List of usage examples for org.apache.solr.search SolrIndexSearcher getDocList
public DocList getDocList(Query query, List<Query> filterList, Sort lsort, int offset, int len, int flags) throws IOException
query and the intersection of the filterList, sorted by sort. From source file:com.searchbox.solr.CategoryLikeThis.java
License:Apache License
@Override public void handleRequestBody(SolrQueryRequest req, SolrQueryResponse rsp) throws Exception { numRequests++;//www. j a v a 2 s. c o m long startTime = System.currentTimeMillis(); if (!keystate) { LOGGER.error( "License key failure, not performing clt query. Please email contact@searchbox.com for more information."); return; } try { SolrParams params = req.getParams(); String senseField = params.get(SenseParams.SENSE_FIELD, SenseParams.DEFAULT_SENSE_FIELD); BooleanQuery catfilter = new BooleanQuery(); // Set field flags ReturnFields returnFields = new SolrReturnFields(req); rsp.setReturnFields(returnFields); int flags = 0; if (returnFields.wantsScore()) { flags |= SolrIndexSearcher.GET_SCORES; } String defType = params.get(QueryParsing.DEFTYPE, QParserPlugin.DEFAULT_QTYPE); String q = params.get(CommonParams.Q); Query query = null; SortSpec sortSpec = null; List<Query> filters = new LinkedList<Query>(); List<RealTermFreqVector> prototypetfs = new LinkedList<RealTermFreqVector>(); try { if (q != null) { QParser parser = QParser.getParser(q, defType, req); query = parser.getQuery(); sortSpec = parser.getSort(true); } String[] fqs = req.getParams().getParams(CommonParams.FQ); if (fqs != null && fqs.length != 0) { for (String fq : fqs) { if (fq != null && fq.trim().length() != 0) { QParser fqp = QParser.getParser(fq, null, req); filters.add(fqp.getQuery()); } } } } catch (Exception e) { numErrors++; throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, e); } SolrIndexSearcher searcher = req.getSearcher(); DocListAndSet cltDocs = null; // Parse Required Params // This will either have a single Reader or valid query Reader reader = null; try { if (q == null || q.trim().length() < 1) { Iterable<ContentStream> streams = req.getContentStreams(); if (streams != null) { Iterator<ContentStream> iter = streams.iterator(); if (iter.hasNext()) { reader = iter.next().getReader(); } if (iter.hasNext()) { numErrors++; throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "SenseLikeThis does not support multiple ContentStreams"); } } } int start = params.getInt(CommonParams.START, 0); int rows = params.getInt(CommonParams.ROWS, 10); // Find documents SenseLikeThis - either with a reader or a query // -------------------------------------------------------------------------------- if (reader != null) { numErrors++; throw new RuntimeException("SLT based on a reader is not yet implemented"); } else if (q != null) { LOGGER.debug("Query for category:\t" + query); DocList match = searcher.getDocList(query, null, null, 0, 10, flags); // get first 10 if (match.size() == 0) { // no docs to make prototype! LOGGER.info("No documents found for prototype!"); rsp.add("response", new DocListAndSet()); return; } HashMap<String, Float> overallFreqMap = new HashMap<String, Float>(); // Create the TF of blah blah blah DocIterator iterator = match.iterator(); while (iterator.hasNext()) { // do a MoreLikeThis query for each document in results int id = iterator.nextDoc(); LOGGER.trace("Working on doc:\t" + id); RealTermFreqVector rtv = new RealTermFreqVector(id, searcher.getIndexReader(), senseField); for (int zz = 0; zz < rtv.getSize(); zz++) { Float prev = overallFreqMap.get(rtv.getTerms()[zz]); if (prev == null) { prev = 0f; } overallFreqMap.put(rtv.getTerms()[zz], rtv.getFreqs()[zz] + prev); } prototypetfs.add(rtv); } List<String> sortedKeys = Ordering.natural().onResultOf(Functions.forMap(overallFreqMap)) .immutableSortedCopy(overallFreqMap.keySet()); int keyiter = Math.min(sortedKeys.size() - 1, BooleanQuery.getMaxClauseCount() - 1); LOGGER.debug("I have this many terms:\t" + sortedKeys.size()); LOGGER.debug("And i'm going to use this many:\t" + keyiter); for (; keyiter >= 0; keyiter--) { TermQuery tq = new TermQuery(new Term(senseField, sortedKeys.get(keyiter))); catfilter.add(tq, BooleanClause.Occur.SHOULD); } } else { numErrors++; throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "CategoryLikeThis requires either a query (?q=) or text to find similar documents."); } LOGGER.debug("document filter is: \t" + catfilter); CategorizationBase model = new CategorizationBase(prototypetfs); CategoryQuery clt = CategoryQuery.CategoryQueryForDocument(catfilter, model, searcher.getIndexReader(), senseField); DocSet filtered = searcher.getDocSet(filters); cltDocs = searcher.getDocListAndSet(clt, filtered, Sort.RELEVANCE, start, rows, flags); } finally { if (reader != null) { reader.close(); } } if (cltDocs == null) { numEmpty++; cltDocs = new DocListAndSet(); // avoid NPE } rsp.add("response", cltDocs.docList); // maybe facet the results if (params.getBool(FacetParams.FACET, false)) { if (cltDocs.docSet == null) { rsp.add("facet_counts", null); } else { SimpleFacets f = new SimpleFacets(req, cltDocs.docSet, params); rsp.add("facet_counts", f.getFacetCounts()); } } } catch (Exception e) { numErrors++; } finally { totalTime += System.currentTimeMillis() - startTime; } }
From source file:com.searchbox.solr.SenseLikeThisHandler.java
License:Apache License
@Override public void handleRequestBody(SolrQueryRequest req, SolrQueryResponse rsp) throws Exception { NamedList<Object> timinginfo = new NamedList<Object>(); numRequests++;/* w ww . j a v a 2 s . c om*/ long startTime = System.currentTimeMillis(); long lstartTime = System.currentTimeMillis(); if (!keystate) { LOGGER.error( "License key failure, not performing sense query. Please email contact@searchbox.com for more information."); return; } boolean fromcache = false; try { SolrParams params = req.getParams(); int start = params.getInt(CommonParams.START, 0); int rows = params.getInt(CommonParams.ROWS, 10); HashSet<String> toIgnore = (new HashSet<String>()); toIgnore.add("start"); toIgnore.add("rows"); toIgnore.add("fl"); toIgnore.add("wt"); toIgnore.add("indent"); SolrCacheKey key = new SolrCacheKey(params, toIgnore); // Set field flags ReturnFields returnFields = new SolrReturnFields(req); rsp.setReturnFields(returnFields); int flags = 0; if (returnFields.wantsScore()) { flags |= SolrIndexSearcher.GET_SCORES; } String defType = params.get(QueryParsing.DEFTYPE, QParserPlugin.DEFAULT_QTYPE); String q = params.get(CommonParams.Q); Query query = null; QueryReductionFilter qr = null; SortSpec sortSpec = null; List<Query> filters = new ArrayList<Query>(); try { if (q != null) { QParser parser = QParser.getParser(q, defType, req); query = parser.getQuery(); sortSpec = parser.getSort(true); } String[] fqs = req.getParams().getParams(CommonParams.FQ); if (fqs != null && fqs.length != 0) { for (String fq : fqs) { if (fq != null && fq.trim().length() != 0) { QParser fqp = QParser.getParser(fq, null, req); filters.add(fqp.getQuery()); } } } } catch (Exception e) { numErrors++; throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, e); } timinginfo.add("Parse Query time", System.currentTimeMillis() - lstartTime); LOGGER.debug("Parsed Query Time:\t" + (System.currentTimeMillis() - lstartTime)); lstartTime = System.currentTimeMillis(); SolrIndexSearcher searcher = req.getSearcher(); SchemaField uniqueKeyField = searcher.getSchema().getUniqueKeyField(); // Parse Required Params // This will either have a single Reader or valid query // Find documents SenseLikeThis - either