1 配置
SQL 客户端启动时可以添加 CLI 选项,具体如下。
./bin/sql-client.sh embedded --help Mode "embedded" submits Flink jobs from the local machine. Syntax: embedded [OPTIONS] "embedded" mode options: -d,--defaults <environment file> The environment properties with which every new session is initialized. Properties might be overwritten by session properties. -e,--environment <environment file> The environment properties to be imported into the session. It might overwrite default environment properties. -h,--help Show the help message with descriptions of all options. -hist,--history <History file path> The file which you want to save the command history into. If not specified, we will auto-generate one under your user's home directory. -j,--jar <JAR file> A JAR file to be imported into the session. The file might contain user-defined classes needed for the execution of statements such as functions, table sources, or sinks. Can be used multiple times. -l,--library <JAR directory> A JAR file directory with which every new session is initialized. The files might contain user-defined classes needed for the execution of statements such as functions, table sources, or sinks. Can be used multiple times. -pyarch,--pyArchives <arg> Add python archive files for job. The archive files will be extracted to the working directory of python UDF worker. Currently only zip-format is supported. For each archive file, a target directory be specified. If the target directory name is specified, the archive file will be extracted to a name can directory with the specified name. Otherwise, the archive file will be extracted to a directory with the same name of the archive file. The files uploaded via this option are accessible via relative path. '#' could be used as the separator of the archive file path and the target directory name. Comma (',') could be used as the separator to specify multiple archive files. This option can be used to upload the virtual environment, the data files used in Python UDF (e.g.: --pyArchives file:///tmp/py37.zip,file:///tmp/data .zip#data --pyExecutable py37.zip/py37/bin/python). The data files could be accessed in Python UDF, e.g.: f = open('data/data.txt', 'r'). -pyexec,--pyExecutable <arg> Specify the path of the python interpreter used to execute the python UDF worker (e.g.: --pyExecutable /usr/local/bin/python3). The python UDF worker depends on Python 3.5+, Apache Beam (version == 2.19.0), Pip (version >= 7.1.0) and SetupTools (version >= 37.0.0). Please ensure that the specified environment meets the above requirements. -pyfs,--pyFiles <pythonFiles> Attach custom python files for job. These files will be added to the PYTHONPATH of both the local client and the remote python UDF worker. The standard python resource file suffixes such as .py/.egg/.zip or directory are all supported. Comma (',') could be used as the separator to specify multiple files (e.g.: --pyFiles file:///tmp/myresource.zip,hdfs:///$n amenode_address/myresource2.zip). -pyreq,--pyRequirements <arg> Specify a requirements.txt file which defines the third-party dependencies. These dependencies will be installed and added to the PYTHONPATH of the python UDF worker. A directory which contains the installation packages of these dependencies could be specified optionally. Use '#' as the separator if the optional parameter exists (e.g.: --pyRequirements file:///tmp/requirements.txt#file:/// tmp/cached_dir). -s,--session <session identifier> The identifier for a session. 'default' is the default identifier. -u,--update <SQL update statement> Experimental (for testing only!): Instructs the SQL Client to immediately execute the given update statement after starting up. The process is shut down after the statement has been submitted to the cluster and returns an appropriate return code. Currently, this feature is only supported for INSERT INTO statements that declare the target sink table.
