Phoenix详细的信息可以查看： 官网 | apache/phoenix | Download | Performance | 支持的SQL语法
CDH6中集成的HBase版本为2.1.0+cdh6.2.0 ,我们从官方的镜像资源下载列表中看到最新的CDH支持到cdh5.14.2的版本，这种版本的直接有一个parcels包，可以通过Cloudera Manager的parcel页面添加资源，来安装组件。
通过查看源码git branch -a可以看到 remotes/origin/5.x-cdh6 分支，但在源码编译安装时这个分支还不完整，编译出来时会缺少一些包，因此我们直接尝试安装v5.0.0-HBase-2.0版本的Phoenix，安装方式有两种，第一种直接下载官方编译好的 apache-phoenix-5.0.0-HBase-2.0包apache-phoenix-5.0.0-HBase-2.0-bin.tar.gz；第二种是通过编译源码的方式来安装。这里我们直接使用源码编译来进行安装。
目录
- 1 Installation
- 1.1 源码编译
- 1.2 安装
- 1.3 Cloudera Manager 的配置
- 2 Using
- 2.1 NoClassDefFoundError解决
- 2.2 原有HBase表的映射同步
- 2.3 查看表信息和插入数据
- 2.4 数据类型
- 2.5 修改表信息
- 2.6 删除库
- 2.7 执行SQL脚本
- 2.8 DDL语法
- 2.9 DML和DQL语法
- 2.9.1 Join
- 2.10 JDBC
- 2.10.1 新建一个Maven工程
- 2.10.2 pom.xml
- 2.10.3 JDBC Client代码
- 3 工具
- 3.1 DBeaver
- 3.2 SQuirrel SQL
- 4 资料
1 Installation
1.1 源码编译
# clone源码
git clone https://github.com/apache/phoenix.git
# 进入源码根目录
cd phoenix/
# 查看tag ,或者查看分支 git branch -a
git tag
# 选择一个版本，进入其分支，或者切换到某个分支：git checkout -b remotes/origin/5.x-cdh6
git checkout tags/v5.0.0-HBase-2.0
# 编译
mvn -T2C install -DskipTests
如果编译时报如下错误：
[INFO] ------------------------------------------------------------------------
[INFO] BUILD FAILURE
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 01:14 min (Wall Clock)
[INFO] Finished at: 2019-07-15T13:57:32+08:00
[INFO] ------------------------------------------------------------------------
[ERROR] Failed to execute goal net.alchim31.maven:scala-maven-plugin:3.2.0:compile (scala-compile-first) on project phoenix-spark: wrap: org.apache.commons.exec.ExecuteException: Process exited with an error: 1 (Exit value: 1) -> [Help 1]
[ERROR]
[ERROR] To see the full stack trace of the errors, re-run Maven with the -e switch.
[ERROR] Re-run Maven using the -X switch to enable full debug logging.
[ERROR]
[ERROR] For more information about the errors and possible solutions, please read the following articles:
[ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/MojoExecutionException
[ERROR]
[ERROR] After correcting the problems, you can resume the build with the command
[ERROR] mvn <goals> -rf :phoenix-spark
可以看到是phoenix-spark模块的net.alchim31.maven:scala-maven-plugin:3.2.0插件编译时发生了错误，我们可以将其原有的scala-maven-plugin注释掉，修改为如下，然后重新编译：
<plugin>
<groupId>net.alchim31.maven</groupId>
<artifactId>scala-maven-plugin</artifactId>
<version>3.2.2</version>
<executions>
<execution>
<goals>
<goal>compile</goal>
<goal>testCompile</goal>
</goals>
</execution>
</executions>
</plugin>
编译成功后如下：
[INFO] ------------------------------------------------------------------------
[INFO] Reactor Summary for Apache Phoenix 5.0.0-HBase-2.0:
[INFO]
[INFO] Apache Phoenix ..................................... SUCCESS [ 9.616 s]
[INFO] Phoenix Core ....................................... SUCCESS [03:34 min]
[INFO] Phoenix - Flume .................................... SUCCESS [01:56 min]
[INFO] Phoenix - Kafka .................................... SUCCESS [01:33 min]
[INFO] Phoenix - Pig ...................................... SUCCESS [04:59 min]
[INFO] Phoenix Query Server Client ........................ SUCCESS [ 48.315 s]
[INFO] Phoenix Query Server ............................... SUCCESS [07:59 min]
[INFO] Phoenix - Pherf .................................... SUCCESS [ 8.362 s]
[INFO] Phoenix - Spark .................................... SUCCESS [09:20 min]
[INFO] Phoenix - Hive ..................................... SUCCESS [57:28 min]
[INFO] Phoenix Client ..................................... SUCCESS [01:20 min]
[INFO] Phoenix Server ..................................... SUCCESS [ 37.906 s]
[INFO] Phoenix Assembly ................................... SUCCESS [ 19.902 s]
[INFO] Phoenix - Tracing Web Application .................. SUCCESS [ 9.255 s]
[INFO] Phoenix Load Balancer .............................. SUCCESS [ 19.176 s]
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 01:01 h (Wall Clock)
[INFO] Finished at: 2019-07-07T11:09:10+08:00
[INFO] ------------------------------------------------------------------------
1.2 安装
将编译成功的phoenix-assembly/target/phoenix-5.0.0-HBase-2.0.tar.gz包拷贝到Phoenix的安装节点，然后解压
cp phoenix-assembly/target/phoenix-5.0.0-HBase-2.0.tar.gz /opt/
cd /opt
tar -zxf phoenix-5.0.0-HBase-2.0.tar.gz
cd phoenix-5.0.0-HBase-2.0
此时为了方便可以配置下环境变量
vim ~/.bash_profile
# 添加如下配置，并保存退出
export PHOENIX_HOME=/opt/phoenix-5.0.0-HBase-2.0
export PATH=$PATH:$PHOENIX_HOME/bin
接下来将Phoenix根目录下的phoenix-5.0.0-HBase-2.0-*.jar 包分发到每个RegionServer的lib下
scp phoenix-5.0.0-HBase-2.0-*.jar root@cdh3:/opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/hbase/lib/
scp phoenix-5.0.0-HBase-2.0-*.jar root@cdh2:/opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/hbase/lib/
scp phoenix-5.0.0-HBase-2.0-*.jar root@cdh1:/opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/hbase/lib/
这里最好也将phoenix-5.0.0-HBase-2.0-client.jar拷贝到Phoenix安装的根目录下
cp phoenix-client/target/phoenix-5.0.0-HBase-2.0-client.jar $PHOENIX_HOME/
1.3 Cloudera Manager 的配置
登陆Cloudera Manager页面，然后修改HBase配置，进入配置页面后搜索hbase-site.xml 的 HBase 服务高级配置代码段，然后添加如下配置，然后重启 HBase 使配置生效。
<property>
<name>hbase.table.sanity.checks</name>
<value>false</value>
</property>
2 Using
进入Phoenix的命令行执行
# 后面跟的参数为HBase配置的ZK的地址
sqlline.py cdh2:2181
2.1 NoClassDefFoundError解决
如果进入Phoenix终端时报如下错误：
Error: org.apache.phoenix.exception.PhoenixIOException: org.apache.hadoop.hbase.DoNotRetryIOException: java.lang.NoClassDefFoundError: org/apache/htrace/Trace
at org.apache.hadoop.hbase.ipc.RpcServer.call(RpcServer.java:469)
at org.apache.hadoop.hbase.ipc.CallRunner.run(CallRunner.java:130)
at org.apache.hadoop.hbase.ipc.RpcExecutor$Handler.run(RpcExecutor.java:324)
at org.apache.hadoop.hbase.ipc.RpcExecutor$Handler.run(RpcExecutor.java:304)
Caused by: java.lang.NoClassDefFoundError: org/apache/htrace/Trace
at org.apache.phoenix.coprocessor.BaseScannerRegionObserver$RegionScannerHolder.overrideDelegate(BaseScannerRegionObserver.java:222)
at org.apache.phoenix.coprocessor.BaseScannerRegionObserver$RegionScannerHolder.nextRaw(BaseScannerRegionObserver.java:273)
at org.apache.hadoop.hbase.regionserver.RSRpcServices.scan(RSRpcServices.java:3134)
at org.apache.hadoop.hbase.regionserver.RSRpcServices.scan(RSRpcServices.java:3383)
at org.apache.hadoop.hbase.shaded.protobuf.generated.ClientProtos$ClientService$2.callBlockingMethod(ClientProtos.java:42002)
at org.apache.hadoop.hbase.ipc.RpcServer.call(RpcServer.java:413)
... 3 more
Caused by: java.lang.ClassNotFoundException: org.apache.htrace.Trace
at java.net.URLClassLoader.findClass(URLClassLoader.java:382)
at java.lang.ClassLoader.loadClass(ClassLoader.java:424)
at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:349)
at java.lang.ClassLoader.loadClass(ClassLoader.java:357)
... 9 more (state=08000,code=101)
可以看到提示缺少org.apache.htrace.Trace类，因此我们需要将htrace-core-3.2.0-incubating.jar拷贝到HBase的lib下：
# CDH环境下已经有这个包了，这里直接将其拷贝到HBase的lib下（每个RegionServer）
# 如果没有可以直接下载这包到HBase的lib下： wget https://repo1.maven.org/maven2/org/apache/htrace/htrace-core/3.2.0-incubating/htrace-core-3.2.0-incubating.jar
cp /opt/cloudera/cm/lib/cdh5/htrace-core-3.2.0-incubating.jar /opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/hbase/lib/
注意： 如果依旧提示缺少这个包，则需要重启HBase服务
再次进入
[root@cdh3 phoenix-5.0.0-HBase-2.0]# bin/sqlline.py cdh2
Setting property: [incremental, false]
Setting property: [isolation, TRANSACTION_READ_COMMITTED]
issuing: !connect jdbc:phoenix:cdh2 none none org.apache.phoenix.jdbc.PhoenixDriver
Connecting to jdbc:phoenix:cdh2
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/opt/phoenix-5.0.0-HBase-2.0/phoenix-5.0.0-HBase-2.0-client.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/jars/slf4j-log4j12-1.7.25.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
Connected to: Phoenix (version 5.0)
Driver: PhoenixEmbeddedDriver (version 5.0)
Autocommit status: true
Transaction isolation: TRANSACTION_READ_COMMITTED
Building list of tables and columns for tab-completion (set fastconnect to true to skip)...
