Hadoop基础教程-第11章 Hive：SQL on Hadoop（11.3 Hive 快速入门）

第11章 Hive：SQL on Hadoop

11.3 Hive快速入门

11.3.1 HQL介绍

Hive查询语言（Hive QL，可以简称HQL）的语法和SQL非常相似，实际上HQL基本上实现了SQL-92标准，并做了一些扩展。但是仍然存在一些差异，比如不支持行级操作，不支持事务处理。HQL更接近MySQL的SQL方言，对于熟悉SQL语言的开发者而言，HQL很容易上手。

11.3.2 进入Hive Shell

Hive Shell即是Hive的CLI交互模式，输入hive回车即可进入。

[root@node3 ~]# hive
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/opt/hive-2.1.1/lib/log4j-slf4j-impl-2.4.1.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/opt/hadoop-2.7.3/share/hadoop/common/lib/slf4j-log4j12-1.7.10.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.apache.logging.slf4j.Log4jLoggerFactory]

Logging initialized using configuration in jar:file:/opt/hive-2.1.1/lib/hive-common-2.1.1.jar!/hive-log4j2.properties Async: true
Hive-on-MR is deprecated in Hive 2 and may not be available in the future versions. Consider using a different execution engine (i.e. spark, tez) or using Hive 1.X releases.
hive>

输入“quit;”即可退出HiveShell

hive> quit;
[root@node3 ~]#

11.3.3 Hive数据库

（1）显示数据库实例

Hive数据库本质上就是表的目录，或者说命名空间。如果创建数据表没有指定数据库，则使用默认数据库default。与MySQL一样，可以通过show databases命令查看已经存在的数据库。

hive> show databases;
OK
default
Time taken: 1.451 seconds, Fetched: 1 row(s)
hive>

（2）创建数据库

使用create database语句创建数据库。

hive> create database test;
OK
Time taken: 0.414 seconds
hive> show databases;
OK
default
test
Time taken: 0.021 seconds, Fetched: 2 row(s)
hive>

（3）查看数据库结构

describe 命令(可以简写desc)来查看数据库定义，包括：数据库名称、数据库在 HDFS 目录、HDFS

hive> desc database test;
OK
test        hdfs://cetc/user/hive/warehouse/test.db root    USER
Time taken: 0.223 seconds, Fetched: 1 row(s)
hive>

Hive会为每个创建的数据库在HDFS上创建一个目录，该数据库的表会以子目录形式存储，表中的数据会以表目录下的文件形式存储。对于default数据库，默认的缺省数据库没有自己的目录，default数据库下的表默认存放在/user/hive/warehouse目录下。

（4）切换当前数据库

hive> use test;
OK
Time taken: 1.217 seconds
hive>

11.3.4 创建表

在默认default数据库下创建表users

hive> create table users(id int,name char(20));
OK
Time taken: 0.481 seconds
hive>

在test数据库下创建表tb1

hive> create table test.tb1(
> id int,
> name string,
> price double
> );
OK
Time taken: 2.044 seconds
hive>

11.3.4 查看表

（1）显示default数据库下的所有表

hive> show tables;
OK
users
Time taken: 0.231 seconds, Fetched: 1 row(s)
hive>

（2）查看表结构

hive> desc users;
OK
id                      int
name                    char(20)
Time taken: 0.171 seconds, Fetched: 2 row(s)
hive>

（3）查看test数据库下的数据表

hive> use test;
OK
Time taken: 0.037 seconds
hive> show tables;
OK
tb1
Time taken: 0.036 seconds, Fetched: 1 row(s)
hive>

也可以通过下面命令实现上面功能

hive> show tables in test;
OK
tb1
Time taken: 0.04 seconds, Fetched: 1 row(s)
hive>

查看tb1表结构

hive> desc tb1;
OK
id                      int
name                    string
price                   double
Time taken: 0.138 seconds, Fetched: 3 row(s)
hive>