1.1 环境配置文件
SQL 查询执行前需要配置相关环境变量。环境配置文件 定义了 catalog、table sources、table sinks、用户自定义函数和其他执行或部署所需属性。
每个环境配置文件是常规的 YAML 文件,例子如下。
# 定义表,如 source、sink、视图或临时表。 tables: - name: MyTableSource type: source-table update-mode: append connector: type: filesystem path: "/path/to/something.csv" format: type: csv fields: - name: MyField1 data-type: INT - name: MyField2 data-type: VARCHAR line-delimiter: " " comment-prefix: "#" schema: - name: MyField1 data-type: INT - name: MyField2 data-type: VARCHAR - name: MyCustomView type: view query: "SELECT MyField2 FROM MyTableSource" # 定义用户自定义函数 functions: - name: myUDF from: class class: foo.bar.AggregateUDF constructor: - 7.6 - false # 定义 catalogs catalogs: - name: catalog_1 type: hive property-version: 1 hive-conf-dir: ... - name: catalog_2 type: hive property-version: 1 default-database: mydb2 hive-conf-dir: ... # 改变表程序基本的执行行为属性。 execution: planner: blink # 可选: 'blink' (默认)或 'old' type: streaming # 必选:执行模式为 'batch' 或 'streaming' result-mode: table # 必选:'table' 或 'changelog' max-table-result-rows: 1000000 # 可选:'table' 模式下可维护的最大行数(默认为 1000000,小于 1 则表示无限制) time-characteristic: event-time # 可选: 'processing-time' 或 'event-time' (默认) parallelism: 1 # 可选:Flink 的并行数量(默认为 1) periodic-watermarks-interval: 200 # 可选:周期性 watermarks 的间隔时间(默认 200 ms) max-parallelism: 16 # 可选:Flink 的最大并行数量(默认 128) min-idle-state-retention: 0 # 可选:表程序的最小空闲状态时间 max-idle-state-retention: 0 # 可选:表程序的最大空闲状态时间 current-catalog: catalog_1 # 可选:当前会话 catalog 的名称(默认为 'default_catalog') current-database: mydb1 # 可选:当前 catalog 的当前数据库名称 # (默认为当前 catalog 的默认数据库) restart-strategy: # 可选:重启策略(restart-strategy) type: fallback # 默认情况下“回退”到全局重启策略 # 用于调整和调优表程序的配置选项。 # 在专用的”配置”页面上可以找到完整的选项列表及其默认值。 configuration: table.optimizer.join-reorder-enabled: true table.exec.spill-compression.enabled: true table.exec.spill-compression.block-size: 128kb # 描述表程序提交集群的属性。 deployment: response-timeout: 5000
上述配置:
- 定义一个从 CSV 文件中读取的 table source
MyTableSource
所需的环境, - 定义了一个视图
MyCustomView
,该视图是用 SQL 查询声明的虚拟表, - 定义了一个用户自定义函数
myUDF
,该函数可以使用类名和两个构造函数参数进行实例化, - 连接到两个 Hive catalogs 并用
catalog_1
来作为当前目录,用mydb1
来作为该目录的当前数据库, - streaming 模式下用 blink planner 来运行时间特征为 event-time 和并行度为 1 的语句,
- 在
table
结果模式下运行试探性的(exploratory)的查询, - 并通过配置选项对联结(join)重排序和溢出进行一些计划调整。
根据使用情况,配置可以被拆分为多个文件。因此,一般情况下(用 --defaults
指定默认环境配置文件)以及基于每个会话(用 --environment
指定会话环境配置文件)来创建环境配置文件。每个 CLI 会话均会被属于 session 属性的默认属性初始化。例如,默认环境配置文件可以指定在每个会话中都可用于查询的所有 table source,而会话环境配置文件仅声明特定的状态保留时间和并行性。启动 CLI 应用程序时,默认环境配置文件和会话环境配置文件都可以被指定。如果未指定默认环境配置文件,则 SQL 客户端将在 Flink 的配置目录中搜索 ./conf/sql-client-defaults.yaml
。
注意 在 CLI 会话中设置的属性(如 SET
命令)优先级最高:
CLI commands > session environment file > defaults environment file
重启策略(Restart Strategies)
重启策略控制 Flink 作业失败时的重启方式。与 Flink 集群的全局重启策略相似,更细精度的重启配置可以在环境配置文件中声明。
Flink 支持以下策略:
