Hadoop 源码详解之 DBOutputFormat 类
1. 类释义
A OutputFormat that sends the reduce output to a SQL table.
一种将Reduce输出到一个SQL表中的输出格式。
DBOutputFormat accepts <key,value> pairs, where key has a type extending DBWritable. Returned RecordWriter writes only the key to the database with a batch SQL query.
DBOuputFormat接收<key,value>对，这其中的key实现了DBWritable接口。返回的RecordWriter使用一个批处理的SQL 查询语句 仅仅把键写到database中。
2. 类源码
3. 方法详解
3.1 setOutput()方法
- 方法释义
Initializes the reduce-part of the job with the appropriate output settings
使用合适的输出设置初始化job的reduce部分。
- 方法源码
/**
* @param job The job
* @param tableName The table to insert data into
* @param fieldNames The field names in the table.
*/
public static void setOutput(Job job, String tableName,
String... fieldNames) throws IOException {
if(fieldNames.length > 0 && fieldNames[0] != null) {
DBConfiguration dbConf = setOutput(job, tableName);
dbConf.setOutputFieldNames(fieldNames);
} else {
if (fieldNames.length > 0) {
setOutput(job, tableName, fieldNames.length);
}
else {
throw new IllegalArgumentException(
"Field names must be greater than 0");
}
}
}
注意上述的输出字段结果可以是多个列，所以其参数使用的是String ... fieldName。
