apache poi解析word（doc）文档成xml及导出成html_sinat_30110061的博客-CSDN博客
//遇到解析word文档的需求，文档格式不定，在网上看了好多文章，大多是简单应用api解析内容，要不就是需要windows环境，还有个在线编辑的第三方的，不太符合本人需求，目前实现了doc的，docx的后续再说吧。
//本来是想把word解析成xml形式的字符串，就 参考apache poi中的 wordToHtmlConverter部分源码，因为它做了很多样式的解析，本人用不到这些，解析过多反而累赘，所以参考源码 +dom4j实现此功能，不过后来就顺便实现了下导出html（只有一两个样式）
//注意修改代码中word文档的地址和导出图片的存放地址（getImgUrl）
privatestatic Document _document_ = null;
static{
_document_ = DocumentHelper.createDocument();
}
publicstaticvoid main(String[] args) {
File file = new File("C:/Users/css/Desktop/1.doc"); //注意文档地址
if(file.exists()){
HWPFDocument doc;
try {
doc = new HWPFDocument(new FileInputStream(file));
Range range = doc.getRange();
wordToHtml(doc);
// printRange(doc);
} catch (FileNotFoundException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
}
}
/**
* word解析并导出html文件
* @param doc
*/
privatestaticvoid wordToHtml(HWPFDocument doc){
Element htmlElement = DocumentHelper.createElement("html");
_document_.setRootElement(htmlElement);
Element headElement = DocumentHelper.createElement("head");
Element charSetElement = DocumentHelper.createElement("meta");
charSetElement.addAttribute("http-equiv", "Content-Type");
charSetElement.addAttribute("content", "text/html; charset=UTF-8");
headElement.add(charSetElement);
htmlElement.add(headElement);
Element bodyElement = DocumentHelper.createElement("body");
Element contentElement = formatRange(doc); //解析word
bodyElement.add(contentElement);
htmlElement.add(bodyElement);
String docString = _document_.asXML();
writeFile(docString, "C:/Users/css/Desktop/12345.html");
System._out_.println(docString);
}
/**
* 导出文件
* @param content
* @param path
*/
private static void writeFile(String content, String path) {
FileOutputStream fos = null;
BufferedWriter bw = null;
try {
File file = new File(path);
fos = new FileOutputStream(file);
bw = new BufferedWriter(new OutputStreamWriter(fos,"UTF-8"));
bw.write(content);
} 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) {
}
}
}
/**
* 解析word文档（包括table表格、图片、text文本内容）
* @param doc
* @return
*/
private static Element formatRange(HWPFDocument doc){
Element contentEle = DocumentHelper.createElement("div");
Range range = doc.getRange();
PicturesTable pt = doc.getPicturesTable();
int pnum = range.numParagraphs();
Paragraph paragraph = null;
String text = null;
Element divElement = null;
CharacterRun run = null;
int numCharacterRuns = 0;
StringBuilder styleStr = null;
for(int i=0;i<pnum;i++){
paragraph = range.getParagraph(i);//段落
styleStr = new StringBuilder();
if(paragraph.isInTable()){//是否为table
Table table = range.getTable(paragraph);
divElement = formatTable(table);//解析table内容
i += table.numParagraphs();//跳过table
i--;
}else if(paragraph.isInList()){ //目前解析的word模板没用到也就没做
System._out_.println("handle paragrah list----------------");
}else{
text = paragraph.text();
if(text != null && !"".equals(text)){
run = paragraph.getCharacterRun(0);
numCharacterRuns = paragraph.numCharacterRuns();
if(run != null && numCharacterRuns>0){
if(run.text().charAt(0)==0x01 && pt.hasPicture(run)){//图片
divElement = formatImg(pt.extractPicture(run, true));//解析图片，创建img节点，并导出图片
}else{//文本内容
