create database db_hive;
show databases;
use db_hibe;
create table student(id int, name string) ROW FORMAT DELIMITED FIELDS TERMINATED BY '\t';
show tables;
desc student;#查看表信息
desc exended student;#详细信息
desc formatted student;#格式化详细信息
load data local inpath '/opt/datas/student.txt'into table db_hive.student ;#加载数据
select * from student;
show functions;# 查看函数
desc function upper;# 查看upper函数的使用方式(转换为大写)
desc function extended upper;# 查看函数的使用案例
在hive cli命令窗口中如何查看hdfs文件系统
hive (default)> dfs -ls / ;
在hive cli命令窗口中如何查看本地文件系统
hive (default)> !ls /opt/datas ;
创建内部表:
create table emp(
empno int,
ename string,
job string,
mgr int,
hiredate string,
sal double,
comm double,
deptno int
)
row format delimited fields terminated by '\t'
加载数据:
load data local inpath '/opt/datas/emp.txt' overwrite into table emp
create table dept(
deptno int,
dname string,
loc string
)
row format delimited fields terminated by '\t'
加载数据:
load data local inpath '/opt/datas/emp.txt' overwrite into table dept
清除一个表的数据:
truncate table dept;
给表改名:
alter table dept_like rename to dept_like_rename;
创建外部表
create EXTERNAL table emp(
empno int,
ename string,
job string,
mgr int,
hiredate string,
sal double,
comm double,
deptno int
)
row format delimited fields terminated by '\t'
location '/user/hive/warehouse/emp'
外部表是指向数据,所以不需要加载数据了
在实际当中,内部表和外部表的区别是:
内部表会删除表数据以及元数据
外部表只会删除元数据,不会删除表数据
创建分区表
create table dept(
deptno int,
dname string,
loc string
)
partitioned by (month string)
row format delimited fields terminated by '\t'
load data local inpath '/opt/datas/emp.txt' overwrite into table dept
partition (month='201710');
查询分区表:
select * from dept where month = '201710'
查询多张分区表(按季度查询)
select * from dept where month = '201709'
union
select * from dept where month = '201708'
union
select * from dept where month = '201707'
;
分区表(二级分区)
create table dept(
deptno int,
dname string,
loc string
)
partitioned by (month string, day String)
row format delimited fields terminated by '\t'
load data local inpath '/opt/datas/emp.txt' overwrite into table dept
partition (month='201710', day='13');
查询二级分区表:
select * from dept where month = '201710' and day = '13'
注意事项:
create table default.dept_nopart(
deptno int,
dname string,
loc string
)
row format delimited fields terminated by '\t'
dfs -put /opt/datas/dept.txt /user/hive/warehouse/dept_nopart;
select * from dept_nopart;
create table default.dept_part(
deptno int,
dname string,
loc string
)
partitioned by (day string)
row format delimited fields terminated by '\t'
# 创建分区表的目录
dfs -mkdir -p /user/hive/warehouse/dept_part/day=201710;
dfs -put /opt/datas/dept.txt /user/hive/warehouse/dept_part;
select * from dept_part;
#这个时候,查询当前的表,会查询不出来数据
我们需要进行一个修复性的操作
第一种修复方式:
msck repair table dept_part;
第二种修复方式:
alter table dept_part add partition(day='201710');
查看表中有几个分区:
show partitions dept_part;
********************************************************
Hive导入数据的六大方式:
load data [local] inpath 'filepath' [overwrite] into table tablename
[partition (partcal1=val1, ...)]
* 原始文件存储的位置
* 本地文件 local
* hdfs
* 对表的数据是否覆盖
* 覆盖 overwrite
* 追加就不加overwrite
* 分区表加载,特殊性
* [partition (partcal1=val1, ...)]