execution: # 退回到 flink-conf.yaml 中定义的全局策略 restart-strategy: type: fallback # 作业直接失败并且不尝试重启 restart-strategy: type: none # 最多重启作业的给定次数 restart-strategy: type: fixed-delay attempts: 3 # 作业被宣告失败前的重试次数(默认:Integer.MAX_VALUE) delay: 10000 # 重试之间的间隔时间,以毫秒为单位(默认:10 秒) # 只要不超过每个时间间隔的最大故障数就继续尝试 restart-strategy: type: failure-rate max-failures-per-interval: 1 # 每个间隔重试的最大次数(默认:1) failure-rate-interval: 60000 # 监测失败率的间隔时间,以毫秒为单位 delay: 10000 # 重试之间的间隔时间,以毫秒为单位(默认:10 秒)
1.2 依赖
SQL 客户端不要求用 Maven 或者 SBT 设置 Java 项目。相反,你可以以常规的 JAR 包给集群提交依赖项。你也可以分别(用 --jar
)指定每一个 JAR 包或者(用 --library
)定义整个 library 依赖库。为连接扩展系统(如 Apache Kafka)和相应的数据格式(如 JSON),Flink提供了开箱即用型 JAR 捆绑包(ready-to-use JAR bundles)。这些 JAR 包各个发行版都可以从 Maven 中央库中下载到。
提供的 SQL JARs 和使用文档的完整清单可以在连接扩展系统页面中找到。
如下例子展示了从 Apache Kafka 中读取 JSON 文件并作为 table source 的环境配置文件。
tables: - name: TaxiRides type: source-table update-mode: append connector: property-version: 1 type: kafka version: "0.11" topic: TaxiRides startup-mode: earliest-offset properties: bootstrap.servers: localhost:9092 group.id: testGroup format: property-version: 1 type: json schema: "ROW<rideId LONG, lon FLOAT, lat FLOAT, rideTime TIMESTAMP>" schema: - name: rideId data-type: BIGINT - name: lon data-type: FLOAT - name: lat data-type: FLOAT - name: rowTime data-type: TIMESTAMP(3) rowtime: timestamps: type: "from-field" from: "rideTime" watermarks: type: "periodic-bounded" delay: "60000" - name: procTime data-type: TIMESTAMP(3) proctime: true
TaxiRide
表的结果格式与绝大多数的 JSON 格式相似。此外,它还添加了 rowtime 属性 rowTime
和 processing-time 属性 procTime
。
connector
和 format
都允许定义属性版本(当前版本为 1
)以便将来向后兼容。
1.3 自定义函数(User-defined Functions)
SQL 客户端允许用户创建用户自定义的函数来进行 SQL 查询。当前,这些自定义函数仅限于 Java/Scala 编写的类以及 Python 文件。
为提供 Java/Scala 的自定义函数,你首先需要实现和编译函数类,该函数继承自 ScalarFunction
、 AggregateFunction
或 TableFunction
(见自定义函数)。一个或多个函数可以打包到 SQL 客户端的 JAR 依赖中。
为提供 Python 的自定义函数,你需要编写 Python 函数并且用装饰器 pyflink.table.udf.udf
或 pyflink.table.udf.udtf
来装饰(见 Python UDFs))。Python 文件中可以放置一个或多个函数。其Python 文件和相关依赖需要通过在环境配置文件中或命令行选项(见 命令行用法)配置中特别指定(见 Python 配置)。
所有函数在被调用之前,必须在环境配置文件中提前声明。functions
列表中每个函数类都必须指定
- 用来注册函数的
name
, - 函数的来源
from
(目前仅限于class
(Java/Scala UDF)或python
(Python UDF)),
Java/Scala UDF 必须指定:
- 声明了全限定名的函数类
class
以及用于实例化的constructor
参数的可选列表。
Python UDF 必须指定:
- 声明全程名称的
fully-qualified-name
,即函数的 “[module name].[object name]”
functions: - name: java_udf # required: name of the function from: class # required: source of the function class: ... # required: fully qualified class name of the function constructor: # optional: constructor parameters of the function class - ... # optional: a literal parameter with implicit type - class: ... # optional: full class name of the parameter constructor: # optional: constructor parameters of the parameter's class - type: ... # optional: type of the literal parameter value: ... # optional: value of the literal parameter - name: python_udf # required: name of the function from: python # required: source of the function fully-qualified-name: ... # required: fully qualified class name of the function
对于 Java/Scala UDF,要确保函数类指定的构造参数顺序和类型都要严格匹配。
构造函数参数
根据用户自定义函数可知,在用到 SQL 语句中之前,有必要将构造参数匹配对应的类型。
如上述示例所示,当声明一个用户自定义函数时,可以使用构造参数来配置相应的类,有以下三种方式:
