HIVE笔记

表关联

内连接（INNER JOIN）

返回两个表中满足关联条件的记录。

SELECT * 
FROM t1 
INNER JOIN t2 
ON t1.col1 = t2.col2;

左连接（LEFT JOIN）

返回左表中的所有记录，以及右表中满足关联条件的记录。

SELECT * 
FROM t1 
LEFT JOIN t2 
ON t1.col1 = t2.col2;

右连接（RIGHT JOIN）

返回右表中的所有记录，以及左表中满足关联条件的记录。

SELECT * 
FROM t1 
RIGHT JOIN t2 
ON t1.col1 = t2.col2;

全连接（FULL OUTER JOIN）

返回左表和右表中的所有记录。

hive full join多表多关联键联合查询

SELECT * 
FROM t1 
FULL OUTER JOIN t2 
ON t1.col1 = t2.col2;

DDL

字段操作

--添加字段
alter table app.table_name add columns(bu_name STRING COMMENT "事业部名称") CASCADE;
--修改字段类型（修改为double）
Alter table tmp.tmp_zp_tablename column columnname  columnname  double;
--调整列位置
alter table app.table_name change bu_name bu_name STRING after col_a;

注意不要直接对有数据的表进行字段顺序调整，会导致历史分区数据错误。

分区操作

--删除分区
alter table tmp.tmp_zp_tablename drop if exists partition(dt='2020-10-24');

常用函数

官网文档

sort_array

sort_array(Array) 只有一个参数
根据自然顺序按升序对输入数组进行排序

SELECT sort_array(array(5, 2, 8, 1, 7)) AS sorted_array;

使用中常和collect函数使用 sort_array(collect_set())

concat_ws

concat_ws(separator, string1, string2, …)
用于将多个字符串连接在一起，中间使用指定的分隔符进行分隔。

SELECT concat_ws(',', 'Hello', 'World') AS result;

常和数组集合函数使用，collect_set collect_list 将数据内容转为字符串
concat_ws(‘,’,collect_set(col) )

collect_set collect_list

collect_set函数可以将指定字段的所有不重复的值，以Set的形式返回。Set是一种无序且不包含重复元素的数据结构。
collect_list函数可以将指定字段的所有值，以List的形式返回。List是一种有序且允许重复元素的数据结构。

SELECT collect_set(name) FROM student;

注意
collect_set和collect_list函数只能应用于对一个字段进行聚合操作，不能对多个字段同时聚合。
collect_set和collect_list函数的性能较差，当数据量较大时，可能会影响查询性能。
collect_set和collect_list函数都是在Reducer阶段进行聚合操作，因此在分布式环境下，需要确保数据被正确分组。

length size

length(string A) Returns the length of the string.
size(Map) Returns the number of elements in the map type.
size(Array) Returns the number of elements in the array type.

TRUNC

TRUNC（number,num_digits）Number需要截尾取整的数字。Num_digits用于指定取整精度的数字，默认值为0。TRUNC()函数截取时不进行四舍五入。

select trunc(123.458) from dual --123
select trunc(123.458,0) from dual --123
select trunc(123.458,1) from dual --123.4
select trunc(123.458,-1) from dual --120
select trunc(123.458,-4) from dual --0
select trunc(123.458,4) from dual --123.458
select trunc(123) from dual --123
select trunc(123,1) from dual --123
select trunc(123,-1) from dual --120

lag/lead

查询每个顾客上次的购买时间

select *,lag(orderdate) over(partition by name order by orderdate) from business;
+----------------+---------------------+----------------+---------------+--+
| business.name  | business.orderdate  | business.cost  | lag_window_0  |
+----------------+---------------------+----------------+---------------+--+
| jack           | 2017-01-01          | 10             | NULL          |
| jack           | 2017-01-05          | 46             | 2017-01-01    |
| jack           | 2017-01-08          | 55             | 2017-01-05    |
| jack           | 2017-02-03          | 23             | 2017-01-08    |
| jack           | 2017-04-06          | 42             | 2017-02-03    |
| mart           | 2017-04-08          | 62             | NULL          |
| mart           | 2017-04-09          | 68             | 2017-04-08    |
| mart           | 2017-04-11          | 75             | 2017-04-09    |
| mart           | 2017-04-13          | 94             | 2017-04-11    |
| neil           | 2017-05-10          | 12             | NULL          |
| neil           | 2017-06-12          | 80             | 2017-05-10    |
| tony           | 2017-01-02          | 15             | NULL          |
| tony           | 2017-01-04          | 29             | 2017-01-02    |
| tony           | 2017-01-07          | 50             | 2017-01-04    |
+----------------+---------------------+----------------+---------------+--+
select *,lag(orderdate,1,"1970-01-01") over(partition by name order by orderdate) from business;
--lag
--lag(col,n,DEFAULT) 第一个参数为列名，第二个参数为往上第n行（可选，默认为1），第三个参数为默认值（当往上第n行为NULL时候，取默认值，如不指定，则为NULL）
--与LAG相反
--LEAD(col,n,DEFAULT) 用于统计窗口内往下第n行值。第一个参数为列名，第二个参数为往下第n行（可选，默认为1），第三个参数为默认值（当往下第n行为NULL时候，取默认值，如不指定，则为NULL）
+----------------+---------------------+----------------+---------------+--+
| business.name  | business.orderdate  | business.cost  | lag_window_0  |
+----------------+---------------------+----------------+---------------+--+
| jack           | 2017-01-01          | 10             | 1970-01-01    |
| jack           | 2017-01-05          | 46             | 2017-01-01    |
| jack           | 2017-01-08          | 55             | 2017-01-05    |
| jack           | 2017-02-03          | 23             | 2017-01-08    |
| jack           | 2017-04-06          | 42             | 2017-02-03    |
| mart           | 2017-04-08          | 62             | 1970-01-01    |
| mart           | 2017-04-09          | 68             | 2017-04-08    |
| mart           | 2017-04-11          | 75             | 2017-04-09    |
| mart           | 2017-04-13          | 94             | 2017-04-11    |
| neil           | 2017-05-10          | 12             | 1970-01-01    |
| neil           | 2017-06-12          | 80             | 2017-05-10    |
| tony           | 2017-01-02          | 15             | 1970-01-01    |
| tony           | 2017-01-04          | 29             | 2017-01-02    |
| tony           | 2017-01-07          | 50             | 2017-01-04    |
+----------------+---------------------+----------------+---------------+--+