with a reader or a query // -------------------------------------------------------------------------------- SenseQuery slt = null; if (q == null) { numErrors++; throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "SenseLikeThis requires either a query (?q=) or text to find similar documents."); } // Matching options boolean includeMatch = params.getBool(MoreLikeThisParams.MATCH_INCLUDE, true); int matchOffset = params.getInt(MoreLikeThisParams.MATCH_OFFSET, 0); // Find the base match DocList match = searcher.getDocList(query, null, null, matchOffset, 1, flags); // only get the first one... if (includeMatch) { rsp.add("match", match); } DocIterator iterator = match.iterator(); if (!iterator.hasNext()) { numErrors++; throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "SenseLikeThis no document found matching request."); } int id = iterator.nextDoc(); timinginfo.add("Find Query Doc", System.currentTimeMillis() - lstartTime); LOGGER.debug("Find Query Doc:\t" + (System.currentTimeMillis() - lstartTime)); lstartTime = System.currentTimeMillis(); SolrCache sc = searcher.getCache("com.searchbox.sltcache"); DocListAndSet sltDocs = null; if (sc != null) { //try to get from cache sltDocs = (DocListAndSet) sc.get(key.getSet()); } else { LOGGER.error("com.searchbox.sltcache not defined, can't cache slt queries"); } sltDocs = (DocListAndSet) sc.get(key.getSet()); if (start + rows > 1000 || sltDocs == null || !params.getBool(CommonParams.CACHE, true)) { //not in cache, need to do search BooleanQuery bq = new BooleanQuery(); Document doc = searcher.getIndexReader().document(id); bq.add(new TermQuery(new Term(uniqueKeyField.getName(), uniqueKeyField.getType().storedToIndexed(doc.getField(uniqueKeyField.getName())))), BooleanClause.Occur.MUST_NOT); filters.add(bq); String[] senseFields = splitList .split(params.get(SenseParams.SENSE_FIELD, SenseParams.DEFAULT_SENSE_FIELD)); String senseField = (senseFields[0] != null) ? senseFields[0] : SenseParams.DEFAULT_SENSE_FIELD; //TODO more intelligent handling of multiple fields , can probably do a boolean junction of multiple sensequeries, but this will be slow long maxlength = -1; for (String possibleField : senseFields) { try { long flength = doc.getField(possibleField).stringValue().length(); if (flength > maxlength) { senseField = possibleField; maxlength = flength; } } catch (Exception e) { System.out.println("Error: " + e.getMessage()); } } LOGGER.debug("Using sense field :\t" + (senseField)); String CKBid = params.get(SenseParams.SENSE_CKB, SenseParams.SENSE_CKB_DEFAULT); RealTermFreqVector rtv = new RealTermFreqVector(id, searcher.getIndexReader(), senseField); timinginfo.add("Make real term freq vector", System.currentTimeMillis() - lstartTime); lstartTime = System.currentTimeMillis(); qr = new QueryReductionFilter(rtv, CKBid, searcher, senseField); qr.setNumtermstouse(params.getInt(SenseParams.SENSE_QR_NTU, SenseParams.SENSE_QR_NTU_DEFAULT)); qr.setThreshold(params.getInt(SenseParams.SENSE_QR_THRESH, SenseParams.SENSE_QR_THRESH_DEFAULT)); qr.setMaxDocSubSet(params.getInt(SenseParams.SENSE_QR_MAXDOC, SenseParams.SENSE_QR_MAXDOC_DEFAULT)); qr.setMinDocSetSizeForFilter( params.getInt(SenseParams.SENSE_MINDOC4QR, SenseParams.SENSE_MINDOC4QR_DEFAULT)); numTermsUsed += qr.getNumtermstouse(); numTermsConsidered += rtv.getSize(); timinginfo.add("Setup SLT query", System.currentTimeMillis() - lstartTime); LOGGER.debug("Setup SLT query:\t" + (System.currentTimeMillis() - lstartTime)); lstartTime = System.currentTimeMillis(); DocList subFiltered = qr.getSubSetToSearchIn(filters); timinginfo.add("Do Query Redux", System.currentTimeMillis() - lstartTime); LOGGER.debug("Do query redux:\t" + (System.currentTimeMillis() - lstartTime)); lstartTime = System.currentTimeMillis(); numFiltered += qr.getFiltered().docList.size(); numSubset += subFiltered.size(); LOGGER.info("Number of documents to search:\t" + subFiltered.size()); slt = new SenseQuery(rtv, senseField, CKBid, params.getFloat(SenseParams.SENSE_WEIGHT, SenseParams.DEFAULT_SENSE_WEIGHT), null); LOGGER.debug("Setup sense query:\t" + (System.currentTimeMillis() - lstartTime)); timinginfo.add("Setup sense query", System.currentTimeMillis() - lstartTime); lstartTime = System.currentTimeMillis(); sltDocs = searcher.getDocListAndSet(slt, subFiltered, Sort.RELEVANCE, 0, 1000, flags); timinginfo.add("Do sense query", System.currentTimeMillis() - lstartTime); lstartTime = System.currentTimeMillis(); LOGGER.debug("Adding this keyto cache:\t" + key.getSet().toString()); searcher.getCache("com.searchbox.sltcache").put(key.getSet(), sltDocs); } else { fromcache = true; timinginfo.add("Getting from cache", System.currentTimeMillis() - lstartTime); LOGGER.debug("Got result from cache"); lstartTime = System.currentTimeMillis(); } if (sltDocs == null) { numEmpty++; sltDocs = new DocListAndSet(); // avoid NPE } rsp.add("response", sltDocs.docList.subset(start, rows)); // maybe facet the results if (params.getBool(FacetParams.FACET, false)) { if (sltDocs.docSet == null) { rsp.add("facet_counts", null); } else { SimpleFacets f = new SimpleFacets(req, sltDocs.docSet, params); rsp.add("facet_counts", f.getFacetCounts()); } } timinginfo.add("Facet parts", System.currentTimeMillis() - lstartTime); LOGGER.debug("Facet parts:\t" + (System.currentTimeMillis() - lstartTime)); // Debug info, not doing it for the moment. boolean dbg = req.getParams().getBool(CommonParams.DEBUG_QUERY, false); boolean dbgQuery = false, dbgResults = false; if (dbg == false) {//if it's true, we are doing everything anyway. String[] dbgParams = req.getParams().getParams(CommonParams.DEBUG); if (dbgParams != null) { for (int i = 0; i < dbgParams.length; i++) { if (dbgParams[i].equals(CommonParams.QUERY)) { dbgQuery = true; } else if (dbgParams[i].equals(CommonParams.RESULTS)) { dbgResults = true; } } } } else { dbgQuery = true; dbgResults = true; } // Copied from StandardRequestHandler... perhaps it should be added to doStandardDebug? if (dbg == true) { try { lstartTime = System.currentTimeMillis(); NamedList<Object> dbgInfo = SolrPluginUtils.doStandardDebug(req, q, slt, sltDocs.docList.subset(start, rows), dbgQuery, dbgResults); dbgInfo.add("Query freqs", slt.getAllTermsasString()); if (null != dbgInfo) { if (null != filters) { dbgInfo.add("filter_queries", req.getParams().getParams(CommonParams.FQ)); List<String> fqs = new ArrayList<String>(filters.size()); for (Query fq : filters) { fqs.add(QueryParsing.toString(fq, req.getSchema())); } dbgInfo.add("parsed_filter_queries", fqs); } if (null != qr) { dbgInfo.add("QueryReduction", qr.getDbgInfo()); } if (null != slt) { dbgInfo.add("SLT", slt.getDbgInfo()); } dbgInfo.add("fromcache", fromcache); rsp.add("debug", dbgInfo); timinginfo.add("Debugging parts", System.currentTimeMillis() - lstartTime); dbgInfo.add("timings", timinginfo); } } catch (Exception e) { SolrException.log(SolrCore.log, "Exception during debug", e); rsp.add("exception_during_debug", SolrException.toStr(e)); } } } catch (Exception e) { numErrors++; e.printStackTrace(); } finally { totalTime += System.currentTimeMillis() - startTime; } }
From source file:com.searchbox.solr.SenseLikeThisHandlerNoReduction.java
License:Apache License