1) 加载本地文件到hive表
local data local inpath 'opt/datas/emp.txt' into table default.emp;
2) 加载hdfs文件到hive表
local data inpath 'opt/datas/emp.txt' into table default.emp;
3) 加载数据覆盖表中已有的数据
local data inpath 'opt/datas/emp.txt' overwrite into table default.emp;
4) 创建表的时候通过insert加载
create table default.emp_ci like emp;
insert into table default.emp_ci select * from default.emp
5) 创建表的时候通过location指定加载
create EXTERNAL table emp(
empno int,
ename string,
job string
)
row format delimited fields terminated by '\t'
location '/user/hive/warehouse/emp'
********************************************************
Hive导出数据的几种方式:
insert overwrite local directory '/opt/datas/hive_exp_emp'
select * from default.emp;
限制格式
insert overwrite local directory '/opt/datas/hive_exp_emp'
row format delimited fields terminated by '\t'
collection items terminated by '\n'
select * from default.emp;
第三种:
bin/hive -e "select * from default.emp;" > /opt/datas/exp_res.txt
第四种(保存在hdfs上):
insert overwrite directory '/opt/datas/hive_exp_emp'
row format delimited fields terminated by '\t'
collection items terminated by '\n'
select * from default.emp;
第五种:
sqoop
hdfs/hive -> rdbms
rdbms -> hdfs/hive/hbase
********************************************************
Hive中的常见查询:
select * from emp
select t.empno from emp t;
select t.empno from emp t where t.sal between 800 and 1500;
select count(*) cnt from emp;
select max(sal) max_sal from emp;
select sum(sal) from emp;
select avg(sal) from emp;
=====================
group by /having
分组
emp表
* 每个部门的平均工资
select t.deptno, avg(t.sal) avg_sal from emp t group by t.deptno;
* 每个部门中每个岗位的最高薪水
select t.deptno, t.job, max(t.sal) max_sal from emp t group by t.deptno, job;
>>> having
* where 是针对单条记录进行筛选
* having 是针对分组结果进行筛选(通常与group by组合)
* 求每个部门的平均薪水大于2000
select deptno,avg(sal) from emp group by deptno having avg_sal > 2000;
join
两个表进行连接
select e.empno,d.deptno from emp e join dept d on e.deptno = d.deptno;
左连接
select e.empno,d.deptno from emp e left join dept d on e.deptno = d.deptno;
右连接
select e.empno,d.deptno from emp e right join dept d on e.deptno = d.deptno;
全连接
full join
select e.empno,d.deptno from emp e full join dept d on e.deptno = d.deptno;
********************************************************
Export
导出,将Hive表中的数据,导出到外部
Import
导入,将外部数据导入Hive表中
EXPORT TABLE emp TO 'export_target_path'
export_target_path:
指的是HDFS上的路径
EXPORT TABLE default.emp TO '/user/hive/export/emp_exp'
create table db_hive.emp like default.emp;
import table db_hive.emp from 'hdfs_exports_location/department';
********************************************************
>> order by
全局排序,一个reduce
select * from emp order by empno desc;
>> sort by
每个reduce内部进行排序,全局不是排序
set mapreduce.job.reduces = 3;
select * from emp sort by empno asc;
>> distribute by
分区partition
类似于mapreduce中分区partition,对数据进行分区,结合sort by进行使用
>> cluster by
结合
当distribute by和sort by字段相同时,可以使用cluster by进行代替
********************************************************
UDF编程
User Definition Function(用户自定义函数)
public calss LowerUDF extends UDF {
public Text evaluate(Text str){
if(null == str.toString()){
return null;
}
return new Text(str.toString().toLowerCase());
}
}
select ename, my_lower(ename) lowername from emp limit 5;
或:
create function self_lower as 'mycalss' using jar 'hdfs:///path/to/jar';
********************************************************
Hive 数据存储
* 按行存储
* 按列存储
TEXTFILE方式(按行存储):
create table page_views(
track_time string,
url string,
ip string,
city_id string
)
row format delimited fields terminated by '\t'
stored as textfile;
local data local inpath 'opt/datas/page_views.data' into table page_views;
ORC方式(按列存储):
create table page_views_orc(
track_time string,
url string,
ip string,
city_id string
)
row format delimited fields terminated by '\t'
stored as orc;
insert into table page_views_orc select * from page_views;
parquet方式(按列存储):
create table page_views_parquet(
track_time string,
url string,
ip string,
city_id string
)
row format delimited fields terminated by '\t'
stored as parquet;
insert into table page_views_parquet select * from page_views;
数据压缩:
create table page_views_orc_snappy(
track_time string,
url string,
ip string,
city_id string
)
row format delimited fields terminated by '\t'
stored as orc tblproperties ("orc.compress"="SNAPPY")
总结:
在实际项目开发中,hive表的数据
* 存储格式
orcfile / parquet
* 数据压缩方式
snappy
********************************************************
Hive 企业优化
大表[拆分]
子表
外部表、分区表
结合使用
多级分区[一般二级]
数据
存储格式[textfile\orcfile\parquet]
数据压缩
snappy
SQL
优化SQL语句
join、filter
MapReduce
Reduct Number
JVM重用
推测执行
转载于:https://my.oschina.net/hehongbo/blog/1547419