List of usage examples for org.apache.solr.client.solrj SolrQuery getFilterQueries
public String[] getFilterQueries()
From source file:at.pagu.soldockr.core.QueryParserTest.java
License:Apache License
@Test public void testWithFilterQuery() { Query query = new SimpleQuery(new Criteria("field_1").is("value_1")) .addFilterQuery(new SimpleFilterQuery(new Criteria("filter_field").is("filter_value"))); SolrQuery solrQuery = queryParser.constructSolrQuery(query); String[] filterQueries = solrQuery.getFilterQueries(); Assert.assertEquals(1, filterQueries.length); Assert.assertEquals("filter_field:filter_value", filterQueries[0]); }
From source file:at.pagu.soldockr.core.QueryParserTest.java
License:Apache License
@Test public void testWithEmptyFilterQuery() { Query query = new SimpleQuery(new Criteria("field_1").is("value_1")).addFilterQuery(new SimpleQuery()); SolrQuery solrQuery = queryParser.constructSolrQuery(query); Assert.assertNull(solrQuery.getFilterQueries()); }
From source file:au.org.ala.biocache.dao.SearchDAOImpl.java
License:Open Source License
/** * Writes the index fields to the supplied output stream in CSV format. * <p>/*from www. j a va2s .c o m*/ * DM: refactored to split the query by month to improve performance. * Further enhancements possible: * 1) Multi threaded * 2) More filtering, by year or decade.. * * @param downloadParams * @param out * @param includeSensitive * @param dd The details of the download * @param checkLimit * @param nextExecutor The ExecutorService to use to process results on different threads * @throws Exception */ @Override public ConcurrentMap<String, AtomicInteger> writeResultsFromIndexToStream( final DownloadRequestParams downloadParams, final OutputStream out, final boolean includeSensitive, final DownloadDetailsDTO dd, boolean checkLimit, final ExecutorService nextExecutor) throws Exception { expandRequestedFields(downloadParams, true); if (dd != null) { dd.resetCounts(); } long start = System.currentTimeMillis(); final ConcurrentMap<String, AtomicInteger> uidStats = new ConcurrentHashMap<>(); getServer(); try { SolrQuery solrQuery = new SolrQuery(); queryFormatUtils.formatSearchQuery(downloadParams); String dFields = downloadParams.getFields(); if (includeSensitive) { //include raw latitude and longitudes if (dFields.contains("decimalLatitude.p")) { dFields = dFields.replaceFirst("decimalLatitude.p", "sensitive_latitude,sensitive_longitude,decimalLatitude.p"); } else if (dFields.contains("decimalLatitude")) { dFields = dFields.replaceFirst("decimalLatitude", "sensitive_latitude,sensitive_longitude,decimalLatitude"); } if (dFields.contains(",locality,")) { dFields = dFields.replaceFirst(",locality,", ",locality,sensitive_locality,"); } if (dFields.contains(",locality.p,")) { dFields = dFields.replaceFirst(",locality.p,", ",locality.p,sensitive_locality,"); } } StringBuilder sb = new StringBuilder(dFields); if (!downloadParams.getExtra().isEmpty()) { sb.append(",").append(downloadParams.getExtra()); } String[] requestedFields = sb.toString().split(","); List<String>[] indexedFields; if (downloadFields == null) { //default to include everything java.util.List<String> mappedNames = new java.util.LinkedList<String>(); for (int i = 0; i < requestedFields.length; i++) mappedNames.add(requestedFields[i]); indexedFields = new List[] { mappedNames, new java.util.LinkedList<String>(), mappedNames, mappedNames, new ArrayList(), new ArrayList() }; } else { indexedFields = downloadFields.getIndexFields(requestedFields, downloadParams.getDwcHeaders(), downloadParams.getLayersServiceUrl()); } //apply custom header String[] customHeader = dd.getRequestParams().getCustomHeader().split(","); for (int i = 0; i + 1 < customHeader.length; i += 2) { for (int j = 0; j < indexedFields[0].size(); j++) { if (customHeader[i].equals(indexedFields[0].get(j))) { indexedFields[2].set(j, customHeader[i + 1]); } } for (int j = 0; j < indexedFields[4].size(); j++) { if (customHeader[i].equals(indexedFields[5].get(j))) { indexedFields[4].set(j, customHeader[i + 1]); } } } if (logger.isDebugEnabled()) { logger.debug("Fields included in download: " + indexedFields[0]); logger.debug("Fields excluded from download: " + indexedFields[1]); logger.debug("The headers in downloads: " + indexedFields[2]); logger.debug("Analysis