点赞数+c*评论数+d*转发数)/新鲜度；

2)新鲜度=最近无播放天数+1，最近无播放天数指的是最后一次播放日期到最近日期之间的天数间隔；

3)当前配置的参数a，b，c，d分别为100、5、3、2；

4)最近播放日期以end_time为准，假设为T，则最近一个月按[T-29, T]闭区间统计；

5)结果中热度保留为整数，并按热度降序排序。

解释：假设最近播放日期为2021-10-03，记作当天日期；近一个月（2021-09-04及之后）发布的视频有2001、2002、2003、2004，不过2004暂时还没有播放记录；视频2001完播率1.0（被播放次数4次，完成播放4次），被点赞3次，评论1次，转发2次，最后一次播放日期为2021-10-03，所以最近无播放天数为0，因此热度为：(100*1.0+5*3+3*1+2*2)/(0+1)=122；同理，视频2003完播率0，被点赞数1，评论和转发均为0，最后一次播放日期为2021-09-30，所以最近无播放天数为3，因此热度为：(100*0+5*1+3*0+2*0)/(3+1)=1（1.2保留为整数）。

-- 建立用户-视频互动表

drop table if exists tb_user_video_log;

create table tb_user_video_log (

    id         int comment '自增ID',

    uid        int comment '用户ID',

    video_id   int comment '视频ID',

    start_time string COMMENT '开始观看时间',

    end_time   string COMMENT '结束观看时间',

    if_follow  int comment '是否关注',

    if_like    int comment '是否点赞',

    if_retweet int comment '是否转发',

    comment_id int comment '评论ID'

) comment '用户-视频互动表'

    row format delimited fields terminated by ',';

-- 建立短视频信息表

drop table if exists tb_video_info;

create table tb_video_info (

    id           int comment '自增ID',

    video_id     int comment '视频ID',

    author       int comment '创作者ID',

    tag          string comment '类别标签',

    duration     int comment '视频时长(秒数)',

    release_time string comment '发布时间'

) comment '短视频信息表'

    row format delimited fields terminated by ',';

-- 插入数据

insert into tb_user_video_log

values (1, 101, 2001, '2021-09-24 10:00:00', '2021-09-24 10:00:30', 1, 1, 1, null),

       (2, 101, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:31', 1, 1, 0, null),

       (3, 102, 2001, '2021-10-01 10:00:00', '2021-10-01 10:00:35', 0, 0, 1, null),

       (4, 103, 2001, '2021-10-03 11:00:50', '2021-10-03 11:01:35', 1, 1, 0, 1732526),

       (5, 106, 2002, '2021-10-02 10:59:05', '2021-10-02 11:00:04', 2, 0, 1, null),

       (6, 107, 2002, '2021-10-02 10:59:05', '2021-10-02 11:00:06', 1, 0, 0, null),

       (7, 108, 2002, '2021-10-02 10:59:05', '2021-10-02 11:00:05', 1, 1, 1, null),

       (8, 109, 2002, '2021-10-03 10:59:05', '2021-10-03 11:00:01', 0, 1, 0, null),

       (9, 105, 2002, '2021-09-25 11:00:00', '2021-09-25 11:00:30', 1, 0, 1, null),

       (10, 101, 2003, '2021-09-26 11:00:00', '2021-09-26 11:00:30', 1, 0, 0, null),

       (11, 101, 2003, '2021-09-30 11:00:00', '2021-09-30 11:00:30', 1, 1, 0, null);

 

insert into tb_video_info

VALUES (1, 2001, 901, '旅游', 30, '2021-09-05 7:00:00'),

       (2, 2002, 901, '旅游', 60, '2021-09-05 7:00:00'),

       (3, 2003, 902, '影视', 90, '2021-09-05 7:00:00'),

       (4, 2004, 902, '影视', 90, '2021-09-05 8:00:00');

参考答案：

-- 1. 获取最后一次的结束时间

select max(date(end_time)) as last_date

from tb_user_video_log;

-- 2. 获取最近一个月的发布的视频

select video_id, duration

from tb_video_info a,

     (select max(date(end_time)) as last_date from tb_user_video_log) b

where datediff(b.last_date, date(a.release_time)) <= 29;

-- 3. 统计每一条视频结束日期到最后日期之间的日期差、播放时长和其他信息

select video_id,

       datediff(last_date, date(end_time))                   as no_play_day,

       unix_timestamp(end_time) - unix_timestamp(start_time) as play_time,

       if_like,

       if(comment_id is null, 0, 1)                          as if_comment,

       if_retweet

from tb_user_video_log a,

     (select max(date(end_time)) as last_date from tb_user_video_log) b;

-- 4. 统计最近一个月发布的每一个视频最近的无播放天数、完播率、总点赞量、总评论数、总转发数

select t1.video_id as video_id,

       min(t1.no_play_day) as no_play_count,

       sum(if(t1.play_time >= t2.duration, 1, 0)) / count(*) as play_rate,

       sum(t1.if_like) as like_count,

       sum(t1.if_comment) as comment_count,

       sum(t1.if_retweet) as retweet_count

from (

    select video_id,

           datediff(last_date, date(end_time))                   as no_play_day,

             unix_timestamp(end_time) - unix_timestamp(start_time) as play_time,

             if_like,

             if(comment_id is null, 0, 1)                          as if_comment,

             if_retweet

      from tb_user_video_log a,

           (select max(date(end_time)) as last_date from tb_user_video_log) b) t1

         left join (select video_id, duration

                    from tb_video_info a,

                         (select max(date(end_time)) as last_date from tb_user_video_log) b

                    where datediff(b.last_date, date(a.release_time)) <= 29) t2 on t1.video_id = t2.video_id

group by t1.video_id;

-- 5. 计算每一个视频的热度

select video_id,

       round((100 * play_rate + 5 * like_count + 3*comment_count + 2 * retweet_count) / (no_play_count + 1)) as hot_index

from (

    select t1.video_id as video_id,

           min(t1.no_play_day) as no_play_count,

           sum(if(t1.play_time >= t2.duration, 1, 0)) / count(*) as play_rate,

           sum(t1.if_like) as like_count,

           sum(t1.if_comment) as comment_count,

           sum(t1.if_retweet) as retweet_count

    from (

        select video_id,

               datediff(last_date, date(end_time))                   as no_play_day,

               unix_timestamp(end_time) - unix_timestamp(start_time) as play_time,

               if_like,

               if(comment_id is null, 0, 1)                          as if_comment,

               if_retweet

        from tb_user_video_log a, (select max(date(end_time)) as last_date from tb_user_video_log) b

    ) t1 left join (

        select video_id, duration from tb_video_info a, (select max(date(end_time)) as last_date from tb_user_video_log) b

                                  where datediff(b.last_date, date(a.release_time)) <= 29

    ) t2 on t1.video_id = t2.video_id group by t1.video_id

) t order by hot_index desc limit 3;