ntile

用于将分组数据按照顺序切分成n片（不是严格等分），返回当前记录所在的切片值。

--查询前20%时间的订单信息
select *,ntile(5) tgroup over(order by orderdate) from business;
+----------------+---------------------+----------------+-----------------+--+
| business.name  | business.orderdate  | business.cost  | ntile_window_0  |
+----------------+---------------------+----------------+-----------------+--+
| jack           | 2017-01-01          | 10             | 1               |
| tony           | 2017-01-02          | 15             | 1               |
| tony           | 2017-01-04          | 29             | 1               |
| jack           | 2017-01-05          | 46             | 2               |
| tony           | 2017-01-07          | 50             | 2               |
| jack           | 2017-01-08          | 55             | 2               |
| jack           | 2017-02-03          | 23             | 3               |
| jack           | 2017-04-06          | 42             | 3               |
| mart           | 2017-04-08          | 62             | 3               |
| mart           | 2017-04-09          | 68             | 4               |
| mart           | 2017-04-11          | 75             | 4               |
| mart           | 2017-04-13          | 94             | 4               |
| neil           | 2017-05-10          | 12             | 5               |
| neil           | 2017-06-12          | 80             | 5               |
+----------------+---------------------+----------------+-----------------+--+

select * from (select *,ntile(5) tgroup over(order by orderdate) from business) t1 where t1.tgroup=1;

persent_rank

分组内当前行的RANK值-1/分组内总行数-1

select *,percent_rank() over(order by orderdate) pr from business;
+----------------+---------------------+----------------+----------------------+--+
| business.name  | business.orderdate  | business.cost  |          pr          |
+----------------+---------------------+----------------+----------------------+--+
| jack           | 2017-01-01          | 10             | 0.0                  |
| tony           | 2017-01-02          | 15             | 0.07692307692307693  |
| tony           | 2017-01-04          | 29             | 0.15384615384615385  |
| jack           | 2017-01-05          | 46             | 0.23076923076923078  |
| tony           | 2017-01-07          | 50             | 0.3076923076923077   |
| jack           | 2017-01-08          | 55             | 0.38461538461538464  |
| jack           | 2017-02-03          | 23             | 0.46153846153846156  |
| jack           | 2017-04-06          | 42             | 0.5384615384615384   |
| mart           | 2017-04-08          | 62             | 0.6153846153846154   |
| mart           | 2017-04-09          | 68             | 0.6923076923076923   |
| mart           | 2017-04-11          | 75             | 0.7692307692307693   |
| mart           | 2017-04-13          | 94             | 0.8461538461538461   |
| neil           | 2017-05-10          | 12             | 0.9230769230769231   |
| neil           | 2017-06-12          | 80             | 1.0                  |
+----------------+---------------------+----------------+----------------------+--+