@Override public void handleRequestBody(SolrQueryRequest req, SolrQueryResponse rsp) throws Exception { SolrParams params = req.getParams(); if (!keystate) { LOGGER.error(/*from w w w .jav a2 s. com*/ "License key failure, not performing sense query. Please email contact@searchbox.com for more information."); return; } int docID; // Set field flags ReturnFields returnFields = new SolrReturnFields(req); rsp.setReturnFields(returnFields); int flags = 0; if (returnFields.wantsScore()) { flags |= SolrIndexSearcher.GET_SCORES; } String defType = params.get(QueryParsing.DEFTYPE, QParserPlugin.DEFAULT_QTYPE); String q = params.get(CommonParams.Q); Query query = null; SortSpec sortSpec = null; List<Query> filters = new ArrayList<Query>(); try { if (q != null) { QParser parser = QParser.getParser(q, defType, req); query = parser.getQuery(); sortSpec = parser.getSort(true); } String[] fqs = req.getParams().getParams(CommonParams.FQ); if (fqs != null && fqs.length != 0) { for (String fq : fqs) { if (fq != null && fq.trim().length() != 0) { QParser fqp = QParser.getParser(fq, null, req); filters.add(fqp.getQuery()); } } } } catch (Exception e) { throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, e); } SolrIndexSearcher searcher = req.getSearcher(); SchemaField uniqueKeyField = searcher.getSchema().getUniqueKeyField(); DocListAndSet sltDocs = null; // Parse Required Params // This will either have a single Reader or valid query Reader reader = null; try { if (q == null || q.trim().length() < 1) { Iterable<ContentStream> streams = req.getContentStreams(); if (streams != null) { Iterator<ContentStream> iter = streams.iterator(); if (iter.hasNext()) { reader = iter.next().getReader(); } if (iter.hasNext()) { throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "SenseLikeThis does not support multiple ContentStreams"); } } } int start = params.getInt(CommonParams.START, 0); int rows = params.getInt(CommonParams.ROWS, 10); // Find documents SenseLikeThis - either with a reader or a query // -------------------------------------------------------------------------------- SenseQuery slt = null; if (reader != null) { throw new RuntimeException("SLT based on a reader is not yet implemented"); } else if (q != null) { // Matching options boolean includeMatch = params.getBool(MoreLikeThisParams.MATCH_INCLUDE, true); int matchOffset = params.getInt(MoreLikeThisParams.MATCH_OFFSET, 0); // Find the base match DocList match = searcher.getDocList(query, null, null, matchOffset, 1, flags); // only get the first one... if (includeMatch) { rsp.add("match", match); } // Get docID DocIterator iterator = match.iterator(); docID = iterator.nextDoc(); BooleanQuery bq = new BooleanQuery(); Document doc = searcher.getIndexReader().document(docID); bq.add(new TermQuery(new Term(uniqueKeyField.getName(), uniqueKeyField.getType().storedToIndexed(doc.getField(uniqueKeyField.getName())))), BooleanClause.Occur.MUST_NOT); filters.add(bq); } else { throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "SenseLikeThis requires either a query (?q=) or text to find similar documents."); } String CKBid = params.get(SenseParams.SENSE_CKB, SenseParams.SENSE_CKB_DEFAULT); String senseField = params.get(SenseParams.SENSE_FIELD, SenseParams.DEFAULT_SENSE_FIELD); slt = new SenseQuery(new RealTermFreqVector(docID, searcher.getIndexReader(), senseField), senseField, CKBid, params.getFloat(SenseParams.SENSE_WEIGHT, SenseParams.DEFAULT_SENSE_WEIGHT), null); //Execute the SLT query //DocSet filtered = searcher.getDocSet(filters); //System.out.println("Number of documents to search:\t" + filtered.size()); //sltDocs = searcher.getDocListAndSet(slt, filtered, Sort.RELEVANCE, start, rows, flags); sltDocs = searcher.getDocListAndSet(slt, filters, Sort.RELEVANCE, start, rows, flags); } finally { if (reader != null) { reader.close(); } } if (sltDocs == null) { sltDocs = new DocListAndSet(); // avoid NPE } rsp.add("response", sltDocs.docList); // maybe facet the results if (params.getBool(FacetParams.FACET, false)) { if (sltDocs.docSet == null) { rsp.add("facet_counts", null); } else { SimpleFacets f = new