List of usage examples for org.apache.solr.common.params HighlightParams FIELD_MATCH
String FIELD_MATCH
To view the source code for org.apache.solr.common.params HighlightParams FIELD_MATCH.
Click Source Link
From source file:com.o19s.solr.swan.highlight.SwanHighlighter.java
License:Apache License
/** * Generates a list of Highlighted query fragments for each item in a list * of documents, or returns null if highlighting is disabled. * * @param docs query results/* www. j a v a2 s .co m*/ * @param query the query * @param req the current request * @param defaultFields default list of fields to summarize * * @return NamedList containing a NamedList for each document, which in * turns contains sets (field, summary) pairs. */ @Override @SuppressWarnings("unchecked") public NamedList<Object> doHighlighting(DocList docs, Query query, SolrQueryRequest req, String[] defaultFields) throws IOException { NamedList fragments = new SimpleOrderedMap(); SolrParams params = req.getParams(); if (!isHighlightingEnabled(params)) return null; SolrIndexSearcher searcher = req.getSearcher(); IndexSchema schema = searcher.getSchema(); String[] fieldNames = getHighlightFields(query, req, defaultFields); Set<String> fset = new HashSet<String>(); { // pre-fetch documents using the Searcher's doc cache Collections.addAll(fset, fieldNames); // fetch unique key if one exists. SchemaField keyField = schema.getUniqueKeyField(); if (null != keyField) fset.add(keyField.getName()); } //CHANGE start // int[] docIds = new int[docs.swordize()]; TreeSet<Integer> docIds = new TreeSet<Integer>(); DocIterator iterator = docs.iterator(); for (int i = 0; i < docs.size(); i++) { docIds.add(iterator.nextDoc()); } // Get Frag list builder String fragListBuilderString = params.get(HighlightParams.FRAG_LIST_BUILDER).toLowerCase(); FragListBuilder fragListBuilder; if (fragListBuilderString.equals("single")) { fragListBuilder = new SingleFragListBuilder(); } else { fragListBuilder = new com.o19s.solr.swan.highlight.SimpleFragListBuilder(); } // get FastVectorHighlighter instance out of the processing loop SpanAwareFastVectorHighlighter safvh = new SpanAwareFastVectorHighlighter( // FVH cannot process hl.usePhraseHighlighter parameter per-field basis params.getBool(HighlightParams.USE_PHRASE_HIGHLIGHTER, true), // FVH cannot process hl.requireFieldMatch parameter per-field basis params.getBool(HighlightParams.FIELD_MATCH, false), fragListBuilder, //new com.o19s.solr.swan.highlight.ScoreOrderFragmentsBuilder(), new WordHashFragmentsBuilder(), // List of docIds to filter spans docIds); safvh.setPhraseLimit(params.getInt(HighlightParams.PHRASE_LIMIT, Integer.MAX_VALUE)); SpanAwareFieldQuery fieldQuery = safvh.getFieldQuery(query, searcher.getIndexReader(), docIds); // Highlight each document for (int docId : docIds) { Document doc = searcher.doc(docId, fset); NamedList docSummaries = new SimpleOrderedMap(); for (String fieldName : fieldNames) { fieldName = fieldName.trim(); if (useFastVectorHighlighter(params, schema, fieldName)) doHighlightingByFastVectorHighlighter(safvh, fieldQuery, req, docSummaries, docId, doc, fieldName); else doHighlightingByHighlighter(query, req, docSummaries, docId, doc, fieldName); } String printId = schema.printableUniqueKey(doc); fragments.add(printId == null ? null : printId, docSummaries); } //CHANGE end return fragments; }
From source file:org.alfresco.solr.highlight.AlfrescoHighlighterTest.java
License:Open Source License
