001/* 002 * This library is part of OpenCms - 003 * the Open Source Content Management System 004 * 005 * Copyright (c) Alkacon Software GmbH & Co. KG (http://www.alkacon.com) 006 * 007 * This library is free software; you can redistribute it and/or 008 * modify it under the terms of the GNU Lesser General Public 009 * License as published by the Free Software Foundation; either 010 * version 2.1 of the License, or (at your option) any later version. 011 * 012 * This library is distributed in the hope that it will be useful, 013 * but WITHOUT ANY WARRANTY; without even the implied warranty of 014 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 015 * Lesser General Public License for more details. 016 * 017 * For further information about Alkacon Software GmbH & Co. KG, please see the 018 * company website: http://www.alkacon.com 019 * 020 * For further information about OpenCms, please see the 021 * project website: http://www.opencms.org 022 * 023 * You should have received a copy of the GNU Lesser General Public 024 * License along with this library; if not, write to the Free Software 025 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA 026 */ 027 028package org.opencms.workplace.tools.content; 029 030import org.opencms.util.CmsHtmlParser; 031import org.opencms.util.CmsHtmlTagRemoveFactory; 032import org.opencms.util.CmsStringUtil; 033 034import java.util.Iterator; 035 036import org.htmlparser.NodeFactory; 037import org.htmlparser.Parser; 038import org.htmlparser.Tag; 039import org.htmlparser.lexer.Lexer; 040import org.htmlparser.lexer.Page; 041import org.htmlparser.util.ParserException; 042 043/** 044 * 045 * Html parser / visitor combination that visits a document and replaces Tag names by using the 046 * replacement configuration of a {@link org.opencms.workplace.tools.content.CmsTagReplaceSettings} 047 * instance. 048 * <p> 049 * 050 * Instances are reusable. 051 * <p> 052 * 053 * @since 6.1.7 054 * 055 */ 056public final class CmsTagReplaceParser extends CmsHtmlParser { 057 058 /** A tag factory that is able to make tags invisible to visitors. */ 059 private final NodeFactory m_nodeFactory; 060 061 /** 062 * Boolean flag that is set to true if during last call to {@link #process(String, String)} 063 * content was changed. 064 */ 065 private boolean m_changedContent; 066 /** 067 * The settings to use for replacing tags. 068 */ 069 private final CmsTagReplaceSettings m_settings; 070 071 /** 072 * Default constructor that turns echo on and uses the settings for replacing tags. 073 * <p> 074 * 075 * @param settings the settings to use for tag replacement. 076 */ 077 public CmsTagReplaceParser(CmsTagReplaceSettings settings) { 078 079 // echo on 080 super(true); 081 m_settings = settings; 082 CmsHtmlTagRemoveFactory nodeFactory = new CmsHtmlTagRemoveFactory(); 083 // add the removals of the settings to the tag factory: 084 Iterator itDeleteTags = m_settings.getDeleteTags().iterator(); 085 while (itDeleteTags.hasNext()) { 086 nodeFactory.addTagRemoval((Tag)itDeleteTags.next()); 087 } 088 m_nodeFactory = nodeFactory; 089 090 } 091 092 /** 093 * Overridden to also return the attributes of the Tag. 094 * <p> 095 * 096 * @see org.opencms.util.CmsHtmlParser#getTagHtml(org.htmlparser.Tag) 097 */ 098 @Override 099 public String getTagHtml(Tag tag) { 100 101 if (CmsStringUtil.isEmpty(tag.getTagName())) { 102 return ""; 103 } 104 StringBuffer result = new StringBuffer(32); 105 result.append('<'); 106 // Tag name is the first "Attribute"... 107 Iterator itAttributes = tag.getAttributesEx().iterator(); 108 while (itAttributes.hasNext()) { 109 result.append(itAttributes.next().toString()); 110 // avoid trailing whitespaces like <H1 > 111 // in 2nd run htmlparser 1.5 would turn the whitespace into an Attribute with null name 112 if (itAttributes.hasNext()) { 113 result.append(' '); 114 } 115 } 116 result.append('>'); 117 return result.toString(); 118 } 119 120 /** 121 * Extracts the text from the given html content, assuming the given html encoding. 122 * <p> 123 * Additionally tags are replaced / removed according to the configuration of this instance. 124 * <p> 125 * 126 * <h3>Please note:</h3> 127 * There are static process methods in the superclass that will not do the replacements / 128 * removals. Don't mix them up with this method. 129 * <p> 130 * 131 * @param html the content to extract the plain text from. 132 * 133 * @param encoding the encoding to use. 134 * 135 * @return the text extracted from the given html content. 136 * 137 * @throws ParserException if something goes wrong. 138 */ 139 @Override 140 public String process(String html, String encoding) throws ParserException { 141 142 // clear from potential previous run: 143 m_result = new StringBuffer(); 144 m_changedContent = false; 145 146 // initialize a parser with the given charset 147 Parser parser = new Parser(); 148 parser.setNodeFactory(m_nodeFactory); 149 Lexer lexer = new Lexer(); 150 Page page = new Page(html, encoding); 151 lexer.setPage(page); 152 parser.setLexer(lexer); 153 // process the page using the given visitor 154 parser.visitAllNodesWith(this); 155 // return the result 156 return getResult(); 157 } 158 159 /** 160 * @see org.opencms.util.CmsHtmlParser#visitEndTag(org.htmlparser.Tag) 161 */ 162 @Override 163 public void visitEndTag(Tag tag) { 164 165 boolean change = m_settings.replace(tag); 166 if (change) { 167 m_changedContent = true; 168 } 169 super.visitEndTag(tag); 170 } 171 172 /** 173 * @see org.opencms.util.CmsHtmlParser#visitTag(org.htmlparser.Tag) 174 */ 175 @Override 176 public void visitTag(Tag tag) { 177 178 boolean change = m_settings.replace(tag); 179 if (change) { 180 m_changedContent = true; 181 } 182 super.visitTag(tag); 183 } 184 185 /** 186 * Returns the changedContent. 187 * <p> 188 * 189 * @return the changedContent 190 */ 191 public boolean isChangedContent() { 192 193 return m_changedContent; 194 } 195 196}