SimpleFacets(req, sltDocs.docSet, params); rsp.add("facet_counts", f.getFacetCounts()); } } // Debug info, not doing it for the moment. boolean dbg = req.getParams().getBool(CommonParams.DEBUG_QUERY, false); boolean dbgQuery = false, dbgResults = false; if (dbg == false) {//if it's true, we are doing everything anyway. String[] dbgParams = req.getParams().getParams(CommonParams.DEBUG); if (dbgParams != null) { for (int i = 0; i < dbgParams.length; i++) { if (dbgParams[i].equals(CommonParams.QUERY)) { dbgQuery = true; } else if (dbgParams[i].equals(CommonParams.RESULTS)) { dbgResults = true; } } } } else { dbgQuery = true; dbgResults = true; } // Copied from StandardRequestHandler... perhaps it should be added to doStandardDebug? if (dbg == true) { try { NamedList<Object> dbgInfo = SolrPluginUtils.doStandardDebug(req, q, query, sltDocs.docList, dbgQuery, dbgResults); if (null != dbgInfo) { if (null != filters) { dbgInfo.add("filter_queries", req.getParams().getParams(CommonParams.FQ)); List<String> fqs = new ArrayList<String>(filters.size()); for (Query fq : filters) { fqs.add(QueryParsing.toString(fq, req.getSchema())); } dbgInfo.add("parsed_filter_queries", fqs); } rsp.add("debug", dbgInfo); } } catch (Exception e) { SolrException.log(SolrCore.log, "Exception during debug", e); rsp.add("exception_during_debug", SolrException.toStr(e)); } } }
From source file:com.sn.solr.plugin.rank.RankEngine.java
License:Apache License
/** * Provides implementation for Dense ranking ["1223"] as identified by the * {@link RankStrategy#LEGACY_DENSE} the difference is that this * implementation is computed without using facet results so this will * noticeably slower than computing rank based on facets * use {@link RankStrategy#DENSE}. Besides this implementation might cause * lot of cache evictions putting stress on memory. * * @see #computeDenseRank(List)/* www . java2s . c om*/ * * @param pairList List of {@link Pair} objects that holds the value of rank * field & respective count. */ @Deprecated public static Map<String, Number> computeLegacyDenseRank(ResponseBuilder rb, String idField, String rankField) throws IOException { SolrIndexSearcher searcher = rb.req.getSearcher(); SolrParams params = rb.req.getParams();// .getParams(FacetParams.FACET_FIELD); String _start = params.get(CommonParams.START); String _rows = params.get(CommonParams.ROWS); int start = 0; int rows = 10; if (_start != null & AppHelper.isInteger(_start)) start = new Integer(_start); if (_rows != null & AppHelper.isInteger(_rows)) rows = new Integer(_rows); LOG.info("Computing rank using strategy: {}", RankStrategy.ORDINAL.getDescription()); FieldSelector fs = new MapFieldSelector(new String[] { idField, rankField }); Map<String, Number> rankMap = new HashMap<String, Number>(); DocList docs = searcher.getDocList(rb.getQuery(), rb.getFilters(), rb.getSortSpec().getSort(), 0, start + rows, 0); int denseRank = 1; int _CurrScore = 0; int _PrevScore = 0; int i = 0; for (DocIterator it = docs.iterator(); it.hasNext();) { Document doc = searcher.doc(it.nextDoc(), fs); _CurrScore = new Integer(doc.get(rankField)); if (i == 0) { _PrevScore = _CurrScore; } if (_PrevScore != _CurrScore) { _PrevScore = _CurrScore; denseRank++; } if (i >= start) { rankMap.put(doc.get(idField), denseRank); } i++; } return rankMap; }
From source file:org.dfdeshom.solr.mlt.MoreLikeThisHandler.java
License:Apache License
