List of usage examples for org.apache.solr.client.solrj SolrQuery addHighlightField
public SolrQuery addHighlightField(String f)
From source file:org.dspace.discovery.SolrServiceImpl.java
License:BSD License
protected SolrQuery resolveToSolrQuery(Context context, DiscoverQuery discoveryQuery, boolean includeUnDiscoverable) { SolrQuery solrQuery = new SolrQuery(); String query = "*:*"; if (discoveryQuery.getQuery() != null) { query = discoveryQuery.getQuery(); }/*from w w w . j a v a 2 s . co m*/ solrQuery.setQuery(query); if (discoveryQuery.isSpellCheck()) { solrQuery.setParam(SpellingParams.SPELLCHECK_Q, query); solrQuery.setParam(SpellingParams.SPELLCHECK_COLLATE, Boolean.TRUE); solrQuery.setParam("spellcheck", Boolean.TRUE); } if (!includeUnDiscoverable) { solrQuery.addFilterQuery("NOT(withdrawn:true)"); solrQuery.addFilterQuery("NOT(discoverable:false)"); } for (int i = 0; i < discoveryQuery.getFilterQueries().size(); i++) { String filterQuery = discoveryQuery.getFilterQueries().get(i); solrQuery.addFilterQuery(filterQuery); } if (discoveryQuery.getDSpaceObjectFilter() != -1) { solrQuery.addFilterQuery("search.resourcetype:" + discoveryQuery.getDSpaceObjectFilter()); } for (int i = 0; i < discoveryQuery.getFieldPresentQueries().size(); i++) { String filterQuery = discoveryQuery.getFieldPresentQueries().get(i); solrQuery.addFilterQuery(filterQuery + ":[* TO *]"); } if (discoveryQuery.getStart() != -1) { solrQuery.setStart(discoveryQuery.getStart()); } if (discoveryQuery.getMaxResults() != -1) { solrQuery.setRows(discoveryQuery.getMaxResults()); } if (discoveryQuery.getSortField() != null) { SolrQuery.ORDER order = SolrQuery.ORDER.asc; if (discoveryQuery.getSortOrder().equals(DiscoverQuery.SORT_ORDER.desc)) order = SolrQuery.ORDER.desc; solrQuery.addSortField(discoveryQuery.getSortField(), order); } for (String property : discoveryQuery.getProperties().keySet()) { List<String> values = discoveryQuery.getProperties().get(property); solrQuery.add(property, values.toArray(new String[values.size()])); } List<DiscoverFacetField> facetFields = discoveryQuery.getFacetFields(); if (0 < facetFields.size()) { //Only add facet information if there are any facets for (DiscoverFacetField facetFieldConfig : facetFields) { String field = transformFacetField(facetFieldConfig, facetFieldConfig.getField(), false); solrQuery.addFacetField(field); // Setting the facet limit in this fashion ensures that each facet can have its own max solrQuery.add("f." + field + "." + FacetParams.FACET_LIMIT, String.valueOf(facetFieldConfig.getLimit())); String facetSort; if (DiscoveryConfigurationParameters.SORT.COUNT.equals(facetFieldConfig.getSortOrder())) { facetSort = FacetParams.FACET_SORT_COUNT; } else { facetSort = FacetParams.FACET_SORT_INDEX; } solrQuery.add("f." + field + "." + FacetParams.FACET_SORT, facetSort); if (facetFieldConfig.getOffset() != -1) { solrQuery.setParam("f." + field + "." + FacetParams.FACET_OFFSET, String.valueOf(facetFieldConfig.getOffset())); } if (facetFieldConfig.getPrefix() != null) { solrQuery.setFacetPrefix(field, facetFieldConfig.getPrefix()); } } List<String> facetQueries = discoveryQuery.getFacetQueries(); for (String facetQuery : facetQueries) { solrQuery.addFacetQuery(facetQuery); } if (discoveryQuery.getFacetMinCount() != -1) { solrQuery.setFacetMinCount(discoveryQuery.getFacetMinCount()); } solrQuery.setParam(FacetParams.FACET_OFFSET, String.valueOf(discoveryQuery.getFacetOffset())); } if (0 < discoveryQuery.getHitHighlightingFields().size()) { solrQuery.setHighlight(true); solrQuery.add(HighlightParams.USE_PHRASE_HIGHLIGHTER, Boolean.TRUE.toString()); for (DiscoverHitHighlightingField highlightingField : discoveryQuery.getHitHighlightingFields()) { solrQuery.addHighlightField(highlightingField.getField() + "_hl"); solrQuery.add("f." + highlightingField.getField() + "_hl." + HighlightParams.FRAGSIZE, String.valueOf(highlightingField.getMaxChars())); solrQuery.add("f." + highlightingField.getField() + "_hl." + HighlightParams.SNIPPETS, String.valueOf(highlightingField.getMaxSnippets())); } } //Add any configured search plugins ! List<SolrServiceSearchPlugin> solrServiceSearchPlugins = new DSpace().getServiceManager() .getServicesByType(SolrServiceSearchPlugin.class); for (SolrServiceSearchPlugin searchPlugin : solrServiceSearchPlugins) { searchPlugin.additionalSearchParameters(context, discoveryQuery, solrQuery); } return solrQuery; }
