List of usage examples for org.apache.solr.client.solrj SolrQuery setShowDebugInfo
public SolrQuery setShowDebugInfo(boolean showDebugInfo)
From source file:org.sleuthkit.autopsy.keywordsearch.AccountsText.java
License:Open Source License
/** * Initialize this object with information about which pages/chunks have * hits. Multiple calls will not change the initial results. *//*from w w w. jav a2s .com*/ synchronized private void loadPageInfo() { if (isPageInfoLoaded) { return; } if (chunkId != null) {//if a chunk is specified, only show that chunk/page this.numberPagesForFile = 1; this.currentPage = chunkId; this.numberOfHitsPerPage.put(chunkId, 0); this.pages.add(chunkId); this.currentHitPerPage.put(chunkId, 0); } else { try { this.numberPagesForFile = solrServer.queryNumFileChunks(this.solrObjectId); } catch (KeywordSearchModuleException | NoOpenCoreException ex) { LOGGER.log(Level.WARNING, "Could not get number pages for content " + this.solrDocumentId, ex); //NON-NLS return; } //if has chunks, get pages with hits TreeSet<Integer> sortedPagesWithHits = new TreeSet<>(); SolrQuery q = new SolrQuery(); q.setShowDebugInfo(DEBUG); //debug q.setQuery(queryString); q.setFields(Server.Schema.ID.toString()); //for this case we only need the document ids q.addFilterQuery( Server.Schema.ID.toString() + ":" + this.solrObjectId + Server.CHUNK_ID_SEPARATOR + "*"); try { QueryResponse response = solrServer.query(q, METHOD.POST); for (SolrDocument resultDoc : response.getResults()) { final String resultDocumentId = resultDoc.getFieldValue(Server.Schema.ID.toString()).toString(); // Put the solr chunk id in the map String resultChunkID = StringUtils.substringAfter(resultDocumentId, Server.CHUNK_ID_SEPARATOR); if (StringUtils.isNotBlank(resultChunkID)) { sortedPagesWithHits.add(Integer.parseInt(resultChunkID)); } else { sortedPagesWithHits.add(0); } } } catch (KeywordSearchModuleException | NoOpenCoreException | NumberFormatException ex) { LOGGER.log(Level.WARNING, "Error executing Solr highlighting query: " + keywords, ex); //NON-NLS } //set page to first page having highlights if (sortedPagesWithHits.isEmpty()) { this.currentPage = 0; } else { this.currentPage = sortedPagesWithHits.first(); } for (Integer page : sortedPagesWithHits) { numberOfHitsPerPage.put(page, 0); //unknown number of matches in the page pages.add(page); currentHitPerPage.put(page, 0); //set current hit to 0th } } isPageInfoLoaded = true; }
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);/* w w w.java 2 s .c o m*/
//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); }/*w ww. jav a 2s. 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 ww.j a va2s .com 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); }/* w w w .j av 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); 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 ""; } }
From source file:org.sleuthkit.autopsy.keywordsearch.LuceneQuery.java
License:Open Source License
/** * Create the query object for the stored keyword * * @param snippets True if query should request snippets * @return//from w w w . j a va 2 s . c o m */ private SolrQuery createAndConfigureSolrQuery(boolean snippets) { SolrQuery q = new SolrQuery(); q.setShowDebugInfo(DEBUG); //debug //set query, force quotes/grouping around all literal queries final String groupedQuery = KeywordSearchUtil.quoteQuery(keywordStringEscaped); String theQueryStr = groupedQuery; if (field != null) { //use the optional field StringBuilder sb = new StringBuilder(); sb.append(field).append(":").append(groupedQuery); theQueryStr = sb.toString(); } q.setQuery(theQueryStr); q.setRows(MAX_RESULTS); if (snippets) { q.setFields(Server.Schema.ID.toString()); } else { q.setFields(Server.Schema.ID.toString()); } for (KeywordQueryFilter filter : filters) { q.addFilterQuery(filter.toString()); } if (snippets) { q.addHighlightField(Server.Schema.TEXT.toString()); //q.setHighlightSimplePre("«"); //original highlighter only //q.setHighlightSimplePost("»"); //original highlighter only q.setHighlightSnippets(1); q.setHighlightFragsize(SNIPPET_LENGTH); //tune the highlighter q.setParam("hl.useFastVectorHighlighter", "on"); //fast highlighter scales