@Override public void handleRequestBody(SolrQueryRequest req, SolrQueryResponse rsp) throws Exception { SolrParams params = req.getParams(); // Set field flags ReturnFields returnFields = new SolrReturnFields(req); rsp.setReturnFields(returnFields);// w w w . j av a 2 s .c o m int flags = 0; if (returnFields.wantsScore()) { flags |= SolrIndexSearcher.GET_SCORES; } String defType = params.get(QueryParsing.DEFTYPE, QParserPlugin.DEFAULT_QTYPE); String q = params.get(CommonParams.Q); Query query = null; SortSpec sortSpec = null; List<Query> filters = null; QParser parser = null; try { if (q != null) { parser = QParser.getParser(q, defType, req); query = parser.getQuery(); sortSpec = parser.getSort(true); } String[] fqs = req.getParams().getParams(CommonParams.FQ); if (fqs != null && fqs.length != 0) { filters = new ArrayList<Query>(); for (String fq : fqs) { if (fq != null && fq.trim().length() != 0) { QParser fqp = QParser.getParser(fq, null, req); filters.add(fqp.getQuery()); } } } } catch (SyntaxError e) { throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, e); } SolrIndexSearcher searcher = req.getSearcher(); MoreLikeThisHelper mlt = new MoreLikeThisHelper(params, searcher); // Hold on to the interesting terms if relevant TermStyle termStyle = TermStyle.get(params.get(MoreLikeThisParams.INTERESTING_TERMS)); List<InterestingTerm> interesting = (termStyle == TermStyle.NONE) ? null : new ArrayList<InterestingTerm>(mlt.mlt.getMaxQueryTerms()); DocListAndSet mltDocs = null; // Parse Required Params // This will either have a single Reader or valid query Reader reader = null; try { if (q == null || q.trim().length() < 1) { Iterable<ContentStream> streams = req.getContentStreams(); if (streams != null) { Iterator<ContentStream> iter = streams.iterator(); if (iter.hasNext()) { reader = iter.next().getReader(); } if (iter.hasNext()) { throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "MoreLikeThis does not support multiple ContentStreams"); } } } int start = params.getInt(CommonParams.START, 0); int rows = params.getInt(CommonParams.ROWS, 10); // Find documents MoreLikeThis - either with a reader or a query // -------------------------------------------------------------------------------- if (reader != null) { mltDocs = mlt.getMoreLikeThis(reader, sortSpec.getSort(), start, rows, filters, interesting, flags); } else if (q != null) { // Matching options boolean includeMatch = params.getBool(MoreLikeThisParams.MATCH_INCLUDE, true); int matchOffset = params.getInt(MoreLikeThisParams.MATCH_OFFSET, 0); // Find the base match DocList match = searcher.getDocList(query, null, null, matchOffset, 1, flags); // only get the first one... if (includeMatch) { rsp.add("match", match); } // This is an iterator, but we only handle the first match DocIterator iterator = match.iterator(); if (iterator.hasNext()) { // do a MoreLikeThis query for each document in results int id = iterator.nextDoc(); mltDocs = mlt.getMoreLikeThis(parser, id, sortSpec.getSort(), start, rows, filters, interesting, flags); } } else { throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "MoreLikeThis requires either a query (?q=) or text to find similar documents."); } } finally { if (reader != null) { reader.close(); } } if (mltDocs == null) { mltDocs = new DocListAndSet(); // avoid NPE } rsp.add("response", mltDocs.docList); if (interesting != null) { if (termStyle == TermStyle.DETAILS) { NamedList<Float> it = new NamedList<Float>(); for (InterestingTerm t : interesting) { it.add(t.term.toString(), t.boost); } rsp.add("interestingTerms", it); } else { List<String> it = new ArrayList<String>(interesting.size()); for (InterestingTerm t : interesting) { it.add(t.term.text()); } rsp.add("interestingTerms", it); } } // maybe facet the results if (params.getBool(FacetParams.FACET, false)) { if (mltDocs.docSet == null) { rsp.add("facet_counts", null); } else { SimpleFacets f = new SimpleFacets(req, mltDocs.docSet, params); rsp.add("facet_counts", f.getFacetCounts()); } } boolean dbg = req.getParams().getBool(CommonParams.DEBUG_QUERY, false); boolean dbgQuery = false, dbgResults = false; if (dbg == false) {//if it's true, we are doing everything