headers: " + indexedFields[4]); logger.debug("Analysis fields: " + indexedFields[5]); } //set the fields to the ones that are available in the index String[] fields = indexedFields[0].toArray(new String[] {}); solrQuery.setFields(fields); StringBuilder qasb = new StringBuilder(); if (!"none".equals(downloadParams.getQa())) { solrQuery.addField("assertions"); if (!"all".equals(downloadParams.getQa()) && !"includeall".equals(downloadParams.getQa())) { //add all the qa fields qasb.append(downloadParams.getQa()); } } solrQuery.addField("institution_uid").addField("collection_uid").addField("data_resource_uid") .addField("data_provider_uid"); solrQuery.setQuery(downloadParams.getFormattedQuery()); solrQuery.setFacetMinCount(1); solrQuery.setFacetLimit(-1); //get the assertion facets to add them to the download fields boolean getAssertionsFromFacets = "all".equals(downloadParams.getQa()) || "includeall".equals(downloadParams.getQa()); SolrQuery monthAssertionsQuery = getAssertionsFromFacets ? solrQuery.getCopy().addFacetField("month", "assertions") : solrQuery.getCopy().addFacetField("month"); if (getAssertionsFromFacets) { //set the order for the facet to be based on the index - this will force the assertions to be returned in the same order each time //based on alphabetical sort. The number of QA's may change between searches so we can't guarantee that the order won't change monthAssertionsQuery.add("f.assertions.facet.sort", "index"); } QueryResponse facetQuery = runSolrQuery(monthAssertionsQuery, downloadParams.getFormattedFq(), 0, 0, "score", "asc"); //set the totalrecords for the download details dd.setTotalRecords(facetQuery.getResults().getNumFound()); //use a separately configured and smaller limit when output will be unzipped final long maxDownloadSize; if (MAX_DOWNLOAD_SIZE > unzippedLimit && out instanceof OptionalZipOutputStream && ((OptionalZipOutputStream) out).getType() == OptionalZipOutputStream.Type.unzipped) { maxDownloadSize = unzippedLimit; } else { maxDownloadSize = MAX_DOWNLOAD_SIZE; } if (checkLimit && dd.getTotalRecords() < maxDownloadSize) { checkLimit = false; } //get the month facets to add them to the download fields get the assertion facets. List<Count> splitByFacet = null; for (FacetField facet : facetQuery.getFacetFields()) { if (facet.getName().equals("assertions") && facet.getValueCount() > 0) { qasb.append(getQAFromFacet(facet)); } if (facet.getName().equals("month") && facet.getValueCount() > 0) { splitByFacet = facet.getValues(); } } if ("includeall".equals(downloadParams.getQa())) { qasb = getAllQAFields(); } String qas = qasb.toString(); //include sensitive fields in the header when the output will be partially sensitive final String[] sensitiveFields; final String[] notSensitiveFields; if (dd.getSensitiveFq() != null) { List<String>[] sensitiveHdr = downloadFields.getIndexFields(sensitiveSOLRHdr, downloadParams.getDwcHeaders(), downloadParams.getLayersServiceUrl()); //header for the output file indexedFields[2].addAll(sensitiveHdr[2]); //lookup for fields from sensitive queries sensitiveFields = org.apache.commons.lang3.ArrayUtils.addAll( indexedFields[0].toArray(new String[] {}), sensitiveHdr[0].toArray(new String[] {})); //use general fields when sensitive data is not permitted notSensitiveFields = org.apache.commons.lang3.ArrayUtils .addAll(indexedFields[0].toArray(new String[] {}), notSensitiveSOLRHdr); } else { sensitiveFields = new String[0]; notSensitiveFields = fields; } //add analysis headers indexedFields[2].addAll(indexedFields[4]); final String[] analysisFields = indexedFields[5].toArray(new String[0]); final String[] qaFields = qas.equals("") ? new String[] {} : qas.split(","); String[] qaTitles = downloadFields.getHeader(qaFields, false, false); String[] header = org.apache.commons.lang3.ArrayUtils.addAll(indexedFields[2].toArray(new String[] {}), qaTitles); //retain output header fields and field names for inclusion of header info in the download StringBuilder infoFields = new StringBuilder("infoFields"); for (String h : indexedFields[3]) infoFields.append(",").append(h); for (String h : qaFields) infoFields.append(",").append(h); StringBuilder infoHeader = new StringBuilder("infoHeaders"); for (String h : header) infoHeader.append(",").append(h); String info = infoFields.toString(); while (info.contains(",,")) info = info.replace(",,", ","); uidStats.put(info, new AtomicInteger(-1)); String hdr = infoHeader.toString(); while (hdr.contains(",,")) hdr = hdr.replace(",,", ","); uidStats.put(hdr, new AtomicInteger(-2)); //construct correct RecordWriter based on the supplied fileType final RecordWriterError rw = downloadParams.getFileType().equals("csv") ? new CSVRecordWriter(out, header, downloadParams.getSep(), downloadParams.getEsc()) : (downloadParams.getFileType().equals("tsv") ? new TSVRecordWriter(out, header) : new ShapeFileRecordWriter(tmpShapefileDir, downloadParams.getFile(), out, (String[]) ArrayUtils.addAll(fields, qaFields))); // Requirement to be able to propagate interruptions to all other threads for this execution // Doing this via this variable final AtomicBoolean interruptFound = dd != null ? dd.getInterrupt() : new AtomicBoolean(false); // Create a fixed length blocking queue for buffering results before they are written // This also creates a push-back effect to throttle the results generating threads // when it fills and offers to it are delayed until the writer consumes elements from the queue final BlockingQueue<String[]> queue = new ArrayBlockingQueue<>(resultsQueueLength); // Create a sentinel that we can check for reference equality to signal the end of the queue final String[] sentinel = new String[0]; // An implementation of RecordWriter that adds to an in-memory queue final RecordWriter concurrentWrapper = new RecordWriter() { private AtomicBoolean finalised = new AtomicBoolean(false); private AtomicBoolean finalisedComplete = new AtomicBoolean(false); @Override public void write(String[] nextLine) { try { if (Thread.currentThread().isInterrupted() || interruptFound.get() || finalised.get()) { finalise(); return; } while (!queue.offer(nextLine, writerTimeoutWaitMillis, TimeUnit.MILLISECONDS)) { if (Thread.currentThread().isInterrupted() || interruptFound.get() || finalised.get()) { finalise(); break; } } } catch (InterruptedException e) { Thread.currentThread().interrupt(); interruptFound.set(true); if (logger.isDebugEnabled()) { logger.debug( "Queue failed to accept the next record due to a thread interrupt, calling finalise the cleanup: ", e); } // If we were interrupted then we should call finalise to cleanup finalise(); } } @Override public void finalise() { if (finalised.compareAndSet(false, true)) { try { // Offer the sentinel at least once, even when the thread is interrupted while (!queue.offer(sentinel, writerTimeoutWaitMillis, TimeUnit.MILLISECONDS)) { // If the thread is interrupted then the queue may not have any active consumers, // so don't loop forever waiting for capacity in this case // The hard shutdown phase will use queue.clear to ensure that the // sentinel gets onto the queue at least once if (Thread.currentThread().isInterrupted() || interruptFound.get()) { break; } } } catch (InterruptedException e) { Thread.currentThread().interrupt(); interruptFound.set(true); if (logger.isDebugEnabled()) { logger.debug( "Queue failed to accept the sentinel in finalise due to a thread interrupt: ", e); } } finally { finalisedComplete.set(true); } } } @Override public boolean finalised() { return finalisedComplete.get(); } }; // A single thread that consumes elements put onto the queue until it sees the sentinel, finalising after the sentinel or an interrupt Runnable writerRunnable = new Runnable() { @Override public void run() { try { long counter = 0; while (true) { counter = counter + 1; if (Thread.currentThread().isInterrupted() || interruptFound.get()) { break; } String[] take = queue.take(); // Sentinel object equality check to see if we are done if (take == sentinel || Thread.currentThread().isInterrupted() || interruptFound.get()) { break; } // Otherwise write to the wrapped record writer rw.write(take); //test for errors. This can contain a flush so only test occasionally if (counter % resultsQueueLength == 0 && rw.hasError()) { throw RecordWriterException.newRecordWriterException(dd, downloadParams, true, rw); } } } catch (RecordWriterException e) { //no trace information is available to print for these errors logger.error(e.getMessage()); interruptFound.set(true); } catch (InterruptedException e) { Thread.currentThread().interrupt(); interruptFound.set(true); } catch (Exception e) { // Reuse interruptFound variable to signal that the writer had issues interruptFound.set(true); logger.error("Download writer failed.", e); } finally { rw.finalise(); } } }; Thread writerThread = new Thread(writerRunnable); writerThread.start(); try { if (rw instanceof ShapeFileRecordWriter) { dd.setHeaderMap(((ShapeFileRecordWriter) rw).getHeaderMappings()); } //order the query by _docid_ for faster paging solrQuery.addSortField("_docid_", ORDER.asc); //for each month create a separate query that pages through 500 records per page List<SolrQuery> queries = new ArrayList<SolrQuery>(); if (splitByFacet != null) { for (Count facet : splitByFacet) { if (facet.getCount() > 0) { SolrQuery splitByFacetQuery; //do not add remainderQuery here if (facet.getName() != null) { splitByFacetQuery = solrQuery.getCopy() .addFilterQuery(facet.getFacetField().getName() + ":" + facet.getName()); splitByFacetQuery.setFacet(false); queries.add(splitByFacetQuery); } } } if (splitByFacet.size() > 0) { SolrQuery remainderQuery = solrQuery.getCopy() .addFilterQuery("-" + splitByFacet.get(0).getFacetField().getName() + ":[* TO *]"); queries.add(0, remainderQuery); } } else { queries.add(0, solrQuery); } //split into sensitive and non-sensitive queries when // - not including all sensitive values // - there is a sensitive fq final List<SolrQuery> sensitiveQ = new ArrayList<SolrQuery>(); if (!includeSensitive && dd.getSensitiveFq() != null) { sensitiveQ.addAll( splitQueries(queries, dd.getSensitiveFq(), sensitiveSOLRHdr, notSensitiveSOLRHdr)); } //Set<Future<Integer>> futures = new HashSet<Future<Integer>>(); final AtomicInteger resultsCount = new AtomicInteger(0); final boolean threadCheckLimit = checkLimit; List<Callable<Integer>> solrCallables = new ArrayList<>(queries.size()); // execute each query, writing the results to stream for (final SolrQuery splitByFacetQuery : queries) { // define a thread Callable<Integer> solrCallable = new Callable<Integer>() { @Override public Integer call() throws Exception { int startIndex = 0; // Randomise the wakeup time so they don't all wakeup on a periodic cycle long localThrottle = throttle + Math.round(Math.random() * throttle); String[] fq = downloadParams.getFormattedFq(); if (splitByFacetQuery.getFilterQueries() != null && splitByFacetQuery.getFilterQueries().length > 0) { if (fq == null) { fq = new String[0]; } fq = org.apache.commons.lang3.ArrayUtils.addAll(fq, splitByFacetQuery.getFilterQueries()); } QueryResponse qr = runSolrQuery(splitByFacetQuery, fq, downloadBatchSize, startIndex, "_docid_", "asc"); AtomicInteger recordsForThread = new AtomicInteger(0); if (logger.isDebugEnabled()) { logger.debug( splitByFacetQuery.getQuery() + " - results: " + qr.getResults().size()); } while (qr != null && !qr.getResults().isEmpty() && !interruptFound.get()) { if (logger.isDebugEnabled()) { logger.debug( "Start index: " + startIndex + ", " + splitByFacetQuery.getQuery()); } int count = 0; if (sensitiveQ.contains(splitByFacetQuery)) { count = processQueryResults(uidStats, sensitiveFields, qaFields, concurrentWrapper, qr, dd, threadCheckLimit, resultsCount, maxDownloadSize, analysisFields); } else { // write non-sensitive values into sensitive fields when not authorised for their sensitive values count = processQueryResults(uidStats, notSensitiveFields, qaFields, concurrentWrapper, qr, dd, threadCheckLimit, resultsCount, maxDownloadSize, analysisFields); } recordsForThread.addAndGet(count); startIndex += downloadBatchSize; // we have already set the Filter query the first time the query was constructed // rerun with the same params but different startIndex if (!threadCheckLimit || resultsCount.get() < maxDownloadSize) { if (!threadCheckLimit) { // throttle the download by sleeping Thread.sleep(localThrottle); } qr = runSolrQuery(splitByFacetQuery, null, downloadBatchSize, startIndex, "_docid_", "asc"); } else { qr = null; } } return recordsForThread.get(); } }; solrCallables.add(solrCallable); } List<Future<Integer>> futures = new ArrayList<>(solrCallables.size()); for (Callable<Integer> nextCallable : solrCallables) { futures.add(nextExecutor.submit(nextCallable)); } // Busy wait because we need to be able to respond to an interrupt on any callable // and propagate it to all of the others for this