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.solr;
029
030import org.opencms.ade.configuration.CmsADEConfigData;
031import org.opencms.file.CmsFile;
032import org.opencms.file.CmsObject;
033import org.opencms.file.CmsResource;
034import org.opencms.main.CmsException;
035import org.opencms.main.CmsLog;
036import org.opencms.main.OpenCms;
037import org.opencms.search.CmsIndexException;
038import org.opencms.search.I_CmsSearchIndex;
039import org.opencms.search.documents.Messages;
040import org.opencms.search.extractors.CmsExtractionResult;
041import org.opencms.search.extractors.I_CmsExtractionResult;
042import org.opencms.xml.containerpage.CmsContainerElementBean;
043import org.opencms.xml.containerpage.CmsContainerPageBean;
044import org.opencms.xml.containerpage.CmsFormatterConfiguration;
045import org.opencms.xml.containerpage.CmsXmlContainerPage;
046import org.opencms.xml.containerpage.CmsXmlContainerPageFactory;
047
048import java.util.ArrayList;
049import java.util.HashMap;
050import java.util.LinkedHashMap;
051import java.util.List;
052import java.util.Locale;
053import java.util.Map;
054
055import org.apache.commons.logging.Log;
056
057/**
058 * Lucene document factory class to extract index data from a resource
059 * of type <code>CmsResourceTypeContainerPage</code>.<p>
060 *
061 * @since 8.5.0
062 */
063public class CmsSolrDocumentContainerPage extends CmsSolrDocumentXmlContent {
064
065    /** The log object for this class. */
066    private static final Log LOG = CmsLog.getLog(CmsSolrDocumentContainerPage.class);
067
068    /** The solr document type name for xml-contents. */
069    public static final String TYPE_CONTAINERPAGE_SOLR = "containerpage-solr";
070
071    /**
072     * Creates a new instance of this lucene document factory.<p>
073     *
074     * @param name name of the document type
075     */
076    public CmsSolrDocumentContainerPage(String name) {
077
078        super(name);
079    }
080
081    /**
082     * Returns the raw text content of a VFS resource of type <code>CmsResourceTypeContainerPage</code>.<p>
083     *
084     * @see org.opencms.search.documents.I_CmsSearchExtractor#extractContent(CmsObject, CmsResource, I_CmsSearchIndex)
085     */
086    @Override
087    public I_CmsExtractionResult extractContent(CmsObject cms, CmsResource resource, I_CmsSearchIndex index)
088    throws CmsException {
089
090        return extractContent(cms, resource, index, null);
091    }
092
093    /**
094     * Extracts the content of a given index resource according to the resource file type and the
095     * configuration of the given index.<p>
096     *
097     * @param cms the cms object
098     * @param resource the resource to extract the content from
099     * @param index the index to extract the content for
100     * @param forceLocale if set, only the content values for the given locale will be extracted
101     *
102     * @return the extracted content of the resource
103     *
104     * @throws CmsException if something goes wrong
105     */
106    public I_CmsExtractionResult extractContent(
107        CmsObject cms,
108        CmsResource resource,
109        I_CmsSearchIndex index,
110        Locale forceLocale)
111    throws CmsException {
112
113        logContentExtraction(resource, index);
114        I_CmsExtractionResult ex = null;
115        try {
116            CmsFile file = readFile(cms, resource);
117            CmsXmlContainerPage containerPage = CmsXmlContainerPageFactory.unmarshal(cms, file);
118
119            List<I_CmsExtractionResult> all = new ArrayList<I_CmsExtractionResult>();
120            CmsContainerPageBean containerBean = containerPage.getContainerPage(cms);
121            if (containerBean != null) {
122                for (CmsContainerElementBean element : containerBean.getElements()) {
123                    // check all elements in this container
124                    // get the formatter configuration for this element
125                    try {
126                        element.initResource(cms);
127                        CmsResource elementResource = element.getResource();
128                        if (!(cms.readProject(index.getProject()).isOnlineProject()
129                            && elementResource.isExpired(System.currentTimeMillis()))) {
130                            CmsADEConfigData adeConfig = OpenCms.getADEManager().lookupConfiguration(
131                                cms,
132                                file.getRootPath());
133                            CmsFormatterConfiguration formatters = adeConfig.getFormatters(cms, element.getResource());
134                            if ((formatters != null)
135                                && (element.getFormatterId() != null)
136                                && formatters.isSearchContent(element.getFormatterId())) {
137                                // the content of this element must be included for the container page
138                                all.add(
139                                    CmsSolrDocumentXmlContent.extractXmlContent(
140                                        cms,
141                                        elementResource,
142                                        index,
143                                        forceLocale));
144                            }
145                        }
146                    } catch (CmsException e) {
147                        LOG.debug(
148                            Messages.get().getBundle().key(
149                                Messages.LOG_SKIPPING_CONTAINERPAGE_ELEMENT_WITH_UNREADABLE_RESOURCE_2,
150                                file.getRootPath(),
151                                element.getId()),
152                            e);
153                    }
154                }
155            }
156            // we have to overwrite the resource and content locales with the one from this container page
157            // TODO: Is this really the wanted behavior? It seems to be done like this before.
158            Map<String, String> fieldMappings = new HashMap<String, String>(1);
159            // Add to each container page the contents in all available locales,
160            // in case one containerpage is used in multiple languages.
161            List<Locale> localesAvailable = OpenCms.getLocaleManager().getAvailableLocales(cms, resource);
162            Map<Locale, LinkedHashMap<String, String>> multilingualValues = new HashMap<Locale, LinkedHashMap<String, String>>(
163                localesAvailable.size());
164            for (Locale localeAvailable : localesAvailable) {
165                multilingualValues.put(localeAvailable, new LinkedHashMap<String, String>());
166            }
167            Locale locale = forceLocale != null
168            ? forceLocale
169            : index.getLocaleForResource(cms, resource, containerPage.getLocales());
170            ex = new CmsExtractionResult(locale, multilingualValues, fieldMappings);
171            ex = ex.merge(all);
172            return ex;
173        } catch (Exception e) {
174            throw new CmsIndexException(
175                Messages.get().container(Messages.ERR_TEXT_EXTRACTION_1, resource.getRootPath()),
176                e);
177        }
178    }
179
180    /**
181     * @see org.opencms.search.documents.I_CmsDocumentFactory#isLocaleDependend()
182     */
183    @Override
184    public boolean isLocaleDependend() {
185
186        return true;
187    }
188
189    /**
190     * @see org.opencms.search.documents.I_CmsDocumentFactory#isUsingCache()
191     */
192    @Override
193    public boolean isUsingCache() {
194
195        return true;
196    }
197}