133/133 (100%) Done
Done
sqlline version 1.2.0
0: jdbc:phoenix:cdh2> !tables
+------------+--------------+-------------+---------------+----------+------------+----------------------------+-----------------+--------------+-----------------+---------------+---------------+-----------------+------------+-------------+----------------+-------------+
| TABLE_CAT | TABLE_SCHEM | TABLE_NAME | TABLE_TYPE | REMARKS | TYPE_NAME | SELF_REFERENCING_COL_NAME | REF_GENERATION | INDEX_STATE | IMMUTABLE_ROWS | SALT_BUCKETS | MULTI_TENANT | VIEW_STATEMENT | VIEW_TYPE | INDEX_TYPE | TRANSACTIONAL | IS_NAMESPAC |
+------------+--------------+-------------+---------------+----------+------------+----------------------------+-----------------+--------------+-----------------+---------------+---------------+-----------------+------------+-------------+----------------+-------------+
| | SYSTEM | CATALOG | SYSTEM TABLE | | | | | | false | null | false | | | | false | false |
| | SYSTEM | FUNCTION | SYSTEM TABLE | | | | | | false | null | false | | | | false | false |
| | SYSTEM | LOG | SYSTEM TABLE | | | | | | true | 32 | false | | | | false | false |
| | SYSTEM | SEQUENCE | SYSTEM TABLE | | | | | | false | null | false | | | | false | false |
| | SYSTEM | STATS | SYSTEM TABLE | | | | | | false | null | false | | | | false | false |
+------------+--------------+-------------+---------------+----------+------------+----------------------------+-----------------+--------------+-----------------+---------------+---------------+-----------------+------------+-------------+----------------+-------------+
2.2 原有HBase表的映射同步
Phoenix对表和列名都是区分大小写的，如果不加双引号默认为大写。通过上面可以看到HBase原有的表并没有列出来，这是因为 Phoenix 无法自动识别 HBase 中原有的表，所以需要将 HBase 中已有的做映射，才能够被 Phoenix 识别并操作。
-- 其实也就是使用CREATE语句重新创建表来映射
-- 注意1：建表时一定注意大小写，如果原表小写，记得用双引号，主键最好大写
-- 注意2：建表时不推荐使用太复杂的数据类型，为了安全的映射放置出错，先使用varchar，
-- 否则会出现精度或者格式上的误差，例如时间日期类，整数型长度，小数型精度等
-- 注意3：记得加上 COLUMN_ENCODED_BYTES 否则可能类名解析无法对应到HBase的类，
-- 详细说明查看 http://phoenix.apache.org/columnencoding.html
-- 注意4：如果是创建的表来映射，在Phoenix删除表，HBase的表和数据也会删除，注意数据不要误删了
--
-- 推荐：我这里映射的是表，如果只用来做查询，最好创建一个视图来来映射：只需要将 TABLE 改为 VIEW
0: jdbc:phoenix:cdh2:2181> CREATE TABLE "m_info" (
. . . . . . . . . . . . .> "ROW" varchar primary key,
. . . . . . . . . . . . .> "cf1"."gc" double ,
. . . . . . . . . . . . .> "cf1"."id" bigint ,
. . . . . . . . . . . . .> "cf1"."jd" double ,
. . . . . . . . . . . . .> "cf1"."name" varchar ,
. . . . . . . . . . . . .> "cf1"."wd" double
. . . . . . . . . . . . .> )column_encoded_bytes=0;
1 row affected (7.282 seconds)
-- 查询数据
0: jdbc:phoenix:cdh2> select * from "m_info" ;
+------+-----------------------+-----------------------+-----------------------+-------+-------------------------+
| ROW | gc | id | jd | name | wd |
+------+-----------------------+-----------------------+-----------------------+-------+-------------------------+
| 2 | 1.1 | 2 | 3.33 | 名字1 | 5.1234 |
+------+-----------------------+-----------------------+-----------------------+-------+-------------------------+
2.3 查看表信息和插入数据
-- 查看表结构
0: jdbc:phoenix:cdh2:2181> !describe "m_info";
+------------+--------------+-------------+--------------+------------+------------+--------------+----------------+-----------------+-----------------+-----------+----------+-------------+----------------+-------------------+--------------------+-------------------+---+
| TABLE_CAT | TABLE_SCHEM | TABLE_NAME | COLUMN_NAME | DATA_TYPE | TYPE_NAME | COLUMN_SIZE | BUFFER_LENGTH | DECIMAL_DIGITS | NUM_PREC_RADIX | NULLABLE | REMARKS | COLUMN_DEF | SQL_DATA_TYPE | SQL_DATETIME_SUB | CHAR_OCTET_LENGTH | ORDINAL_POSITION | I |
+------------+--------------+-------------+--------------+------------+------------+--------------+----------------+-----------------+-----------------+-----------+----------+-------------+----------------+-------------------+--------------------+-------------------+---+
| | | m_info | ROW | 12 | VARCHAR | null | null | null | null | 0 | | | null | null | null | 1 | f |
| | | m_info | gc | 8 | DOUBLE | null | null | null | null | 1 | | | null | null | null | 2 | t |
| | | m_info | id | -5 | BIGINT | null | null | null | null | 1 | | | null | null | null | 3 | t |
| | | m_info | jd | 8 | DOUBLE | null | null | null | null | 1 | | | null | null | null | 4 | t |
| | | m_info | name | 12 | VARCHAR | null | null | null | null | 1 | | | null | null | null | 5 | t |
| | | m_info | wd | 8 | DOUBLE | null | null | null | null | 1 | | | null | null | null | 6 | t |
+------------+--------------+-------------+--------------+------------+------------+--------------+----------------+-----------------+-----------------+-----------+----------+-------------+----------------+-------------------+--------------------+-------------------+---+
-- 插入数据。这里注意的是value中的值如果是字符串，必须用单引号
0: jdbc:phoenix:cdh2:2181> UPSERT INTO "m_info"("ROW", "gc", "id", "jd", "name", "wd") VALUES('2', 1.1, 2, 3.33, '名字1', 5.1234);
1 row affected (1.074 seconds)
2.4 数据类型
| 字段类型 | 映射的类 | 值的范围 | 占用字节 |
|---|---|---|---|
| INTEGER | java.lang.Integer |
-2147483648 到 2147483647 | 4字节 |
| UNSIGNED_INT | java.lang.Integer |
0 到 2147483647 | 4字节 |
| BIGINT | java.lang.Long |
-9223372036854775808 到 9223372036854775807 | 8字节 |
| UNSIGNED_LONG | java.lang.Long |
0 到 9223372036854775807 | 8字节 |
| TINYINT | java.lang.Byte |
-128 到 127 | 1字节 |
| UNSIGNED_TINYINT | java.lang.Byte |
0 到 127 | 1字节 |
| SMALLINT | java.lang.Short |
-32768 到 32767 | 2字节 |
| UNSIGNED_SMALLINT | java.lang.Short |
0 到 32767 | 2字节 |
| FLOAT | java.lang.Float |
-3.402823466 E + 38 到 3.402823466 E + 38 | 4字节 |
| UNSIGNED_FLOAT | java.lang.Float |
0到3.402823466 E + 38 | 4字节 |
| DOUBLE | java.lang.Double |
-1.7976931348623158 E + 308 到 1.7976931348623158 E + 308 | 8字节 |
| UNSIGNED_DOUBLE | java.lang.Double |
0 到 1.7976931348623158 E + 308 | 8字节 |
| DECIMAL 或 DECIMAL(precisionInt , scaleInt) | java.math.BigDecimal |
最大精度为38位。可变长度 | |
| BOOLEAN | java.lang.Boolean |
0表示false，1表示true | 1字节 |
| TIME | java.sql.Time |
自纪元以来的毫秒数（基于时间GMT） | 8字节 |
| UNSIGNED_TIME | java.sql.Time |
格式为 yyyy-MM-dd hh:mm:ss | 8字节 |
| DATE | java.sql.Date |
自纪元以来的毫秒数（基于时间GMT） | 8字节 |
| UNSIGNED_DATE | java.sql.Date |
格式为 yyyy-MM-dd hh:mm:ss | 8字节 |
| TIMESTAMP | java.sql.Timestamp |
自纪元以来的毫秒数（基于时间GMT） | 12字节（纪元时间长8字节加纳秒的4字节整数） |
| UNSIGNED_TIMESTAMP | java.sql.Timestamp |
格式为 yyyy-MM-dd hh:mm:ss[.nnnnnnnnn] | 12字节 |
| VARCHAR 或 VARCHAR(precisionInt) | java.lang.String |
可选的最大字节长度 | |
| CHAR(precisionInt) | java.lang.String |
固定长度字符串 | |
| BINARY(precisionInt) | byte[] |
原始固定长度字节数组 | |
| VARBINARY | byte[] |
原始可变长度字节数组 | |
| ARRAY 或 ARRAY[dimensionInt] | java.sql.Array |
2.5 修改表信息
首先Phoenix对字段的映射的元数据信息如下，下面修改字段类型时可以由此进行修改
| DATA_TYPE | TYPE_NAME |
|---|---|