From source file:org.eclipse.rdf4j.sail.solr.SolrIndex.java
License:Open Source License
/** * Parse the passed query./*from w w w .j a v a2 s . co m*/ * * @param query * string * @return the parsed query * @throws ParseException * when the parsing brakes */ @Override protected Iterable<? extends DocumentScore> query(Resource subject, String query, URI propertyURI, boolean highlight) throws MalformedQueryException, IOException { SolrQuery q = prepareQuery(propertyURI, new SolrQuery(query)); if (highlight) { q.setHighlight(true); String field = (propertyURI != null) ? SearchFields.getPropertyField(propertyURI) : "*"; q.addHighlightField(field); q.setHighlightSimplePre(SearchFields.HIGHLIGHTER_PRE_TAG); q.setHighlightSimplePost(SearchFields.HIGHLIGHTER_POST_TAG); q.setHighlightSnippets(2); } QueryResponse response; if (q.getHighlight()) { q.addField("*"); } else { q.addField(SearchFields.URI_FIELD_NAME); } q.addField("score"); try { if (subject != null) { response = search(subject, q); } else { response = search(q); } } catch (SolrServerException e) { throw new IOException(e); } SolrDocumentList results = response.getResults(); final Map<String, Map<String, List<String>>> highlighting = response.getHighlighting(); return Iterables.transform(results, new Function<SolrDocument, DocumentScore>() { @Override public DocumentScore apply(SolrDocument document) { SolrSearchDocument doc = new SolrSearchDocument(document); Map<String, List<String>> docHighlighting = (highlighting != null) ? highlighting.get(doc.getId()) : null; return new SolrDocumentScore(doc, docHighlighting); } }); }
From source file:org.gbif.common.search.builder.SolrQueryBuilder.java
License:Apache License
/** * Helper method that sets the highlighting parameters. * //from w w w .jav a2s . c om * @param searchRequest the searchRequest used to extract the parameters * @param solrQuery this object is modified by adding the facets parameters */ private void setHighLightParams(SearchRequest<P> searchRequest, SolrQuery solrQuery) { solrQuery.setHighlight(searchRequest.isHighlight()); solrQuery.setHighlightSnippets(NUM_HL_SNIPPETS); solrQuery.setHighlightFragsize(HL_FRAGMENT_SIZE); if (searchRequest.isHighlight()) { for (String hlField : queryBuilder.getHighlightedFields()) { solrQuery.addHighlightField(hlField); } } }
From source file:org.ofbiz.solr.SolrProductSearch.java
License:Apache License
/** * Runs a query on the Solr Search Engine and returns the results. * <p>/* w w w . jav a 2s. c om*/ * This function only returns an object of type QueryResponse, so it is probably not a good idea to call it directly from within the * groovy files (As a decent example on how to use it, however, use keywordSearch instead). */ public static Map<String, Object> runSolrQuery(DispatchContext dctx, Map<String, Object> context) { // get Connection HttpSolrServer server = null; Map<String, Object> result; Integer viewIndex = (Integer) context.get("viewIndex"); Integer viewSize = (Integer) context.get("viewSize"); if (viewIndex < 1) { viewIndex = 1; } try { server = new HttpSolrServer(SolrUtil.solrUrl); // create Query Object SolrQuery solrQuery = new SolrQuery(); solrQuery.setQuery((String) context.get("query")); // solrQuery.setQueryType("dismax"); boolean faceted = (Boolean) context.get("facet"); if (faceted) { solrQuery.setFacet(faceted); //solrQuery.addFacetField("manu"); solrQuery.addFacetField("features"); solrQuery.addFacetField("cat"); solrQuery.setFacetMinCount(1); solrQuery.setFacetLimit(8); solrQuery.addFacetQuery("listPrice:[0 TO 50]"); solrQuery.addFacetQuery("listPrice:[50 TO 100]"); solrQuery.addFacetQuery("listPrice:[100 TO 250]"); solrQuery.addFacetQuery("listPrice:[250 TO 500]"); solrQuery.addFacetQuery("listPrice:[500 TO 1000]"); solrQuery.addFacetQuery("listPrice:[1000 TO 2500]"); solrQuery.addFacetQuery("listPrice:[2500 TO 5000]"); solrQuery.addFacetQuery("listPrice:[5000 TO 10000]"); solrQuery.addFacetQuery("listPrice:[10000 TO 50000]"); solrQuery.addFacetQuery("listPrice:[50000 TO *]"); } boolean spellCheck = (Boolean) context.get("spellcheck"); if (spellCheck) { solrQuery.setParam("spellcheck", spellCheck); } boolean highLight = (Boolean) context.get("highlight"); if (highLight) { solrQuery.setHighlight(highLight); solrQuery.setHighlightSimplePre("<span class=\"highlight\">"); solrQuery.addHighlightField("description"); solrQuery.setHighlightSimplePost("</span>"); solrQuery.setHighlightSnippets(2); } // Set additional Parameter // SolrQuery.ORDER order = SolrQuery.ORDER.desc; if (viewIndex != null && viewIndex > 0) { solrQuery.setStart((viewIndex - 1) * viewSize); } if (viewSize != null && viewSize > 0) { solrQuery.setRows(viewSize); } // if ((List) context.get("queryFilter") != null && ((ArrayList<SolrDocument>) context.get("queryFilter")).size() > 0) { // List filter = (List) context.get("queryFilter"); // String[] tn = new String[filter.size()]; // Iterator it = filter.iterator(); // for (int i = 0; i < filter.size(); i++) { // tn[i] = (String) filter.get(i); // } // solrQuery.setFilterQueries(tn); // } String queryFilter = (String) context.get("queryFilter"); if (UtilValidate.isNotEmpty(queryFilter)) solrQuery.setFilterQueries(queryFilter.split(" ")); if ((String) context.get("returnFields") != null) { solrQuery.setFields((String) context.get("returnFields")); } // if((Boolean)context.get("sortByReverse"))order.reverse(); if ((String) context.get("sortBy") != null && ((String) context.get("sortBy")).length() > 0) { SolrQuery.ORDER order; if (!((Boolean) context.get("sortByReverse"))) order = SolrQuery.ORDER.asc; else order = SolrQuery.ORDER.desc; solrQuery.setSort(((String) context.get("sortBy")).replaceFirst("-", ""), order); } if ((String) context.get("facetQuery") != null) { solrQuery.addFacetQuery((String) context.get("facetQuery")); } QueryResponse rsp = server.query(solrQuery); result = ServiceUtil.returnSuccess(); result.put("queryResult", rsp); } catch (Exception e) { Debug.logError(e, e.getMessage(), module); result = ServiceUtil.returnError(e.toString()); } return result; }
From source file:org.segrada.search.solr.SolrSearchEngine.java
License:Apache License
@Override public PaginationInfo<SearchHit> search(String searchTerm, Map<String, String> filters) { // to avoid NPEs if (filters == null) filters = new HashMap<>(); // set defaults int page = 1; int entriesPerPage = 20; try {/*ww w .jav a2s . com*/ // Parse a simple query that searches for "text": MultiFieldQueryParser parser; String[] containFields; // do we have a filter to contain to certain fields? if (filters.containsKey("fields")) { String fields = filters.get("fields"); if (fields.isEmpty()) containFields = new String[] { this.title, this.subTitles, this.content }; else if (fields.equalsIgnoreCase(this.title)) containFields = new String[] { this.title }; else if (fields.equalsIgnoreCase(this.subTitles)) containFields = new String[] { this.subTitles }; else if (fields.equalsIgnoreCase(this.content)) containFields = new String[] { this.content }; else if (fields.equalsIgnoreCase("allTitles")) containFields = new String[] { this.title, this.subTitles }; else throw new RuntimeException("fields-Filter " + fields + " is not known."); } else containFields = new