java读取大容量excel之一

　　最近在用poi读取大容量excel,发现只要是excel文件大于2M左右，便会出现OOM(out of memory)，经过查询得知，原来poi读取excel的原理是如下：

org.apache.poi.xssf.usermodel.XSSFWorkbook.XSSFWorkbook(InputStream is) throws IOException

采用usermodel,这种方式是以dom方式读取excel，好处是读取方便，不足是一次性将文件加载到内存中，容易造成OOM；第二种模型，eventusermodel，这种方式采用事件驱动的方法解析xml，在遇到文件内容时，事件会触发，这种做法可以大大降低内存的消耗。以下是我整理的读取大容量excel2003(.xls)和excel2007(.xlsx)的例子，仅供学习

其他资料参考：

http://tool.oschina.net/apidocs/apidoc?api=apache-POI

 

package com.speed.excel;

import java.io.FileInputStream;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;

import org.apache.poi.hssf.eventusermodel.EventWorkbookBuilder.SheetRecordCollectingListener;
import org.apache.poi.hssf.eventusermodel.FormatTrackingHSSFListener;
import org.apache.poi.hssf.eventusermodel.HSSFEventFactory;
import org.apache.poi.hssf.eventusermodel.HSSFListener;
import org.apache.poi.hssf.eventusermodel.HSSFRequest;
import org.apache.poi.hssf.eventusermodel.MissingRecordAwareHSSFListener;
import org.apache.poi.hssf.eventusermodel.dummyrecord.LastCellOfRowDummyRecord;
import org.apache.poi.hssf.eventusermodel.dummyrecord.MissingCellDummyRecord;
import org.apache.poi.hssf.model.HSSFFormulaParser;
import org.apache.poi.hssf.record.BOFRecord;
import org.apache.poi.hssf.record.BlankRecord;
import org.apache.poi.hssf.record.BoolErrRecord;
import org.apache.poi.hssf.record.BoundSheetRecord;
import org.apache.poi.hssf.record.FormulaRecord;
import org.apache.poi.hssf.record.LabelRecord;
import org.apache.poi.hssf.record.LabelSSTRecord;
import org.apache.poi.hssf.record.NumberRecord;
import org.apache.poi.hssf.record.Record;
import org.apache.poi.hssf.record.SSTRecord;
import org.apache.poi.hssf.record.StringRecord;
import org.apache.poi.hssf.usermodel.HSSFWorkbook;
import org.apache.poi.poifs.filesystem.POIFSFileSystem;

/**
* 抽象Excel2003读取器，通过实现HSSFListener监听器，采用事件驱动模式解析excel2003
* 中的内容，遇到特定事件才会触发，大大减少了内存的使用。
*
*/
public class Excel2003Reader implements HSSFListener{
private int minColumns = -1;
private POIFSFileSystem fs;
private int lastRowNumber;
private int lastColumnNumber;

/** Should we output the formula, or the value it has? */
private boolean outputFormulaValues = true;

/** For parsing Formulas */
private SheetRecordCollectingListener workbookBuildingListener;
//excel2003工作薄
private HSSFWorkbook stubWorkbook;

// Records we pick up as we process
private SSTRecord sstRecord;
private FormatTrackingHSSFListener formatListener;

//表索引
private int sheetIndex = -1;
private BoundSheetRecord[] orderedBSRs;
@SuppressWarnings("unchecked")
private ArrayList boundSheetRecords = new ArrayList();

// For handling formulas with string results
private int nextRow;
private int nextColumn;
private boolean outputNextStringRecord;
//当前行
private int curRow = 0;
//存储行记录的容器
private List<String> rowlist = new ArrayList<String>();;
@SuppressWarnings( "unused")
private String sheetName;

private IRowReader rowReader;

public void setRowReader(IRowReader rowReader){
this.rowReader = rowReader;
}

/**
* 遍历excel下所有的sheet
* @throws IOException
*/
public void process(String fileName) throws IOException {
this.fs = new POIFSFileSystem(new FileInputStream(fileName));
MissingRecordAwareHSSFListener listener = new MissingRecordAwareHSSFListener(
this);
formatListener = new FormatTrackingHSSFListener(listener);
HSSFEventFactory factory = new HSSFEventFactory();
HSSFRequest request = new HSSFRequest();
if (outputFormulaValues) {
request.addListenerForAllRecords(formatListener);
} else {
workbookBuildingListener = new SheetRecordCollectingListener(
formatListener);
request.addListenerForAllRecords(workbookBuildingListener);
}
factory.processWorkbookEvents(request, fs);
}