| INTEGER | 4 |
| UNSIGNED_INT | 4 |
| BIGINT | -5 |
| UNSIGNED_LONG | -5 |
| TINYINT | -6 |
| UNSIGNED_TINYINT | -6 |
| SMALLINT | 5 |
| UNSIGNED_SMALLINT | 5 |
| FLOAT | 6 |
| UNSIGNED_FLOAT | 6 |
| DOUBLE | 8 |
| UNSIGNED_DOUBLE | 8 |
| DECIMAL | 3 |
| BOOLEAN | 16 |
| TIME | 92 |
| UNSIGNED_TIME | 92 |
| DATE | 91 |
| UNSIGNED_DATE | 91 |
| TIMESTAMP | 93 |
| UNSIGNED_TIMESTAMP | 93 |
| VARCHAR | 12 |
| CHAR | 1 |
| BINARY | -2 |
| VARBINARY | -3 |
0: jdbc:phoenix:cdh2> CREATE TABLE "test" (
. . . . . . . . . . > "ROW" varchar primary key,
. . . . . . . . . . > "info1"."age" UNSIGNED_TINYINT ,
. . . . . . . . . . > "info1"."name" varchar ,
. . . . . . . . . . > "info1"."property" double
. . . . . . . . . . > )column_encoded_bytes=0;
3 rows affected (8.962 seconds)
-- 查看表中的字段信息
0: jdbc:phoenix:cdh2> select TENANT_ID,TABLE_SCHEM,TABLE_NAME,COLUMN_NAME,COLUMN_FAMILY,DATA_TYPE,COLUMN_SIZE,DECIMAL_DIGITS from SYSTEM.CATALOG where TABLE_NAME='test';
+------------+--------------+-------------+--------------+----------------+------------+--------------+-----------------+
| TENANT_ID | TABLE_SCHEM | TABLE_NAME | COLUMN_NAME | COLUMN_FAMILY | DATA_TYPE | COLUMN_SIZE | DECIMAL_DIGITS |
+------------+--------------+-------------+--------------+----------------+------------+--------------+-----------------+
| | | test | | | null | null | null |
| | | test | ROW | | 12 | null | null |
| | | test | age | info1 | -6 | null | null |
| | | test | name | info1 | 12 | null | null |
| | | test | property | info1 | 8 | null | null |
+------------+--------------+-------------+--------------+----------------+------------+--------------+-----------------+
5 rows selected (0.064 seconds)
-- 将name的varchar类型的长度调整为 32，前4个字段为主键
5 rows selected (0.064 seconds)
0: jdbc:phoenix:cdh2> upsert into SYSTEM.CATALOG (TENANT_ID,TABLE_SCHEM,TABLE_NAME,COLUMN_NAME,COLUMN_FAMILY,COLUMN_SIZE) values('','','test','name','info1',32);
1 row affected (0.493 seconds)
0: jdbc:phoenix:cdh2> upsert into SYSTEM.CATALOG (TENANT_ID,TABLE_SCHEM,TABLE_NAME,COLUMN_NAME,COLUMN_FAMILY,DATA_TYPE) values('','','test','age','info1',-5);
1 row affected (0.019 seconds)
0: jdbc:phoenix:cdh2> select TENANT_ID,TABLE_SCHEM,TABLE_NAME,COLUMN_NAME,COLUMN_FAMILY,DATA_TYPE,COLUMN_SIZE,DECIMAL_DIGITS from SYSTEM.CATALOG where TABLE_NAME='test';
+------------+--------------+-------------+--------------+----------------+------------+--------------+-----------------+
| TENANT_ID | TABLE_SCHEM | TABLE_NAME | COLUMN_NAME | COLUMN_FAMILY | DATA_TYPE | COLUMN_SIZE | DECIMAL_DIGITS |
+------------+--------------+-------------+--------------+----------------+------------+--------------+-----------------+
| | | test | | | null | null | null |
| | | test | ROW | | 12 | null | null |
| | | test | age | info1 | -5 | null | null |
| | | test | name | info1 | 12 | 32 | null |
| | | test | property | info1 | 8 | null | null |
+------------+--------------+-------------+--------------+----------------+------------+--------------+-----------------+
5 rows selected (0.039 seconds)
2.6 删除库
-- 如果HBase的数据和Phoenix的数据是同步的，则可以直接删除SCHEM
0: jdbc:phoenix:cdh3> DROP SCHEMA IF EXISTS my_schema;
-- 如果不能删除，可以先删除HBase的表信息
hbase(main):003:0> disable "test.Person"
Took 0.5061 seconds
hbase(main):004:0> drop "test.Person"
Took 0.2915 seconds
-- 再清除Phoenix的数据
-- 查看CATALOG信息
0: jdbc:phoenix:cdh3> select TENANT_ID,TABLE_SCHEM,TABLE_NAME,COLUMN_NAME,COLUMN_FAMILY,TABLE_SEQ_NUM,TABLE_TYPE,PK_NAME from SYSTEM.CATALOG;
-- 删除里面的表信息
| TENANT_ID | TABLE_SCHEM | TABLE_NAME | COLUMN_NAME | COLUMN_FAMILY | TABLE_SEQ_NUM | TABLE_TYPE | PK_NAME |
+------------+--------------+-------------+-------------------------------------+----------------+----------------+-------------+----------+
| | …… | …… | …… | | null | | |
| | test | Person | | | 0 | u | |
| | test | Person | | 0 | null | | |
| | test | Person | age | 0 | null | | |
| | test | Person | idcard_num | | null | | |
| | test | Person | name | 0 | null | | |
+------------+--------------+-------------+------------------------+----------------+----------------+-------------+----------+
-- 删除SCHEM信息
0: jdbc:phoenix:cdh3> DELETE FROM SYSTEM.CATALOG WHERE TABLE_SCHEM='test';
5 rows affected (0.352 seconds)
-- 再次使用 !tables 查看表，发现库以及表均已经成功删除
2.7 执行SQL脚本
我们使用Phoenix自带的样例数据演示，数据在 $PHOENIX_HOME/examples，具体执行如下：
-- 1 查看数据和脚本，其中 $PHOENIX_HOME 是Phoenix安装的目录。
-- 可以看到有STOCK_SYMBOL数据文件和sql脚本、WEB_STAT数据文件和脚本文件
[root@cdh3 ~]# ls $PHOENIX_HOME/examples
pig STOCK_SYMBOL.csv STOCK_SYMBOL.sql WEB_STAT.csv WEB_STAT_QUERIES.sql WEB_STAT.sql
-- 1.1 查看 STOCK_SYMBOL.sql。可以看到有三个SQL，分别为建表SQL、更新插入一条数据SQL、查询表数据SQL
[root@cdh3 ~]# cat $PHOENIX_HOME/examples/STOCK_SYMBOL.sql
CREATE TABLE IF NOT EXISTS STOCK_SYMBOL (SYMBOL VARCHAR NOT NULL PRIMARY KEY, COMPANY VARCHAR);
UPSERT INTO STOCK_SYMBOL VALUES ('CRM','SalesForce.com');
SELECT * FROM STOCK_SYMBOL;
-- 1.2 查看 WEB_STAT.sql。可以看到这里是一个建表SQL
[root@cdh3 ~]# cat $PHOENIX_HOME/examples/WEB_STAT.sql
CREATE TABLE IF NOT EXISTS WEB_STAT (
HOST CHAR(2) NOT NULL,
DOMAIN VARCHAR NOT NULL,
FEATURE VARCHAR NOT NULL,
DATE DATE NOT NULL,
USAGE.CORE BIGINT,
USAGE.DB BIGINT,
STATS.ACTIVE_VISITOR INTEGER
CONSTRAINT PK PRIMARY KEY (HOST, DOMAIN, FEATURE, DATE)
);
-- 1.3 查看 WEB_STAT_QUERIES.sql。可以看到有三个查询 SQL
[root@cdh3 ~]# cat $PHOENIX_HOME/examples/WEB_STAT_QUERIES.sql
SELECT DOMAIN, AVG(CORE) Average_CPU_Usage, AVG(DB) Average_DB_Usage
FROM WEB_STAT
GROUP BY DOMAIN
ORDER BY DOMAIN DESC;
-- Sum, Min and Max CPU usage by Salesforce grouped by day
SELECT TRUNC(DATE,'DAY') DAY, SUM(CORE) TOTAL_CPU_Usage, MIN(CORE) MIN_CPU_Usage, MAX(CORE) MAX_CPU_Usage
FROM WEB_STAT
WHERE DOMAIN LIKE 'Salesforce%'
GROUP BY TRUNC(DATE,'DAY');
-- list host and total active users when core CPU usage is 10X greater than DB usage
SELECT HOST, SUM(ACTIVE_VISITOR) TOTAL_ACTIVE_VISITORS
FROM WEB_STAT
WHERE DB > (CORE * 10)
GROUP BY HOST;
-- 2 执行自带脚本
[root@cdh3 ~]# $PHOENIX_HOME/bin/sqlline.py localhost $PHOENIX_HOME/examples/STOCK_SYMBOL.sql
148/148 (100%) Done
Done
1/4 CREATE TABLE IF NOT EXISTS STOCK_SYMBOL (SYMBOL VARCHAR NOT NULL PRIMARY KEY, COMPANY VARCHAR);
No rows affected (2.349 seconds)
2/4 UPSERT INTO STOCK_SYMBOL VALUES ('CRM','SalesForce.com');
1 row affected (0.162 seconds)
3/4 SELECT * FROM STOCK_SYMBOL;
+---------+-----------------+
| SYMBOL | COMPANY |
+---------+-----------------+
| CRM | SalesForce.com |
+---------+-----------------+
1 row selected (0.038 seconds)
4/4
Closing: org.apache.phoenix.jdbc.PhoenixConnection
sqlline version 1.2.0
-- 也可以指定csv数据文件和执行sql脚本
[root@cdh3 ~]# psql.py localhost $PHOENIX_HOME/examples/WEB_STAT.sql $PHOENIX_HOME/examples/WEB_STAT.csv $PHOENIX_HOME/examples/WEB_STAT_QUERIES.sql
no rows upserted
Time: 2.709 sec(s)
csv columns from database.
