dataX案例 从Oracle中读取数据（自定义sql语句）写入到hdfs上

OracleReader 插件官方详细说明
https://github.com/alibaba/DataX/blob/master/oraclereader/doc/oraclereader.md

HdfsWriter插件官方文档，有介绍各个参数的说明
https://github.com/alibaba/DataX/blob/master/hdfswriter/doc/hdfswriter.md

代码示例，编写oracle_hdfs.json文件

{
    "job": {
        "content": [
            {
                "reader": {
                    "name": "oraclereader",
                    "parameter": {
                        "connection": [
                            {
                                "jdbcUrl": ["jdbc:oracle:thin:@15.18.40.126:1521:PRD"],
								"querySql": ["SELECT aa.BWKEY,aa.MATNR,aa.BDATJ , aa.POPER, aa.actual_unit_cost FROM(SELECT CKMLHD.BWKEY,CKMLHD.MATNR,CKMLCR.BDATJ,CKMLCR.POPER,CKMLCR.PVPRS/CKMLCR.PEINH AS actual_unit_cost FROM SAPSR3.CKMLHD JOIN SAPSR3.CKMLCR ON CKMLCR.KALNR=CKMLHD.KALNR AND CKMLCR.MANDT=CKMLHD.MANDT WHERE CKMLCR.PVPRS > 0 AND CKMLHD.BWKEY IN('2800','5100','9100') AND CONCAT(CKMLCR.BDATJ,SUBSTR(CKMLCR.POPER,2)) <> to_char(SYSDATE,'YYYYMM')) aa"]
                            }
                        ],
                        "password": "52525",
                        "username": "root",
                        "where": ""
                    }
                },
                "writer": {
                    "name": "hdfswriter",
                    "parameter": {
                        "column": [
								{"name":"BWKEY","type":"string"},
								{"name":"MATNR","type":"string"},
								{"name":"BDATJ","type":"string"},
								{"name":"POPER","type":"string"},
								{"name":"ACTUAL_UNIT_COST","type":"double"}            
                              ],
                        "defaultFS": "hdfs://kncloud02:8020",
                        "path": "/user/oozie/oozie_test/dir10",
						"fileName": "company_cost",
                        "fileType": "text",
						"fieldDelimiter": ",",
						"compress": "gzip",
                        "writeMode": "append"
                    }
                }
            }
        ],
        "setting": {
            "speed": {
                "channel": "5"
            }
        }
    }
}

进入datax目录，运行datax命令，开始执行
python bin/datax.py job/oracle_hdfs.json