poi实现word文档的导入（针对.doc .docx rtf）

public void importUnitInfo() throws IOException{
String filePath = this.getHttpServletRequest().getParameter("docWord");
File file = new File(filePath);
POIFSFileSystem pfs = null;
List<String> list = new ArrayList<String>();
organization = new Organization();
String info="";
Boolean sign =true;
FileInputStream ins = new FileInputStream(filePath);// 载入文档
// WordExtractor extractor = new WordExtractor(ins);
// // 对DOC文件进行提取
// String text = extractor.getText();
XWPFWordExtractor docx = null;
int index = filePath.lastIndexOf(".");
String fileType =filePath.substring(index);
try {
if(fileType.equals(".docx")){

docx = new XWPFWordExtractor(POIXMLDocument.openPackage(filePath));//对docx文档的操作

}else if(fileType.equals(".doc")){

pfs = new POIFSFileSystem(ins); // 对doc文档的操作
}
} catch (Exception e) {
if(pfs==null&&docx==null){
sign=false;

}
}
if(sign==true){//导入的文件格式是word(doc或者docx)
try {
if(fileType.equals(".doc")){
HWPFDocument hwpf = new HWPFDocument(pfs);
Range range = hwpf.getRange();// 得到文档的读取范围
TableIterator it = new TableIterator(range);// 迭代文档中的表格
String cellString="";
if (it.hasNext()) {
TableRow tr = null;
TableCell td = null;
org.apache.poi.hwpf.usermodel.Paragraph para = null;
org.apache.poi.hwpf.usermodel.Table tb = it.next();
// 迭代行，从第1行开始
for (int i = 0; i < tb.numRows(); i++) {
tr = tb.getRow(i);
for (int j = 0; j < tr.numCells(); j++) {
td = tr.getCell(j);// 取得单元格
// 取得单元格的内容
for (int k = 0; k < td.numParagraphs(); k++) {
para = td.getParagraph(k);
cellString = para.text();
boolean flag = true;

if (cellString != null && cellString.compareTo("") != 0&&flag==true) {
// 如果不trim，取出的内容后会有一个乱码字符
cellString = cellString.trim();
}
info+= cellString;
}
}

}
}
}else if(fileType.equals(".docx")){
info = docx.getText();
info =info.replaceAll("\n", "");
info = info.replaceAll("\r", "");

}
info = info.replaceAll("\\s*", "");
.......（获得word中的内容（info）后，对具体内容操作很简单就不列出了） this.getHttpServletResponse().getWriter().write("success");
} catch (Exception e) {
this.getHttpServletResponse().getWriter().write("fail");
}
}else {//导入的文件是rtf格式
try {
RTFEditorKit rtf = new RTFEditorKit();
DefaultStyledDocument styledDoc = new DefaultStyledDocument();
FileInputStream in = new FileInputStream(filePath);
rtf.read(in, styledDoc, 0);
info = new String(styledDoc.getText(0,styledDoc.getLength()).getBytes("GBK")); //提取文本，读取中文需要使用ISO8859_1编码，否则会出现乱码
info = info.replaceAll("\n", "");
info = info.replaceAll("\\s*", "");
.......（获得word中的内容（info）后，对具体内容操作很简单就不列出了） [/color][color=black][size=xx-small][/size] this.getHttpServletResponse().getWriter().write("success");
} catch (IOException e) {
this.getHttpServletResponse().getWriter().write("fail");
e.printStackTrace();
} catch (BadLocationException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}

}