CSV Upsert complete. 39 rows upserted
Time: 19.653 sec(s)
DOMAIN AVERAGE_CPU_USAGE AVERAGE_DB_USAGE
----------------------------- ------------------------------- --------------------------
Salesforce.com 260.727 257.636
Google.com 212.875 213.75
Apple.com 114.111 119.556
Time: 0.127 sec(s)
DAY TOTAL_CPU_USAGE MIN_CPU_USAGE MAX_CPU_USAGE
----------------------- -------------------- -------------------------- ----------------------------
2013-01-01 00:00:00.000 35 35 35
2013-01-02 00:00:00.000 150 25 125
2013-01-03 00:00:00.000 88 88 88
2013-01-04 00:00:00.000 26 3 23
2013-01-05 00:00:00.000 550 75 475
2013-01-06 00:00:00.000 12 12 12
2013-01-08 00:00:00.000 345 345 345
2013-01-09 00:00:00.000 390 35 355
2013-01-10 00:00:00.000 345 345 345
2013-01-11 00:00:00.000 335 335 335
2013-01-12 00:00:00.000 5 5 5
2013-01-13 00:00:00.000 355 355 355
2013-01-14 00:00:00.000 5 5 5
2013-01-15 00:00:00.000 720 65 655
2013-01-16 00:00:00.000 785 785 785
2013-01-17 00:00:00.000 1590 355 1235
Time: 0.045 sec(s)
HO TOTAL_ACTIVE_VISITORS
-- ----------------------------------------
EU 150
NA 1
Time: 0.022 sec(s)
-- 3 登陆 Phoenix。也可以使用Squirrel工具
[root@cdh3 ~]# $PHOENIX_HOME/bin/sqlline.py localhost
2.8 DDL语法
数据库定义语言DDL（Data Definition Language）是我们在创建表的时候用到的一些SQL，比如说：CREATE、ALTER、DROP等。DDL主要是用在定义或改变表的结构，数据类型，表之间的链接和约束等初始化工作上。
-- 1 建表，建表的语句可以看 执行SQL脚本 中的如下两个文件的建表语句
-- 基本上和一般的SQL建表差不多，支持主键，支持 NOT NULL约束，这里又不同的是，建表时可以在字段前加上列族名以区分字段信息
$PHOENIX_HOME/examples/STOCK_SYMBOL.sql
$PHOENIX_HOME/examples/WEB_STAT.sql
-- 2 ALTER一个表，例如在STOCK_SYMBOL添加一列 name
0: jdbc:phoenix:localhost> ALTER TABLE STOCK_SYMBOL ADD "name" varchar(64);
No rows affected (0.749 seconds)
-- 再次查看表数据，发现有多了一列name
0: jdbc:phoenix:localhost> SELECT * FROM STOCK_SYMBOL;
+---------+-----------------+-------+
| SYMBOL | COMPANY | name |
+---------+-----------------+-------+
| CRM | SalesForce.com | |
+---------+-----------------+-------+
3 rows selected (0.15 seconds)
-- 3 ALTER一个表，删除一列，例如将上一步在STOCK_SYMBOL表添加的name一列删除
0: jdbc:phoenix:localhost> ALTER TABLE STOCK_SYMBOL DROP COLUMN "name";
No rows affected (0.605 seconds)
-- 再次查看表数据，发现name这一列已经删除
0: jdbc:phoenix:localhost> SELECT * FROM STOCK_SYMBOL;
+---------+-----------------+
| SYMBOL | COMPANY |
+---------+-----------------+
| CRM | SalesForce.com |
+---------+-----------------+
3 rows selected (0.095 seconds)
-- 如果demo2 表是一个视图，也可以将视图中的某个列字段删去，语法如下
0: jdbc:phoenix:localhost> ALTER VIEW "demo2" DROP COLUMN "comm";
-- 4 删除库
-- 4.1 如果HBase的数据和Phoenix的数据是同步的，则可以直接删除SCHEM
0: jdbc:phoenix:cdh3> DROP SCHEMA IF EXISTS my_schema;
-- 4.2 如果不能删除，可以先删除HBase的表信息
hbase(main):003:0> disable "test.Person"
Took 0.5061 seconds
hbase(main):004:0> drop "test.Person"
Took 0.2915 seconds
-- 再清除Phoenix的数据
-- 4.3 删除SCHEM信息
0: jdbc:phoenix:cdh3> DELETE FROM SYSTEM.CATALOG WHERE TABLE_SCHEM='test';
5 rows affected (0.352 seconds)
-- 再次使用 !tables 查看表，发现库以及表均已经成功删除
2.8.1 加盐表
在密码学中，加盐是指在散列之前将散列内容（例如：密码）的任意固定位置插入特定的字符串。这个在散列中加入字符串的方式称为“加盐”。其作用是让加盐后的散列结果和没有加盐的结果不相同，在不同的应用情景中，这个处理可以增加额外的安全性。而Phoenix中加盐是指对pk对应的byte数组插入特定的byte数据。
加盐能解决HBASE读写热点问题，例如:单调递增rowkey数据的持续写入，使得负载集中在某一个RegionServer上引起的热点问题。在创建表的时候指定属性值：SALT_BUCKETS，其值表示所分buckets(region)数量， 范围是1~256。
-- 加盐表的创建格式如下
CREATE TABLE mytable (my_key VARCHAR PRIMARY KEY, col VARCHAR) SALT_BUCKETS = 8;
2.8.2 自增ID
现在有一个需求是这样的，有一批书籍，需要对其进行编号存储，要求编号要唯一，建表语句如下：
0: jdbc:phoenix:localhost> CREATE TABLE books(
. . . . . . . . . . . . .> id INTEGER NOT NULL PRIMARY KEY,
. . . . . . . . . . . . .> name VARCHAR,
. . . . . . . . . . . . .> author VARCHAR
. . . . . . . . . . . . .> )SALT_BUCKETS = 8;
No rows affected (3.250 seconds)
我们接下来需要使用自增ID，来生成上表的id，但是使用这样的自增ID，容易造成集群的热点问题，因此这里在建表时通过加盐的方式解决这个潜在的问题。下面是Phoenix中自增ID的语法，如下：
-- 自增ID的语法如下：
-- START 用于指定第一个值。如果不指定默认为1.
