OFFICE文档使用POI控件，PDF可以使用PDFBOX0.7.3控件，完全支持中文，用XPDF也行，不过感觉PDFBOX比较好，而且作者也在更新。水平有限，万望各位指正
WORD:
import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.apache.poi.hwpf.extractor.WordExtractor;
import java.io.File;
import java.io.InputStream;
import java.io.FileInputStream;
import com.search.code.Index;
public Document getDocument(Index index, String url, String title, InputStream is) throws DocCenterException {
String bodyText = null;
try {
WordExtractor ex = new WordExtractor(is);//is是WORD文件的InputStream
bodyText = ex.getText();
if(!bodyText.equals("")){
index.AddIndex(url, title, bodyText);
}
}catch (DocCenterException e) {
throw new DocCenterException("无法从该Mocriosoft Word文档中提取内容", e);
}catch(Exception e){
e.printStackTrace();
}
}
return null;
}
Excel:
import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.apache.poi.hwpf.extractor.WordExtractor;
import org.apache.poi.hssf.usermodel.HSSFWorkbook;
import org.apache.poi.hssf.usermodel.HSSFSheet;
import org.apache.poi.hssf.usermodel.HSSFRow;
import org.apache.poi.hssf.usermodel.HSSFCell;
import java.io.File;
import java.io.InputStream;
import java.io.FileInputStream;
import com.search.code.Index;
public Document getDocument(Index index, String url, String title, InputStream is) throws DocCenterException {
StringBuffer content = new StringBuffer();
try{
HSSFWorkbook workbook = new HSSFWorkbook(is);//创建对Excel工作簿文件的引用
for (int numSheets = 0; numSheets < workbook.getNumberOfSheets(); numSheets++) {
if (null != workbook.getSheetAt(numSheets)) {
HSSFSheet aSheet = workbook.getSheetAt(numSheets);//获得一个sheet
for (int rowNumOfSheet = 0; rowNumOfSheet <= aSheet.getLastRowNum(); rowNumOfSheet++) {
if (null != aSheet.getRow(rowNumOfSheet)) {
HSSFRow aRow = aSheet.getRow(rowNumOfSheet); //获得一个行
for (short cellNumOfRow = 0; cellNumOfRow <= aRow.getLastCellNum(); cellNumOfRow++) {
if (null != aRow.getCell(cellNumOfRow)) {
HSSFCell aCell = aRow.getCell(cellNumOfRow);//获得列值
content.append(aCell.getStringCellValue());
}
}
}
}
}
}
if(!content.equals("")){
index.AddIndex(url, title, content.toString());
}
}catch (DocCenterException e) {
throw new DocCenterException("无法从该Mocriosoft Word文档中提取内容", e);
}catch(Exception e) {
System.out.println("已运行xlRead() : " + e );
}
return null;
}
PowerPoint:
import java.io.InputStream;
import org.apache.lucene.document.Document;
import org.apache.poi.hslf.HSLFSlideShow;
import org.apache.poi.hslf.model.TextRun;
import org.apache.poi.hslf.model.Slide;
import org.apache.poi.hslf.usermodel.SlideShow;
public Document getDocument(Index index, String url, String title, InputStream is)
throws DocCenterException {
StringBuffer content = new StringBuffer("");
try{
责任编辑：小草
您现在的位置： 