List of usage examples for org.apache.solr.client.solrj SolrQuery setFacet
public SolrQuery setFacet(boolean b)
From source file:at.pagu.soldockr.core.QueryParser.java
License:Apache License
private void appendFacetingOnFields(SolrQuery solrQuery, FacetQuery query) { FacetOptions facetOptions = query.getFacetOptions(); if (facetOptions == null || !facetOptions.hasFields()) { return;//from w w w . j a v a 2 s. co m } solrQuery.setFacet(true); solrQuery.addFacetField(convertFieldListToStringArray(facetOptions.getFacetOnFields())); solrQuery.setFacetMinCount(facetOptions.getFacetMinCount()); solrQuery.setFacetLimit(facetOptions.getPageable().getPageSize()); if (facetOptions.getPageable().getPageNumber() > 0) { solrQuery.set(FacetParams.FACET_OFFSET, facetOptions.getPageable().getOffset()); } if (FacetOptions.FacetSort.INDEX.equals(facetOptions.getFacetSort())) { solrQuery.setFacetSort(FacetParams.FACET_SORT_INDEX); } }
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 w w w . j ava 2s. c om * 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
/** * Note - this method extracts from CASSANDRA rather than the Index. *///from w w w . j a va 2 s . c om public ConcurrentMap<String, AtomicInteger> writeResultsToStream(DownloadRequestParams downloadParams, OutputStream out, int i, boolean includeSensitive, DownloadDetailsDTO dd, boolean limit) throws Exception { expandRequestedFields(downloadParams, false); int resultsCount = 0; ConcurrentMap<String, AtomicInteger> uidStats = new ConcurrentHashMap<>(); //stores the remaining limit for data resources that have a download limit Map<String, Integer> downloadLimit = new HashMap<>(); try { SolrQuery solrQuery = initSolrQuery(downloadParams, false, null); //ensure that the qa facet is being ordered alphabetically so that the order is consistent. boolean getAssertionsFromFacets = "all".equals(downloadParams.getQa()) || "includeall".equals(downloadParams.getQa()); 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 solrQuery.add("f.assertions.facet.sort", "index"); } queryFormatUtils.formatSearchQuery(downloadParams); if (logger.isInfoEnabled()) { logger.info("search query: " + downloadParams.getFormattedQuery()); } 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"); String dFields = downloadParams.getFields(); if (includeSensitive) { //include raw latitude and longitudes dFields = dFields .replaceFirst("decimalLatitude.p", "decimalLatitude,decimalLongitude,decimalLatitude.p") .replaceFirst(",locality,", ",locality,sensitive_locality,"); } StringBuilder sb = new StringBuilder(dFields); if (downloadParams.getExtra().length() > 0) { sb.append(",").append(downloadParams.getExtra()); } StringBuilder qasb = new StringBuilder(); solrQuery.setFacet(true); QueryResponse qr = runSolrQuery(solrQuery, downloadParams.getFormattedFq(), 0, 0, "_docid_", "asc"); dd.setTotalRecords(qr.getResults().getNumFound()); //get the assertion facets to add them to the download fields List<FacetField> facets = qr.getFacetFields(); for (FacetField facet : facets) { if (facet.getName().equals("assertions") && facet.getValueCount() > 0) { qasb.append(getQAFromFacet(facet)); } else if (facet.getName().equals("data_resource_uid") && checkDownloadLimits) { //populate the download limit initDownloadLimits(downloadLimit, facet); } } if ("includeall".equals(downloadParams.getQa())) { qasb = getAllQAFields(); } //Write the header line String qas = qasb.toString(); List<String>[] indexedFields = downloadFields.getIndexFields(downloadParams.getFields().split(","), false, downloadParams.getLayersServiceUrl()); String[] fields = sb.toString().split(","); //avoid analysis field duplicates for (String s : indexedFields[5]) fields = (String[]) ArrayUtils.removeElement(fields, s); String[] qaFields = qas.equals("") ? new String[] {} : qas.split(","); String[] qaTitles = downloadFields.getHeader(qaFields, false, false); String[] titles = downloadFields.getHeader(fields, true, downloadParams.getDwcHeaders()); String[] analysisHeaders = indexedFields[4].toArray(new String[0]); String[] analysisFields = indexedFields[5].toArray(new String[0]); //apply custom header String[] customHeader = dd.getRequestParams().getCustomHeader().split(","); for (i = 0; i + 1 < customHeader.length; i += 2) { for (int j = 0; j < analysisFields.length; j++) { if (customHeader[i].equals(analysisFields[j])) { analysisFields[j] = customHeader[i + 1]; } } for (int j = 0; j < qaFields.length; j++) { if (customHeader[i].equals(qaFields[j])) { qaTitles[j] = customHeader[i + 1]; } } for (int j = 0; j < fields.length; j++) { if (customHeader[i].equals(fields[j])) { titles[j] = customHeader[i + 1]; } } } //append sensitive fields for the header only if (!includeSensitive && dd.getSensitiveFq() != null) { //sensitive headers do not have a DwC name, always set getIndexFields dwcHeader=false List<String>[] sensitiveHdr = downloadFields.getIndexFields(sensitiveSOLRHdr, false, downloadParams.getLayersServiceUrl()); titles = org.apache.commons.lang3.ArrayUtils.addAll(titles, sensitiveHdr[2].toArray(new String[] {})); } String[] header = org.apache.commons.lang3.ArrayUtils .addAll(org.apache.commons.lang3.ArrayUtils.addAll(titles, qaTitles), analysisHeaders); //Create the Writer that will be used to format the records //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))); try { if (rw instanceof ShapeFileRecordWriter) { dd.setHeaderMap(((ShapeFileRecordWriter) rw).getHeaderMappings()); } //retain output header fields and field names for inclusion of header info in the download StringBuilder infoFields = new StringBuilder("infoFields,"); for (String h : fields) infoFields.append(",").append(h); for (String h : analysisFields) 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)); //download the records that have limits first... if (downloadLimit.size() > 0) { String[] originalFq = downloadParams.getFormattedFq(); StringBuilder fqBuilder = new StringBuilder("-("); for (String dr : downloadLimit.keySet()) { //add another fq to the search for data_resource_uid downloadParams.setFq((String[]) ArrayUtils.add(originalFq, "data_resource_uid:" + dr)); resultsCount = downloadRecords(downloadParams, rw, downloadLimit, uidStats, fields, qaFields, resultsCount, dr, includeSensitive, dd, limit, analysisFields); if (fqBuilder.length() > 2) { fqBuilder.append(" OR "); } fqBuilder.append("data_resource_uid:").append(dr); } fqBuilder.append(")"); //now include the rest of the data resources //add extra fq for the remaining records downloadParams.setFq((String[]) ArrayUtils.add(originalFq, fqBuilder.toString())); resultsCount = downloadRecords(downloadParams, rw, downloadLimit, uidStats, fields, qaFields, resultsCount, null, includeSensitive, dd, limit, analysisFields); } else { //download all at once downloadRecords(downloadParams, rw, downloadLimit, uidStats, fields, qaFields, resultsCount, null, includeSensitive, dd, limit, analysisFields); } } finally { rw.finalise(); } } catch (SolrServerException ex) { logger.error("Problem communicating with SOLR server. " + ex.getMessage(), ex); } return uidStats; }
From source file:au.org.ala.biocache.dao.SearchDAOImpl.java
License:Open Source License
private List<OccurrencePoint> getPoints(SpatialSearchRequestParams searchParams, PointType pointType, int max) throws Exception { List<OccurrencePoint> points = new ArrayList<OccurrencePoint>(); // new OccurrencePoint(PointType.POINT); queryFormatUtils.formatSearchQuery(searchParams); if (logger.isInfoEnabled()) { logger.info("search query: " + searchParams.getFormattedQuery()); }/*from ww w. j a va 2 s . co m*/ SolrQuery solrQuery = new SolrQuery(); solrQuery.setQueryType("standard"); solrQuery.setQuery(searchParams.getFormattedQuery()); solrQuery.setRows(0); solrQuery.setFacet(true); solrQuery.addFacetField(pointType.getLabel()); solrQuery.setFacetMinCount(1); solrQuery.setFacetLimit(max); // unlimited = -1 QueryResponse qr = runSolrQuery(solrQuery, searchParams.getFormattedFq(), 1, 0, "score", "asc"); List<FacetField> facets = qr.getFacetFields(); if (facets != null) { for (FacetField facet : facets) { List<FacetField.Count> facetEntries = facet.getValues(); if (facet.getName().contains(pointType.getLabel()) && (facetEntries != null) && (facetEntries.size() > 0)) { for (FacetField.Count fcount : facetEntries) { if (StringUtils.isNotEmpty(fcount.getName()) && fcount.getCount() > 0) { OccurrencePoint point = new OccurrencePoint(pointType); point.setCount(fcount.getCount()); String[] pointsDelimited = StringUtils.split(fcount.getName(), ','); List<Float> coords = new ArrayList<Float>(); for (String coord : pointsDelimited) { try { Float decimalCoord = Float.parseFloat(coord); coords.add(decimalCoord); } catch (NumberFormatException numberFormatException) { logger.warn("Error parsing Float for Lat/Long: " + numberFormatException.getMessage(), numberFormatException); } } if (!coords.isEmpty()) { Collections.reverse(coords); // must be long, lat order point.setCoordinates(coords); points.add(point); } } } } } } return points; }
From source file:au.org.ala.biocache.dao.SearchDAOImpl.java
License:Open Source License
/** * @see au.org.ala.biocache.dao.SearchDAO#getFacetPointsShort(au.org.ala.biocache.dto.SpatialSearchRequestParams, String) *//* ww w . j a v a 2 s .c om*/ @Override public FacetField getFacetPointsShort(SpatialSearchRequestParams searchParams, String pointType) throws Exception { queryFormatUtils.formatSearchQuery(searchParams); if (logger.isInfoEnabled()) { logger.info("search query: " + searchParams.getFormattedQuery()); } SolrQuery solrQuery = new SolrQuery(); solrQuery.setRequestHandler("standard"); solrQuery.setQuery(searchParams.getFormattedQuery()); solrQuery.setRows(0); solrQuery.setFacet(true); solrQuery.addFacetField(pointType); solrQuery.setFacetMinCount(1); solrQuery.setFacetLimit(searchParams.getFlimit());//MAX_DOWNLOAD_SIZE); // unlimited = -1 QueryResponse qr = runSolrQuery(solrQuery, searchParams.getFormattedFq(), 0, 0, "_docid_", "asc"); List<FacetField> facets = qr.getFacetFields(); //return first facet, there should only be 1 if (facets != null && facets.size() > 0) { return facets.get(0); } return null; }
From source file:au.org.ala.biocache.dao.SearchDAOImpl.java
License:Open Source License
/** * @see au.org.ala.biocache.dao.SearchDAO#getOccurrences(au.org.ala.biocache.dto.SpatialSearchRequestParams, au.org.ala.biocache.dto.PointType, String) *//*from w w w. j a v a 2 s . c o m*/ @Override public List<OccurrencePoint> getOccurrences(SpatialSearchRequestParams searchParams, PointType pointType, String colourBy) throws Exception { List<OccurrencePoint> points = new ArrayList<OccurrencePoint>(); searchParams.setPageSize(100); String queryString = ""; queryFormatUtils.formatSearchQuery(searchParams); queryString = searchParams.getFormattedQuery(); if (logger.isInfoEnabled()) { logger.info("search query: " + queryString); } SolrQuery solrQuery = new SolrQuery(); solrQuery.setQueryType("standard"); solrQuery.setQuery(queryString); solrQuery.setRows(0); solrQuery.setFacet(true); solrQuery.addFacetField(pointType.getLabel()); solrQuery.setFacetMinCount(1); solrQuery.setFacetLimit(MAX_DOWNLOAD_SIZE); // unlimited = -1 QueryResponse qr = runSolrQuery(solrQuery, searchParams); SearchResultDTO searchResults = processSolrResponse(searchParams, qr, solrQuery, OccurrenceIndex.class); List<OccurrenceIndex> ocs = searchResults.getOccurrences(); if (!ocs.isEmpty() && ocs.size() > 0) { for (OccurrenceIndex oc : ocs) { List<Float> coords = new ArrayList<Float>(); coords.add(oc.getDecimalLongitude().floatValue()); coords.add(oc.getDecimalLatitude().floatValue()); OccurrencePoint point = new OccurrencePoint(); point.setCoordinates(coords); point.setOccurrenceUid(oc.getUuid()); points.add(point); } } return points; }