@Test public void highlightingRequiredFieldsTest() { logger.info("######### requireFieldMatch ###########"); SolrServletRequest req = areq(// w w w. j a va 2s .c o m params("q", "name:long", "qt", "/afts", "start", "0", "rows", "5", HighlightParams.HIGHLIGHT, "true", HighlightParams.Q, "long", HighlightParams.FIELDS, "name,title", HighlightParams.SIMPLE_PRE, "{", HighlightParams.SIMPLE_POST, "}"), "{\"locales\":[\"en\"], \"tenants\": [ \"\" ]}"); assertQ(req, "*[count(//lst[@name='highlighting']/lst)=2]", "*[count(//lst[@name='highlighting']/lst/arr[@name='title'])=1]", "//lst[@name='highlighting']/lst[1]/arr[@name='title']/str[.='title1 is very {long}']"); //add name req = areq(params("q", "name:long OR title:long", "qt", "/afts", "start", "0", "rows", "5", HighlightParams.HIGHLIGHT, "true", HighlightParams.Q, "title:long", HighlightParams.FIELDS, "name,title", HighlightParams.FIELD_MATCH, "true", HighlightParams.SIMPLE_PRE, "{", HighlightParams.SIMPLE_POST, "}"), "{\"locales\":[\"en\"], \"tenants\": [ \"\" ]}"); assertQ(req, "*[count(//lst[@name='highlighting']/lst)=2]", "*[count(//lst[@name='highlighting']/lst/arr[@name='title'])=1]", "*[count(//lst[@name='highlighting']/lst/arr[@name='name'])=0]", "//lst[@name='highlighting']/lst[1]/arr[@name='title']/str[.='title1 is very {long}']"); logger.info("######### MultiTerm ###########"); }
From source file:org.opencms.search.solr.CmsSolrIndex.java
License:Open Source License
/** * Performs the actual search.<p>/*from w ww. ja v a 2 s .c o m*/ * * @param cms the current OpenCms context * @param ignoreMaxRows <code>true</code> to return all all requested rows, <code>false</code> to use max rows * @param query the OpenCms Solr query * @param response the servlet response to write the query result to, may also be <code>null</code> * @param ignoreSearchExclude if set to false, only contents with search_exclude unset or "false" will be found - typical for the the non-gallery case * @param filter the resource filter to use * * @return the found documents * * @throws CmsSearchException if something goes wrong * * @see #search(CmsObject, CmsSolrQuery, boolean) */ @SuppressWarnings("unchecked") public CmsSolrResultList search(CmsObject cms, final CmsSolrQuery query, boolean ignoreMaxRows, ServletResponse response, boolean ignoreSearchExclude, CmsResourceFilter filter) throws CmsSearchException { // check if the user is allowed to access this index checkOfflineAccess(cms); if (!ignoreSearchExclude) { query.addFilterQuery(CmsSearchField.FIELD_SEARCH_EXCLUDE + ":\"false\""); } int previousPriority = Thread.currentThread().getPriority(); long startTime = System.currentTimeMillis(); // remember the initial query SolrQuery initQuery = query.clone(); query.setHighlight(false); LocalSolrQueryRequest solrQueryRequest = null; try { // initialize the search context CmsObject searchCms = OpenCms.initCmsObject(cms); // change thread priority in order to reduce search impact on overall system performance if (getPriority() > 0) { Thread.currentThread().setPriority(getPriority()); } // the lists storing the found documents that will be returned List<CmsSearchResource> resourceDocumentList = new ArrayList<CmsSearchResource>(); SolrDocumentList solrDocumentList = new SolrDocumentList(); // Initialize rows, offset, end and the current page. int rows = query.getRows() != null ? query.getRows().intValue() : CmsSolrQuery.DEFAULT_ROWS.intValue(); if (!ignoreMaxRows && (rows > ROWS_MAX)) { rows = ROWS_MAX; } int start = query.getStart() != null ? query.getStart().intValue() : 0; int end = start + rows; int page = 0; if (rows > 0) { page = Math.round(start / rows) + 1; } // set the start to '0' and expand the rows before performing the query query.setStart(new Integer(0)); query.setRows(new Integer((5 * rows * page) + start)); // perform the Solr query and remember the original Solr response QueryResponse queryResponse = m_solr.query(query); long solrTime = System.currentTimeMillis() - startTime; // initialize the counts long hitCount = queryResponse.getResults().getNumFound(); start = -1; end = -1; if ((rows > 0) && (page > 0) && (hitCount > 0)) { // calculate the final size of the search result start = rows * (page - 1); end = start + rows; // ensure that both i and n are inside the range of foundDocuments.size() start = new Long((start > hitCount) ? hitCount : start).intValue(); end = new Long((end > hitCount) ? hitCount : end).intValue(); } else { // return all found documents in the search result start = 0; end = new Long(hitCount).intValue(); } long visibleHitCount = hitCount; float maxScore = 0; // If we're using a postprocessor, (re-)initialize it before using it if (m_postProcessor != null) { m_postProcessor.init(); } // process found documents List<CmsSearchResource> allDocs = new ArrayList<CmsSearchResource>(); int cnt = 0; for (int i = 0; (i < queryResponse.getResults().size()) && (cnt < end); i++) { try { SolrDocument doc = queryResponse.getResults().get(i); CmsSolrDocument searchDoc = new CmsSolrDocument(doc); if (needsPermissionCheck(searchDoc)) { // only if the document is an OpenCms internal resource perform the permission check CmsResource resource = filter == null ? getResource(searchCms, searchDoc) : getResource(searchCms, searchDoc, filter); if (resource != null) { // permission check performed successfully: the user has read permissions! if (cnt >= start) { if (m_postProcessor != null) { doc = m_postProcessor.process(searchCms, resource, (SolrInputDocument) searchDoc.getDocument()); } resourceDocumentList.add(new CmsSearchResource(resource, searchDoc)); if (null != doc) { solrDocumentList.add(doc); } maxScore = maxScore < searchDoc.getScore() ? searchDoc.getScore() : maxScore; } allDocs.add(new CmsSearchResource(resource, searchDoc)); cnt++; } else { visibleHitCount--; } } else { // if permission check is not required for this index, // add a pseudo resource together with document to the results resourceDocumentList.add(new CmsSearchResource(PSEUDO_RES, searchDoc)); solrDocumentList.add(doc); maxScore = maxScore < searchDoc.getScore() ? searchDoc.getScore() : maxScore; cnt++; } } catch (Exception e) { // should not happen, but if it does we want to go on with the next result nevertheless LOG.warn(Messages.get().getBundle().key(Messages.LOG_SOLR_ERR_RESULT_ITERATION_FAILED_0), e); } } // the last documents were all secret so let's take the last found docs if (resourceDocumentList.isEmpty() && (allDocs.size() > 0)) { page = Math.round(allDocs.size() / rows) + 1; int showCount = allDocs.size() % rows; showCount = showCount == 0 ? rows : showCount; start = allDocs.size() - new Long(showCount).intValue(); end = allDocs.size(); if (allDocs.size() > start) { resourceDocumentList = allDocs.subList(start, end); for (CmsSearchResource r : resourceDocumentList) { maxScore = maxScore < r.getDocument().getScore() ? r.getDocument().getScore() : maxScore; solrDocumentList.add(((CmsSolrDocument) r.getDocument()).getSolrDocument()); } } } long processTime = System.currentTimeMillis() - startTime - solrTime; // create and return the result solrDocumentList.setStart(start); solrDocumentList.setMaxScore(new Float(maxScore)); solrDocumentList.setNumFound(visibleHitCount); queryResponse.getResponse().setVal(queryResponse.getResponse().indexOf(QUERY_RESPONSE_NAME, 0), solrDocumentList); queryResponse.getResponseHeader().setVal(queryResponse.getResponseHeader().indexOf(QUERY_TIME_NAME, 0), new Integer(new Long(System.currentTimeMillis() - startTime).intValue())); long highlightEndTime = System.currentTimeMillis(); SolrCore core = m_solr instanceof EmbeddedSolrServer ? ((EmbeddedSolrServer) m_solr).getCoreContainer().getCore(getCoreName()) : null; CmsSolrResultList result = null; try { SearchComponent highlightComponenet = null; if (core != null) { highlightComponenet = core.getSearchComponent("highlight"); solrQueryRequest = new LocalSolrQueryRequest(core, queryResponse.getResponseHeader()); } SolrQueryResponse solrQueryResponse = null; if (solrQueryRequest != null) { // create and initialize the solr response solrQueryResponse = new SolrQueryResponse(); solrQueryResponse.setAllValues(queryResponse.getResponse()); int paramsIndex = queryResponse.getResponseHeader().indexOf(HEADER_PARAMS_NAME, 0); NamedList<Object> header = null; Object o = queryResponse.getResponseHeader().getVal(paramsIndex); if (o instanceof