001 /* 002 * Copyright (c) 2007 Henri Sivonen 003 * 004 * Permission is hereby granted, free of charge, to any person obtaining a 005 * copy of this software and associated documentation files (the "Software"), 006 * to deal in the Software without restriction, including without limitation 007 * the rights to use, copy, modify, merge, publish, distribute, sublicense, 008 * and/or sell copies of the Software, and to permit persons to whom the 009 * Software is furnished to do so, subject to the following conditions: 010 * 011 * The above copyright notice and this permission notice shall be included in 012 * all copies or substantial portions of the Software. 013 * 014 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 015 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 016 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL 017 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 018 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING 019 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER 020 * DEALINGS IN THE SOFTWARE. 021 */ 022 023 package nu.validator.htmlparser.test; 024 025 import java.io.File; 026 import java.io.FileInputStream; 027 import java.io.IOException; 028 import java.io.OutputStreamWriter; 029 030 import nu.validator.htmlparser.sax.HtmlParser; 031 032 import org.xml.sax.InputSource; 033 import org.xml.sax.SAXException; 034 035 public class TreePrinter { 036 037 public static void main(String[] args) throws SAXException, IOException { 038 TreeDumpContentHandler treeDumpContentHandler = new TreeDumpContentHandler(new OutputStreamWriter(System.out, "UTF-8")); 039 HtmlParser htmlParser = new HtmlParser(); 040 htmlParser.setContentHandler(treeDumpContentHandler); 041 htmlParser.setLexicalHandler(treeDumpContentHandler); 042 htmlParser.setErrorHandler(new SystemErrErrorHandler()); 043 File file = new File(args[0]); 044 InputSource is = new InputSource(new FileInputStream(file)); 045 is.setSystemId(file.toURI().toASCIIString()); 046 htmlParser.parse(is); 047 } 048 }