【数仓】DataX 通过SpringBoot项目自动生成 job.json 文件

相关文章

• 【数仓】基本概念、知识普及、核心技术
• 【数仓】数据分层概念以及相关逻辑
• 【数仓】Hadoop软件安装及使用（集群配置）
• 【数仓】Hadoop集群配置常用参数说明
• 【数仓】zookeeper软件安装及集群配置
• 【数仓】kafka软件安装及集群配置
• 【数仓】flume软件安装及配置
• 【数仓】flume常见配置总结，以及示例
• 【数仓】Maxwell软件安装及配置，采集mysql数据
• 【数仓】通过Flume+kafka采集日志数据存储到Hadoop
• 【数仓】DataX软件安装及配置，从mysql同步到hdfs

DataX的任务脚本job.json格式基本类似，而且我们在实际同步过程中通常都是一个表对应一个job，那么如果需要同步的表非常多的话，需要编写的job.json文件也非常多。既然是类似文件结构，那么我们就有办法通过程序自动生成相关的job.json文件。

居于以上考虑，有了下面的SpringBoot项目自动生成job.json的程序！

一、job 配置说明

DataX的job配置中的reader、writer和setting是构成数据同步任务的关键组件。

1、reader

reader是数据同步任务中的数据源读取配置部分，用于指定从哪个数据源读取数据以及如何读取数据。它通常包含以下关键信息：

• name: 读取插件的名称，如mysqlreader、hdfsreader等，用于指定从哪种类型的数据源读取数据。
• parameter: 具体的读取参数配置，包括数据源连接信息、读取的表或文件路径、字段信息等。

示例：
假设要从MySQL数据库读取数据，reader的配置可能如下：

"reader": {"name": "mysqlreader","parameter": {"username": "root","password": "password","column": ["id", "name", "age"],"connection": [{"jdbcUrl": "jdbc:mysql://localhost:3306/test_db","table": ["test_table"]}]}
}

2、writer

writer是数据同步任务中的目标数据源写入配置部分，用于指定将数据写入哪个目标数据源以及如何写入数据。它通常包含以下关键信息：

• name: 写入插件的名称，如mysqlwriter、hdfswriter等，用于指定将数据写入哪种类型的数据源。
• parameter: 具体的写入参数配置，包括目标数据源连接信息、写入的表或文件路径、字段映射等。

示例：
假设要将数据写入HDFS，writer的配置可能如下：

"writer": {"name": "hdfswriter","parameter": {"writeMode": "append","fieldDelimiter": ",","compress": "gzip","column": [{"name": "id", "type": "int"}, {"name": "name", "type": "string"}, {"name": "age", "type": "int"}],"connection": [{"hdfsUrl": "hdfs://localhost:9000","file": ["/user/hive/warehouse/test_table"]}]}
}

3、setting

setting是数据同步任务的全局设置部分，用于配置影响整个任务行为的参数。它通常包含以下关键信息：

• speed: 控制数据同步的速度和并发度，包括通道数（channel）和每个通道的数据传输速度（如byte）。
• errorLimit: 设置数据同步过程中的错误容忍度，包括允许出错的记录数（record）和错误率（percentage）。

示例：
一个典型的setting配置可能如下：

"setting": {"speed": {"channel": 3, // 并发通道数"byte": 1048576 // 每个通道的数据传输速度，单位是字节（1MB）},"errorLimit": {"record": 0, // 允许出错的记录数"percentage": 0.02 // 允许出错的记录数占总记录数的百分比}
}

综上所述，reader、writer和setting三个部分共同构成了DataX数据同步任务的配置文件。通过合理配置这些部分，用户可以灵活地定义数据源、目标数据源以及数据同步的行为和性能。在实际应用中，用户应根据具体的数据源类型、目标数据源类型和数据同步需求来填写和调整这些配置。