11.3.5 插入数据

（1）显式字段插入

hive> insert into users(id,name) values(1,'aa');
WARNING: Hive-on-MR is deprecated in Hive 2 and may not be available in the future versions. Consider using a different execution engine (i.e. spark, tez) or using Hive 1.X releases.
Query ID = root_20170810094706_d75cb718-9742-4b25-864c-44a63c27d064
Total jobs = 3
Launching Job 1 out of 3
Number of reduce tasks is set to 0 since there's no reduce operator
Starting Job = job_1502371586068_0001, Tracking URL = http://node1:8088/proxy/application_1502371586068_0001/ Kill Command = /opt/hadoop-2.7.3/bin/hadoop job  -kill job_1502371586068_0001
Hadoop job information for Stage-1: number of mappers: 1; number of reducers: 0
2017-08-10 09:47:34,602 Stage-1 map = 0%,  reduce = 0%
2017-08-10 09:47:50,211 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 2.36 sec
MapReduce Total cumulative CPU time: 2 seconds 360 msec
Ended Job = job_1502371586068_0001
Stage-4 is selected by condition resolver.
Stage-3 is filtered out by condition resolver.
Stage-5 is filtered out by condition resolver.
Moving data to directory hdfs://cetc/user/hive/warehouse/users/.hive-staging_hive_2017-08-10_09-47-06_012_378591815767107640-1/-ext-10000
Loading data to table default.users
MapReduce Jobs Launched:
Stage-Stage-1: Map: 1   Cumulative CPU: 2.36 sec   HDFS Read: 4225 HDFS Write: 92 SUCCESS
Total MapReduce CPU Time Spent: 2 seconds 360 msec
OK
Time taken: 46.216 seconds
hive>

（2）隐式字段插入

hive> insert into users values(2,'bb');
WARNING: Hive-on-MR is deprecated in Hive 2 and may not be available in the future versions. Consider using a different execution engine (i.e. spark, tez) or using Hive 1.X releases.
Query ID = root_20170810095011_cf3ab969-c1bb-459b-a1eb-f9b411ec0614
Total jobs = 3
Launching Job 1 out of 3
Number of reduce tasks is set to 0 since there's no reduce operator
Starting Job = job_1502371586068_0002, Tracking URL = http://node1:8088/proxy/application_1502371586068_0002/ Kill Command = /opt/hadoop-2.7.3/bin/hadoop job  -kill job_1502371586068_0002
Hadoop job information for Stage-1: number of mappers: 1; number of reducers: 0
2017-08-10 09:50:28,186 Stage-1 map = 0%,  reduce = 0%
2017-08-10 09:50:41,103 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 2.34 sec
MapReduce Total cumulative CPU time: 2 seconds 340 msec
Ended Job = job_1502371586068_0002
Stage-4 is selected by condition resolver.
Stage-3 is filtered out by condition resolver.
Stage-5 is filtered out by condition resolver.
Moving data to directory hdfs://cetc/user/hive/warehouse/users/.hive-staging_hive_2017-08-10_09-50-11_441_8940366110116441398-1/-ext-10000
Loading data to table default.users
MapReduce Jobs Launched:
Stage-Stage-1: Map: 1   Cumulative CPU: 2.34 sec   HDFS Read: 4228 HDFS Write: 92 SUCCESS
Total MapReduce CPU Time Spent: 2 seconds 340 msec
OK
Time taken: 32.393 seconds
hive>