String[] { this.title, this.subTitles, this.content }; parser = new MultiFieldQueryParser(Version.LUCENE_47, containFields, analyzer); // which operator do we use? parser.setDefaultOperator(QueryParser.Operator.AND); if (filters.containsKey("operator")) { String operator = filters.get("operator"); if (operator.equalsIgnoreCase("or")) parser.setDefaultOperator(QueryParser.Operator.OR); else if (!operator.isEmpty() && !operator.equalsIgnoreCase("and")) throw new RuntimeException("operator-Filter " + operator + " is not and/or."); } // filters for query SolrQuery query = new SolrQuery(); // class filter if (filters.containsKey("class") && !filters.get("class").isEmpty()) { // multiple classes? String[] classes = filters.get("class").split(","); // single class if (classes.length <= 1) { query.addFilterQuery(this.className, filters.get("class")); } else { // multiple classes StringBuilder chained = new StringBuilder("("); for (int i = 0; i < classes.length; i++) { if (i > 0) chained.append(" OR "); chained.append("className:").append(classes[i].trim()); } query.addFilterQuery(this.className, chained + ")"); } } // tag filter if (filters.containsKey("tags") && !filters.get("tags").isEmpty()) { // split tags into array String[] tags = filters.get("tags").split(","); BooleanQuery booleanQuery = new BooleanQuery(); for (String tagLocal : tags) { booleanQuery.add(new TermQuery(new Term("tag", tagLocal.trim())), BooleanClause.Occur.SHOULD); } query.addFilterQuery(this.tag, booleanQuery.toString()); } // define query Query queryTerm = null; if (searchTerm != null) queryTerm = parser.parse(searchTerm); if (queryTerm == null) queryTerm = new MatchAllDocsQuery(); // fallback to match all documents query.setQuery(queryTerm.toString()); // get hits per page if (filters.containsKey("limit")) { try { entriesPerPage = Integer.valueOf(filters.get("limit")); if (entriesPerPage <= 0 || entriesPerPage > 1000) entriesPerPage = 20; } catch (NumberFormatException e) { logger.warn("Could not parse limit " + filters.get("limit") + " to integer", e); } } // get page number if (filters.containsKey("page")) { try { page = Integer.valueOf(filters.get("page")); } catch (NumberFormatException e) { logger.warn("Could not parse page " + filters.get("page") + " to integer", e); } } // calculate start index int startIndex = (page - 1) * entriesPerPage; query.setStart(startIndex); query.setRows(entriesPerPage); query.setFields("*", "score"); // define highlighting query.setHighlight(true); query.addHighlightField(this.content); query.setHighlightFragsize(18); query.setHighlightSnippets(10); query.setHighlightSimplePre("<b>"); query.setHighlightSimplePost("</b>"); // do query QueryResponse response = solr.query(query); SolrDocumentList results = response.getResults(); // how many pages do we have? int pages = (int) (results.getNumFound() / entriesPerPage + 1); // cycle trough hits List<SearchHit> hits = new ArrayList<>(); for (SolrDocument doc : results) { SearchHit searchHit = createHitFromDocument(doc); // add score Object score = doc.get("score"); if (score != null && score instanceof Float) searchHit.setRelevance((float) score); // get highlighted components if (searchTerm != null && response.getHighlighting().get(searchHit.getId()) != null) { List<String> fragments = response.getHighlighting().get(searchHit.getId()).get(this.content); if (fragments != null) { String[] bestFragments = new String[fragments.size() > 10 ? 10 : fragments.size()]; for (int i = 0; i < bestFragments.length; i++) bestFragments[i] = fragments.get(i); searchHit.setHighlightText(bestFragments); } } // add hit hits.add(searchHit); } // return pagination info return new PaginationInfo<>(page, pages, (int) results.getNumFound(), entriesPerPage, hits); } catch (Throwable e) { logger.error("Error in search.", e); } // return empty list result in order to avoid NPEs return new PaginationInfo<>(page, 1, 0, entriesPerPage, new ArrayList<>()); }
