合并Word文件（包含doc和docx文件）
我遇到的问题是下载并合并附件，这里的附件大多是doc文件，也包含少量的docx文件，但是文件路径是从数据库中读取出来的，均不带后缀名，传统的xwpfdocument和hwpfdocument不能完全解决我的问题；尤其是将文件合并，不能轻易办到，需要对文档进行解析。
这里我采用的方法是将Word文件转换为HTML文件，把Word的合并转化为HTML的合并；这样一来就减少了难度，不过就是还需要再把HTML文件转化为doc文件或者docx文件（此时，你就可以指定是哪种文件了）。
在转换的时候，分两个方向，一个是doc文件转换，另一个是docx文件转换；这里的转换时必须包含文档中的格式的（图片和表格我这里没有进行测试）。如果是没有文件后缀，那么就需要先判断是doc文件还是docx文件，这里用到了一个工具类，就是通过文件的文件头来判断文件类型，因为我这里只是为了区别doc和docx，所以就比较了前四位的16进制数，按照if和else来走两条转换路线。（具体文件的文件头可上网查资料，各种文件都有）。
转换doc文件的时候，是按照字符读取的，判断每个字符的字体颜色和样式，将其转换为HTML的代码，最后应该是整个稳当的HTML的字符串的累加，因为我这里是合并，所以我使用for循环进行了文件主体的叠加，最后在循环的外面加上HTML的头部和尾部信息即可。除此之外，根据需求，需要在不同文档之间插入分页符，分页符用HTML代码可以表示，"<br clear=all style='page-break-before:always' mce_style='page-break-before:always'> "；需要的直接添加即可。
然后是docx文件的转换，这里的就不像doc文件可以按照每个字符来读了，而是将整个文档直接转换为HTML文件，通过打印字符串可以得知，转换出来的HTML代码就是一个大的div，如果直接使用这个代码，合并的时候格式就不统一了，所以需要将图中的style样式去掉，直接用字符串截取，并加上一个空的<div>即可。
最后，再将HTML文件转为doc或者docx，可以将路径放在服务器上面，并且实现下载就行（建议每次下载完之后将该文件清空，可以循环利用）。详细代码如下。
package com.landray.kmss.km.doc.util;
import java.io.BufferedWriter;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileNotFoundException;
import java.io.FileOutputStream;
import java.io.IOException;
import java.io.OutputStream;
import java.io.OutputStreamWriter;
import java.util.ArrayList;
import java.util.List;
import org.apache.poi.hwpf.HWPFDocument;
import org.apache.poi.hwpf.model.PicturesTable;
import org.apache.poi.hwpf.usermodel.CharacterRun;
import org.apache.poi.hwpf.usermodel.Paragraph;
import org.apache.poi.hwpf.usermodel.Picture;
import org.apache.poi.hwpf.usermodel.Range;
import org.apache.poi.hwpf.usermodel.Table;
import org.apache.poi.hwpf.usermodel.TableCell;
import org.apache.poi.hwpf.usermodel.TableIterator;
import org.apache.poi.hwpf.usermodel.TableRow;
import org.apache.poi.xwpf.usermodel.XWPFDocument;
public class WordExcelToHtml {
/**
* 回车符ASCII码
*/
private static final short ENTER_ASCII = 13;
/**
* 空格符ASCII码
*/
private static final short SPACE_ASCII = 32;
/**
* 水平制表符ASCII码
*/
private static final short TABULATION_ASCII = 9;
// public static String htmlText = "";
public static String mainText = "";
public static String htmlTextTbl = "";
public static int counter = 0;
public static int beginPosi = 0;
public static int endPosi = 0;
public static int beginArray[];
public static int endArray[];
public static String htmlTextArray[];
public static boolean tblExist = false;
public static void main(String argv[]) {
try {
String htmlText = "<html><head><meta http-equiv='Content-Type' content='text/html; charset=utf-8' />"
+ "</head><body>"; //将每一个Word中的主体部分拿出来，合并之后加上HTML的头和尾，但是要注意编码
List<String> list = new ArrayList<String>();
String file1 = "D://file8";
String file2 = "D://file9";
String file3 = "D://file11";
list.add(file1);
list.add(file2);
list.add(file3);
// String mainText1 = "";
for (int i = 0; i < list.size(); i++) {
htmlText += getWordAndStyle(list.get(i))
+ "<br clear=all style='page-break-before:always' mce_style='page-break-before:always'> ";
//每一个文档读取完之后，加上一个分页符，继续累加
}
htmlText += "</body></html>";
String filePath = "D://1.html";
writeFile(htmlText, filePath);
new HtmlToDoc().writeWordFile(filePath, "D://file10.doc");
} catch (Exception e) {
e.printStackTrace();
}
}
/**
* 读取每个文字样式
*
* @param fileName
* @throws Exception
*/
public static String getWordAndStyle(String fileName) throws Exception {
String htmlText = "";
FileInputStream in = new FileInputStream(new File(fileName));
//根据文本内容判断是doc还是docx
byte[] b = new byte[4];
in.read(b, 0, b.length);