From source file:au.org.ala.biocache.dao.SearchDAOImpl.java
License:Open Source License
/** * http://ala-biocache1.vm.csiro.au:8080/solr/select?q=*:*&rows=0&facet=true&facet.field=data_provider_id&facet.field=data_provider&facet.sort=data_provider_id * * @see au.org.ala.biocache.dao.SearchDAO#getDataProviderCounts() *///w ww. jav a 2 s . c o m //IS THIS BEING USED BY ANYTHING?? @Override public List<DataProviderCountDTO> getDataProviderCounts() throws Exception { List<DataProviderCountDTO> dpDTOs = new ArrayList<DataProviderCountDTO>(); // new OccurrencePoint(PointType.POINT); SolrQuery solrQuery = new SolrQuery(); solrQuery.setQueryType("standard"); solrQuery.setQuery("*:*"); solrQuery.setRows(0); solrQuery.setFacet(true); solrQuery.addFacetField("data_provider_uid"); solrQuery.addFacetField("data_provider"); solrQuery.setFacetMinCount(1); QueryResponse qr = runSolrQuery(solrQuery, null, 1, 0, "data_provider", "asc"); List<FacetField> facets = qr.getFacetFields(); if (facets != null && facets.size() == 2) { FacetField dataProviderIdFacet = facets.get(0); FacetField dataProviderNameFacet = facets.get(1); List<FacetField.Count> dpIdEntries = dataProviderIdFacet.getValues(); List<FacetField.Count> dpNameEntries = dataProviderNameFacet.getValues(); if (dpIdEntries != null) { for (int i = 0; i < dpIdEntries.size(); i++) { FacetField.Count dpIdEntry = dpIdEntries.get(i); FacetField.Count dpNameEntry = dpNameEntries.get(i); String dataProviderId = dpIdEntry.getName(); String dataProviderName = dpNameEntry.getName(); long count = dpIdEntry.getCount(); if (count > 0) { DataProviderCountDTO dto = new DataProviderCountDTO(dataProviderId, dataProviderName, count); dpDTOs.add(dto); } } } } if (logger.isInfoEnabled()) { logger.info("Find data providers = " + dpDTOs.size()); } return dpDTOs; }
From source file:au.org.ala.biocache.dao.SearchDAOImpl.java
License:Open Source License
/** * @see au.org.ala.biocache.dao.SearchDAO#findRecordByStateFor(java.lang.String) * IS THIS BEGIN USED OR NECESSARY/*from w ww . j a va 2s. c o m*/ */ @Override public List<FieldResultDTO> findRecordByStateFor(String query) throws Exception { List<FieldResultDTO> fDTOs = new ArrayList<FieldResultDTO>(); // new OccurrencePoint(PointType.POINT); SolrQuery solrQuery = new SolrQuery(); solrQuery.setQueryType("standard"); solrQuery.setQuery(query); solrQuery.setRows(0); solrQuery.setFacet(true); solrQuery.addFacetField("state"); solrQuery.setFacetMinCount(1); QueryResponse qr = runSolrQuery(solrQuery, null, 1, 0, "data_provider", "asc"); List<FacetField> facets = qr.getFacetFields(); FacetField ff = qr.getFacetField("state"); if (ff != null) { for (Count count : ff.getValues()) { //only start adding counts when we hit a decade with some results. if (count.getCount() > 0) { FieldResultDTO f = new FieldResultDTO(count.getName(), count.getCount()); fDTOs.add(f); } } } return fDTOs; }
From source file:au.org.ala.biocache.dao.SearchDAOImpl.java
License:Open Source License
/** * Calculates the breakdown of the supplied query based on the supplied params *//*ww w .java2 s . co m*/ public TaxaRankCountDTO calculateBreakdown(BreakdownRequestParams queryParams) throws Exception { if (logger.isDebugEnabled()) { logger.debug("Attempting to find the counts for " + queryParams); } TaxaRankCountDTO trDTO = null; SolrQuery solrQuery = new SolrQuery(); solrQuery.setQueryType("standard"); queryFormatUtils.formatSearchQuery(queryParams); solrQuery.setQuery(queryParams.getFormattedQuery()); queryParams.setPageSize(0); solrQuery.setFacet(true); solrQuery.setFacetMinCount(1); solrQuery.setFacetSort("count"); solrQuery.setFacetLimit(-1); //add the rank:name as a fq if necessary if (StringUtils.isNotEmpty(queryParams.getName()) && StringUtils.isNotEmpty(queryParams.getRank())) { queryParams.setFormattedFq((String[]) ArrayUtils.addAll(queryParams.getFormattedFq(), new