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.search.extractors; 029 030import org.opencms.xml.CmsXmlGenericWrapper; 031 032import java.io.InputStream; 033import java.util.Iterator; 034import java.util.List; 035import java.util.zip.ZipEntry; 036import java.util.zip.ZipInputStream; 037 038import org.dom4j.Document; 039import org.dom4j.Node; 040import org.dom4j.io.SAXReader; 041 042/** 043 * Extracts the text from OpenOffice documents (.ods, .odf).<p> 044 * 045 * @since 7.0.4 046 */ 047public final class CmsExtractorOpenOffice extends A_CmsTextExtractor { 048 049 /** Static member instance of the extractor. */ 050 private static final CmsExtractorOpenOffice INSTANCE = new CmsExtractorOpenOffice(); 051 052 /** 053 * Hide the public constructor.<p> 054 */ 055 private CmsExtractorOpenOffice() { 056 057 // noop 058 } 059 060 /** 061 * Returns an instance of this text extractor.<p> 062 * 063 * @return an instance of this text extractor 064 */ 065 public static I_CmsTextExtractor getExtractor() { 066 067 return INSTANCE; 068 } 069 070 /** 071 * @see org.opencms.search.extractors.A_CmsTextExtractor#extractText(java.io.InputStream, java.lang.String) 072 */ 073 @Override 074 public I_CmsExtractionResult extractText(InputStream in, String encoding) throws Exception { 075 076 try (ZipInputStream zin = new ZipInputStream(in)) { 077 ZipEntry ze; 078 boolean FOUND_CONTENT = false; 079 String result = ""; 080 while (!FOUND_CONTENT) { 081 ze = zin.getNextEntry(); 082 FOUND_CONTENT = ze.getName().equalsIgnoreCase("content.xml"); 083 if (FOUND_CONTENT) { 084 result = readContent(zin); 085 } 086 } 087 result = removeControlChars(result); 088 return new CmsExtractionResult(result); 089 } 090 } 091 092 /** 093 * Internal routine that parses the specific content.xml part of 094 * an odf document.<p> 095 * 096 * @param in the input stream spooled to the start of the content.xml part 097 * 098 * @return the extracted content 099 * 100 * @throws Exception if sth goes wrong 101 */ 102 private String readContent(java.io.InputStream in) throws Exception { 103 104 StringBuffer resultBuffer = new StringBuffer(); 105 SAXReader reader = new SAXReader(); 106 Document doc = reader.read(in); 107 List<Node> textlist = CmsXmlGenericWrapper.selectNodes(doc, "//text:p[@*] | //text:span[@*]"); 108 Iterator<Node> li = textlist.iterator(); 109 while (li.hasNext()) { 110 Node textNode = li.next(); 111 String text = textNode.getText(); 112 if (text.length() > 1) { 113 text = " " + text + " "; 114 resultBuffer.append(text); 115 } 116 } 117 return resultBuffer.toString(); 118 } 119 120}