Java解析excel文件大數據量時，報OOM異常問題。
在解析EXCEL，JAVA 通常使用poi包下的 XSSFWorkbook 對象，但是，遇到海量數據（比如十幾萬條 或者 JVM堆設置了內存比較小時），就會拋出OOM異常，下面就放出解決方法。
直接上代碼：
package com.xxx.xxx.xxx.common.util;
import java.io.IOException;
import java.io.InputStream;
import java.util.ArrayList;
import java.util.List;
import org.apache.poi.openxml4j.exceptions.InvalidFormatException;
import org.apache.poi.openxml4j.exceptions.OpenXML4JException;
import org.apache.poi.openxml4j.opc.OPCPackage;
import org.apache.poi.xssf.eventusermodel.ReadOnlySharedStringsTable;
import org.apache.poi.xssf.eventusermodel.XSSFReader;
import org.apache.poi.xssf.eventusermodel.XSSFSheetXMLHandler;
import org.apache.poi.xssf.eventusermodel.XSSFSheetXMLHandler.SheetContentsHandler;
import org.apache.poi.xssf.model.StylesTable;
import org.apache.poi.xssf.usermodel.XSSFComment;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.stereotype.Component;
import org.xml.sax.InputSource;
import org.xml.sax.SAXException;
import org.xml.sax.XMLReader;
import org.xml.sax.helpers.XMLReaderFactory;
/**
* @author Jimmy Shan
* @date 2021-04-09
* @desc 解析大數據量excel文件，避免OOM發生
*/
@Component
public class ExcelParserBigData {
private static final Logger logger = LoggerFactory.getLogger(ExcelParserBigData.class);
private ISheetContentHandler contentHandler = new DefaultSheetHandler(); //表格默認處理器
private List<String[]> datas = new ArrayList<String[]>(); //讀取數據
/**
* @desc 轉換表格,默認為轉換第一個表格
*/
public ExcelParserBigData parse(InputStream stream)
throws InvalidFormatException, IOException, ParseException {
return parse(stream, 1);
}
/**
* @desc 解析方法
*/
public synchronized ExcelParserBigData parse(InputStream stream, int sheetId)
throws InvalidFormatException, IOException, ParseException {
// 每次轉換前都清空數據
datas.clear();
// 打開表格文件輸入流
OPCPackage pkg = OPCPackage.open(stream);
try {
// 創建表閱讀器
XSSFReader reader;
try {
reader = new XSSFReader(pkg);
} catch (OpenXML4JException e) {
logger.error("讀取表格出錯");
throw new ParseException(e.fillInStackTrace());
}
// 轉換指定單元表
InputStream shellStream = reader.getSheet("rId" + sheetId);
try {
InputSource sheetSource = new InputSource(shellStream);
StylesTable styles = reader.getStylesTable();
ReadOnlySharedStringsTable strings = new ReadOnlySharedStringsTable(pkg);
getContentHandler().init(datas);// 設置讀取出的數據
// 獲取轉換器
XMLReader parser = getSheetParser(styles, strings);
parser.parse(sheetSource);
} catch (SAXException e) {
logger.error("讀取表格出錯");
throw new ParseException(e.fillInStackTrace());
} finally {
shellStream.close();
}
} finally {
pkg.close();
}
return this;
}
/**
* @desc 獲取表格讀取數據,獲取數據前，需要先轉換數據，此方法不會獲取第一行數據，表格讀取數據
*/
public List<String[]> getDatas() {
return getDatas(true);
}
/**
* @desc 獲取表格讀取數據,獲取數據前，需要先轉換數據
*/
public List<String[]> getDatas(boolean dropFirstRow) {
if (dropFirstRow && datas.size() > 0) {
datas.remove(0);// 刪除表頭
}
return datas;
}
/**
* @desc 獲取讀取表格的轉換器
*/
protected XMLReader getSheetParser(StylesTable styles, ReadOnlySharedStringsTable strings) throws SAXException {
XMLReader parser = XMLReaderFactory.createXMLReader();
parser.setContentHandler(new XSSFSheetXMLHandler(styles, strings, getContentHandler(), false));
return parser;
}
public ISheetContentHandler getContentHandler() {
return contentHandler;
}
public void setContentHandler(ISheetContentHandler contentHandler) {
this.contentHandler = contentHandler;
}
/**
* @desc 表格轉換錯誤
*/
public class ParseException extends Exception {