-- INCREMENT指定每次调用NEXT VALUE FOR后自增大小。 如果不指定默认为1。
-- MINVALUE和MAXVALUE一般与CYCLE连用, 让自增数据形成一个环，从最小值到最大值，再从最大值到最小值。
-- CACHE默认为100, 表示server端生成100个自增序列缓存在客户端，可以减少rpc次数。
-- 此值也可以通过phoenix.sequence.cacheSize来配置。
CREATE SEQUENCE [IF NOT EXISTS] SCHEMA.SEQUENCE_NAME
[START WITH number]
[INCREMENT BY number]
[MINVALUE number]
[MAXVALUE number]
[CYCLE]
[CACHE number]
现在我们使用这个语法创建自增序列，我们定初始值为10000，自增间隔为1，缓存大小为1000，实现如下：
-- 1 创建自增序列
0: jdbc:phoenix:localhost> CREATE SEQUENCE book_sequence START WITH 10000 INCREMENT BY 1 CACHE 1000;
No rows affected (0.174 seconds)
-- 2 通过自增序列，写入数据信息。这里插入两条数据，通过NEXT VALUE FOR 调用自增序列
0: jdbc:phoenix:localhost> UPSERT INTO books(id, name, author) VALUES( NEXT VALUE FOR
. . . . . . . . . . . . .> book_sequence,'Effective JavaScript', 'David Herman');
1 row affected (0.085 seconds)
-- 插入第二条数据
0: jdbc:phoenix:localhost> UPSERT INTO books(id, name, author) VALUES( NEXT VALUE FOR
. . . . . . . . . . . . .> book_sequence,'Speaking Javascript','Dr.Axel Rauschmayer');
1 row affected (0.014 seconds)
-- 3 查看书籍信息表。books没有加引号时默认为大写。发现数据已成功插入，且其ID为从10000以1位步长自增
0: jdbc:phoenix:localhost> SELECT * FROM BOOKS;
+--------+-----------------------+----------------------+
| ID | NAME | AUTHOR |
+--------+-----------------------+----------------------+
| 10001 | Speaking Javascript | Dr.Axel Rauschmayer |
| 10000 | Effective JavaScript | David Herman |
+--------+-----------------------+----------------------+
2 rows selected (0.245 seconds)
2.9 DML和DQL语法
数据操纵语言DML（Data Manipulation Language）主要对表中的数据进行插入、删除、修改的操作，在Phoenix对应的操作就是UPSERT和DELETE。数据查询语言DQL（Data Query Language）主要用于数据的查询，在Phoenix对应的操作就是SELETE。
-- 1 SELECT 。查询 STOCK_SYMBOL 表数据，取其3条
0: jdbc:phoenix:localhost> SELECT * FROM WEB_STAT LIMIT 3;
+-------+-------------+------------+--------------------------+-------+------+-----------------+
| HOST | DOMAIN | FEATURE | DATE | CORE | DB | ACTIVE_VISITOR |
+-------+-------------+------------+--------------------------+-------+------+-----------------+
| EU | Apple.com | Mac | 2013-01-01 01:01:01.000 | 35 | 22 | 34 |
| EU | Apple.com | Store | 2013-01-03 01:01:01.000 | 345 | 722 | 170 |
| EU | Google.com | Analytics | 2013-01-13 08:06:01.000 | 25 | 2 | 6 |
+-------+-------------+------------+--------------------------+-------+------+-----------------+
3 rows selected (0.132 seconds)
-- 2 UPSERT。这个是需要注意的，因为HBase本身的架构特点，没有update语法，Phoenix的插入和修改统一为 UPSERT 关键字。
-- 2.1 插入一条数据，不指定字段，直接填写插入的值，建表时有多少字段就要一一对应，不能漏
0: jdbc:phoenix:localhost> UPSERT INTO STOCK_SYMBOL VALUES('AAPL', 'APPLE Inc.');
1 row affected (0.355 seconds)
-- 2.2 插入一条数据，可以只指定我们需要插入数据的字段，其它没有约束时使用默认值
0: jdbc:phoenix:localhost> UPSERT INTO STOCK_SYMBOL(SYMBOL, COMPANY) VALUES('GOOG', 'Google');
1 row affected (0.009 seconds)
-- 2.3 更新一条数据。结合主键使用UPSERT实现。
-- 例如更新 WEB_STAT ，从建表语句可以看到其主键为 HOST、DOMAIN、FEATURE、DATE，
-- 将HOST='EU', DOMAIN='Apple.com', FEATURE='Mac', DATE='2013-01-01 01:01:01.000'的 ACTIVE_VISITOR改为33
0: jdbc:phoenix:localhost> UPSERT INTO WEB_STAT(HOST, DOMAIN, FEATURE, DATE, ACTIVE_VISITOR) VALUES('EU', 'Apple.com', 'Mac', '2013-01-01 01:01:01.000', 33);
1 row affected (0.191 seconds)
-- 查看结果。发现第一行数据的 ACTIVE_VISITOR 已经更新为33了
0: jdbc:phoenix:localhost> SELECT * FROM WEB_STAT LIMIT 3;
+-------+-------------+------------+--------------------------+-------+------+-----------------+
| HOST | DOMAIN | FEATURE | DATE | CORE | DB | ACTIVE_VISITOR |
+-------+-------------+------------+--------------------------+-------+------+-----------------+
| EU | Apple.com | Mac | 2013-01-01 01:01:01.000 | 35 | 22 | 33 |
| EU | Apple.com | Store | 2013-01-03 01:01:01.000 | 345 | 722 | 170 |
| EU | Google.com | Analytics | 2013-01-13 08:06:01.000 | 25 | 2 | 6 |
+-------+-------------+------------+--------------------------+-------+------+-----------------+
3 rows selected (0.059 seconds)
-- 3统计不同域名（DOMAIN）下访客数（ACTIVE_VISITOR）大于100的数据量
0: jdbc:phoenix:localhost> SELECT DOMAIN,COUNT(1) FROM WEB_STAT
. . . . . . . . . . . . .> WHERE ACTIVE_VISITOR>100 GROUP BY DOMAIN;
+-----------------+-----------+
| DOMAIN | COUNT(1) |
+-----------------+-----------+
| Apple.com | 3 |
| Google.com | 1 |
| Salesforce.com | 4 |
+-----------------+-----------+
3 rows selected (0.041 seconds)
-- 4 CASE，将常见的域名转成中文，其它则返回原始内容
0: jdbc:phoenix:localhost> SELECT (CASE DOMAIN WHEN 'Apple.com' THEN '苹果公司'
. . . . . . . . . . . . .> WHEN 'Google.com' THEN '谷歌'
. . . . . . . . . . . . .> ELSE DOMAIN END ) AS "域名" FROM WEB_STAT GROUP BY DOMAIN;
+-----------------+
| 域名 |
+-----------------+
| 苹果公司 |
| 谷歌 |
| Salesforce.com |
+-----------------+
3 rows selected (0.084 seconds)
-- 5 分页查询。结合order by ,排序字段最好为主键。LIMIT语法如下
-- [ LIMIT { count } ]
-- [ OFFSET start [ ROW | ROWS ] ]
-- [ FETCH { FIRST | NEXT } [ count ] { ROW | ROWS } ONLY ]
-- 例如接下来，我们对WEB_STAT表的数据按照活跃的访客数（ACTIVE_VISITOR）倒序进行分页，
-- 每页显示20条数据，从起始的第一条数据开始显示。
-- LIMIT限制每页显示20条数据， OFFSET 为第一次数据的起始偏移量
0: jdbc:phoenix:localhost> SELECT * FROM WEB_STAT ORDER BY ACTIVE_VISITOR DESC
. . . . . . . . . . . . .> LIMIT 20 OFFSET 0;
+-------+-----------------+------------+--------------------------+-------+-------+-----------------+
| HOST | DOMAIN | FEATURE | DATE | CORE | DB | ACTIVE_VISITOR |
+-------+-----------------+------------+--------------------------+-------+-------+-----------------+
| NA | Salesforce.com | Reports | 2013-01-09 17:36:01.000 | 355 | 432 | 315 |
| NA | Salesforce.com | Login | 2013-01-17 02:20:01.000 | 1235 | 2422 | 243 |
| EU | Google.com | Search | 2013-01-09 01:01:01.000 | 395 | 922 | 190 |
| EU | Apple.com | Store | 2013-01-03 01:01:01.000 | 345 | 722 | 170 |
| NA | Apple.com | Mac | 2013-01-02 04:01:01.000 | 345 | 255 | 155 |
| NA | Salesforce.com | Login | 2013-01-10 01:01:01.000 | 345 | 252 | 150 |
| EU | Salesforce.com | Login | 2013-01-12 01:01:01.000 | 5 | 62 | 150 |
| NA | Apple.com | Login | 2013-01-04 01:01:01.000 | 135 | 2 | 110 |
| NA | Google.com | Search | 2013-01-10 01:05:01.000 | 835 | 282 | 80 |
| NA | Salesforce.com | Login | 2013-01-16 01:01:01.000 | 785 | 782 | 80 |
| NA | Google.com | Analytics | 2013-01-07 06:01:01.000 | 23 | 1 | 57 |
| NA | Google.com | Analytics | 2013-01-14 01:01:01.000 | 65 | 252 | 56 |
| EU | Salesforce.com | Reports | 2013-01-05 04:14:12.000 | 475 | 252 | 53 |
| NA | Google.com | Search | 2013-01-08 08:01:01.000 | 345 | 242 | 46 |
| NA | Salesforce.com | Reports | 2013-01-15 07:09:01.000 | 655 | 426 | 46 |
| NA | Salesforce.com | Login | 2013-01-04 11:01:11.000 | 23 | 56 | 45 |
| NA | Salesforce.com | Dashboard | 2013-01-03 11:01:01.000 | 88 | 66 | 44 |
| EU | Salesforce.com | Dashboard | 2013-01-06 05:04:05.000 | 12 | 22 | 43 |