particular query // Because the executor service is shared to prevent too many concurrent threads being run, // this requires a busy wait loop on the main thread to monitor state boolean waitAgain = false; do { waitAgain = false; for (Future<Integer> future : futures) { if (!future.isDone()) { // Wait again even if an interrupt flag is set, as it may have been set partway through the iteration // The calls to future.cancel will occur next time if the interrupt is setup partway through an iteration waitAgain = true; // If one thread finds an interrupt it is propagated to others using the interruptFound AtomicBoolean if (interruptFound.get()) { future.cancel(true); } } } // Don't trigger the timeout interrupt if we don't have to wait again as we are already done at this point if (waitAgain && (System.currentTimeMillis() - start) > downloadMaxTime) { interruptFound.set(true); break; } if (waitAgain) { Thread.sleep(downloadCheckBusyWaitSleep); } } while (waitAgain); AtomicInteger totalDownload = new AtomicInteger(0); for (Future<Integer> future : futures) { if (future.isDone()) { totalDownload.addAndGet(future.get()); } else { // All incomplete futures that survived the loop above are cancelled here future.cancel(true); } } long finish = System.currentTimeMillis(); long timeTakenInSecs = (finish - start) / 1000; if (timeTakenInSecs <= 0) timeTakenInSecs = 1; if (logger.isInfoEnabled()) { logger.info("Download of " + resultsCount + " records in " + timeTakenInSecs + " seconds. Record/sec: " + resultsCount.intValue() / timeTakenInSecs); } } finally { try { // Once we get here, we need to finalise starting at the concurrent wrapper, // as there are no more non-sentinel records to be added to the queue // This eventually triggers finalisation of the underlying writer when the queue empties // This is a soft shutdown, and hence we wait below for this stage to complete in normal circumstances // Note, this blocks for writerTimeoutWaitMillis trying to legitimately add the sentinel to the end of the queue // We force the sentinel to be added in the hard shutdown phase below concurrentWrapper.finalise(); } finally { try { // Track the current time right now so we can abort after downloadMaxCompletionTime milliseconds in this phase final long completionStartTime = System.currentTimeMillis(); // Busy wait check for finalised to be called in the RecordWriter or something is interrupted // By this stage, there are at maximum download.internal.queue.size items remaining (default 1000) while (writerThread.isAlive() && !writerThread.isInterrupted() && !interruptFound.get() && !Thread.currentThread().isInterrupted() && !rw.finalised() && !((System.currentTimeMillis() - completionStartTime) > downloadMaxCompletionTime)) { Thread.sleep(downloadCheckBusyWaitSleep); } } finally { try { // Attempt all actions that could trigger the writer thread to finalise, as by this stage we are in hard shutdown mode // Signal that we are in hard shutdown mode interruptFound.set(true); // Add the sentinel or clear the queue and try again until it gets onto the queue // We are in hard shutdown mode, so only priority is that the queue either // gets the sentinel or the thread is interrupted to clean up resources while (!queue.offer(sentinel)) { queue.clear(); } // Interrupt the single writer thread writerThread.interrupt(); // Explicitly call finalise on the RecordWriter as a backup // In normal circumstances it is called via the sentinel or the interrupt // This will not block if finalise has been called previously in the current three implementations rw.finalise(); } finally { if (rw != null && rw.hasError()) { throw RecordWriterException.newRecordWriterException(dd, downloadParams, true, rw); } else { // Flush whatever output was still pending for more deterministic debugging out.flush(); } } } } } } catch (SolrServerException ex) { logger.error("Problem communicating with SOLR server while processing download. " + ex.getMessage(), ex); } return uidStats; }
From source file:au.org.ala.biocache.dao.SearchDAOImpl.java
License:Open Source License
