public classReadHtml2 { publicstatic void main(String[] args) throws IOException {//String strUrl=" /?reg&tpl=mn"; /// URL url=new URL(strUrl); File f=new File("fortest.htm"); //輸入流 //InputStreamReader isr=newInputStreamReader(url.openStream()); InputStreamReader isr1=newInputStreamReader(new FileInputStream(f)); BufferedReader br=new BufferedReader(isr1); //獲取html轉換成String String s; String AllContent=""; while((s=br.readLine())!=null) { AllContent=AllContent+s; } //使用後HTML Parser 控件 Parser myParser; NodeList nodeList = null; myParser =Parser.createParser(AllContent, "utf-8"); NodeFilter tableFilter = newNodeClassFilter(TableTag.class); OrFilter lastFilter = newOrFilter(); lastFilter.setPredicates(newNodeFilter[] { tableFilter }); try { //獲取標簽為table的節點列表 nodeList =myParser.parse(lastFilter); //循環讀取每個table for (int i = 0; i <=nodeList.size(); i++) { if (nodeList.elementAt(i)instanceof TableTag) { TableTag tag = (TableTag)nodeList.elementAt(i); TableRow[] rows =tag.getRows(); System.out.println("----------------------table "+i+"--------------------------------"); //循環讀取每壹行 for (int j = 0; j <rows.length; j++) { TableRow tr =(TableRow) rows[j]; TableColumn[] td =tr.getColumns(); //讀取每行的單元格內容 for (int k = 0; k< td.length; k++) { System.out.println(td[k].getStringText());//（按照自己需要的格式輸出） } } } } } catch (ParserException e) { e.printStackTrace(); }}}
上一篇:Awtk源代碼下一篇:QQ飛車刷車方法