NamedList) { header = (NamedList<Object>) o; header.setVal(header.indexOf(CommonParams.ROWS, 0), new Integer(rows)); header.setVal(header.indexOf(CommonParams.START, 0), new Long(start)); } // set the OpenCms Solr query as parameters to the request solrQueryRequest.setParams(initQuery); // perform the highlighting if ((header != null) && (initQuery.getHighlight()) && (highlightComponenet != null)) { header.add(HighlightParams.HIGHLIGHT, "on"); if ((initQuery.getHighlightFields() != null) && (initQuery.getHighlightFields().length > 0)) { header.add(HighlightParams.FIELDS, CmsStringUtil.arrayAsString(initQuery.getHighlightFields(), ",")); } String formatter = initQuery.getParams(HighlightParams.FORMATTER) != null ? initQuery.getParams(HighlightParams.FORMATTER)[0] : null; if (formatter != null) { header.add(HighlightParams.FORMATTER, formatter); } if (initQuery.getHighlightFragsize() != 100) { header.add(HighlightParams.FRAGSIZE, new Integer(initQuery.getHighlightFragsize())); } if (initQuery.getHighlightRequireFieldMatch()) { header.add(HighlightParams.FIELD_MATCH, new Boolean(initQuery.getHighlightRequireFieldMatch())); } if (CmsStringUtil.isNotEmptyOrWhitespaceOnly(initQuery.getHighlightSimplePost())) { header.add(HighlightParams.SIMPLE_POST, initQuery.getHighlightSimplePost()); } if (CmsStringUtil.isNotEmptyOrWhitespaceOnly(initQuery.getHighlightSimplePre())) { header.add(HighlightParams.SIMPLE_PRE, initQuery.getHighlightSimplePre()); } if (initQuery.getHighlightSnippets() != 1) { header.add(HighlightParams.SNIPPETS, new Integer(initQuery.getHighlightSnippets())); } ResponseBuilder rb = new ResponseBuilder(solrQueryRequest, solrQueryResponse, Collections.singletonList(highlightComponenet)); try { rb.doHighlights = true; DocListAndSet res = new DocListAndSet(); SchemaField idField = OpenCms.getSearchManager().getSolrServerConfiguration() .getSolrSchema().getUniqueKeyField(); int[] luceneIds = new int[rows]; int docs = 0; for (SolrDocument doc : solrDocumentList) { String idString = (String) doc.getFirstValue(CmsSearchField.FIELD_ID); int id = solrQueryRequest.getSearcher().getFirstMatch( new Term(idField.getName(), idField.getType().toInternal(idString))); luceneIds[docs++] = id; } res.docList = new DocSlice(0, docs, luceneIds, null, docs, 0); rb.setResults(res); rb.setQuery(QParser.getParser(initQuery.getQuery(), null, solrQueryRequest).getQuery()); rb.setQueryString(initQuery.getQuery()); highlightComponenet.prepare(rb); highlightComponenet.process(rb); highlightComponenet.finishStage(rb); } catch (Exception e) { LOG.error(e.getMessage() + " in query: " + initQuery, new Exception(e)); } // Make highlighting also available via the CmsSolrResultList queryResponse.setResponse(solrQueryResponse.getValues()); highlightEndTime = System.currentTimeMillis(); } } result = new CmsSolrResultList(initQuery, queryResponse, solrDocumentList, resourceDocumentList, start, new Integer(rows), end, page, visibleHitCount, new Float(maxScore), startTime, highlightEndTime); if (LOG.isDebugEnabled()) { Object[] logParams = new Object[] { new Long(System.currentTimeMillis() - startTime), new Long(result.getNumFound()), new Long(solrTime), new Long(processTime), new Long(result.getHighlightEndTime() != 0 ? result.getHighlightEndTime() - startTime : 0) }; LOG.debug(query.toString() + "\n" + Messages.get().getBundle().key(Messages.LOG_SOLR_SEARCH_EXECUTED_5, logParams)); } if (response != null) { writeResp(response, solrQueryRequest, solrQueryResponse); } } finally { if (solrQueryRequest != null) { solrQueryRequest.close(); } if (core != null) { core.close(); } } return result; } catch (Exception e) { throw new CmsSearchException(Messages.get().container(Messages.LOG_SOLR_ERR_SEARCH_EXECUTION_FAILD_1, CmsEncoder.decode(query.toString()), e), e); } finally { if (solrQueryRequest != null) { solrQueryRequest.close(); } // re-set thread to previous priority Thread.currentThread().setPriority(previousPriority); } }