二、示例，从mysql同步到hdfs

该配置文件定义了从一个 MySQL 数据库读取数据，并将这些数据写入到 HDFS 的过程。

{"job": {"content": [{"reader": {"name": "mysqlreader", "parameter": {"column": ["id","name","msg","create_time","status","last_login_time"], "connection": [{"jdbcUrl": ["jdbc:mysql://192.168.56.1:3306/user?characterEncoding=UTF-8&useUnicode=true&useSSL=false&tinyInt1isBit=false&allowPublicKeyRetrieval=true&serverTimezone=Asia/Shanghai"], "table": ["t_user"]}], "password": "password", "username": "test", "where": "id>3"}}, "writer": {"name": "hdfswriter", "parameter": {"column": [{"name":"id","type":"bigint"},{"name":"name","type":"string"},{"name":"msg","type":"string"},{"name":"create_time","type":"date"},{"name":"status","type":"string"},{"name":"last_login_time","type":"date"}], "compress": "gzip", "defaultFS": "hdfs://hadoop131:9000", "fieldDelimiter": "\t", "fileName": "mysql2hdfs01", "fileType": "text", "path": "/mysql2hdfs", "writeMode": "append"}}}], "setting": {"speed": {"channel": "1"}}}
}

• 参考 mysqlreader
• 参考 hdfswriter

三、通过SpringBoot项目自动生成job文件

本例使用SpringBoot 3.0 结合 JDBC 读取mysql数据库表结构信息，生成job.json文件

1、创建SpringBoot项目，添加pom依赖以及配置

1）增加pom.xml依赖jar包

<!-- Spring Boot JDBC Starter -->
<dependency><groupId>org.springframework.boot</groupId><artifactId>spring-boot-starter-jdbc</artifactId>
</dependency>
<!-- MySQL JDBC Driver -->
<dependency><groupId>com.mysql</groupId><artifactId>mysql-connector-j</artifactId>
</dependency>
<dependency><groupId>cn.hutool</groupId><artifactId>hutool-all</artifactId><version>5.8.25</version>
</dependency>

2）增加application.properties配置项

server.port=8080
# mysql 数据库链接
spring.datasource.url=jdbc:mysql://127.0.0.1:3306/user?characterEncoding=UTF-8&useUnicode=true&useSSL=false&tinyInt1isBit=false&allowPublicKeyRetrieval=true&serverTimezone=Asia/Shanghai
spring.datasource.username=test
spring.datasource.password=password
spring.datasource.driver-class-name=com.mysql.cj.jdbc.Driver# datax 相关配置，在生成文件时使用
datax.hdfs.defaultFS=hdfs://hadoop131:9000
datax.hdfs.path=/origin_data
# 需要生成job文件的表，多个用逗号隔开
datax.mysql.tables=t_user,t_user_test,t_sys_dict
# job文件存储位置
datax.savepath=d:/temp/

2、按照job.json格式创建好各个 vo

1）基础结构vo

@Data
public class DataxJobRoot {private Job job;
}
@Data
public class Job {private List<Content> content;private Setting setting = new Setting();
}
@Data
public class Content {private Reader reader;private Writer writer;
}
@Data
public class Setting {private Speed speed = new Speed();@Datapublic static class Speed {private String channel = "1";}
}
@Data
public class Reader {private String name;private Parameter parameter;
}
@Data
public class Writer {private String name;private Parameter parameter;@Datapublic static class MysqlParameter {private List<String> column;private List<Connection> connection;private String password;private String username;private String writeMode = "replace";}@Datapublic static class Connection {private String jdbcUrl;private List<String> table;}
}public class Parameter {
}

2）mysql2hdfs的vo实现类

@EqualsAndHashCode(callSuper = true)
@Data
public class MysqlReader extends Reader {public String getName() {return "mysqlreader";}@EqualsAndHashCode(callSuper = true)@Datapublic static class MysqlParameter extends Parameter {private List<String> column;private List<Connection> connection;private String password;private String username;private String where;}@Datapublic static class Connection {private List<String> jdbcUrl;private List<String> table;}
}@EqualsAndHashCode(callSuper = true)
@Data
public class HdfsWriter extends Writer {public String getName() {return "hdfswriter";}@EqualsAndHashCode(callSuper = true)@Datapublic static class HdfsParameter extends Parameter {private List<Column> column;private String compress = "gzip";private String encoding = "UTF-8";private String defaultFS;private String fieldDelimiter = "\t";private String fileName;private String fileType = "text";private String path;private String writeMode = "append";}@Datapublic static class Column {String name;String type;}
}