| EU | Salesforce.com | Reports | 2013-01-02 14:32:01.000 | 125 | 131 | 42 |
| NA | Apple.com | Login | 2013-01-01 01:01:01.000 | 35 | 22 | 40 |
+-------+-----------------+------------+--------------------------+-------+-------+-----------------+
20 rows selected (0.24 seconds)
-- 第二页，LIMIT限制每页显示20条数据， OFFSET 指定为20表示第二页的数据的起始偏移量
0: jdbc:phoenix:localhost> SELECT * FROM WEB_STAT ORDER BY ACTIVE_VISITOR DESC
. . . . . . . . . . . . .> LIMIT 20 OFFSET 20;
+-------+-----------------+------------+--------------------------+-------+------+-----------------+
| HOST | DOMAIN | FEATURE | DATE | CORE | DB | ACTIVE_VISITOR |
+-------+-----------------+------------+--------------------------+-------+------+-----------------+
| EU | Apple.com | Mac | 2013-01-01 01:01:01.000 | 35 | 22 | 34 |
| NA | Salesforce.com | Login | 2013-01-17 01:01:01.000 | 355 | 242 | 33 |
| NA | Salesforce.com | Dashboard | 2013-01-11 01:01:01.000 | 335 | 32 | 30 |
| NA | Apple.com | iPad | 2013-01-05 01:01:01.000 | 85 | 2 | 18 |
| NA | Salesforce.com | Reports | 2013-01-09 16:33:01.000 | 35 | 42 | 15 |
| NA | Apple.com | iPad | 2013-01-06 01:01:01.000 | 35 | 22 | 10 |
| NA | Salesforce.com | Login | 2013-01-08 14:11:01.000 | 345 | 242 | 10 |
| NA | Salesforce.com | Login | 2013-01-01 01:01:01.000 | 35 | 42 | 10 |
| NA | Apple.com | Mac | 2013-01-08 01:01:01.000 | 3 | 2 | 10 |
| NA | Salesforce.com | Dashboard | 2013-01-14 04:07:01.000 | 5 | 2 | 9 |
| NA | Apple.com | iPad | 2013-01-07 01:01:01.000 | 9 | 27 | 7 |
| NA | Google.com | Analytics | 2013-01-11 01:02:01.000 | 7 | 2 | 7 |
| NA | Salesforce.com | Reports | 2013-01-15 04:09:01.000 | 65 | 26 | 6 |
| NA | Google.com | Search | 2013-01-12 01:01:01.000 | 8 | 7 | 6 |
| EU | Google.com | Analytics | 2013-01-13 08:06:01.000 | 25 | 2 | 6 |
| EU | Salesforce.com | Reports | 2013-01-13 08:04:04.000 | 355 | 52 | 5 |
| EU | Salesforce.com | Reports | 2013-01-05 03:11:12.000 | 75 | 22 | 3 |
| EU | Salesforce.com | Reports | 2013-01-02 12:02:01.000 | 25 | 11 | 2 |
| NA | Salesforce.com | Login | 2013-01-04 06:01:21.000 | 3 | 52 | 1 |
+-------+-----------------+------------+--------------------------+-------+------+-----------------+
19 rows selected (0.18 seconds)
-- 6 删除数据。删除时可以添加条件，也可以模糊匹配删除
-- 为了模拟，我们先添加两条数据
0: jdbc:phoenix:localhost> UPSERT INTO STOCK_SYMBOL VALUES('WAG', 'Walgreens');
1 row affected (0.116 seconds)
0: jdbc:phoenix:localhost> UPSERT INTO STOCK_SYMBOL VALUES('WMT', 'Walmart');
1 row affected (0.008 seconds)
-- 6.1 删除 SYMBOL='CRM'的一条数据
0: jdbc:phoenix:localhost> DELETE FROM STOCK_SYMBOL WHERE SYMBOL='CRM';
1 row affected (0.089 seconds)
-- 发现已成功删除
0: jdbc:phoenix:localhost> SELECT * FROM STOCK_SYMBOL;
+---------+-------------+
| SYMBOL | COMPANY |
+---------+-------------+
| AAPL | APPLE Inc. |
| GOOG | Google |
| WAG | Walgreens |
| WMT | Walmart |
+---------+-------------+
4 rows selected (0.022 seconds)
-- 6.2 删除 COMPANY 为Wal开头的所有数据
0: jdbc:phoenix:localhost> DELETE FROM STOCK_SYMBOL WHERE COMPANY LIKE 'Wal%';
2 rows affected (0.034 seconds)
-- 发现有两条满足条件的数据已被删除
0: jdbc:phoenix:localhost> SELECT * FROM STOCK_SYMBOL;
+---------+-------------+
| SYMBOL | COMPANY |
+---------+-------------+
| AAPL | APPLE Inc. |
| GOOG | Google |
+---------+-------------+
2 rows selected (0.017 seconds)
-- 6.3 删除demo表，如果demo表不是视图并，对一个的HBase的表和数据也会被删除。
0: jdbc:phoenix:cdh3> DELETE TABLE "demo";
2.9.1 Join
Phoenix本身提供了Join功能，但比较适用于大表join 小表之类的简单join，如果要做大表join大表的复杂分析可以借助Spark来做。
官方文档介绍中Phoenix支持Join类型如右图所示有两种，一种是内连接：INNER JOIN，一种是外连接：OUTER JOIN，其中OUTER JOIN支持LEFT和RIGHT两种方式。但同时其还支持UNION ALL语法，可以合并来自多个SELECT语句的行数据。
关于这几种连接文氏图（Venn diagram）如下所示，INNER JOIN 一般被译作内连接，内连接查询能将左表（表 A）和右表（表 B）中能关联起来的数据连接后返回；LEFT JOIN 一般被译作左连接，也写作 LEFT OUTER JOIN ,左连接查询会返回左表（表 A）中所有记录，不管右表（表 B）中有没有关联的数据,在右表中找到的关联数据列也会被一起返回；RIGHT JOIN 一般被译作右连接，也写作 RIGHT OUTER JOIN ,右连接查询会返回右表（表 B）中所有记录，不管左表（表 A）中有没有关联的数据,在左表中找到的关联数据列也会被一起返回。FULL OUTER JOIN 一般被译作外连接、全连接，实际查询语句中可以写作 FULL OUTER JOIN 或 FULL JOIN ,外连接查询能返回左右表里的所有记录，其中左右表里能关联起来的记录被连接后返回。
现在有两张表，一个是网站数据表websites，主要记录了网站主键id、名称、连接、网站流量排名、国家等信息。另一张表为访问日志表access_log，主要记录的有每个网站每日的访问数量，两个表通过id来关联，即websites.id=access_log.site_id ,现在我们通过Phoenix来分别演示INNER JOIN、LEFT [OUTER] JOIN、RIGHT [OUTER] JOIN、FULL [OUTER] JOIN（Phoenix虽然不支持，但可以通过UNION ALL 来实现），对两个表进行关联，并根据访问数量降序排，输出标识id、名称、数量、日期。
-- 1 建表
-- 1.1 Websites 表
0: jdbc:phoenix:> CREATE TABLE IF NOT EXISTS TEST."websites" (
. . . . . . . . > "id" UNSIGNED_LONG NOT NULL PRIMARY KEY,
. . . . . . . . > "name" varchar(255) ,
. . . . . . . . > "url" varchar(255) ,
. . . . . . . . > "alexa" UNSIGNED_INT, /*'网站流量排名'*/
. . . . . . . . > "country" varchar(32)
. . . . . . . . > );
No rows affected (0.954 seconds)
-- 1.2 access_log 表
0: jdbc:phoenix:> CREATE TABLE IF NOT EXISTS TEST."access_log" (
. . . . . . . . > "aid" UNSIGNED_LONG NOT NULL,
. . . . . . . . > "site_id" UNSIGNED_LONG NOT NULL,
. . . . . . . . > "count" UNSIGNED_INT,
. . . . . . . . > "access_date" DATE ,
. . . . . . . . > CONSTRAINT PK PRIMARY KEY ("aid", "site_id")
. . . . . . . . > );
No rows affected (0.393 seconds)
-- 1.3 查看创建的表。可以看到本次创建了SCHEM为TEST的两张表access_log、websites
0: jdbc:phoenix:> !tables
+------------+--------------+-------------+---------------+----------+------------+----------------------+-----------------+--------------+-----------------+-------------
| TABLE_CAT | TABLE_SCHEM | TABLE_NAME | TABLE_TYPE | REMARKS | TYPE_NAME | SELF_REFERENCING_COL_NAME | REF_GENERATION | INDEX_STATE | IMMUTABLE_ROWS | SALT_BUCKETS | MULTI_TENANT | VIEW_STATEMENT | VIEW_TYPE | INDEX_TYPE | TRANSACTIONAL | IS_NAMESPAC |
+------------+--------------+-------------+---------------+----------+------------+----------------------------+-----------------+--------------+-----------------+--------
| | SYSTEM | CATALOG | SYSTEM TABLE | | | | | | false | null |
| | SYSTEM | FUNCTION | SYSTEM TABLE | | | | | | false | null |
| | SYSTEM | LOG | SYSTEM TABLE | | | | | | true | 32 |
| | SYSTEM | SEQUENCE | SYSTEM TABLE | | | | | | false | null |
| | SYSTEM | STATS | SYSTEM TABLE | | | | | | false | null |
| | | BOOKS | TABLE | | | | | | false | 8 |
| | | STOCK_SYMBOL | TABLE | | | | | | false | null |
| | | WEB_STAT | TABLE | | | | | | false | null |
| | | demo | TABLE | | | | | | false | null |
| | TEST | access_log | TABLE | | | | | | false | null |
| | TEST | websites | TABLE | | | | | | false | null |
+------------+--------------+-------------+---------------+----------+------------+----------------------------+-----------------+--------------+-----------------+--------
-- 2 插入数据。注意：字符串一定用 单引号
-- 2.1 插入 websites 表测试数据
0: jdbc:phoenix:> UPSERT INTO TEST."websites" VALUES(1, 'Google', 'https://www.google.cm/', 1, 'USA');