开窗函数

示例表：

+----------------+---------------------+----------------+--+
| business.name  | business.orderdate  | business.cost  |
+----------------+---------------------+----------------+--+
| jack           | 2017-01-01          | 10             |
| tony           | 2017-01-02          | 15             |
| jack           | 2017-02-03          | 23             |
| tony           | 2017-01-04          | 29             |
| jack           | 2017-01-05          | 46             |
| jack           | 2017-04-06          | 42             |
| tony           | 2017-01-07          | 50             |
| jack           | 2017-01-08          | 55             |
| mart           | 2017-04-08          | 62             |
| mart           | 2017-04-09          | 68             |
| neil           | 2017-05-10          | 12             |
| mart           | 2017-04-11          | 75             |
| neil           | 2017-06-12          | 80             |
| mart           | 2017-04-13          | 94             |
+----------------+---------------------+----------------+--+```

```sql
select name,orderdate,cost, 
sum(cost) over() as sample1,--所有行相加 
sum(cost) over(partition by name) as sample2,--按name分组，组内数据相加 
sum(cost) over(partition by name order by orderdate) as sample3,--按name分组，组内数据累加 
sum(cost) over(partition by name order by orderdate rows between UNBOUNDED PRECEDING and current row ) as sample4 ,--和sample3一样,由起点到当前行的聚合 
sum(cost) over(partition by name order by orderdate rows between 1 PRECEDING and current row) as sample5, --当前行和前面一行做聚合 
sum(cost) over(partition by name order by orderdate rows between 1 PRECEDING AND 1 FOLLOWING ) as sample6,--当前行和前边一行及后面一行 
sum(cost) over(partition by name order by orderdate rows between current row and UNBOUNDED FOLLOWING ) as sample7 --当前行及后面所有行 
from business;

其中sample3和sample4是一样的，都是按name分组，组内数据累加。上面总共开了7个窗口函数，select执行完了之后（select不需要执行MapReduce程序），每多一个窗口，就多一个MapReduce执行函数，但是这个前提是窗口开的不一样，只有窗口开的不一样才有额外的MapReduce，sample3~sample7的窗口都是一样的，只不过他们各自加的行的范围不一样而已，所以窗口都是一个窗口。

排序函数

排序函数有rank()、dense_rank()、row_number()，下面对比差异。

给定下表：

+-------------+----------------+--------------+--+
| score.name  | score.subject  | score.score  |
+-------------+----------------+--------------+--+
| 孙悟空         | 语文             | 87           |
| 孙悟空         | 数学             | 95           |
| 孙悟空         | 英语             | 68           |
| 大海          | 语文             | 94           |
| 大海          | 数学             | 56           |
| 大海          | 英语             | 84           |
| 宋宋          | 语文             | 64           |
| 宋宋          | 数学             | 86           |
| 宋宋          | 英语             | 84           |
| 婷婷          | 语文             | 65           |
| 婷婷          | 数学             | 85           |
| 婷婷          | 英语             | 78           |
+-------------+----------------+--------------+--+
select *,rank() over(partition by subject order by score desc) r, 
dense_rank() over(partition by subject order by score desc) dr,
row_number() over(partition by subject order by score desc) rr
from score;
+-------------+----------------+--------------+----+-----+-----+--+
| score.name  | score.subject  | score.score  | r  | dr  | rr  |
+-------------+----------------+--------------+----+-----+-----+--+
| 孙悟空         | 数学             | 95           | 1  | 1   | 1   |
| 宋宋          | 数学             | 86           | 2  | 2   | 2   |
| 婷婷          | 数学             | 85           | 3  | 3   | 3   |
| 大海          | 数学             | 56           | 4  | 4   | 4   |
| 宋宋          | 英语             | 84           | 1  | 1   | 1   |
| 大海          | 英语             | 84           | 1  | 1   | 2   |
| 婷婷          | 英语             | 78           | 3  | 2   | 3   |
| 孙悟空         | 英语             | 68           | 4  | 3   | 4   |
| 大海          | 语文             | 94           | 1  | 1   | 1   |
| 孙悟空         | 语文             | 87           | 2  | 2   | 2   |
| 婷婷          | 语文             | 65           | 3  | 3   | 3   |
| 宋宋          | 语文             | 64           | 4  | 4   | 4   |
+-------------+----------------+--------------+----+-----+-----+--+

注：排序还可以用累加至当前行实现，效果和row_number()相同

count(1) over(partition by subject order by score desc rows between unbounded preceding and current row) as rank

时间函数

months_between

	MONTHS_BETWEEN (date1, date2)用于计算date1和date2之间有几个月。如果date1在日历中比date2晚，那么MONTHS_BETWEEN()就返回一个正数。如果date1在日历中比date2早，那么MONTHS_BETWEEN()就返回一个负数。如果date1和date2日期一样，那MONTHS_BETWEEN()就返回一个0。

hive> select months_between('2020-10-21','2020-08-20');
OK
2.03225806
Time taken: 0.995 seconds, Fetched: 1 row(s)
hive> select months_between('2020-08-20','2020-10-21');
OK
-2.03225806
Time taken: 0.076 seconds, Fetched: 1 row(s)
hive> select months_between('2020-08-20','2020-08-20');
OK
0.0
Time taken: 0.056 seconds, Fetched: 1 row(s)




# 行专列/列转行
https://zhuanlan.zhihu.com/p/115913870
https://blog.csdn.net/jiantianming2/article/details/79189672


## Hive Map Reduce个数如何设置? 来自面试官的10大连环拷问
https://zhuanlan.zhihu.com/p/270002498