String[] { queryParams.getRank() + ":" + queryParams.getName() })); } //add the ranks as facets if (queryParams.getLevel() == null) { List<String> ranks = queryParams.getRank() != null ? searchUtils.getNextRanks(queryParams.getRank(), queryParams.getName() == null) : searchUtils.getRanks(); for (String r : ranks) { solrQuery.addFacetField(r); } } else { //the user has supplied the "exact" level at which to perform the breakdown solrQuery.addFacetField(queryParams.getLevel()); } QueryResponse qr = runSolrQuery(solrQuery, queryParams); if (queryParams.getMax() != null && queryParams.getMax() > 0) { //need to get the return level that the number of facets are <=max ranks need to be processed in reverse order until max is satisfied if (qr.getResults().getNumFound() > 0) { List<FacetField> ffs = qr.getFacetFields(); //reverse the facets so that they are returned in rank reverse order species, genus, family etc Collections.reverse(ffs); for (FacetField ff : ffs) { //logger.debug("Handling " + ff.getName()); trDTO = new TaxaRankCountDTO(ff.getName()); if (ff.getValues() != null && ff.getValues().size() <= queryParams.getMax()) { List<FieldResultDTO> fDTOs = new ArrayList<FieldResultDTO>(); for (Count count : ff.getValues()) { if (count.getCount() > 0) { FieldResultDTO f = new FieldResultDTO(count.getName(), count.getCount()); fDTOs.add(f); } } trDTO.setTaxa(fDTOs); break; } } } } else if (queryParams.getRank() != null || queryParams.getLevel() != null) { //just want to process normally the rank to facet on will start with the highest rank and then go down until one exists for if (qr.getResults().getNumFound() > 0) { List<FacetField> ffs = qr.getFacetFields(); for (FacetField ff : ffs) { trDTO = new TaxaRankCountDTO(ff.getName()); if (ff != null && ff.getValues() != null) { List<Count> counts = ff.getValues(); if (counts.size() > 0) { List<FieldResultDTO> fDTOs = new ArrayList<FieldResultDTO>(); for (Count count : counts) { if (count.getCount() > 0) { FieldResultDTO f = new FieldResultDTO(count.getName(), count.getCount()); fDTOs.add(f); } } trDTO.setTaxa(fDTOs); break; } } } } } return trDTO; }
From source file:au.org.ala.biocache.dao.SearchDAOImpl.java
License:Open Source License
/** * @see au.org.ala.biocache.dao.SearchDAO#findTaxonCountForUid(au.org.ala.biocache.dto.BreakdownRequestParams, String) * @deprecated use {@link #calculateBreakdown(BreakdownRequestParams)} instead *//*from w w w .j ava2 s .c o m*/ @Deprecated public TaxaRankCountDTO findTaxonCountForUid(BreakdownRequestParams breakdownParams, String query) throws Exception { TaxaRankCountDTO trDTO = null; List<String> ranks = breakdownParams.getLevel() == null ? searchUtils.getNextRanks(breakdownParams.getRank(), breakdownParams.getName() == null) : new ArrayList<String>(); if (breakdownParams.getLevel() != null) ranks.add(breakdownParams.getLevel()); if (ranks != null && ranks.size() > 0) { SolrQuery solrQuery = new SolrQuery(); solrQuery.setQueryType("standard"); solrQuery.setQuery(query); solrQuery.setRows(0); solrQuery.setFacet(true); solrQuery.setFacetMinCount(1); solrQuery.setFacetSort("count"); solrQuery.setFacetLimit(-1); //we want all facets for (String r : ranks) { solrQuery.addFacetField(r); } QueryResponse qr = runSolrQuery(solrQuery, queryFormatUtils.getQueryContextAsArray(breakdownParams.getQc()), 1, 0, breakdownParams.getRank(), "asc"); if (qr.getResults().size() > 0) { for (String r : ranks) { trDTO = new TaxaRankCountDTO(r); FacetField ff = qr.getFacetField(r); if (ff != null && ff.getValues() != null) { List<Count> counts = ff.getValues(); if (counts.size() > 0) { List<FieldResultDTO> fDTOs = new ArrayList<FieldResultDTO>(); for (Count count : counts) { FieldResultDTO f = new FieldResultDTO(count.getName(), count.getCount()); fDTOs.add(f); } trDTO.setTaxa(fDTOs); break; } } } } } return trDTO; }