better than standard one NON-NLS q.setParam("hl.tag.pre", "«"); //makes sense for FastVectorHighlighter only NON-NLS q.setParam("hl.tag.post", "«"); //makes sense for FastVectorHighlighter only NON-NLS q.setParam("hl.fragListBuilder", "simple"); //makes sense for FastVectorHighlighter only NON-NLS //Solr bug if fragCharSize is smaller than Query string, StringIndexOutOfBoundsException is thrown. q.setParam("hl.fragCharSize", Integer.toString(theQueryStr.length())); //makes sense for FastVectorHighlighter only NON-NLS //docs says makes sense for the original Highlighter only, but not really //analyze all content SLOW! consider lowering q.setParam("hl.maxAnalyzedChars", Server.HL_ANALYZE_CHARS_UNLIMITED); //NON-NLS } return q; }
From source file:org.sleuthkit.autopsy.keywordsearch.LuceneQuery.java
License:Open Source License
/** * return snippet preview context//www.ja v a 2 s . c om * * @param query the keyword query for text to highlight. Lucene special * cahrs should already be escaped. * @param solrObjectId Solr object id associated with the hit * @param chunkID chunk id associated with the content hit, or 0 if no * chunks * @param isRegex whether the query is a regular expression (different Solr * fields are then used to generate the preview) * @param group whether the query should look for all terms grouped together * in the query order, or not * @return */ public static String querySnippet(String query, long solrObjectId, int chunkID, boolean isRegex, boolean group) throws NoOpenCoreException { Server solrServer = KeywordSearch.getServer(); String highlightField; if (isRegex) { highlightField = LuceneQuery.HIGHLIGHT_FIELD_REGEX; } else { highlightField = LuceneQuery.HIGHLIGHT_FIELD_LITERAL; } SolrQuery q = new SolrQuery(); String queryStr; if (isRegex) { StringBuilder sb = new StringBuilder(); sb.append(highlightField).append(":"); if (group) { sb.append("\""); } sb.append(query); if (group) { sb.append("\""); } queryStr = sb.toString(); } else { //simplify query/escaping and use default field //always force grouping/quotes queryStr = KeywordSearchUtil.quoteQuery(query); } q.setQuery(queryStr); String contentIDStr; if (chunkID == 0) { contentIDStr = Long.toString(solrObjectId); } else { contentIDStr = Server.getChunkIdString(solrObjectId, chunkID); } String idQuery = Server.Schema.ID.toString() + ":" + KeywordSearchUtil.escapeLuceneQuery(contentIDStr); q.setShowDebugInfo(DEBUG); //debug q.addFilterQuery(idQuery); q.addHighlightField(highlightField); //q.setHighlightSimplePre("«"); //original highlighter only //q.setHighlightSimplePost("»"); //original highlighter only q.setHighlightSnippets(1); q.setHighlightFragsize(SNIPPET_LENGTH); //tune the highlighter q.setParam("hl.useFastVectorHighlighter", "on"); //fast highlighter scales better than standard one NON-NLS q.setParam("hl.tag.pre", "«"); //makes sense for FastVectorHighlighter only NON-NLS q.setParam("hl.tag.post", "«"); //makes sense for FastVectorHighlighter only NON-NLS q.setParam("hl.fragListBuilder", "simple"); //makes sense for FastVectorHighlighter only NON-NLS //Solr bug if fragCharSize is smaller than Query string, StringIndexOutOfBoundsException is thrown. q.setParam("hl.fragCharSize", Integer.toString(queryStr.length())); //makes sense for FastVectorHighlighter only NON-NLS //docs says makes sense for the original Highlighter only, but not really //analyze all content SLOW! consider lowering 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 ""; } List<String> contentHighlights = responseHighlightID.get(highlightField); if (contentHighlights == null) { return ""; } else { // extracted content is HTML-escaped, but snippet goes in a plain text field return EscapeUtil.unEscapeHtml(contentHighlights.get(0)).trim(); } } catch (NoOpenCoreException ex) { logger.log(Level.WARNING, "Error executing Lucene Solr Query: " + query, ex); //NON-NLS throw ex; } catch (KeywordSearchModuleException ex) { logger.log(Level.WARNING, "Error executing Lucene Solr Query: " + query, ex); //NON-NLS return ""; } }
From source file:org.sleuthkit.autopsy.keywordsearch.TermComponentQuery.java
License:Open Source License