/** * Downloads the records for the supplied query. Used to break up the download into components * 1) 1 call for each data resource that has a download limit (supply the data resource uid as the argument dataResource) * 2) 1 call for the remaining records/*from ww w . ja va 2 s. c o m*/ * * @param downloadParams * @param downloadLimit * @param uidStats * @param fields * @param qaFields * @param resultsCount * @param dataResource The dataResource being download. This should be null if multiple data resource are being downloaded. * @return * @throws Exception */ private int downloadRecords(DownloadRequestParams downloadParams, RecordWriterError writer, Map<String, Integer> downloadLimit, ConcurrentMap<String, AtomicInteger> uidStats, String[] fields, String[] qaFields, int resultsCount, String dataResource, boolean includeSensitive, DownloadDetailsDTO dd, boolean limit, String[] analysisLayers) throws Exception { if (logger.isInfoEnabled()) { logger.info("download query: " + downloadParams.getQ()); } SolrQuery solrQuery = initSolrQuery(downloadParams, false, null); solrQuery.setRows(limit ? MAX_DOWNLOAD_SIZE : -1); queryFormatUtils.formatSearchQuery(downloadParams); solrQuery.setQuery(downloadParams.getFormattedQuery()); //Only the fields specified below will be included in the results from the SOLR Query solrQuery.setFields("row_key", "institution_uid", "collection_uid", "data_resource_uid", "data_provider_uid"); if (dd != null) { dd.resetCounts(); } //get coordinates for analysis layer intersection if (analysisLayers.length > 0) { if (!includeSensitive && dd.getSensitiveFq() != null) { for (String s : sensitiveSOLRHdr) solrQuery.addField(s); } else { for (String s : notSensitiveSOLRHdr) solrQuery.addField(s); } } int pageSize = downloadBatchSize; StringBuilder sb = new StringBuilder(downloadParams.getFields()); if (downloadParams.getExtra().length() > 0) { sb.append(",").append(downloadParams.getExtra()); } List<SolrQuery> queries = new ArrayList<SolrQuery>(); queries.add(solrQuery); //split into sensitive and non-sensitive queries when // - not including all sensitive values // - there is a sensitive fq List<SolrQuery> sensitiveQ = new ArrayList<SolrQuery>(); if (!includeSensitive && dd.getSensitiveFq() != null) { sensitiveQ = splitQueries(queries, dd.getSensitiveFq(), null, null); } final String[] sensitiveFields; final String[] notSensitiveFields; if (!includeSensitive && dd.getSensitiveFq() != null) { //lookup for fields from sensitive queries sensitiveFields = org.apache.commons.lang3.ArrayUtils.addAll(fields, sensitiveCassandraHdr); //use general fields when sensitive data is not permitted notSensitiveFields = org.apache.commons.lang3.ArrayUtils.addAll(fields, notSensitiveCassandraHdr); } else { sensitiveFields = new String[0]; notSensitiveFields = fields; } for (SolrQuery q : queries) { int startIndex = 0; String[] fq = downloadParams.getFormattedFq(); if (q.getFilterQueries() != null && q.getFilterQueries().length > 0) { if (fq == null) { fq = new String[0]; } fq = org.apache.commons.lang3.ArrayUtils.addAll(fq, q.getFilterQueries()); } QueryResponse qr = runSolrQuery(q, fq, pageSize, startIndex, "_docid_", "asc"); List<String> uuids = new ArrayList<String>(); List<String[]> intersectionAll = intersectResults(dd.getRequestParams().getLayersServiceUrl(), analysisLayers, qr.getResults()); while (qr.getResults().size() > 0 && (!limit || resultsCount < MAX_DOWNLOAD_SIZE) && shouldDownload(dataResource, downloadLimit, false)) { if (logger.isDebugEnabled()) { logger.debug("Start index: " + startIndex); } Map<String, String[]> dataToInsert = new HashMap<String, String[]>(); //cycle through the results adding them to the list that will be sent to cassandra int row = 0; for (SolrDocument sd : qr.getResults()) { if (sd.getFieldValue("data_resource_uid") != null) { String druid = sd.getFieldValue("data_resource_uid").toString(); if (shouldDownload(druid, downloadLimit, true) && (!limit || resultsCount < MAX_DOWNLOAD_SIZE)) { resultsCount++; uuids.add(sd.getFieldValue("row_key").toString()); //include analysis layer intersections if (intersectionAll.size() > row + 1) dataToInsert.put(sd.getFieldValue("row_key").toString(), (String[]) ArrayUtils.subarray(intersectionAll.get(row + 1), 2, intersectionAll.get(row + 1).length)); //increment the counters.... incrementCount(uidStats, sd.getFieldValue("institution_uid")); incrementCount(uidStats, sd.getFieldValue("collection_uid")); incrementCount(uidStats, sd.getFieldValue("data_provider_uid")); incrementCount(uidStats, druid); } } row++; } String[] newMiscFields; if (sensitiveQ.contains(q)) { newMiscFields = au.org.ala.biocache.Store.writeToWriter(writer, uuids.toArray(new String[] {}), sensitiveFields, qaFields, true, (dd.getRequestParams() != null ? dd.getRequestParams().getIncludeMisc() : false), dd.getMiscFields(), dataToInsert); } else { newMiscFields = au.org.ala.biocache.Store.writeToWriter(writer, uuids.toArray(new String[] {}), notSensitiveFields, qaFields, includeSensitive, (dd.getRequestParams() != null ? dd.getRequestParams().getIncludeMisc() : false), dd.getMiscFields(), dataToInsert); } //test for errors if (writer.hasError()) { throw RecordWriterException.newRecordWriterException(dd, downloadParams, false, writer); } dd.setMiscFields(newMiscFields); startIndex += pageSize; uuids.clear(); dd.updateCounts(qr.getResults().size()); if (!limit || resultsCount < MAX_DOWNLOAD_SIZE) { //we have already set the Filter query the first time the query was constructed rerun with he same params but different startIndex qr = runSolrQuery(q, null, pageSize, startIndex, "_docid_", "asc"); } } } return resultsCount; }
From source file:bamboo.trove.services.CdxRestrictionServiceTest.java
License:Apache License
@Test public void testRule1144() throws CdxRestrictionService.RulesOutOfDateException, ParseException { // Rule 1144 was just odd. has a left-anchored wildcard plus an empty string. SimpleDateFormat sdf = new SimpleDateFormat(RestrictionsDAO.DATE_FORMAT); Document doc = new Document(); setUrl(doc, "http://blocked.adultshop.com.au"); // Rule 953) Multiple URL patterns from a Pandora conversion. 2nd is a broad wildcard for a whole domain // *.adultshop.com.au // Embargoed for 70 years!!! // Fake 'today' is => "2014-01-01T00:00:00+1100" Date MATCH_DATE = sdf.parse("1945-01-01T00:00:00+1100"); // - 69 years Date NOT_MATCH_DATE = sdf.parse("1943-01-01T00:00:00+1100"); // - 71 years doc.setDate(MATCH_DATE);//ww w. ja v a 2s.c o m assertEquals(DocumentStatus.RESTRICTED_FOR_BOTH, service.filterDocument(doc).getIndexerPolicy()); doc.setDate(NOT_MATCH_DATE); assertEquals(DocumentStatus.ACCEPTED, service.filterDocument(doc).getIndexerPolicy()); // Let's eyeball what that empty String will do RuleChangeUpdateManager manager = new RuleChangeUpdateManager(); SolrQuery query = manager.convertRuleToSearch(ruleSet.getRules().get(1144L), ""); boolean found = false; // We do not want to see *:* on the end String target = "(" + SolrEnum.URL_TOKENIZED + ":pandora.nla.gov.au\\/pan\\/35531\\/) OR (" + SolrEnum.URL_TOKENIZED + ":adultshop.com.au\\/)"; for (String fq : query.getFilterQueries()) { if (target.equals(fq)) { found = true; } } if (!found) { // This behaviour is only ok when a single URL pattern was provided and it was empty. Assert.fail("An empty URL pattern should not be part of a query when there are other valid patterns"); } }
From source file:com.idealista.solrmeter.QueryServiceSolrJImplTestCase.java
License:Apache License
public void testCreateQuery() throws QueryException { QueryServiceSolrJSpy service = new QueryServiceSolrJSpy(); SolrQuery query = service.createQuery("some query", "name:tomas", "/dismax", false, "name, surname", null, null, 10, 0, ""); assertEquals("some query", query.get("q")); assertEquals(new String[] { "name:tomas" }, query.getFilterQueries()); //assertEquals("/dismax", query.getQueryType()); assertEquals(new String[] { "name", "surname" }, query.getFacetFields()); service.createQuery(null, "name:tomas", "/dismax", false, "name, surname", null, null, 10, 0, ""); service.createQuery("", null, "/dismax", false, "name, surname", null, null, 10, 0, ""); service.createQuery("", "name:tomas", null, false, "name, surname", null, null, 10, 0, ""); service.createQuery("", "name:tomas", "/dismax", false, null, null, null, 10, 0, ""); service.createQuery("", "name:tomas", "/dismax", false, "name, surname", "name", "desc", 10, 0, ""); service.createQuery("", "name:tomas", "/dismax", false, "name, surname", "name", "asc", 10, 0, ""); service.createQuery("", "name:tomas", "/dismax", false, "name, surname", "name", "asc", 10, 0, null); service.createQuery("", "name:tomas", "/dismax", false, "name, surname", "name", "asc", null, 0, ""); service.createQuery("", "name:tomas", "/dismax", false, "name, surname", "name", "asc", 10, null, ""); }