1 row affected (0.212 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."websites" VALUES(2, '淘宝', 'https://www.taobao.com/', 4689, 'CN');
1 row affected (0.036 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."websites" VALUES(3, '豆瓣', 'https://www.douban.com/', 325, 'CN');
1 row affected (0.02 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."websites" VALUES(4, '微博', 'http://weibo.com/', 20, 'CN');
1 row affected (0.039 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."websites" VALUES(5, 'Facebook', 'https://www.facebook.com/', 3, 'USA');
1 row affected (0.027 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."websites" VALUES(7, 'stackoverflow', 'http://stackoverflow.com/', 0, 'IND');
1 row affected (0.016 seconds)
-- 2.2 插入 access_log 表测试数据
0: jdbc:phoenix:> UPSERT INTO TEST."access_log" VALUES(1, 1, 45, '2016-05-10');
1 row affected (0.049 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."access_log" VALUES(2, 3, 100, '2016-05-13');
1 row affected (0.064 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."access_log" VALUES(3, 1, 230, '2016-05-14');
1 row affected (0.054 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."access_log" VALUES(4, 2, 10, '2016-05-14');
1 row affected (0.017 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."access_log" VALUES(5, 5, 205, '2016-05-14');
1 row affected (0.021 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."access_log" VALUES(6, 4, 13, '2016-05-15');
1 row affected (0.018 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."access_log" VALUES(7, 3, 220, '2016-05-15');
1 row affected (0.038 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."access_log" VALUES(8, 5, 545, '2016-05-16');
1 row affected (0.03 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."access_log" VALUES(9, 3, 201, '2016-05-17');
1 row affected (0.019 seconds)
0: jdbc:phoenix:> UPSERT INTO TEST."access_log" VALUES(10, 6, 419, '2018-11-21');
1 row affected (0.02 seconds)
-- 3 查看表中的数据
-- 3.1 查看 websites 表数据
0: jdbc:phoenix:> SELECT * FROM TEST."websites";
+-----+----------------+----------------------------+--------+----------+
| id | name | url | alexa | country |
+-----+----------------+----------------------------+--------+----------+
| 1 | Google | https://www.google.cm/ | 1 | USA |
| 2 | 淘宝 | https://www.taobao.com/ | 4689 | CN |
| 3 | 豆瓣 | https://www.douban.com/ | 325 | CN |
| 4 | 微博 | http://weibo.com/ | 20 | CN |
| 5 | Facebook | https://www.facebook.com/ | 3 | USA |
| 7 | stackoverflow | http://stackoverflow.com/ | 0 | IND |
+-----+----------------+----------------------------+--------+----------+
6 rows selected (0.139 seconds)
-- 3.2 查看 access_log 表数据
0: jdbc:phoenix:> SELECT * FROM TEST."access_log";
+------+----------+--------+--------------------------+
| aid | site_id | count | access_date |
+------+----------+--------+--------------------------+
| 1 | 1 | 45 | 2016-05-10 00:00:00.000 |
| 2 | 3 | 100 | 2016-05-13 00:00:00.000 |
| 3 | 1 | 230 | 2016-05-14 00:00:00.000 |
| 4 | 2 | 10 | 2016-05-14 00:00:00.000 |
| 5 | 5 | 205 | 2016-05-14 00:00:00.000 |
| 6 | 4 | 13 | 2016-05-15 00:00:00.000 |
| 7 | 3 | 220 | 2016-05-15 00:00:00.000 |
| 8 | 5 | 545 | 2016-05-16 00:00:00.000 |
| 9 | 3 | 201 | 2016-05-17 00:00:00.000 |
| 10 | 6 | 419 | 2018-11-21 00:00:00.000 |
+------+----------+--------+--------------------------+
10 rows selected (0.123 seconds)
-- 4 INNER JOIN
-- 可以看到 inner join 返回的是满足条件的A表和B表都都存在的数据
0: jdbc:phoenix:> SELECT A."name",B."count",B."access_date"
. . . . . . . . > FROM TEST."websites" AS A
. . . . . . . . > INNER JOIN TEST."access_log" AS B
. . . . . . . . > ON A."id"=B."site_id"
. . . . . . . . > ORDER BY B."count" DESC;
+-----------+----------+--------------------------+
| A.name | B.count | B.access_date |
+-----------+----------+--------------------------+
| Facebook | 545 | 2016-05-16 00:00:00.000 |
| Google | 230 | 2016-05-14 00:00:00.000 |
| 豆瓣 | 220 | 2016-05-15 00:00:00.000 |
| Facebook | 205 | 2016-05-14 00:00:00.000 |
| 豆瓣 | 201 | 2016-05-17 00:00:00.000 |
| 豆瓣 | 100 | 2016-05-13 00:00:00.000 |
| Google | 45 | 2016-05-10 00:00:00.000 |
| 微博 | 13 | 2016-05-15 00:00:00.000 |
| 淘宝 | 10 | 2016-05-14 00:00:00.000 |
+-----------+----------+--------------------------+
9 rows selected (0.276 seconds)
-- 5 LEFT JOIN
-- 可以看到 A.name=stackoverflow 在A表存在，但是B表不存在，依然返回作为结果。
-- 而 B.count=419 的数据A表没有匹配到，因此就没有返回。 总之以左表A为准
-- 注意：这里和MySQL不一样的地方是，null值排在最前面（无论是排序是 ASC 还是 DESC）
0: jdbc:phoenix:> SELECT A."name",B."count",B."access_date"
. . . . . . . . > FROM TEST."websites" AS A
. . . . . . . . > LEFT JOIN TEST."access_log" AS B
. . . . . . . . > ON A."id"=B."site_id"
. . . . . . . . > ORDER BY B."count" DESC;
+----------------+----------+--------------------------+
| A.name | B.count | B.access_date |
+----------------+----------+--------------------------+
| stackoverflow | null | |
| Facebook | 545 | 2016-05-16 00:00:00.000 |
| Google | 230 | 2016-05-14 00:00:00.000 |
| 豆瓣 | 220 | 2016-05-15 00:00:00.000 |
| Facebook | 205 | 2016-05-14 00:00:00.000 |
| 豆瓣 | 201 | 2016-05-17 00:00:00.000 |
| 豆瓣 | 100 | 2016-05-13 00:00:00.000 |
| Google | 45 | 2016-05-10 00:00:00.000 |
| 微博 | 13 | 2016-05-15 00:00:00.000 |
| 淘宝 | 10 | 2016-05-14 00:00:00.000 |
+----------------+----------+--------------------------+
10 rows selected (0.169 seconds)
-- 6 RIGHT JOIN
-- 可以看到 B.count=419 在B表存在，但是A表不存在，依然返回作为结果。
-- 而 A.name=stackoverflow 的数据B表没有匹配到，因此就没有返回。 总之以右表B为准
0: jdbc:phoenix:> SELECT A."name",B."count",B."access_date"
. . . . . . . . > FROM TEST."websites" AS A
. . . . . . . . > RIGHT JOIN TEST."access_log" AS B
. . . . . . . . > ON A."id"=B."site_id"
. . . . . . . . > ORDER BY B."count" DESC;
+-----------+----------+--------------------------+
| A.name | B.count | B.access_date |
+-----------+----------+--------------------------+
| Facebook | 545 | 2016-05-16 00:00:00.000 |
| | 419 | 2018-11-21 00:00:00.000 |
| Google | 230 | 2016-05-14 00:00:00.000 |
| 豆瓣 | 220 | 2016-05-15 00:00:00.000 |
| Facebook | 205 | 2016-05-14 00:00:00.000 |
| 豆瓣 | 201 | 2016-05-17 00:00:00.000 |
| 豆瓣 | 100 | 2016-05-13 00:00:00.000 |
| Google | 45 | 2016-05-10 00:00:00.000 |
| 微博 | 13 | 2016-05-15 00:00:00.000 |
| 淘宝 | 10 | 2016-05-14 00:00:00.000 |
+-----------+----------+--------------------------+
10 rows selected (0.205 seconds)
-- 7 UNION ALL。Phoenix不支持 FULL OUTER JOIN ，但可以通过 UNION ALL 实现
-- FULL OUTER JOIN 等价于 (LEFT OUTER JOIN) UNION ALL (RIGHT OUTER JOIN )
-- 可以看到返回左表和右表都有的所有数据.
-- 注意：目前还不支持 UNION ALL 的左右 SELECT 语句中有 ORDER BY ,否则SQL解析时会报如下错误
-- ERROR 602 (42P00): Syntax error. Missing "EOF" at line 5, column 1.
