二百七十二、Kettle——ClickHouse中增量导入数据重复性统计表数据（1天1次）

一、目的

在数据质量模块，需要对原始数据的重复性进行统计

Hive中原有SQL语句和ClickHouse现有SQL语句很大不同

二、Hive中原有代码

2.1 表结构

--41、八大类基础数据重复性统计表  事件+事件资源不需要重复
create  table  if not exists  hurys_db.dwd_data_duplicate(
    data_type      int        comment '1:转向比,2:统计,3:评价,4:区域,5:过车,6:静态排队,7:动态排队,8:轨迹,9:事件数据,10:事件资源',
    device_no      string     comment '设备编号',
    data_duplicate float      comment '数据重复率'
)
comment '数据重复性统计表'
partitioned by (day string)
stored as orc
;

2.2 SQL代码

insert  overwrite  table  hurys_db.dwd_data_duplicate partition(day)
select
       '6' data_type,
       device_no,
       round(sum(num)/count_num,2)  data_duplicate,
       day
from (select
       device_no,
       create_time,
       lane_no,
       count(1) num,
       count_num,
       day
from (select device_no,
             create_time,
             lane_no,
             count(device_no) over (partition by device_no,day) count_num,
             day
      from hurys_db.ods_queue
      where day = '2024-09-04'
    ) as t1
group by device_no, create_time, lane_no, count_num, day
having count(1) > 1
) as t3
group by device_no, count_num, day;

三、ClickHouse中现有代码

3.1 表结构

--41、八大类基础数据重复性统计表(长期存储)
create  table  if not exists  hurys_jw.dwd_data_duplicate(
    data_type      Int32            comment '1:转向比,2:统计,3:评价,4:区域,5:过车,6:静态排队,7:动态排队,8:轨迹,9:事件数据,10:事件资源',
    device_no      String           comment '设备编号',
    data_duplicate Decimal(10, 2)   comment '数据重复率',
    day            Date             comment '日期'
)
ENGINE = MergeTree
PARTITION BY day
PRIMARY KEY day
ORDER BY day
SETTINGS index_granularity = 8192;

3.2 SQL代码

select
       '6' data_type,
       device_no,
       round(sum(num)/count_num,2)  data_duplicate,
       day
from (select
       device_no,
       create_time,
       lane_no,
       count(1) num,
       count_num,
       day
from (select device_no,
             create_time,
             lane_no,
             count(device_no) over (partition by device_no,DATE(create_time)) AS count_num,
             DATE(create_time) day
      from hurys_jw.ods_queue
      where day = '2024-10-22' -- where day > ?
    ) as t1
group by device_no, create_time, lane_no, count_num, day
having count(1) > 1
) as t3
group by device_no, count_num, day;

3.3 Kettle任务

3.3.1 newtime

3.3.2 替换NULL值

3.3.3 clickhouse输入1

select
'6' data_type,
device_no,
round(sum(num)/count_num,2) data_duplicate,
cast(day as String) day
from (select
device_no,
create_time,
lane_no,
count(1) num,
count_num,
day
from (select device_no,
create_time,
lane_no,
count(device_no) over (partition by device_no,DATE(create_time)) AS count_num,
DATE(create_time) day
from hurys_jw.ods_queue
where day > ?
) as t1
group by device_no, create_time, lane_no, count_num, day
having count(1) > 1
) as t3
group by device_no, count_num, day
;