From source file:com.plugtree.solrmeter.QueryServiceSolrJImplTestCase.java
License:Apache License
public void testCreateQuery() throws QueryException { QueryServiceSolrJSpy service = new QueryServiceSolrJSpy(); SolrQuery query = service.createQuery("some query", "name:tomas", "/dismax", false, "name, surname", null, null, 10, 0, ""); assertEquals("some query", query.get("q")); assertEquals(new String[] { "name:tomas" }, query.getFilterQueries()); assertEquals("/dismax", query.getQueryType()); assertEquals(new String[] { "name", "surname" }, query.getFacetFields()); service.createQuery(null, "name:tomas", "/dismax", false, "name, surname", null, null, 10, 0, ""); service.createQuery("", null, "/dismax", false, "name, surname", null, null, 10, 0, ""); service.createQuery("", "name:tomas", null, false, "name, surname", null, null, 10, 0, ""); service.createQuery("", "name:tomas", "/dismax", false, null, null, null, 10, 0, ""); service.createQuery("", "name:tomas", "/dismax", false, "name, surname", "name", "desc", 10, 0, ""); service.createQuery("", "name:tomas", "/dismax", false, "name, surname", "name", "asc", 10, 0, ""); service.createQuery("", "name:tomas", "/dismax", false, "name, surname", "name", "asc", 10, 0, null); service.createQuery("", "name:tomas", "/dismax", false, "name, surname", "name", "asc", null, 0, ""); service.createQuery("", "name:tomas", "/dismax", false, "name, surname", "name", "asc", 10, null, ""); }
From source file:com.ushahidi.swiftriver.core.solr.repository.DropDocumentRepositoryImplTest.java
License:Open Source License
@Test public void findInRiverWithBoundingBox() throws Exception { QueryResponse mockQueryResponse = mock(QueryResponse.class); List<DropDocument> foundDocuments = new ArrayList<DropDocument>(); when(mockSolrServer.query(any(SolrQuery.class))).thenReturn(mockQueryResponse); when(mockQueryResponse.getBeans(DropDocument.class)).thenReturn(foundDocuments); Pageable pageRequest = new PageRequest(0, 20); DropFilter dropFilter = new DropFilter(); dropFilter.setBoundingBox("36.8,-122.75,37.8,-121.75"); dropSearchRepository.findInRiver(1L, dropFilter, pageRequest); ArgumentCaptor<SolrQuery> solrQueryArgument = ArgumentCaptor.forClass(SolrQuery.class); verify(mockSolrServer).query(solrQueryArgument.capture()); SolrQuery solrQuery = solrQueryArgument.getValue(); String[] expectedFilterQuery = new String[] { "riverId:(1)", "geo:[36.8,-122.75 TO 37.8,-121.75]" }; String[] actualFilterQuery = solrQuery.getFilterQueries(); assertEquals("*:*", solrQuery.getQuery()); assertEquals(2, actualFilterQuery.length); assertEquals(expectedFilterQuery[0], actualFilterQuery[0]); assertEquals(expectedFilterQuery[1], actualFilterQuery[1]); }
From source file:com.ushahidi.swiftriver.core.solr.repository.DropDocumentRepositoryImplTest.java
License:Open Source License
@Test public void findInBucketWithBoundingBox() throws Exception { QueryResponse mockQueryResponse = mock(QueryResponse.class); List<DropDocument> foundDocuments = new ArrayList<DropDocument>(); when(mockSolrServer.query(any(SolrQuery.class))).thenReturn(mockQueryResponse); when(mockQueryResponse.getBeans(DropDocument.class)).thenReturn(foundDocuments); Pageable pageRequest = new PageRequest(0, 20); DropFilter dropFilter = new DropFilter(); dropFilter.setBoundingBox("40,-74,41,-73"); dropSearchRepository.findInBucket(20L, dropFilter, pageRequest); ArgumentCaptor<SolrQuery> solrQueryArgument = ArgumentCaptor.forClass(SolrQuery.class); verify(mockSolrServer).query(solrQueryArgument.capture()); SolrQuery solrQuery = solrQueryArgument.getValue(); String[] expectedFilterQuery = new String[] { "bucketId:(20)", "geo:[40.0,-74.0 TO 41.0,-73.0]" }; String[] actualFilterQuery = solrQuery.getFilterQueries(); assertEquals("*:*", solrQuery.getQuery()); assertEquals(2, actualFilterQuery.length); assertEquals(expectedFilterQuery[0], actualFilterQuery[0]); assertEquals(expectedFilterQuery[1], actualFilterQuery[1]); }
From source file:ddf.catalog.source.solr.SolrFilterDelegateTest.java
License:Open Source License
@Test public void testXpathExists() { String xpath = "//root/sub/@attribute"; String expectedQuery = "{!xpath}xpath:\"" + xpath + "\""; SolrQuery xpathQuery = toTest.xpathExists(xpath); assertThat(xpathQuery.getFilterQueries()[0], is(expectedQuery)); }