/**
* HSSFListener 监听方法，处理 Record
*/
@SuppressWarnings("unchecked")
public void processRecord(Record record) {
int thisRow = -1;
int thisColumn = -1;
String thisStr = null;
String value = null;
switch (record.getSid()) {
case BoundSheetRecord.sid:
boundSheetRecords.add(record);
break;
case BOFRecord.sid:
BOFRecord br = (BOFRecord) record;
if (br.getType() == BOFRecord.TYPE_WORKSHEET) {
// 如果有需要，则建立子工作薄
if (workbookBuildingListener != null && stubWorkbook == null) {
stubWorkbook = workbookBuildingListener
.getStubHSSFWorkbook();
}

sheetIndex++;
if (orderedBSRs == null) {
orderedBSRs = BoundSheetRecord
.orderByBofPosition(boundSheetRecords);
}
sheetName = orderedBSRs[sheetIndex].getSheetname();
}
break;

case SSTRecord.sid:
sstRecord = (SSTRecord) record;
break;

case BlankRecord.sid:
BlankRecord brec = (BlankRecord) record;
thisRow = brec.getRow();
thisColumn = brec.getColumn();
thisStr = "";
rowlist.add(thisColumn, thisStr);
break;
case BoolErrRecord.sid: //单元格为布尔类型
BoolErrRecord berec = (BoolErrRecord) record;
thisRow = berec.getRow();
thisColumn = berec.getColumn();
thisStr = berec.getBooleanValue()+"";
rowlist.add(thisColumn, thisStr);
break;

case FormulaRecord.sid: //单元格为公式类型
FormulaRecord frec = (FormulaRecord) record;
thisRow = frec.getRow();
thisColumn = frec.getColumn();
if (outputFormulaValues) {
if (Double.isNaN(frec.getValue())) {
// Formula result is a string
// This is stored in the next record
outputNextStringRecord = true;
nextRow = frec.getRow();
nextColumn = frec.getColumn();
} else {
thisStr = formatListener.formatNumberDateCell(frec);
}
} else {
thisStr = '"' + HSSFFormulaParser.toFormulaString(stubWorkbook,
frec.getParsedExpression()) + '"';
}
rowlist.add(thisColumn,thisStr);
break;
case StringRecord.sid://单元格中公式的字符串
if (outputNextStringRecord) {
// String for formula
StringRecord srec = (StringRecord) record;
thisStr = srec.getString();
thisRow = nextRow;
thisColumn = nextColumn;
outputNextStringRecord = false;
}
break;
case LabelRecord.sid:
LabelRecord lrec = (LabelRecord) record;
curRow = thisRow = lrec.getRow();
thisColumn = lrec.getColumn();
value = lrec.getValue().trim();
value = value.equals("")?" ":value;
this.rowlist.add(thisColumn, value);
break;
case LabelSSTRecord.sid: //单元格为字符串类型
LabelSSTRecord lsrec = (LabelSSTRecord) record;
curRow = thisRow = lsrec.getRow();
thisColumn = lsrec.getColumn();
if (sstRecord == null) {
rowlist.add(thisColumn, " ");
} else {
value = sstRecord
.getString(lsrec.getSSTIndex()).toString().trim();
value = value.equals("")?" ":value;
rowlist.add(thisColumn,value);
}
break;
case NumberRecord.sid: //单元格为数字类型
NumberRecord numrec = (NumberRecord) record;
curRow = thisRow = numrec.getRow();
thisColumn = numrec.getColumn();
value = formatListener.formatNumberDateCell(numrec).trim();
value = value.equals("")?" ":value;
// 向容器加入列值
rowlist.add(thisColumn, value);
break;
default:
break;
}

// 遇到新行的操作
if (thisRow != -1 && thisRow != lastRowNumber) {
lastColumnNumber = -1;
}

// 空值的操作
if (record instanceof MissingCellDummyRecord) {
MissingCellDummyRecord mc = (MissingCellDummyRecord) record;
curRow = thisRow = mc.getRow();
thisColumn = mc.getColumn();
rowlist.add(thisColumn," ");
}

// 更新行和列的值
if (thisRow > -1)
lastRowNumber = thisRow;
if (thisColumn > -1)
lastColumnNumber = thisColumn;

// 行结束时的操作
if (record instanceof LastCellOfRowDummyRecord) {
if (minColumns > 0) {
// 列值重新置空
if (lastColumnNumber == -1) {
lastColumnNumber = 0;
}
}
lastColumnNumber = -1;
// 每行结束时， 调用getRows() 方法
rowReader.getRows(sheetIndex,curRow, rowlist);