3）hdfs2mysql的vo实现类

@EqualsAndHashCode(callSuper = true)
@Data
public class HdfsReader extends Reader {@Overridepublic String getName() {return "hdfsreader";}public HdfsParameter getParameter() {return new HdfsParameter();}@EqualsAndHashCode(callSuper = true)@Datapublic static class HdfsParameter extends Parameter {private List<String> column = Collections.singletonList("*");private String compress = "gzip";private String encoding = "UTF-8";private String defaultFS;private String fieldDelimiter = "\t";private String fileName;private String fileType = "text";private String path;private String nullFormat = "\\N";}
}
@EqualsAndHashCode(callSuper = true)
@Data
public class MysqlWriter extends Writer {public String getName() {return "mysqlwriter";}public MysqlParameter getParameter() {return new MysqlParameter();}@EqualsAndHashCode(callSuper = true)@Datapublic static class MysqlParameter extends Parameter {private List<String> column;private List<Connection> connection;private String password;private String username;private String writeMode = "replace";}@Datapublic static class Connection {private String jdbcUrl;private List<String> table;}
}

3、创建Repository、Service类读取数据库表结构

@Repository
public class DatabaseInfoRepository {private final JdbcTemplate jdbcTemplate;@Autowiredpublic DatabaseInfoRepository(JdbcTemplate jdbcTemplate) {this.jdbcTemplate = jdbcTemplate;}// 获取所有表名public List<String> getAllTableNames() {String sql = "SHOW TABLES";return jdbcTemplate.queryForList(sql, String.class);}// 根据表名获取字段信息public List<Map<String, Object>> getTableColumns(String tableName) {String sql = "SHOW FULL COLUMNS FROM " + tableName;return jdbcTemplate.queryForList(sql);}
}

@Service
public class DatabaseInfoService {private final DatabaseInfoRepository databaseInfoRepository;@Autowiredpublic DatabaseInfoService(DatabaseInfoRepository databaseInfoRepository) {this.databaseInfoRepository = databaseInfoRepository;}public void printAllTablesAndColumns() {// 获取所有表名List<String> tableNames = databaseInfoRepository.getAllTableNames();// 遍历表名，获取并打印每个表的字段信息for (String tableName : tableNames) {System.out.println("Table: " + tableName);// 获取当前表的字段信息List<Map<String, Object>> columns = databaseInfoRepository.getTableColumns(tableName);// 遍历字段信息并打印for (Map<String, Object> column : columns) {System.out.println("  Column: " + column.get("Field") + " (Type: " + column.get("Type") + ")" + " (Comment: " + column.get("Comment") + ")");}System.out.println(); // 打印空行作为分隔}}/** 查询指定表的所有字段列表 */public List<String> getColumns(String tableName) {List<String> list = new ArrayList<>();// 获取当前表的字段信息List<Map<String, Object>> columns = databaseInfoRepository.getTableColumns(tableName);// 遍历字段信息并打印for (Map<String, Object> column : columns) {list.add(column.get("Field").toString());}return list;}/** 查询指定表的所有字段列表，封装成HdfsWriter格式 */public List<HdfsWriter.Column> getHdfsColumns(String tableName) {List<HdfsWriter.Column> list = new ArrayList<>();// 获取当前表的字段信息List<Map<String, Object>> columns = databaseInfoRepository.getTableColumns(tableName);// 遍历字段信息并打印for (Map<String, Object> column : columns) {String name = column.get("Field").toString();String typeDb = column.get("Type").toString();String type = "string";if (typeDb.equals("bigint")) {type = "bigint";} else if (typeDb.startsWith("varchar")) {type = "string";} else if (typeDb.startsWith("date") || typeDb.endsWith("timestamp")) {type = "date";}HdfsWriter.Column columnHdfs = new HdfsWriter.Column();columnHdfs.setName(name);columnHdfs.setType(type);list.add(columnHdfs);}return list;}
}