@Override public QueryResults performQuery() throws NoOpenCoreException { final SolrQuery q = createQuery(); q.setShowDebugInfo(DEBUG); q.setTermsLimit(MAX_TERMS_RESULTS);// w w w .j av a 2s .c o m logger.log(Level.INFO, "Query: {0}", q.toString()); //NON-NLS terms = executeQuery(q); QueryResults results = new QueryResults(this, keywordList); int resultSize = 0; for (Term term : terms) { final String termStr = KeywordSearchUtil.escapeLuceneQuery(term.getTerm()); LuceneQuery filesQuery = new LuceneQuery(keywordList, new Keyword(termStr, true)); //filesQuery.setField(TERMS_SEARCH_FIELD); for (KeywordQueryFilter filter : filters) { //set filter //note: we can't set filter query on terms query //but setting filter query on terms results query will yield the same result filesQuery.addFilter(filter); } try { QueryResults subResults = filesQuery.performQuery(); Set<KeywordHit> filesResults = new HashSet<>(); for (Keyword key : subResults.getKeywords()) { List<KeywordHit> keyRes = subResults.getResults(key); resultSize += keyRes.size(); filesResults.addAll(keyRes); } results.addResult(new Keyword(term.getTerm(), false), new ArrayList<>(filesResults)); } catch (NoOpenCoreException e) { logger.log(Level.WARNING, "Error executing Solr query,", e); //NON-NLS throw e; } catch (RuntimeException e) { logger.log(Level.WARNING, "Error executing Solr query,", e); //NON-NLS } } //TODO limit how many results we store, not to hit memory limits logger.log(Level.INFO, "Regex # results: {0}", resultSize); //NON-NLS return results; }
From source file:org.sleuthkit.autopsy.keywordsearch.TermsComponentQuery.java
License:Open Source License
/** * Executes the regex query as a two step operation. In the first step, the * Solr terms component is used to find any terms in the index that match * the regex. In the second step, term queries are executed for each matched * term to produce the set of keyword hits for the regex. * * @return A QueryResult object or null. * * @throws NoOpenCoreException// w w w. ja v a2 s . com */ @Override public QueryResults performQuery() throws KeywordSearchModuleException, NoOpenCoreException { /* * Do a query using the Solr terms component to find any terms in the * index that match the regex. */ final SolrQuery termsQuery = new SolrQuery(); termsQuery.setRequestHandler(SEARCH_HANDLER); termsQuery.setTerms(true); termsQuery.setTermsRegexFlag(CASE_INSENSITIVE); termsQuery.setTermsRegex(searchTerm); termsQuery.addTermsField(SEARCH_FIELD); termsQuery.setTimeAllowed(TERMS_SEARCH_TIMEOUT); termsQuery.setShowDebugInfo(DEBUG_FLAG); termsQuery.setTermsLimit(MAX_TERMS_QUERY_RESULTS); List<Term> terms = KeywordSearch.getServer().queryTerms(termsQuery).getTerms(SEARCH_FIELD); /* * Do a term query for each term that matched the regex. */ QueryResults results = new QueryResults(this, keywordList); for (Term term : terms) { /* * If searching for credit card account numbers, do a Luhn check on * the term and discard it if it does not pass. */ if (keyword.getArtifactAttributeType() == ATTRIBUTE_TYPE.TSK_CARD_NUMBER) { Matcher matcher = CREDIT_CARD_NUM_PATTERN.matcher(term.getTerm()); matcher.find(); final String ccn = CharMatcher.anyOf(" -").removeFrom(matcher.group("ccn")); if (false == CREDIT_CARD_NUM_LUHN_CHECK.isValid(ccn)) { continue; } } /* * Do an ordinary query with the escaped term and convert the query * results into a single list of keyword hits without duplicates. * * Note that the filters field appears to be unused. There is an old * comment here, what does it mean? "Note: we can't set filter query * on terms query but setting filter query on fileResults query will * yield the same result." The filter is NOT being added to the term * query. */ String escapedTerm = KeywordSearchUtil.escapeLuceneQuery(term.getTerm()); LuceneQuery termQuery = new LuceneQuery(keywordList, new Keyword(escapedTerm, true)); filters.forEach(termQuery::addFilter); // This appears to be unused QueryResults termQueryResult = termQuery.performQuery(); Set<KeywordHit> termHits = new HashSet<>(); for (Keyword word : termQueryResult.getKeywords()) { termHits.addAll(termQueryResult.getResults(word)); } results.addResult(new Keyword(term.getTerm(), false), new ArrayList<>(termHits)); } return results; }