From source file:org.segrada.search.solr.SolrSearchEngine.java
License:Apache License
@Override public String[] searchInDocument(String searchTerm, String id) { // sanity check if (searchTerm == null || id == null || searchTerm.isEmpty() || id.isEmpty()) return new String[] {}; try {/*w w w. j a va2s . co m*/ // only search content MultiFieldQueryParser parser = new MultiFieldQueryParser(Version.LUCENE_47, new String[] { "content" }, analyzer); SolrQuery query = new SolrQuery(); // set operator and contain by id parser.setDefaultOperator(QueryParser.Operator.AND); query.setQuery(parser.parse(searchTerm).toString()); // filter by id query.addFilterQuery("id:" + id); query.setRows(1); // define highlighting query.setHighlight(true); query.addHighlightField(this.content); query.setHighlightFragsize(100); query.setHighlightSnippets(100); query.setHighlightSimplePre("<b>"); query.setHighlightSimplePost("</b>"); // do query QueryResponse response = solr.query(query); SolrDocumentList results = response.getResults(); if (!results.isEmpty() && response.getHighlighting().get(id) != null) { List<String> fragments = response.getHighlighting().get(id).get(this.content); String[] bestFragments = new String[fragments.size() > 100 ? 100 : fragments.size()]; for (int i = 0; i < bestFragments.length; i++) bestFragments[i] = fragments.get(i); return bestFragments; } } catch (Throwable e) { logger.error("Error in search.", e); } return new String[] {}; }
From source file:org.sleuthkit.autopsy.keywordsearch.AccountsText.java
License:Open Source License
@Override
@NbBundle.Messages({ "AccountsText.getMarkup.noMatchMsg="
+ "<html><pre><span style\\\\='background\\\\:yellow'>There were no keyword hits on this page. <br />"
+ "The keyword could have been in the file name."
+ " <br />Advance to another page if present, or to view the original text, choose File Text"
+ " <br />in the drop down menu to the right...</span></pre></html>",
"AccountsText.getMarkup.queryFailedMsg="
+ "<html><pre><span style\\\\='background\\\\:yellow'>Failed to retrieve keyword hit results."
+ " <br />Confirm that Autopsy can connect to the Solr server. "
+ "<br /></span></pre></html>" })
public String getText() {
loadPageInfo(); //inits once
SolrQuery q = new SolrQuery();
q.setShowDebugInfo(DEBUG); //debug
q.addHighlightField(HIGHLIGHT_FIELD);
q.setQuery(queryString);/*from w w w . j a va 2 s. com*/
//set the documentID filter
String queryDocumentID = this.solrObjectId + Server.CHUNK_ID_SEPARATOR + this.currentPage;
q.addFilterQuery(Server.Schema.ID.toString() + ":" + queryDocumentID);
//configure the highlighter
q.setParam("hl.useFastVectorHighlighter", "true"); //fast highlighter scales better than standard one NON-NLS
q.setParam("hl.tag.pre", HIGHLIGHT_PRE); //makes sense for FastVectorHighlighter only NON-NLS
q.setParam("hl.tag.post", HIGHLIGHT_POST); //makes sense for FastVectorHighlighter only NON-NLS
q.setParam("hl.fragListBuilder", "single"); //makes sense for FastVectorHighlighter only NON-NLS
q.setParam("hl.maxAnalyzedChars", Server.HL_ANALYZE_CHARS_UNLIMITED); //docs says makes sense for the original Highlighter only, but not really //NON-NLS
try {
//extract highlighting and bail early on null responses
Map<String, Map<String, List<String>>> highlightingPerDocument = solrServer.query(q, METHOD.POST)
.getHighlighting();
Map<String, List<String>> highlightingPerField = highlightingPerDocument.get(queryDocumentID);
if (highlightingPerField == null) {
return Bundle.AccountsText_getMarkup_noMatchMsg();
}
List<String> highlights = highlightingPerField.get(HIGHLIGHT_FIELD);
if (highlights == null) {
return Bundle.AccountsText_getMarkup_noMatchMsg();
}
//There should only be one item
String highlighting = highlights.get(0).trim();
/*
* use regex matcher to iterate over occurences of HIGHLIGHT_PRE,
* and prepend them with an anchor tag.