private static final long serialVersionUID = -2451526411018517607L;
public ParseException(Throwable t) {
super("表格轉換錯誤", t);
}
}
public interface ISheetContentHandler extends SheetContentsHandler {
/**
* @desc 設置轉換后的數據集，用于存放轉換結果
*/
void init(List<String[]> datas);
}
/**
* @desc 默認表格解析handder
*/
class DefaultSheetHandler implements ISheetContentHandler {
/**
* @desc 讀取數據
*/
private List<String[]> datas;
private int columsLength;
private String[] readRow;
private ArrayList<String> fristRow = new ArrayList<String>();
@Override
public void init(List<String[]> datas) {
this.datas = datas;
//this.columsLength = columsLength;
}
@Override
public void startRow(int rowNum) {
if (rowNum != 0) {
readRow = new String[columsLength];
}
}
@Override
public void endRow(int rowNum) {
//將Excel第一行表頭的列數當做數組的長度，要保證后續的行的列數不能超過這個長度，這是個約定。
if (rowNum == 0) {
columsLength = fristRow.size();
readRow = fristRow.toArray(new String[fristRow.size()]);
}else {
readRow = fristRow.toArray(new String[columsLength]);
}
datas.add(readRow.clone());
readRow = null;
fristRow.clear();
}
@Override
public void cell(String cellReference, String formattedValue, XSSFComment comment) {
int index = getCellIndex(cellReference);//轉換A1,B1,C1等表格位置為真實索引位置
try {
fristRow.set(index, formattedValue);
} catch (IndexOutOfBoundsException e) {
int size = fristRow.size();
for (int i = index - size+1;i>0;i--){
fristRow.add(null);
}
fristRow.set(index,formattedValue);
}
}
@Override
public void headerFooter(String text, boolean isHeader, String tagName) {
}
/**
* @desc 轉換表格引用為列編號
*/
public int getCellIndex(String cellReference) {
String ref = cellReference.replaceAll("\\d+", "");
int num = 0;
int result = 0;
for (int i = 0; i < ref.length(); i++) {
char ch = cellReference.charAt(ref.length() - i - 1);
num = (int) (ch - 'A' + 1);
num *= Math.pow(26, i);
result += num;
}
return result - 1;
}
}
}
調用方：
package com.xxx.xxx.xxx.service.impl;
import com.xxx.xxx.xxx.common.util.ExcelParserBigData;
import com.xxx.xxx.xxx.service.DemoBaseService;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Service;
import java.io.FileInputStream;
import java.util.List;
/**
* @author Jimmy Shan
* @date 2021-04-07
* @desc 示例 服務實現
*/
@Service("demoBaseService")
public class DemoBaseServiceImpl implements DemoBaseService {
private static final Logger LOGGER = LoggerFactory.getLogger(DemoBaseServiceImpl.class);
@Autowired
private ExcelParserBigData excelParserBigData;
/**
* @desc 解析大數據量excel文件
*/
@Override
public void parseBigXlsx(String upPathName) {
long start = System.currentTimeMillis();
FileInputStream inf = null;
try {
inf = new FileInputStream(upPathName);
ExcelParserBigData parseBigData = excelParserBigData.parse(inf);
List<String[]> dataList = parseBigData.getDatas(); //這里的List泛型是String數組，解析后的數據是以 數組形式存放的。
for (int i = 0; i < dataList.size(); i++) {
String[] str = dataList.get(i);
LOGGER.info("第 {} 行", (i +1));
LOGGER.info("oppoId = {}, oppoName = {}", str[0], str[1]);
}
} catch (Exception e) {
e.printStackTrace();
} finally {
try {
if(inf != null) {
inf.close();
}
LOGGER.info("解析總耗時：{} 毫秒", (System.currentTimeMillis() - start));
} catch (Exception e) {
e.printStackTrace();
}
}
}
}
接口定義：
package com.xxx.xxx.xxx.service;
/**
* @author Jimmy Shan
* @date 2021-04-07
* @desc 示例 服務
*/
public interface DemoBaseService {
/**
* @desc 解析大數據量excel文件
*/
void parseBigXlsx(String upPathName);
}
OK，記錄到此。
浙公網安備 33010602011771號