org.apache.commons

commons-vfs2

2.0

org.scannotation

scannotation

1.0.3

dom4j

dom4j

1.6.1

pentaho-kettle

kettle-vfs

5.2.0.0

pentaho

pentaho-kettle

kettle-engine

5.2.0.0

pentaho-kettle

kettle-core

5.2.0.0

OK，准备工作完成后开始编写Kettle工具类：

//

//Source code recreated from a .class file by IntelliJ IDEA//(powered by Fernflower decompiler)//import java.io.File;

import java.io.FileInputStream;

import java.io.IOException;

import java.io.InputStream;

import java.io.OutputStream;

import java.io.PrintWriter;

import java.net.URL;

import java.net.URLDecoder;

import javax.servlet.ServletOutputStream;

import javax.servlet.http.HttpServletResponse;

import org.apache.struts2.ServletActionContext;

import org.pentaho.di.core.KettleEnvironment;

import org.pentaho.di.trans.Trans;

import org.pentaho.di.trans.TransMeta;public classKettle {publicKettle() {

}public staticInputStream export(String sql, String savePath, String hostIp, String dbName, String port, String userName, String password) {

FileInputStreamin = null;try{

HttpServletResponse response=ServletActionContext.getResponse();

URL url= Kettle.class.getResource("/");

String paths= url.getFile().substring(0, url.getFile().indexOf("classes"));

paths=URLDecoder.decode(paths);

Long times=System.currentTimeMillis();

File path= newFile(savePath);if (!path.exists() && !path.isDirectory()) {

path.mkdir();

path.canWrite();

}

KettleEnvironment.init();

TransMeta tm= new TransMeta(paths + "classes/" + "export.ktr");

Trans trans= newTrans(tm);

trans.setVariable("hostIp", hostIp);

trans.setVariable("dbName", dbName);

trans.setVariable("port", port);

trans.setVariable("userName", userName);

trans.setVariable("password", password);

trans.setVariable("sql", sql);

trans.setVariable("savePath", savePath + "/" +times);

trans.prepareExecution((String[])null);

trans.startThreads();

trans.waitUntilFinished();

File file= new File(savePath + "/" + times + ".xlsx");

File file2= new File(savePath + "/" + times + ".xls");

ServletOutputStreamout;if(file.exists()) {in = newFileInputStream(file);

response.reset();

response.setContentType("application/x-download;charset=UTF-8");

response.addHeader("Content-Disposition", "attachment;filename='" + times + ".xlsx'");

response.addHeader("Content-Length", String.valueOf(file.length()));out =response.getOutputStream();

writeFile(out, in);

file.delete();

}else if(file2.exists()) {in = newFileInputStream(file2);

response.reset();

response.setContentType("application/x-download;charset=UTF-8");

response.addHeader("Content-Disposition", "attachment;filename='" + times + ".xls'");

response.addHeader("Content-Length", String.valueOf(file2.length()));out =response.getOutputStream();

writeFile(out, in);

file2.delete();

}else{

response.setContentType("text/html;charset=UTF-8");

PrintWriter writer=response.getWriter();

writer.write("");

}

}catch(Exception var18) {

var18.printStackTrace();

}return in;

}public static voidwriteFile(OutputStream fos, InputStream fis) throws IOException {byte[] buffer = new byte[1024];

boolean var3= false;intlen;while((len = fis.read(buffer)) > 0) {

fos.write(buffer,0, len);

fos.flush();

}

fis.close();

fos.close();

}

}

以上工具类要读取一个export.ktr的文件，我们把它放在\src\main\java下(当然你也可以放在你想要的路径下，修改以上程序能获取到就是了)，记得打包不要排除了，不然找不到