*/
Matcher m = ANCHOR_DETECTION_PATTERN.matcher(highlighting);
StringBuffer sb = new StringBuffer(highlighting.length());
int count = 0;
while (m.find()) {
count++;
m.appendReplacement(sb, INSERT_PREFIX + count + INSERT_POSTFIX);
}
m.appendTail(sb);
//store total hits for this page, now that we know it
this.numberOfHitsPerPage.put(this.currentPage, count);
if (this.currentItem() == 0 && this.hasNextItem()) {
this.nextItem();
}
// extracted content (minus highlight tags) is HTML-escaped
return "<html><pre>" + sb.toString() + "</pre></html>"; //NON-NLS
} catch (Exception ex) {
LOGGER.log(Level.WARNING, "Error executing Solr highlighting query: " + keywords, ex); //NON-NLS
return Bundle.AccountsText_getMarkup_queryFailedMsg();
}
}
From source file:org.sleuthkit.autopsy.keywordsearch.HighlightedMatchesSource.java
License:Open Source License
@Override public String getMarkup() { init(); //inits once String highLightField = null; String highlightQuery = keywordHitQuery; if (isRegex) { highLightField = LuceneQuery.HIGHLIGHT_FIELD_REGEX; //escape special lucene chars if not already escaped (if not a compound query) //TODO a better way to mark it a compound highlight query final String findSubstr = LuceneQuery.HIGHLIGHT_FIELD_REGEX + ":"; if (!highlightQuery.contains(findSubstr)) { highlightQuery = KeywordSearchUtil.escapeLuceneQuery(highlightQuery); }/*from www . ja v a2s . c o m*/ } else { highLightField = LuceneQuery.HIGHLIGHT_FIELD_LITERAL; //escape special lucene chars always for literal queries query highlightQuery = KeywordSearchUtil.escapeLuceneQuery(highlightQuery); } SolrQuery q = new SolrQuery(); q.setShowDebugInfo(DEBUG); //debug String queryStr = null; if (isRegex) { StringBuilder sb = new StringBuilder(); sb.append(highLightField).append(":"); if (group) { sb.append("\""); } sb.append(highlightQuery); if (group) { sb.append("\""); } queryStr = sb.toString(); } else { //use default field, simplifies query //always force grouping/quotes queryStr = KeywordSearchUtil.quoteQuery(highlightQuery); } q.setQuery(queryStr); final long contentId = content.getId(); String contentIdStr = Long.toString(contentId); if (hasChunks) { contentIdStr += "_" + Integer.toString(this.currentPage); } final String filterQuery = Server.Schema.ID.toString() + ":" + contentIdStr; q.addFilterQuery(filterQuery); q.addHighlightField(highLightField); //for exact highlighting, try content_ws field (with stored="true" in Solr schema) //q.setHighlightSimplePre(HIGHLIGHT_PRE); //original highlighter only //q.setHighlightSimplePost(HIGHLIGHT_POST); //original highlighter only q.setHighlightFragsize(0); // don't fragment the highlight, works with original highlighter, or needs "single" list builder with FVH //tune the highlighter q.setParam("hl.useFastVectorHighlighter", "on"); //fast highlighter scales better than standard one q.setParam("hl.tag.pre", HIGHLIGHT_PRE); //makes sense for FastVectorHighlighter only q.setParam("hl.tag.post", HIGHLIGHT_POST); //makes sense for FastVectorHighlighter only q.setParam("hl.fragListBuilder", "single"); //makes sense for FastVectorHighlighter only //docs says makes sense for the original Highlighter only, but not really q.setParam("hl.maxAnalyzedChars", Server.HL_ANALYZE_CHARS_UNLIMITED); try { QueryResponse response = solrServer.query(q, METHOD.POST); Map<String, Map<String, List<String>>> responseHighlight = response.getHighlighting(); Map<String, List<String>> responseHighlightID = responseHighlight.get(contentIdStr); if (responseHighlightID == null) { return NO_MATCHES; } List<String> contentHighlights = responseHighlightID.get(highLightField); if (contentHighlights == null) { return NO_MATCHES; } else { // extracted content (minus highlight tags) is HTML-escaped String highlightedContent = contentHighlights.get(0).trim(); highlightedContent = insertAnchors(highlightedContent); return "<html><pre>" + highlightedContent + "</pre></html>"; } } catch (NoOpenCoreException ex) { logger.log(Level.WARNING, "Couldn't query markup for page: " + currentPage, ex); return ""; } catch (KeywordSearchModuleException ex) { logger.log(Level.WARNING, "Could not query markup for page: " + currentPage, ex); return ""; } }
From source file:org.sleuthkit.autopsy.keywordsearch.HighlightedText.java
License:Open Source License
@Override public String getText() { loadPageInfo(); //inits once String highLightField = null; if (isRegex) { highLightField = LuceneQuery.HIGHLIGHT_FIELD_REGEX; } else {// w w w . ja va 2 s . c o m highLightField = LuceneQuery.HIGHLIGHT_FIELD_LITERAL; } SolrQuery q = new SolrQuery(); q.setShowDebugInfo(DEBUG); //debug // input query has already been properly constructed and escaped q.setQuery(keywordHitQuery); String contentIdStr = Long.toString(this.objectId); if (hasChunks) { contentIdStr += "_" + Integer.toString(this.currentPage); } final String filterQuery = Server.Schema.ID.toString() + ":" + KeywordSearchUtil.escapeLuceneQuery(contentIdStr); q.addFilterQuery(filterQuery); q.addHighlightField(highLightField); //for exact highlighting, try content_ws field (with stored="true" in Solr schema) //q.setHighlightSimplePre(HIGHLIGHT_PRE); //original highlighter only //q.setHighlightSimplePost(HIGHLIGHT_POST); //original highlighter only q.setHighlightFragsize(0); // don't fragment the highlight, works with original highlighter, or needs "single" list builder with FVH //tune the highlighter q.setParam("hl.useFastVectorHighlighter", "on"); //fast highlighter scales better than standard one NON-NLS q.setParam("hl.tag.pre", HIGHLIGHT_PRE); //makes sense for FastVectorHighlighter only NON-NLS q.setParam("hl.tag.post", HIGHLIGHT_POST); //makes sense for FastVectorHighlighter only NON-NLS q.setParam("hl.fragListBuilder", "single"); //makes sense for FastVectorHighlighter only NON-NLS //docs says makes sense for the original Highlighter only, but not really q.setParam("hl.maxAnalyzedChars", Server.HL_ANALYZE_CHARS_UNLIMITED); //NON-NLS try { QueryResponse response = solrServer.query(q, METHOD.POST); Map<String, Map<String, List<String>>> responseHighlight = response.getHighlighting(); Map<String, List<String>> responseHighlightID = responseHighlight.get(contentIdStr); if (responseHighlightID == null) { return NbBundle.getMessage(this.getClass(), "HighlightedMatchesSource.getMarkup.noMatchMsg"); } List<String> contentHighlights = responseHighlightID.get(highLightField); if (contentHighlights == null) { return NbBundle.getMessage(this.getClass(), "HighlightedMatchesSource.getMarkup.noMatchMsg"); } else { // extracted content (minus highlight tags) is HTML-escaped String highlightedContent = contentHighlights.get(0).trim(); highlightedContent = insertAnchors(highlightedContent); return "<html><pre>" + highlightedContent + "</pre></html>"; //NON-NLS } } catch (Exception ex) { logger.log(Level.WARNING, "Error executing Solr highlighting query: " + keywordHitQuery, ex); //NON-NLS return NbBundle.getMessage(this.getClass(), "HighlightedMatchesSource.getMarkup.queryFailedMsg"); } }
From source file:org.sleuthkit.autopsy.keywordsearch.HighlightedTextMarkup.java
License:Open Source License