divElement = DocumentHelper.createElement("div");
divElement.addText(text);
if(run.isBold()){//加粗样式
styleStr.append("font-weight:bold;");
}
if(styleStr.length()>0){
divElement.addAttribute("style", styleStr.toString());
}
}
}
}
}
if(divElement != null){//添加段落节点
contentEle.add(divElement);
}
}
return contentEle;
}
/**
* 解析table表格
* @param table
* @return
*/
private static Element formatTable(Table table){
Element tableElement = DocumentHelper.createElement("table");
Element theadElement = DocumentHelper.createElement("thead");
Element tbodyElement = DocumentHelper.createElement("tbody");
int[] tableCellEdges = buildTableCellEdgesArray(table); //单元格边界
int rownum = table.numRows();
int maxColumns = Integer._MIN_VALUE_;
for ( int r = 0; r < rownum; r++ ){
maxColumns = Math.max( maxColumns, table.getRow( r ).numCells() );
}
Element rowElement = null;
TableRow row = null;
int cellnum = 0;
Element cellElement = null;
TableCell cell = null;
int rowSpan = 0;
int colSpan = 0;
tableElement.addAttribute("border", "1");
tableElement.addAttribute("style", "border-spacing:0");
for(int i=0;i<rownum;i++){
row = table.getRow(i);
if(row != null){
rowElement = DocumentHelper.createElement("tr");
cellnum = row.numCells();
int currentEdgeIndex = 0;
for(int j=0;j<cellnum;j++){
cell = row.getCell(j);
if(cell.isVerticallyMerged() && !cell.isFirstVerticallyMerged())
{
currentEdgeIndex += getNumberColumnsSpanned(tableCellEdges, currentEdgeIndex, cell );
continue;
}
if(cell != null){
if(row.isTableHeader()){
cellElement = DocumentHelper.createElement("th");
}else{
cellElement = DocumentHelper.createElement("td");
}
colSpan = getNumberColumnsSpanned( tableCellEdges, currentEdgeIndex, cell );//取得列合并数
currentEdgeIndex += colSpan;
if ( colSpan == 0 ){
continue;
}
if ( colSpan != 1 ){
cellElement.addAttribute( "colspan", String.valueOf( colSpan ) );
}
rowSpan = getNumberRowsSpanned( table, tableCellEdges, i, j, cell );//取得行合并数
if ( rowSpan > 1 ){
cellElement.addAttribute( "rowspan", String.valueOf( rowSpan ) );
}
cellElement.addText(cell.text());
}
if(cellElement != null){
rowElement.add(cellElement);
}
}
}
if(row.isTableHeader()){
theadElement.add(rowElement);
}else{
tbodyElement.add(rowElement);
}
}
if(theadElement.hasContent()){
tableElement.add(theadElement);
}
if(tbodyElement.hasContent()){
tableElement.add(tbodyElement);
}
return tableElement;
http://sina.lt/cb6e
http://sina.lt/cb6f
http://sina.lt/cb6g
http://sina.lt/cb6h
http://sina.lt/cb6j
http://sina.lt/cb6k
http://sina.lt/cb6m
http://sina.lt/cb6p
http://sina.lt/cb6q
http://sina.lt/cb6r
http://sina.lt/cb6s
http://sina.lt/cb6t
http://sina.lt/cb6u
http://sina.lt/cb6v
http://sina.lt/cb6w
http://sina.lt/cb6x
http://sina.lt/cb6y
http://sina.lt/cb6z
http://sina.lt/cb62
http://sina.lt/cb63
http://sina.lt/cb64
http://sina.lt/cb65
http://sina.lt/cb66
http://sina.lt/cb67
http://sina.lt/cb68
http://sina.lt/cb69
http://sina.lt/cb6A
http://sina.lt/cb6B
http://sina.lt/cb6C
}
/**
* 解析table单元格边界
* @param table
* @return
*/
private static int[] buildTableCellEdgesArray( Table table )
{
Set<Integer> edges = new TreeSet<Integer>();
for ( int r = 0; r < table.numRows(); r++ )
{