（3）插入多条记录

hive> insert into users values(3,'cc'),(4,'dd'),(5,'ee');
WARNING: Hive-on-MR is deprecated in Hive 2 and may not be available in the future versions. Consider using a different execution engine (i.e. spark, tez) or using Hive 1.X releases.
Query ID = root_20170810095507_48fc0ce3-7789-4427-b35c-3809586f208f
Total jobs = 3
Launching Job 1 out of 3
Number of reduce tasks is set to 0 since there's no reduce operator
Starting Job = job_1502371586068_0003, Tracking URL = http://node1:8088/proxy/application_1502371586068_0003/ Kill Command = /opt/hadoop-2.7.3/bin/hadoop job  -kill job_1502371586068_0003
Hadoop job information for Stage-1: number of mappers: 1; number of reducers: 0
2017-08-10 09:55:24,363 Stage-1 map = 0%,  reduce = 0%
2017-08-10 09:55:36,294 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 2.35 sec
MapReduce Total cumulative CPU time: 2 seconds 350 msec
Ended Job = job_1502371586068_0003
Stage-4 is selected by condition resolver.
Stage-3 is filtered out by condition resolver.
Stage-5 is filtered out by condition resolver.
Moving data to directory hdfs://cetc/user/hive/warehouse/users/.hive-staging_hive_2017-08-10_09-55-07_839_6570192849374755881-1/-ext-10000
Loading data to table default.users
MapReduce Jobs Launched:
Stage-Stage-1: Map: 1   Cumulative CPU: 2.35 sec   HDFS Read: 4238 HDFS Write: 138 SUCCESS
Total MapReduce CPU Time Spent: 2 seconds 350 msec
OK
Time taken: 31.315 seconds
hive>

可见，Hive每次向表中插入一条记录，都会转换为MapReduce程序，效率低下。

11.3.6 简单查询

（1）字段查询

hive> select id,name from users;
OK
1   aa
2   bb
3   cc
4   dd
5   ee
Time taken: 0.224 seconds, Fetched: 5 row(s)
hive>

（2）条件查看

hive> select * from users where id=1;
OK
1   aa
Time taken: 0.399 seconds, Fetched: 1 row(s)
hive> select name from users where name like '%a';
OK
aa
Time taken: 0.214 seconds, Fetched: 1 row(s)
hive>

（3）统计查询

hive> select count(*) from users;
WARNING: Hive-on-MR is deprecated in Hive 2 and may not be available in the future versions. Consider using a different execution engine (i.e. spark, tez) or using Hive 1.X releases.
Query ID = root_20170810095858_fa356df4-8b18-4a5b-835a-643427dd99f6
Total jobs = 1
Launching Job 1 out of 1
Number of reduce tasks determined at compile time: 1
In order to change the average load for a reducer (in bytes):
set hive.exec.reducers.bytes.per.reducer=<number>
In order to limit the maximum number of reducers:
set hive.exec.reducers.max=<number>
In order to set a constant number of reducers:
set mapreduce.job.reduces=<number>
Starting Job = job_1502371586068_0004, Tracking URL = http://node1:8088/proxy/application_1502371586068_0004/ Kill Command = /opt/hadoop-2.7.3/bin/hadoop job  -kill job_1502371586068_0004
Hadoop job information for Stage-1: number of mappers: 1; number of reducers: 1
2017-08-10 09:59:21,494 Stage-1 map = 0%,  reduce = 0%
2017-08-10 09:59:31,484 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 2.02 sec
2017-08-10 09:59:43,424 Stage-1 map = 100%,  reduce = 100%, Cumulative CPU 4.11 sec
MapReduce Total cumulative CPU time: 4 seconds 110 msec
Ended Job = job_1502371586068_0004
MapReduce Jobs Launched:
Stage-Stage-1: Map: 1  Reduce: 1   Cumulative CPU: 4.11 sec   HDFS Read: 8068 HDFS Write: 101 SUCCESS
Total MapReduce CPU Time Spent: 4 seconds 110 msec
OK
5
Time taken: 46.061 seconds, Fetched: 1 row(s)
hive>

11.3.7 不支持删除修改记录

hive> delete from users where id=1;
FAILED: SemanticException [Error 10294]: Attempt to do update or delete using transaction manager that does not support these operations.
hive> update users set name='xx' where id=1;
FAILED: SemanticException [Error 10294]: Attempt to do update or delete using transaction manager that does not support these operations.
hive>

表中数据不支持删除修改操作，但是可以截断表和删除整张表。

hive> truncate table users;
OK
Time taken: 0.273 seconds
hive> select * from users;
OK
Time taken: 0.243 seconds
hive>

hive> drop table users;
OK
Time taken: 2.025 seconds
hive>