export.ktr文件内容(拷贝我的就行，如果不能使用自己去网上找个，百度下里面参数意思，数据库我用的mysql)：

export

Normal

0

/

ID_BATCH

Y

ID_BATCH

CHANNEL_ID

Y

CHANNEL_ID

TRANSNAME

Y

TRANSNAME

STATUS

Y

STATUS

LINES_READ

Y

LINES_READ

LINES_WRITTEN

Y

LINES_WRITTEN

LINES_UPDATED

Y

LINES_UPDATED

LINES_INPUT

Y

LINES_INPUT

LINES_OUTPUT

Y

LINES_OUTPUT

LINES_REJECTED

Y

LINES_REJECTED

ERRORS

Y

ERRORS

STARTDATE

Y

STARTDATE

ENDDATE

Y

ENDDATE

LOGDATE

Y

LOGDATE

DEPDATE

Y

DEPDATE

REPLAYDATE

Y

REPLAYDATE

LOG_FIELD

Y

LOG_FIELD

EXECUTING_SERVER

N

EXECUTING_SERVER

EXECUTING_USER

N

EXECUTING_USER

CLIENT

N

CLIENT

ID_BATCH

Y

ID_BATCH

SEQ_NR

Y

SEQ_NR

LOGDATE

Y

LOGDATE

TRANSNAME

Y

TRANSNAME

STEPNAME

Y

STEPNAME

STEP_COPY

Y

STEP_COPY

LINES_READ

Y

LINES_READ

LINES_WRITTEN

Y

LINES_WRITTEN

LINES_UPDATED

Y

LINES_UPDATED

LINES_INPUT

Y

LINES_INPUT

LINES_OUTPUT

Y

LINES_OUTPUT

LINES_REJECTED

Y

LINES_REJECTED

ERRORS

Y

ERRORS

INPUT_BUFFER_ROWS

Y

INPUT_BUFFER_ROWS

OUTPUT_BUFFER_ROWS

Y

OUTPUT_BUFFER_ROWS

ID_BATCH

Y

ID_BATCH

CHANNEL_ID

Y

CHANNEL_ID

LOG_DATE

Y

LOG_DATE

LOGGING_OBJECT_TYPE

Y

LOGGING_OBJECT_TYPE

OBJECT_NAME

Y

OBJECT_NAME

OBJECT_COPY

Y

OBJECT_COPY

REPOSITORY_DIRECTORY

Y

REPOSITORY_DIRECTORY

FILENAME

Y

FILENAME

OBJECT_ID

Y

OBJECT_ID

OBJECT_REVISION

Y

OBJECT_REVISION

PARENT_CHANNEL_ID

Y

PARENT_CHANNEL_ID

ROOT_CHANNEL_ID

Y

ROOT_CHANNEL_ID

ID_BATCH

Y

ID_BATCH

CHANNEL_ID

Y

CHANNEL_ID

LOG_DATE

Y

LOG_DATE

TRANSNAME

Y

TRANSNAME

STEPNAME

Y

STEPNAME

STEP_COPY

Y

STEP_COPY

LINES_READ

Y

LINES_READ

LINES_WRITTEN

Y

LINES_WRITTEN

LINES_UPDATED

Y

LINES_UPDATED

LINES_INPUT

Y

LINES_INPUT

LINES_OUTPUT

Y

LINES_OUTPUT

LINES_REJECTED

Y

LINES_REJECTED

ERRORS

Y

ERRORS

LOG_FIELD

N

LOG_FIELD

ID_BATCH

Y

ID_BATCH

CHANNEL_ID

Y

CHANNEL_ID

LOG_DATE

Y

LOG_DATE

METRICS_DATE

Y

METRICS_DATE

METRICS_CODE

Y

METRICS_CODE

METRICS_DESCRIPTION

Y

METRICS_DESCRIPTION

METRICS_SUBJECT

Y

METRICS_SUBJECT

METRICS_TYPE

Y

METRICS_TYPE

METRICS_VALUE

Y

METRICS_VALUE

0.0

0.0

10000

50

50

N

Y

50000

Y

N

1000

100

-

2016/03/01 17:21:47.083

-

2016/09/01 09:29:12.684

H4sIAAAAAAAAAAMAAAAAAAAAAAA=

N

111

${hostIp}

MYSQL

Native

${dbName}

${port}

${userName}

${password}

EXTRA_OPTION_MYSQL.defaultFetchSize

500

EXTRA_OPTION_MYSQL.useCursorFetch

true

FORCE_IDENTIFIERS_TO_LOWERCASE

N

FORCE_IDENTIFIERS_TO_UPPERCASE

N

IS_CLUSTERED

N

PORT_NUMBER

${port}

PRESERVE_RESERVED_WORD_CASE

N

QUOTE_ALL_FIELDS

N

STREAM_RESULTS

Y

SUPPORTS_BOOLEAN_DATA_TYPE

Y

SUPPORTS_TIMESTAMP_DATA_TYPE

Y

USE_POOLING

N

表输入

Excel输出

Y

Excel输出

ExcelOutput

Y

1

none

Y

N

UTF-8

N

Y

${savePath}

xls

N

N

N

N

N

N

Sheet1

Y

N

N

Encrypted

0

N

N

N

C:\Users\47\Desktop\通用模板.xls

arial

10

Y

N

no

horizontal

white

dark_teal

320

center

arial

10

black

none

544

176

Y

表输入

TableInput

N

1

none

111

${sql}

0

N

Y

N

272

176

Y

N

主方法里调用：

Kettle.export(sql, SystemConstant.EXPORT_PATH, "数据库IP", "数据库名字", "数据库端口", "数据库账号", "数据库密码");

你也可以用Springboot整合，有对应maven及配置，自行查资料。

效率测试：

*导出10w行记录

执行时间: 1133ms

执行时间: 1082ms

执行时间: 1096ms

* 导出100w行记录

执行时间: 39784ms

执行时间: 8566ms

执行时间: 8622ms

* Excel 2007行数极限 1048575 执行时间: 9686ms

第一次导数据要加载kettle组件运行稍慢，后面几次再导数据速度就飞快了，更多结果有兴趣的可以去试试。