4、创建Service生成job.json文件

@Service
public class GenHdfs2mysqlJsonService {@Value("${spring.datasource.url}")private String url;@Value("${spring.datasource.password}")private String password;@Value("${spring.datasource.username}")private String username;@Value("${datax.mysql.tables}")private String tables;@Value("${datax.hdfs.defaultFS}")private String defaultFS;@Value("${datax.hdfs.path}")private String path;@Value("${datax.savepath}")private String savepath;@Autowiredprivate DatabaseInfoService databaseInfoService;/*** 生成 hdfs2mysql的job.json* @param table*/public void genHdfs2mysqlJson(String table) {DataxJobRoot root = new DataxJobRoot();Job job = new Job();root.setJob(job);Content content = new Content();HdfsReader reader = new HdfsReader();MysqlWriter writer = new MysqlWriter();content.setReader(reader);content.setWriter(writer);job.setContent(Collections.singletonList(content));HdfsReader.HdfsParameter hdfsParameter = reader.getParameter();hdfsParameter.setPath(path);hdfsParameter.setFileName(table + "_hdfs");hdfsParameter.setDefaultFS(defaultFS);MysqlWriter.MysqlParameter mysqlParameter = writer.getParameter();mysqlParameter.setPassword(password);mysqlParameter.setUsername(username);List<String> columns = databaseInfoService.getColumns(table);mysqlParameter.setColumn(columns);MysqlWriter.Connection connection = new MysqlWriter.Connection();connection.setJdbcUrl(url);connection.setTable(Collections.singletonList(table));mysqlParameter.setConnection(Collections.singletonList(connection));String jsonStr = JSONUtil.parse(root).toJSONString(2);System.out.println(jsonStr);File file = FileUtil.file(savepath, table + "_h2m.json");FileUtil.appendString(jsonStr, file, "utf-8");}/*** 生成 mysql2hdfs 的job.json* @param table*/public void genMysql2HdfsJson(String table) {DataxJobRoot root = new DataxJobRoot();Job job = new Job();root.setJob(job);Content content = new Content();HdfsWriter writer = new HdfsWriter();MysqlReader reader = new MysqlReader();content.setReader(reader);content.setWriter(writer);job.setContent(Collections.singletonList(content));HdfsWriter.HdfsParameter hdfsParameter = new HdfsWriter.HdfsParameter();writer.setParameter(hdfsParameter);hdfsParameter.setPath(path);hdfsParameter.setFileName(table + "_hdfs");hdfsParameter.setDefaultFS(defaultFS);List<HdfsWriter.Column> lstColumns = databaseInfoService.getHdfsColumns(table);hdfsParameter.setColumn(lstColumns);MysqlReader.MysqlParameter mysqlParameter = new MysqlReader.MysqlParameter();reader.setParameter(mysqlParameter);mysqlParameter.setPassword(password);mysqlParameter.setUsername(username);List<String> columns = databaseInfoService.getColumns(table);mysqlParameter.setColumn(columns);MysqlReader.Connection connection = new MysqlReader.Connection();connection.setJdbcUrl(Collections.singletonList(url));connection.setTable(Collections.singletonList(table));mysqlParameter.setConnection(Collections.singletonList(connection));String jsonStr = JSONUtil.parse(root).toJSONString(2);System.out.println(jsonStr);File file = FileUtil.file(savepath, table + "_m2h.json");FileUtil.appendString(jsonStr, file, "utf-8");}public void genAllTable() {Splitter.on(",").split(tables).forEach(this::genMysql2HdfsJson);}}

5、执行测试

调用genAllTable()方法，在配置的存储目录中自动生成每个表的job.json文件，结构示例如下：

{"job": {"content": [{"reader": {"name": "mysqlreader","parameter": {"column": ["id","name","msg","create_time","last_login_time","status"],"connection": [{"jdbcUrl": ["jdbc:mysql://127.0.0.1:3306/user?characterEncoding=UTF-8&useUnicode=true&useSSL=false&tinyInt1isBit=false&allowPublicKeyRetrieval=true&serverTimezone=Asia/Shanghai"],"table": ["t_user"]}],"password": "password","username": "test"}},"writer": {"name": "hdfswriter","parameter": {"column": [{"name": "id","type": "bigint"},{"name": "name","type": "string"},{"name": "msg","type": "string"},{"name": "create_time","type": "date"},{"name": "last_login_time","type": "date"},{"name": "status","type": "bigint"}],"compress": "gzip","encoding": "UTF-8","defaultFS": "hdfs://hadoop131:9000","fieldDelimiter": "\t","fileName": "t_user_hdfs","fileType": "text","path": "/origin_data","writeMode": "append"}}}],"setting": {"speed": {"channel": "1"}}}
}

至此，通过SpringBoot项目自动生成DataX的job.json文件，功能完成！

其中细节以及其他的reader\writer转换可以按照例子实现。

参考

• 【数仓】DataX软件安装及配置，从mysql同步到hdfs
• https://github.com/alibaba/DataX/blob/master/userGuid.md