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.documents;
029
030import org.opencms.file.CmsFile;
031import org.opencms.file.CmsObject;
032import org.opencms.file.CmsResource;
033import org.opencms.file.types.CmsResourceTypeXmlContent;
034import org.opencms.file.types.I_CmsResourceType;
035import org.opencms.main.CmsException;
036import org.opencms.main.OpenCms;
037import org.opencms.search.CmsIndexException;
038import org.opencms.search.I_CmsSearchDocument;
039import org.opencms.search.I_CmsSearchIndex;
040import org.opencms.search.extractors.CmsExtractionResult;
041import org.opencms.search.extractors.I_CmsExtractionResult;
042import org.opencms.util.CmsStringUtil;
043import org.opencms.xml.A_CmsXmlDocument;
044import org.opencms.xml.CmsXmlContentDefinition;
045import org.opencms.xml.content.CmsXmlContentFactory;
046import org.opencms.xml.content.I_CmsXmlContentHandler;
047import org.opencms.xml.types.I_CmsXmlContentValue;
048
049import java.util.ArrayList;
050import java.util.Iterator;
051import java.util.LinkedHashMap;
052import java.util.List;
053import java.util.Locale;
054
055/**
056 * Lucene document factory class to extract index data from an OpenCms VFS resource
057 * of type <code>CmsResourceTypeXmlContent</code>.<p>
058 *
059 * All XML nodes from the content for all locales will be stored separately in the item map
060 * which you can access using {@link CmsExtractionResult#getContentItems()}. The XML elements will be
061 * accessible using their xpath. The xpath will have the form like for example
062 * <code>Text[1]</code> or <code>Nested[1]/Text[1]</code>.<p>
063 *
064 * @since 6.0.0
065 */
066public class CmsDocumentXmlContent extends A_CmsVfsDocument {
067
068    /**
069     * Creates a new instance of this lucene document factory.<p>
070     *
071     * @param name name of the document type
072     */
073    public CmsDocumentXmlContent(String name) {
074
075        super(name);
076    }
077
078    /**
079     *
080     * @see org.opencms.search.documents.A_CmsVfsDocument#createDocument(org.opencms.file.CmsObject, org.opencms.file.CmsResource, org.opencms.search.I_CmsSearchIndex)
081     */
082    @Override
083    public I_CmsSearchDocument createDocument(CmsObject cms, CmsResource resource, I_CmsSearchIndex index)
084    throws CmsException {
085
086        CmsXmlContentDefinition def = CmsXmlContentDefinition.getContentDefinitionForResource(cms, resource);
087        if (def.getContentHandler().isContainerPageOnly()) {
088            return null;
089        }
090        return super.createDocument(cms, resource, index);
091    }
092
093    /**
094     * Returns the raw text content of a given VFS resource of type <code>CmsResourceTypeXmlContent</code>.<p>
095     *
096     * All XML nodes from the content for all locales will be stored separately in the item map
097     * which you can access using {@link CmsExtractionResult#getContentItems()}. The XML elements will be
098     * accessible using their xpath. The xpath will have the form like for example
099     * <code>Text[1]</code> or <code>Nested[1]/Text[1]</code>.<p>
100     *
101     * @see org.opencms.search.documents.I_CmsSearchExtractor#extractContent(CmsObject, CmsResource, I_CmsSearchIndex)
102     */
103    public I_CmsExtractionResult extractContent(CmsObject cms, CmsResource resource, I_CmsSearchIndex index)
104    throws CmsException {
105
106        logContentExtraction(resource, index);
107        try {
108            CmsFile file = readFile(cms, resource);
109            A_CmsXmlDocument xmlContent = CmsXmlContentFactory.unmarshal(cms, file);
110            I_CmsXmlContentHandler handler = xmlContent.getHandler();
111            Locale locale = index.getLocaleForResource(cms, resource, xmlContent.getLocales());
112            List<String> elements = xmlContent.getNames(locale);
113            StringBuffer content = new StringBuffer();
114            LinkedHashMap<String, String> items = new LinkedHashMap<String, String>();
115            for (Iterator<String> i = elements.iterator(); i.hasNext();) {
116                String xpath = i.next();
117                // xpath will have the form "Text[1]" or "Nested[1]/Text[1]"
118                I_CmsXmlContentValue value = xmlContent.getValue(xpath, locale);
119                if (handler.isSearchable(value)) {
120                    // the content value is searchable
121                    String extracted = value.getPlainText(cms);
122                    if (CmsStringUtil.isNotEmptyOrWhitespaceOnly(extracted)) {
123                        items.put(xpath, extracted);
124                        content.append(extracted);
125                        content.append('\n');
126                    }
127                }
128            }
129            return new CmsExtractionResult(content.toString(), items);
130        } catch (Exception e) {
131            throw new CmsIndexException(
132                Messages.get().container(Messages.ERR_TEXT_EXTRACTION_1, resource.getRootPath()),
133                e);
134        }
135    }
136
137    /**
138     * @see org.opencms.search.documents.I_CmsDocumentFactory#getDocumentKeys(java.util.List, java.util.List)
139     */
140    @Override
141    public List<String> getDocumentKeys(List<String> resourceTypes, List<String> mimeTypes) throws CmsException {
142
143        if (resourceTypes.contains("*")) {
144            // we need to find all configured XML content types
145            List<String> allTypes = new ArrayList<String>();
146            for (Iterator<I_CmsResourceType> i = OpenCms.getResourceManager().getResourceTypes().iterator(); i.hasNext();) {
147                I_CmsResourceType resourceType = i.next();
148                if ((resourceType instanceof CmsResourceTypeXmlContent)
149                    // either we need a configured schema, or another class name (which must then contain an inline schema)
150                    && (((CmsResourceTypeXmlContent)resourceType).getConfiguration().containsKey(
151                        CmsResourceTypeXmlContent.CONFIGURATION_SCHEMA)
152                        || !CmsResourceTypeXmlContent.class.equals(resourceType.getClass()))) {
153                    // add the XML content resource type name
154                    allTypes.add(resourceType.getTypeName());
155                }
156            }
157            resourceTypes = allTypes;
158        }
159
160        return super.getDocumentKeys(resourceTypes, mimeTypes);
161    }
162
163    /**
164     * @see org.opencms.search.documents.I_CmsDocumentFactory#isLocaleDependend()
165     */
166    public boolean isLocaleDependend() {
167
168        return true;
169    }
170
171    /**
172     * @see org.opencms.search.documents.I_CmsDocumentFactory#isUsingCache()
173     */
174    public boolean isUsingCache() {
175
176        return true;
177    }
178}