in.close();
FileInputStream in1 = new FileInputStream(new File(fileName));
System.out.println(bytesToHexString(b) + ";;;");
if (bytesToHexString(b).equalsIgnoreCase("d0cf11e0")) {//"d0cf11e0"代表的是doc文件
HWPFDocument doc = new HWPFDocument(in1);
Range rangetbl = doc.getRange();// 得到文档的读取范围
TableIterator it = new TableIterator(rangetbl);
int num = 0;
beginArray = new int[num];
endArray = new int[num];
htmlTextArray = new String[num];
// 取得文档中字符的总数
int length = doc.characterLength();
// 创建图片容器;
PicturesTable pTable = doc.getPicturesTable();
int cur = 0;
String tempString = "";
for (int i = 0; i < length - 1; i++) {
// 整篇文章的字符通过一个个字符的来判断,range为得到文档的范围
Range range = new Range(i, i + 1, doc);
CharacterRun cr = range.getCharacterRun(0);
if (tblExist && cur < beginArray.length) {
if (i == beginArray[cur]) {
htmlText += tempString + htmlTextArray[cur];
tempString = "";
i = endArray[cur] - 1;
cur++;
continue;
}
}
if (pTable.hasPicture(cr)) {
htmlText += tempString;
// 读写图片
// readPicture(pTable, cr);
tempString = "";
} else {
Range range2 = new Range(i + 1, i + 2, doc);
// 第二个字符
CharacterRun cr2 = range2.getCharacterRun(0);
char c = cr.text().charAt(0);
// 判断是否为回车符
if (c == ENTER_ASCII) {
tempString += "<br/>";
}
// 判断是否为空格符
else if (c == SPACE_ASCII)
tempString += " ";
// 判断是否为水平制表符
else if (c == TABULATION_ASCII)
tempString += " ";
// 比较前后2个字符是否具有相同的格式
boolean flag = compareCharStyle(cr, cr2);
String fontStyle = "<span class='text' style=\"font-family:"
+ cr.getFontName()
+ ";font-size:"
+ cr.getFontSize()
/ 2
+ "pt;color:"
+ ColorUtils.getHexColor(cr.getIco24()) + ";";
if (cr.isBold())
fontStyle += "font-weight:bold;";
if (cr.isItalic())
fontStyle += "font-style:italic;";
htmlText += fontStyle + "\" mce_style=\"font-family:"
+ cr.getFontName() + ";font-size:"
+ cr.getFontSize() / 2 + "pt;";
if (cr.isBold())
fontStyle += "font-weight:bold;";
if (cr.isItalic())
fontStyle += "font-style:italic;";
htmlText += fontStyle + "\">" + tempString + cr.text()
+ "</span>";
tempString = "";
}
}
htmlText += tempString;
return htmlText;
} else {
Word2007ToHtml w = new Word2007ToHtml();
String filepath = "";
String fileName1 = fileName;
String htmlName = "D://3.html";
w.Word2007ToHtml(fileName1, htmlName);
String result = w.readFileByBytes(htmlName);
int i = result.indexOf('>');
String realreasult = "<div>"+result.substring(i+1);
System.out.println(realreasult);
htmlText += realreasult;
return htmlText;
}
}
/**
* 读写文档中的表格
*
* @param pTable
* @param cr
* @throws Exception
*/
public static void readTable(TableIterator it, Range rangetbl)
throws Exception {
htmlTextTbl = "";
// 迭代文档中的表格
counter = -1;
while (it.hasNext()) {
tblExist = true;
htmlTextTbl = "";
Table tb = (Table) it.next();
beginPosi = tb.getStartOffset();
endPosi = tb.getEndOffset();
System.out.println("............" + beginPosi + "...." + endPosi);
counter = counter + 1;
// 迭代行，默认从0开始
beginArray[counter] = beginPosi;
endArray[counter] = endPosi;
htmlTextTbl += "<table border='1'>";
for (int i = 0; i < tb.numRows(); i++) {
TableRow tr = tb.getRow(i);
htmlTextTbl += "<tr >";
// 迭代列，默认从0开始
for (int j = 0; j < tr.numCells(); j++) {
TableCell td = tr.getCell(j);// 取得单元格
int cellWidth = td.getWidth();
// 取得单元格的内容
for (int k = 0; k < td.numParagraphs(); k++) {
Paragraph para = td.getParagraph(k);
String s = para.text().toString().trim();
if (s == "") {
s = " ";
}
System.out.println(s);
htmlTextTbl += "<td class='text' width=" + cellWidth
+ ">" + s + "</td>";
System.out.println(i + ":" + j + ":" + cellWidth + ":"