TableRow tableRow = table.getRow( r );
for ( int c = 0; c < tableRow.numCells(); c++ )
{
TableCell tableCell = tableRow.getCell( c );
edges.add( Integer.valueOf( tableCell.getLeftEdge() ) );
edges.add( Integer.valueOf( tableCell.getLeftEdge()
+ tableCell.getWidth() ) );
}
}
Integer[] sorted = edges.toArray( new Integer[edges.size()] );
int[] result = new int[sorted.length];
for ( int i = 0; i < sorted.length; i++ )
{
result[i] = sorted[i].intValue();
}
return result;
}
/**
* 解析table列合并数
* @param tableCellEdges
* @param currentEdgeIndex
* @param tableCell
* @return
*/
private static int getNumberColumnsSpanned( int[] tableCellEdges,
int currentEdgeIndex, TableCell tableCell )
{
int nextEdgeIndex = currentEdgeIndex;
int colSpan = 0;
int cellRightEdge = tableCell.getLeftEdge() + tableCell.getWidth();
while ( tableCellEdges[nextEdgeIndex] < cellRightEdge )
{
colSpan++;
nextEdgeIndex++;
}
return colSpan;
}
/**
* 解析table行合并数
* @param table
* @param tableCellEdges
* @param currentRowIndex
* @param currentColumnIndex
* @param tableCell
* @return
*/
private static int getNumberRowsSpanned( Table table,
final int[] tableCellEdges, int currentRowIndex,
int currentColumnIndex, TableCell tableCell )
{
if ( !tableCell.isFirstVerticallyMerged() )
return 1;
final int numRows = table.numRows();
int count = 1;
for ( int r1 = currentRowIndex + 1; r1 < numRows; r1++ )
{
TableRow nextRow = table.getRow( r1 );
if ( currentColumnIndex >= nextRow.numCells() )
break;
// we need to skip row if he don't have cells at all
boolean hasCells = false;
int currentEdgeIndex = 0;
for ( int c = 0; c < nextRow.numCells(); c++ )
{
TableCell nextTableCell = nextRow.getCell( c );
if ( !nextTableCell.isVerticallyMerged()
|| nextTableCell.isFirstVerticallyMerged() )
{
int colSpan = getNumberColumnsSpanned( tableCellEdges,
currentEdgeIndex, nextTableCell );
currentEdgeIndex += colSpan;
if ( colSpan != 0 )
{
hasCells = true;
break;
}
}
else
{
currentEdgeIndex += getNumberColumnsSpanned(
tableCellEdges, currentEdgeIndex, nextTableCell );
}
}
if ( !hasCells )
continue;
TableCell nextCell = nextRow.getCell( currentColumnIndex );
if ( !nextCell.isVerticallyMerged()
|| nextCell.isFirstVerticallyMerged() )
break;
count++;
}
return count;
}
/**
* 解析图片，创建img节点，并导出图片
* @param picture
* @return
*/
private static Element formatImg(Picture picture){
Element imgElement = null;
if(picture != null){
String imgUrl = getImgUrl(picture.suggestFullFileName());
exportImg(picture, imgUrl);
imgElement = DocumentHelper.createElement("img");
imgElement.addAttribute("src", imgUrl);
}
return imgElement;
}
/**
* 获取图片保存位置
* @param suggestedName
* @return
*/
private static String getImgUrl(String suggestedName){
return "C:/Users/css/Desktop/"+suggestedName; //注意图片导出地址
}
/**
* 导出图片
* @param picture
* @param expUrl
*/
private static void exportImg(Picture picture,String expUrl){
if(picture != null && expUrl != null && !"".equals(expUrl))
try {
picture.writeImageContent(new FileOutputStream(expUrl));
} catch (IOException e) {
e.printStackTrace();
}
}
原网址: 访问
创建于: 2023-05-08 16:42:40
目录: default
标签: 无
未标明原创文章均为采集，版权归作者所有，转载无需和我联系，请注明原出处，南摩阿彌陀佛，知识，不只知道，要得到
最新评论