隐式类型的文本值:SQL 客户端将自动根据文本推导对应的类型。目前,只支持 BOOLEAN
、INT
、 DOUBLE
和 VARCHAR
。
如果自动推导的类型与期望不符(例如,你需要 VARCHAR 类型的 false
),可以改用显式类型。
- true # -> BOOLEAN (case sensitive) - 42 # -> INT - 1234.222 # -> DOUBLE - foo # -> VARCHAR
显式类型的文本值:为保证类型安全,需明确声明 type
和 value
属性的参数。
- type: DECIMAL value: 11111111111111111
下表列出支持的 Java 参数类型和与之相对应的 SQL 类型。
Java 类型 | SQL 类型 |
---|---|
java.math.BigDecimal |
DECIMAL |
java.lang.Boolean |
BOOLEAN |
java.lang.Byte |
TINYINT |
java.lang.Double |
DOUBLE |
java.lang.Float |
REAL , FLOAT |
java.lang.Integer |
INTEGER , INT |
java.lang.Long |
BIGINT |
java.lang.Short |
SMALLINT |
java.lang.String |
VARCHAR |
其他类型 (例如 TIMESTAMP
和 ARRAY
)、原始类型和 null
目前还不支持。
(嵌套)类实例:除了文本值外,还可以通过指定构造参数的 class
和 constructor
属性来创建(嵌套)类实例。这个过程可以递归执行,直到最后的构造参数是用文本值来描述的。
- class: foo.bar.paramClass constructor: - StarryName - class: java.lang.Integer constructor: - class: java.lang.String constructor: - type: VARCHAR value: 3
2 扩展
============================================================================== **Table Sources** ============================================================================== Define table sources here. See the Table API & SQL documentation for details. tables: - name: Rides --表名 type: source --表类型 soruce为读入型源表,sink为写入型目标表(source表不存储真实的数据,sink表存储真实数据存储在外部依赖如mysql,kafka等) update-mode: append --更新方式 append 或者 update(Update 流只能写入支持更新的外部存储,如 MySQL, HBase。Append 流可以写入任意地存储,不过一般写入日志类型的系统,如 Kafka。) schema: --映射 目标表的字段及类型,此处字段和类型与format处的字段对应 - name: rideId type: LONG - name: taxiId type: LONG - name: isStart type: BOOLEAN - name: lon type: FLOAT - name: lat type: FLOAT - name: rideTime -- 输出字段由eventTime变更为rideTime ,依据timestamp类型字段将其设为时间属性rowTime type: TIMESTAMP rowtime: timestamps: type: "from-field" --时间戳字段获取方式 :来自源表字段 from: "eventTime" --时间戳字段 :源表的时间戳字段 watermarks: --水印 type: "periodic-bounded" --定义周期性水印 delay: "60000" --最大延迟 - name: psgCnt type: INT connector: --连接器 property-version: 1 type: kafka --连接kafka version: universal --0.11版本以上选择 universal topic: Rides --消费的topic名称 startup-mode: earliest-offset --消费方式 earliest-offset从头开始消费数据 latest-offset消费最新数据 properties: --设置zk,kafka端口及IP地址 - key: zookeeper.connect value: zookeeper:2181 - key: bootstrap.servers value: kafka:9092 - key: group.id --设置消费者组 value: testGroup format: --解析数据格式化 property-version: 1 type: json --此处解析数据类型是json格式,与上面字段映射一样 schema: "ROW(rideId LONG, isStart BOOLEAN, eventTime TIMESTAMP, lon FLOAT, lat FLOAT, psgCnt INT, taxiId LONG)" - name: Fares type: source update-mode: append schema: - name: rideId type: LONG - name: payTime type: TIMESTAMP rowtime: timestamps: type: "from-field" from: "eventTime" watermarks: type: "periodic-bounded" delay: "60000" - name: payMethod type: STRING - name: tip type: FLOAT - name: toll type: FLOAT - name: fare type: FLOAT connector: property-version: 1 type: kafka version: universal topic: Fares startup-mode: earliest-offset properties: - key: zookeeper.connect value: zookeeper:2181 - key: bootstrap.servers value: kafka:9092 - key: group.id value: testGroup format: property-version: 1 type: json schema: "ROW(rideId LONG, eventTime TIMESTAMP, payMethod STRING, tip FLOAT, toll FLOAT, fare FLOAT)" - name: DriverChanges type: source update-mode: append schema: - name: taxiId type: LONG - name: driverId type: LONG - name: usageStartTime type: TIMESTAMP rowtime: timestamps: type: "from-field" from: "eventTime" watermarks: type: "periodic-bounded" delay: "60000" connector: property-version: 1 type: kafka version: universal topic: DriverChanges startup-mode: earliest-offset properties: - key: zookeeper.connect value: zookeeper:2181 - key: bootstrap.servers value: kafka:9092 - key: group.id value: testGroup format: property-version: 1 type: json schema: "ROW(eventTime TIMESTAMP, taxiId LONG, driverId LONG)" - name: Drivers type: temporal-table history-table: DriverChanges primary-key: taxiId time-attribute: usageStartTime - name: Sink_TenMinPsgCnt -- 表名(外部存储系统 如kakfa的topic,或者mysql的表名 type: sink-table -- 表类型 soruce为读入型源表,sink为写入型目标表 schema: - name: cntStart --要输出的目标字段名称 类型 type: STRING - name: cntEnd type: STRING - name: cnt type: INT update-mode: append connector: property-version: 1 type: kafka version: universal topic: Sink_TenMinPsgCnt -- 输出的topic名称 properties: - key: zookeeper.connect value: zookeeper:2181 - key: bootstrap.servers value: kafka:9092 - key: group.id value: testGroup format: property-version: 1 type: json schema: "ROW(cntStart STRING,cntEnd STRING,cnt INT)" -- 此处为输出的kafka的字段,中间的字段由sql加工别名转换为输出字段,注:字段个数,类型,顺序要与上面schema一摸一样 functions: -- 函数定义 - name: isInNYC from: class class: com.ververica.sql_training.udfs.IsInNYC - name: toAreaId from: class class: com.ververica.sql_training.udfs.ToAreaId - name: toCoords from: class class: com.ververica.sql_training.udfs.ToCoords ============================================================================== **Execution properties** ============================================================================== Execution properties allow for changing the behavior of a table program. execution: planner: blink # using the Blink planner type: streaming # 'batch' or 'streaming' execution result-mode: table # 'changelog' or 'table' presentation of results parallelism: 1 # parallelism of the program max-parallelism: 128 # maximum parallelism min-idle-state-retention: 0 # minimum idle state retention in ms max-idle-state-retention: 0 # maximum idle state retention in ms ============================================================================== **Deployment properties** ============================================================================== Deployment properties allow for describing the cluster to which table programs are submitted to. deployment: type: standalone # only the 'standalone' deployment is supported response-timeout: 5000 # general cluster communication timeout in ms gateway-address: "" # (optional) address from cluster to gateway gateway-port: 0 # (optional) port from cluster to gateway