+ s);
} // end for
} // end for
} // end for
htmlTextTbl += "</table>";
htmlTextArray[counter] = htmlTextTbl;
} // end while
}
/**
* 读写文档中的图片
*
* @param pTable
* @param cr
* @throws Exception
*/
public static void readPicture(PicturesTable pTable, CharacterRun cr)
throws Exception {
// 提取图片
Picture pic = pTable.extractPicture(cr, false);
// 返回POI建议的图片文件名
String afileName = pic.suggestFullFileName();
OutputStream out = new FileOutputStream(new File("e://test"
+ File.separator + afileName));
pic.writeImageContent(out);
// htmlText += "<img src=\"e://test//" + afileName
// + "\" mce_src=\"e://test//" + afileName + "\"/>";
}
public static boolean compareCharStyle(CharacterRun cr1, CharacterRun cr2) {
boolean flag = false;
if (cr1.isBold() == cr2.isBold() && cr1.isItalic() == cr2.isItalic()
&& cr1.getFontName().equals(cr2.getFontName())
&& cr1.getFontSize() == cr2.getFontSize()) {
flag = true;
}
return flag;
}
/**
* 写文件
*
* @param s
*/
public static void writeFile(String s, String filePath) {
FileOutputStream fos = null;
BufferedWriter bw = null;
try {
File file = new File(filePath);
fos = new FileOutputStream(file);
bw = new BufferedWriter(new OutputStreamWriter(fos));
bw.write(s);
} catch (FileNotFoundException fnfe) {
fnfe.printStackTrace();
} catch (IOException ioe) {
ioe.printStackTrace();
} finally {
try {
if (bw != null)
bw.close();
if (fos != null)
fos.close();
} catch (IOException ie) {
}
}
}
// 判断文件类型
public static String bytesToHexString(byte[] src) {
StringBuilder stringBuilder = new StringBuilder();
if (src == null || src.length <= 0) {
return null;
}
for (int i = 0; i < src.length; i++) {
int v = src[i] & 0xFF;
String hv = Integer.toHexString(v);
if (hv.length() < 2) {
stringBuilder.append(0);
}
stringBuilder.append(hv);
}
return stringBuilder.toString();
}
}
获取字体颜色的工具类：
package com.landray.kmss.km.doc.util;
public class ColorUtils {
public static int red(int c) {
return c & 0XFF;
}
public static int green(int c) {
return (c >> 8) & 0XFF;
}
public static int blue(int c) {
return (c >> 16) & 0XFF;
}
public static int rgb(int c) {
return (red(c) << 16) | (green(c) <<8) | blue(c);
}
public static String rgbToSix(String rgb) {
int length = 6 - rgb.length();
String str = "";
while(length > 0){
str += "0";
length--;
}
return str + rgb;
}
public static String getHexColor(int color) {
color = color == -1 ? 0 : color;
int rgb = rgb(color);
return "#" + rgbToSix(Integer.toHexString(rgb));
}
}
将HTML文件转换为doc文件：
package com.landray.kmss.km.doc.util;
import java.io.BufferedReader;
import java.io.ByteArrayInputStream;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileOutputStream;
import java.io.IOException;
import java.io.InputStreamReader;
import java.nio.charset.Charset;
import org.apache.poi.poifs.filesystem.DirectoryEntry;
import org.apache.poi.poifs.filesystem.DocumentEntry;
import org.apache.poi.poifs.filesystem.POIFSFileSystem;
//将docx文件转为HTML
package com.landray.kmss.km.doc.util;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileOutputStream;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.io.OutputStream;
import java.io.Reader;
import org.apache.poi.xwpf.usermodel.XWPFDocument;
import org.apache.poi.xwpf.converter.core.FileImageExtractor;
import org.apache.poi.xwpf.converter.core.FileURIResolver;
import org.apache.poi.xwpf.converter.xhtml.XHTMLConverter;
import org.apache.poi.xwpf.converter.xhtml.XHTMLOptions;
import org.junit.Test;
public class Word2007ToHtml{
@Test