-- 如果要实现排序，可以用子查询或者视图来实现
0: jdbc:phoenix:> SELECT A."name",B."count",B."access_date"
. . . . . . . . > FROM TEST."websites" AS A
. . . . . . . . > LEFT JOIN TEST."access_log" AS B
. . . . . . . . > ON A."id"=B."site_id"
. . . . . . . . > UNION ALL
. . . . . . . . > SELECT A."name",B."count",B."access_date"
. . . . . . . . > FROM TEST."websites" AS A
. . . . . . . . > RIGHT JOIN TEST."access_log" AS B
. . . . . . . . > ON A."id"=B."site_id";
+----------------+--------+--------------------------+
| NAME | COUNT | ACCESS_DATE |
+----------------+--------+--------------------------+
| Google | 45 | 2016-05-10 00:00:00.000 |
| Google | 230 | 2016-05-14 00:00:00.000 |
| 淘宝 | 10 | 2016-05-14 00:00:00.000 |
| 豆瓣 | 100 | 2016-05-13 00:00:00.000 |
| 豆瓣 | 220 | 2016-05-15 00:00:00.000 |
| 豆瓣 | 201 | 2016-05-17 00:00:00.000 |
| 微博 | 13 | 2016-05-15 00:00:00.000 |
| Facebook | 205 | 2016-05-14 00:00:00.000 |
| Facebook | 545 | 2016-05-16 00:00:00.000 |
| stackoverflow | null | |
| Google | 45 | 2016-05-10 00:00:00.000 |
| 豆瓣 | 100 | 2016-05-13 00:00:00.000 |
| Google | 230 | 2016-05-14 00:00:00.000 |
| 淘宝 | 10 | 2016-05-14 00:00:00.000 |
| Facebook | 205 | 2016-05-14 00:00:00.000 |
| 微博 | 13 | 2016-05-15 00:00:00.000 |
| 豆瓣 | 220 | 2016-05-15 00:00:00.000 |
| Facebook | 545 | 2016-05-16 00:00:00.000 |
| 豆瓣 | 201 | 2016-05-17 00:00:00.000 |
| | 419 | 2018-11-21 00:00:00.000 |
+----------------+--------+--------------------------+
20 rows selected (0.258 seconds)
-- 7.1 子查询方式实现 UNION ALL 排序
0: jdbc:phoenix:> SELECT * FROM (
. . . . . . . . > SELECT A."name",B."count",B."access_date"
. . . . . . . . > FROM TEST."websites" AS A
. . . . . . . . > LEFT JOIN TEST."access_log" AS B
. . . . . . . . > ON A."id"=B."site_id"
. . . . . . . . > UNION ALL
. . . . . . . . > SELECT A."name",B."count",B."access_date"
. . . . . . . . > FROM TEST."websites" AS A
. . . . . . . . > RIGHT JOIN TEST."access_log" AS B
. . . . . . . . > ON A."id"=B."site_id"
. . . . . . . . > ) ORDER BY COUNT DESC;
+----------------+--------+--------------------------+
| NAME | COUNT | ACCESS_DATE |
+----------------+--------+--------------------------+
| stackoverflow | null | |
| Facebook | 545 | 2016-05-16 00:00:00.000 |
| Facebook | 545 | 2016-05-16 00:00:00.000 |
| | 419 | 2018-11-21 00:00:00.000 |
| Google | 230 | 2016-05-14 00:00:00.000 |
| Google | 230 | 2016-05-14 00:00:00.000 |
| 豆瓣 | 220 | 2016-05-15 00:00:00.000 |
| 豆瓣 | 220 | 2016-05-15 00:00:00.000 |
| Facebook | 205 | 2016-05-14 00:00:00.000 |
| Facebook | 205 | 2016-05-14 00:00:00.000 |
| 豆瓣 | 201 | 2016-05-17 00:00:00.000 |
| 豆瓣 | 201 | 2016-05-17 00:00:00.000 |
| 豆瓣 | 100 | 2016-05-13 00:00:00.000 |
| 豆瓣 | 100 | 2016-05-13 00:00:00.000 |
| Google | 45 | 2016-05-10 00:00:00.000 |
| Google | 45 | 2016-05-10 00:00:00.000 |
| 微博 | 13 | 2016-05-15 00:00:00.000 |
| 微博 | 13 | 2016-05-15 00:00:00.000 |
| 淘宝 | 10 | 2016-05-14 00:00:00.000 |
| 淘宝 | 10 | 2016-05-14 00:00:00.000 |
+----------------+--------+--------------------------+
20 rows selected (0.291 seconds)
-- 7.2 视图方式实现 UNION ALL 排序
-- 如果删除视图，语法格式为：DROP VIEW IF EXISTS my_schema.my_view
-- 注意1：目前视图中还不能包含 AS 关键字
-- 注意2：目前子句中不能选择字段，只能用*号，否则包如下错误：
-- ERROR 604 (42P00): Syntax error. Mismatched input. Expecting "ASTERISK", got "TEST" at line 2, column 8.
-- Debug：目前版本存在问题。当前视图支持的语法比较有限，SELECT中不能有复杂的语法，否则解析时会无法判断结束标记。可以反馈社区等后期修复
-- CREATE VIEW IF EXISTS TEST."websites_union_access_log" AS
-- SELECT * FROM TEST."websites" LEFT JOIN TEST."access_log"
-- ON TEST."websites"."id"=TEST."access_log"."site_id"
-- UNION ALL
-- SELECT * FROM TEST."websites"
-- RIGHT JOIN TEST."access_log"
-- ON TEST."websites"."id"=TEST."access_log"."site_id";
-- SELECT AB."name",AB."count",AB."access_date" FROM TEST."websites_union_access_log"
-- AS AB ORDER BY AB."count" DESC;
2.10 JDBC
使用JDBC方式连接Phoenix，对数据进行处理
2.10.1 新建一个Maven工程
在IDEA中新建一个Maven项目
2.10.2 pom.xml
<dependencies>
<!-- https://mvnrepository.com/artifact/org.apache.phoenix/phoenix-core -->
<dependency>
<groupId>org.apache.phoenix</groupId>
<artifactId>phoenix-core</artifactId>
<version>5.0.0-HBase-2.0</version>
</dependency>
<!-- https://mvnrepository.com/artifact/org.apache.hadoop/hadoop-common -->
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-common</artifactId>
<version>3.0.0</version>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<version>4.12</version>
</dependency>
</dependencies>
2.10.3 JDBC Client代码
3 工具
以下两款常用的工具都支持Win和Mac OS
3.1 DBeaver
在创建Phoenix连接，编辑驱动设那里点击 添加文件(F) ，选中前面我们编译出来的phoenix-5.0.0-HBase-2.0-client.jar包即可。
3.2 SQuirrel SQL
3.2.1 下载
进入这个页面下载 http://squirrel-sql.sourceforge.net/，这里可以下载压缩包版的 Plain zips the latest release for Windows/Linux/MacOS X/others。有三个版本，选在一个版本下载即可，例如这里下载squirrelsql-3.9.1-optional.zip
3.2.2 解压和配置
解压到安装目录，然后将phoenix-5.0.0-HBase-2.0-client.jar和phoenix-core-5.0.0-HBase-2.0.jar下载或者软连接到解压的包中的lib目录中。
Su-yuexi:squirrelsql-3.9.1-optional yoreyuan$ cd lib
Su-yuexi:lib yoreyuan$ ln -s ~/.dbeaver-drivers/phoenix-5.0.0-HBase-2.0-client.jar phoenix-5.0.0-HBase-2.0-client.jar
Su-yuexi:lib yoreyuan$ ln -s ~/soft/maven/repository/org/apache/phoenix/phoenix-core/5.0.0-HBase-2.0/phoenix-core-5.0.0-HBase-2.0.jar phoenix-core-5.0.0-HBase-2.0.jar
Su-yuexi:lib yoreyuan$ cd ../
Su-yuexi:squirrelsql-3.9.1-optional yoreyuan$ chmod +x squirrel-sql-mac.sh
3.2.3 启动
执行启动脚本，此时会弹出SQuirrel SQL Client Help和SQuirrel SQL Client窗口。如果安装或其它操作也可以查看帮助文档。
./squirrel-sql-mac.sh
3.2.4 配置 Phoenix 驱动
- 添加Driver。点击
Driver - Driver Name: 填写一个名字
- Example URL:
jdbc:phoenix:cdh2,cdh3:2181 - 在
Java Class Path选在我们前面添加的lib/phoenix-core-5.0.0-HBase-2.0.jar - 点击
List Drivers，等一会会自动在Class Name：补上org.apache.phoenix.jdbc.PhoenixDriver - OK
- 配置成功后会在左侧的
Drivers区域添加上去我们刚配置的Phoenix连接。
3.2.5 连接Phoenix
- 点击
Aliases中的➕ - Name: 连接的别名
- Driver: 选择上一步配置好的驱动。URL会自动填入。
- Test。User Name 和 Password 可以不用填写。
- OK
3.2.6 查询
执行SQL查询数据，如上图右，这里需要注意的如果我们建表时没有指定Schema，则创建的表都会在一个空的Schema下（DBeaver工具目前的版本无法显示空Schema下的表,但是可以正产查询），点击点击SQL，就可以编写SQL，进行数据的查询等操作，例如查询表空schema的表数据：SELECT * FROM WEB_stat;，如果有schema在表前面加上用英文点号分割，例如查询test下的person表：SELECT * FROM test.person;。
4 资料
这里推荐一份 云栖社区的一份资料，可以参考
