Hive与Oracle表关联语句对比

最新推荐文章于 2023-02-28 12:39:37 发布

置顶

龙纹披风么么哒

最新推荐文章于 2023-02-28 12:39:37 发布

阅读量2.7k

点赞数 3

分类专栏： Hadoop 文章标签：数据库 ORACLE HIVE 表关联大数据

本文链接：https://blog.csdn.net/HENRYfabrgeas/article/details/24689935

版权

本文探讨了在从ORACLE迁移至HIVE过程中遇到的表关联语法差异，详细比较了内关联、左关联、右关联、全外关联和笛卡尔积的实现方式。在HIVE中，关联条件不能用where表达，左、右关联和全外关联需明确指定，且不存在类似ORACLE中的某些语法结构。

摘要由CSDN通过智能技术生成

在将ORACLE存储过程迁移到HIVE平台时，不可避免地会遇到表关联的相应语法问题。

本文详细对比了ORALCE和HIVE的各种表关联语法，包括内关联，左，右关联，全外关联和笛卡尔积。

一.创建表

ORACLE：

create table a
(
a1  number(10),
a2 varchar2(50)
);

create table b
(
b1  number(10),
b2 varchar2(50)
);

HIVE：

CREATE TABLE IF NOT EXISTS a (
a1 STRING,
a2 STRING)
COMMENT 'TABLE A'
ROW FORMAT DELIMITED
FIELDS TERMINATED BY '|'
LINES TERMINATED BY '\n'
STORED AS TEXTFILE
TBLPROPERTIES ( 'created_at'='2014-04-28','creator'='HENRY' );

二.插入数据

ORACLE：

insert into a(a1,a2) values(1,'X');
insert into a(a1,a2) values(2,'Y');
insert into a(a1,a2) values(3,'Z');

insert into b(b1,b2) values(1,'X');
insert into b(b1,b2) values(2,'Y');
insert into b(b1,b2) values(4,'Z');

HIVE：

hive (default)> load data local inpath './data1' into table a;
Copying data from file:/home/hadoop/roger/sql/renguihe/data
Copying file: file:/home/hadoop/roger/sql/renguihe/data
Loading data to table default.a
Table default.a stats: [num_partitions: 0, num_files: 1, num_rows: 0, total_size: 12, raw_data_size: 0]
OK
Time taken: 1.961 seconds
hive (default)> load data local inpath './data1' into table b;
Copying data from file:/home/hadoop/roger/sql/renguihe/data
Copying file: file:/home/hadoop/roger/sql/renguihe/data
Loading data to table default.b
Table default.b stats: [num_partitions: 0, num_files: 1, num_rows: 0, total_size: 12, raw_data_size: 0]
OK
Time taken: 0.392 seconds

其中data1数据文件内容为：

1|X
2|Y
3|Z

data2数据文件内容为：

1|X
2|Y
4|Z

三.等值关联

ORACLE：

select * from a,b where a.a1 = b.b1;

或：

select * from a join b on a.a1 = b.b1;

结果如下图所示：

HIVE：

select * from a join b on a.a1 = b.b1;

注意HIVE中不能使用where来表示关联条件。

执行过程及结果如下图所示：

hive (default)> select * from a join b on a.a1 = b.b1;        
Total MapReduce jobs = 1
setting HADOOP_USER_NAME        hadoop
Execution log at: /tmp/hadoop/.log
2014-04-29 09:13:27     Starting to launch local task to process map join;      maximum memory = 1908932608
2014-04-29 09:13:27     Processing rows:        3       Hashtable size: 3       Memory usage:   110981704       rate:   0.058
2014-04-29 09:13:27     Dump the hashtable into file: file:/tmp/hadoop/hive_2014-04-29_09-13-25_273_8486588204512196396/-local-10002/HashTable-Stage-3/MapJoin-mapfile00--.hashtable
2014-04-29 09:13:27     Upload 1 File to: file:/tmp/hadoop/hive_2014-04-29_09-13-25_273_8486588204512196396/-local-10002/HashTable-Stage-3/MapJoin-mapfile00--.hashtable File size: 438
2014-04-29 09:13:27     End of local task; Time Taken: 0.339 sec.
Execution completed successfully
Mapred Local Task Succeeded . Convert the Join into MapJoin
Mapred Local Task Succeeded . Convert the Join into MapJoin
Launching Job 1 out of 1
Number of reduce tasks is set to 0 since there's no reduce operator
Starting Job = job_201404251509_0131, Tracking URL =