public void Word2007ToHtml(String fileName,String htmlName) throws IOException {
final String file = fileName;
File f = new File(file);
if (!f.exists()) {
System.out.println("Sorry File does not Exists!");
} else {
// ) 加载word文档生成 XWPFDocument对象
InputStream in = new FileInputStream(f);
XWPFDocument document = new XWPFDocument(in);
// ) 解析 XHTML配置 (这里设置IURIResolver来设置图片存放的目录)
File imageFolderFile = new File("D://");
XHTMLOptions options = XHTMLOptions.create().URIResolver(new FileURIResolver(imageFolderFile));
options.setExtractor(new FileImageExtractor(imageFolderFile));
options.setIgnoreStylesIfUnused(false);
options.setFragment(true);
// ) 将 XWPFDocument转换成XHTML
File file1 = new File(htmlName);
OutputStream out = new FileOutputStream(file1);
XHTMLConverter.getInstance().convert(document, out, options);
//也可以使用字符数组流获取解析的内容
// ByteArrayOutputStream baos = new ByteArrayOutputStream();
// XHTMLConverter.getInstance().convert(document, baos, options);
// String content = baos.toString();
// System.out.println(content);
// baos.close();
}
}
public static void main(String[] args) throws IOException {
Word2007ToHtml w = new Word2007ToHtml();
String fileName = "D://file1.docx";
String htmlName = "D://3.html";
w.Word2007ToHtml(fileName,htmlName);
String result = readFileByBytes(htmlName);
System.out.println(result);
}
public static String readFileByBytes(String fileName) {
String s="";
File file = new File(fileName);
Reader reader = null;
try {
//System.out.println("以字符为单位读取文件内容，一次读一个字节：");
// 一次读一个字符
reader = new InputStreamReader(new FileInputStream(file),"utf-8");
int tempchar;
while ((tempchar = reader.read()) != -1) {
// 对于windows下，\r\n这两个字符在一起时，表示一个换行。
// 但如果这两个字符分开显示时，会换两次行。
// 因此，屏蔽掉\r，或者屏蔽\n。否则，将会多出很多空行。
if (((char) tempchar) != '\r') {
s +=(char) tempchar;
}
}
reader.close();
} catch (Exception e) {
e.printStackTrace();
}
return s;
}
}
/** * 将html文档转为doc * @author soildwang * */
public class HtmlToDoc {
/**
* 读取html文件到word *
* @param filepath html文件的路径 * @return
* * @throws Exception */
public boolean writeWordFile(String filepath,String outfile) throws Exception {
boolean flag = false;
ByteArrayInputStream bais = null;
FileOutputStream fos = null;
//String outfile = "D://file8.doc"; //根据实际情况写路径
try {
if (!"".equals(outfile)) {
File fileDir = new File(outfile);
if (fileDir.exists()) {
String content = readFile(filepath);
byte b[] = content.getBytes();
bais = new ByteArrayInputStream(b);
POIFSFileSystem poifs = new POIFSFileSystem();
DirectoryEntry directory = poifs.getRoot();
DocumentEntry documentEntry = directory.createDocument("WordDocument", bais);
fos = new FileOutputStream(outfile);
poifs.writeFilesystem(fos);
bais.close();
fos.close();
}
}
} catch (IOException e) {
e.printStackTrace();
} finally {
if(fos != null) fos.close();
if(bais != null) bais.close();
} return flag;
}
/**
* * 读取html文件到字符串 * @param filename
* * @return * @throws Exception
* */
public String readFile(String filename) throws Exception {
StringBuffer buffer = new StringBuffer("");
BufferedReader br = null;
try {
br = new BufferedReader(new InputStreamReader(new FileInputStream(new File(filename)),Charset.forName("utf-8")));
buffer = new StringBuffer();
while (br.ready())
buffer.append((char) br.read());
} catch (Exception e) {
e.printStackTrace();
} finally {
if(br!=null) br.close();
}
return buffer.toString();
}
//局部测试
public static void main(String[] args) throws Exception {
new HtmlToDoc().writeWordFile("d://1.html","D://file8.doc");//根据实际情况写文件路径
}
}