// 清空容器
rowlist.clear();
}
}

}

package com.speed.excel;

import java.io.InputStream;
import java.util.ArrayList;
import java.util.Iterator;
import java.util.List;

import org.apache.poi.openxml4j.opc.OPCPackage;
import org.apache.poi.xssf.eventusermodel.XSSFReader;
import org.apache.poi.xssf.model.SharedStringsTable;
import org.apache.poi.xssf.usermodel.XSSFRichTextString;
import org.xml.sax.Attributes;
import org.xml.sax.InputSource;
import org.xml.sax.SAXException;
import org.xml.sax.XMLReader;
import org.xml.sax.helpers.DefaultHandler;
import org.xml.sax.helpers.XMLReaderFactory;

/**
* 抽象Excel2007读取器，excel2007的底层数据结构是xml文件，采用SAX的事件驱动的方法解析
* xml，需要继承DefaultHandler，在遇到文件内容时，事件会触发，这种做法可以大大降低
* 内存的耗费，特别使用于大数据量的文件。
*
*/
public class Excel2007Reader extends DefaultHandler {
//共享字符串表
private SharedStringsTable sst;
//上一次的内容
private String lastContents;
private boolean nextIsString;

private int sheetIndex = -1;
private List<String> rowlist = new ArrayList<String>();
//当前行
private int curRow = 0;
//当前列
private int curCol = 0;

private IRowReader rowReader;

public void setRowReader(IRowReader rowReader){
this.rowReader = rowReader;
}

/**只遍历一个电子表格，其中sheetId为要遍历的sheet索引，从1开始，1-3
* @param filename
* @param sheetId
* @throws Exception
*/
public void processOneSheet(String filename,int sheetId) throws Exception {
OPCPackage pkg = OPCPackage.open(filename);
XSSFReader r = new XSSFReader(pkg);
SharedStringsTable sst = r.getSharedStringsTable();
XMLReader parser = fetchSheetParser(sst);

// 根据 rId# 或 rSheet# 查找sheet
InputStream sheet2 = r.getSheet("rId"+sheetId);
sheetIndex++;
InputSource sheetSource = new InputSource(sheet2);
parser.parse(sheetSource);
sheet2.close();
}

/**
* 遍历工作簿中所有的电子表格
* @param filename
* @throws Exception
*/
public void process(String filename) throws Exception {
try {
OPCPackage pkg = OPCPackage.open(filename);
XSSFReader r = new XSSFReader(pkg);
SharedStringsTable sst = r.getSharedStringsTable();
XMLReader parser = fetchSheetParser(sst);
Iterator<InputStream> sheets = r.getSheetsData();
while (sheets.hasNext()) {
curRow = 0;
sheetIndex++;
InputStream sheet = sheets.next();
InputSource sheetSource = new InputSource(sheet);
parser.parse(sheetSource);
sheet.close();
}
} catch (Exception e) {
throw new IllegalArgumentException("读取文件失败，此文件可能已损坏，请参照模板重新上传！");
}
}

public XMLReader fetchSheetParser(SharedStringsTable sst)
throws SAXException {
XMLReader parser = XMLReaderFactory
.createXMLReader("org.apache.xerces.parsers.SAXParser");
this.sst = sst;
parser.setContentHandler(this);
return parser;
}

public void startElement(String uri, String localName, String name,
Attributes attributes) throws SAXException {

// c => 单元格
if ("c".equals(name)) {
// 如果下一个元素是 SST 的索引，则将nextIsString标记为true
String cellType = attributes.getValue("t");
if ("s".equals(cellType)) {
nextIsString = true;
} else {
nextIsString = false;
}
}
// 置空
lastContents = "";
}

public void endElement(String uri, String localName, String name)
throws SAXException {

// 根据SST的索引值的到单元格的真正要存储的字符串
// 这时characters()方法可能会被调用多次
if (nextIsString) {
try {
int idx = Integer.parseInt(lastContents);
lastContents = new XSSFRichTextString(sst.getEntryAt(idx))
.toString();
} catch (Exception e) {

}
}
if ("v".equals(name)) {
String value = lastContents.trim();
value = value.equals("")?" ":value;
rowlist.add(curCol, value);
curCol++;
}else {
//如果标签名称为 row ，这说明已到行尾，调用 optRows() 方法
if (name.equals("row")) {
rowReader.getRows(sheetIndex,curRow,rowlist);
rowlist.clear();
curRow++;
curCol = 0;
}
}
}

public void characters(char[] ch, int start, int length)
throws SAXException {
//得到单元格内容的值
lastContents += new String(ch, start, length);
}
}