@Override public String getMarkup() { loadPageInfo(); //inits once String highLightField = null; String highlightQuery = keywordHitQuery; if (isRegex) { highLightField = LuceneQuery.HIGHLIGHT_FIELD_REGEX; //escape special lucene chars if not already escaped (if not a compound query) //TODO a better way to mark it a compound highlight query final String findSubstr = LuceneQuery.HIGHLIGHT_FIELD_REGEX + ":"; if (!highlightQuery.contains(findSubstr)) { highlightQuery = KeywordSearchUtil.escapeLuceneQuery(highlightQuery); }//from w w w . jav a 2s. com } else { highLightField = LuceneQuery.HIGHLIGHT_FIELD_LITERAL; //escape special lucene chars always for literal queries query highlightQuery = KeywordSearchUtil.escapeLuceneQuery(highlightQuery); } SolrQuery q = new SolrQuery(); q.setShowDebugInfo(DEBUG); //debug String queryStr = null; if (isRegex) { StringBuilder sb = new StringBuilder(); sb.append(highLightField).append(":"); if (group) { sb.append("\""); } sb.append(highlightQuery); if (group) { sb.append("\""); } queryStr = sb.toString(); } else { //use default field, simplifies query //always force grouping/quotes queryStr = KeywordSearchUtil.quoteQuery(highlightQuery); } q.setQuery(queryStr); String contentIdStr = Long.toString(this.objectId); if (hasChunks) { contentIdStr += "_" + Integer.toString(this.currentPage); } final String filterQuery = Server.Schema.ID.toString() + ":" + KeywordSearchUtil.escapeLuceneQuery(contentIdStr); q.addFilterQuery(filterQuery); q.addHighlightField(highLightField); //for exact highlighting, try content_ws field (with stored="true" in Solr schema) //q.setHighlightSimplePre(HIGHLIGHT_PRE); //original highlighter only //q.setHighlightSimplePost(HIGHLIGHT_POST); //original highlighter only q.setHighlightFragsize(0); // don't fragment the highlight, works with original highlighter, or needs "single" list builder with FVH //tune the highlighter q.setParam("hl.useFastVectorHighlighter", "on"); //fast highlighter scales better than standard one NON-NLS q.setParam("hl.tag.pre", HIGHLIGHT_PRE); //makes sense for FastVectorHighlighter only NON-NLS q.setParam("hl.tag.post", HIGHLIGHT_POST); //makes sense for FastVectorHighlighter only NON-NLS q.setParam("hl.fragListBuilder", "single"); //makes sense for FastVectorHighlighter only NON-NLS //docs says makes sense for the original Highlighter only, but not really q.setParam("hl.maxAnalyzedChars", Server.HL_ANALYZE_CHARS_UNLIMITED); //NON-NLS try { QueryResponse response = solrServer.query(q, METHOD.POST); Map<String, Map<String, List<String>>> responseHighlight = response.getHighlighting(); Map<String, List<String>> responseHighlightID = responseHighlight.get(contentIdStr); if (responseHighlightID == null) { return NbBundle.getMessage(this.getClass(), "HighlightedMatchesSource.getMarkup.noMatchMsg"); } List<String> contentHighlights = responseHighlightID.get(highLightField); if (contentHighlights == null) { return NbBundle.getMessage(this.getClass(), "HighlightedMatchesSource.getMarkup.noMatchMsg"); } else { // extracted content (minus highlight tags) is HTML-escaped String highlightedContent = contentHighlights.get(0).trim(); highlightedContent = insertAnchors(highlightedContent); return "<html><pre>" + highlightedContent + "</pre></html>"; //NON-NLS } } catch (NoOpenCoreException ex) { logger.log(Level.WARNING, "Couldn't query markup for page: " + currentPage, ex); //NON-NLS return ""; } catch (KeywordSearchModuleException ex) { logger.log(Level.WARNING, "Could not query markup for page: " + currentPage, ex); //NON-NLS return ""; } }