anyway. String[] dbgParams = req.getParams().getParams(CommonParams.DEBUG); if (dbgParams != null) { for (int i = 0; i < dbgParams.length; i++) { if (dbgParams[i].equals(CommonParams.QUERY)) { dbgQuery = true; } else if (dbgParams[i].equals(CommonParams.RESULTS)) { dbgResults = true; } } } } else { dbgQuery = true; dbgResults = true; } // Copied from StandardRequestHandler... perhaps it should be added to doStandardDebug? if (dbg == true) { try { NamedList<Object> dbgInfo = SolrPluginUtils.doStandardDebug(req, q, mlt.getRawMLTQuery(), mltDocs.docList, dbgQuery, dbgResults); if (null != dbgInfo) { if (null != filters) { dbgInfo.add("filter_queries", req.getParams().getParams(CommonParams.FQ)); List<String> fqs = new ArrayList<String>(filters.size()); for (Query fq : filters) { fqs.add(QueryParsing.toString(fq, req.getSchema())); } dbgInfo.add("parsed_filter_queries", fqs); } rsp.add("debug", dbgInfo); } } catch (Exception e) { SolrException.log(SolrCore.log, "Exception during debug", e); rsp.add("exception_during_debug", SolrException.toStr(e)); } } }
From source file:org.dice.solrenhancements.morelikethis.DiceMoreLikeThisHandler.java
License:Apache License
private MLTResult getMoreLikeTheseFromQuery(SolrQueryResponse rsp, SolrParams params, int flags, String q, Query query, SortSpec sortSpec, List<Query> targetFqFilters, List<Query> mltFqFilters, SolrIndexSearcher searcher, MoreLikeThisHelper mlt, int start, int rows) throws IOException, SyntaxError { boolean includeMatch = params.getBool(MoreLikeThisParams.MATCH_INCLUDE, true); int matchOffset = params.getInt(MoreLikeThisParams.MATCH_OFFSET, 0); // Find the base match DocList match = searcher.getDocList(query, targetFqFilters, null, matchOffset, 10000, flags); // only get the first one... if (match.matches() == 0) { throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, String.format("MoreLikeThis was unable to find any documents matching the query: '%s'.", q)); }//from ww w . ja v a 2s . co m if (includeMatch) { rsp.add("match", match); } // This is an iterator, but we only handle the first match DocIterator iterator = match.iterator(); if (iterator.hasNext()) { // do a MoreLikeThis query for each document in results return mlt.getMoreLikeTheseFromDocs(iterator, start, rows, mltFqFilters, flags, sortSpec.getSort()); } return null; }
From source file:org.dice.solrenhancements.unsupervisedfeedback.DiceUnsupervisedFeedbackHandler.java
License:Apache License
private DocListAndSet expandQueryAndReExecute(SolrQueryResponse rsp, SolrParams params, int maxDocumentsToMatch, int flags, String q, Query seedQuery, SortSpec sortSpec, List<Query> targetFqFilters, List<Query> mltFqFilters, SolrIndexSearcher searcher, UnsupervisedFeedbackHelper uff, List<InterestingTerm> interesting, DocListAndSet mltDocs, int start, int rows) throws IOException, SyntaxError { boolean includeMatch = params.getBool(UnsupervisedFeedbackParams.MATCH_INCLUDE, true); int matchOffset = params.getInt(UnsupervisedFeedbackParams.MATCH_OFFSET, 0); // Find the base match DocList match = searcher.getDocList(seedQuery, targetFqFilters, null, matchOffset, maxDocumentsToMatch, flags); // only get the first one... if (match.matches() == 0) { throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, String.format( "Unsupervised feedback handler was unable to find any documents matching the seed query: '%s'.", q));//from w ww. j a v a 2 s .co m } if (includeMatch) { rsp.add("match", match); } // This is an iterator, but we only handle the first match DocIterator iterator = match.iterator(); if (iterator.hasNext()) { // do a MoreLikeThis query for each document in results mltDocs = uff.expandQueryAndReExecute(iterator, seedQuery, start, rows, mltFqFilters, interesting, flags, sortSpec.getSort()); } return mltDocs; }