private void handleFilterExclusions() throws IOException {
List<String> excludeTags = freq.domain.excludeTags;
if (excludeTags == null || excludeTags.size() == 0) {
return;
}
// TODO: somehow remove responsebuilder dependency
ResponseBuilder rb = SolrRequestInfo.getRequestInfo().getResponseBuilder();
Map tagMap = (Map) rb.req.getContext().get("tags");
if (tagMap == null) {
// no filters were tagged
return;
}
IdentityHashMap<Query, Boolean> excludeSet = new IdentityHashMap<>();
for (String excludeTag : excludeTags) {
Object olst = tagMap.get(excludeTag);
// tagMap has entries of List<String,List<QParser>>, but subject to change in the future
if (!(olst instanceof Collection)) continue;
for (Object o : (Collection<?>) olst) {
if (!(o instanceof QParser)) continue;
QParser qp = (QParser) o;
try {
excludeSet.put(qp.getQuery(), Boolean.TRUE);
} catch (SyntaxError syntaxError) {
// This should not happen since we should only be retrieving a previously parsed query
throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, syntaxError);
}
}
}
if (excludeSet.size() == 0) return;
List<Query> qlist = new ArrayList<>();
// add the base query
if (!excludeSet.containsKey(rb.getQuery())) {
qlist.add(rb.getQuery());
}
// add the filters
if (rb.getFilters() != null) {
for (Query q : rb.getFilters()) {
if (!excludeSet.containsKey(q)) {
qlist.add(q);
}
}
}
// now walk back up the context tree
// TODO: we lose parent exclusions...
for (FacetContext curr = fcontext; curr != null; curr = curr.parent) {
if (curr.filter != null) {
qlist.add(curr.filter);
}
}
// recompute the base domain
fcontext.base = fcontext.searcher.getDocSet(qlist);
}
protected DocSet computeDocSet(DocSet baseDocSet, List<String> excludeTagList)
throws SyntaxError, IOException {
Map<?, ?> tagMap = (Map<?, ?>) req.getContext().get("tags");
// rb can be null if facets are being calculated from a RequestHandler e.g. MoreLikeThisHandler
if (tagMap == null || rb == null) {
return baseDocSet;
}
IdentityHashMap<Query, Boolean> excludeSet = new IdentityHashMap<>();
for (String excludeTag : excludeTagList) {
Object olst = tagMap.get(excludeTag);
// tagMap has entries of List<String,List<QParser>>, but subject to change in the future
if (!(olst instanceof Collection)) continue;
for (Object o : (Collection<?>) olst) {
if (!(o instanceof QParser)) continue;
QParser qp = (QParser) o;
excludeSet.put(qp.getQuery(), Boolean.TRUE);
}
}
if (excludeSet.size() == 0) return baseDocSet;
List<Query> qlist = new ArrayList<>();
// add the base query
if (!excludeSet.containsKey(rb.getQuery())) {
qlist.add(rb.getQuery());
}
// add the filters
if (rb.getFilters() != null) {
for (Query q : rb.getFilters()) {
if (!excludeSet.containsKey(q)) {
qlist.add(q);
}
}
}
// get the new base docset for this facet
DocSet base = searcher.getDocSet(qlist);
if (rb.grouping() && rb.getGroupingSpec().isTruncateGroups()) {
Grouping grouping = new Grouping(searcher, null, rb.getQueryCommand(), false, 0, false);
grouping.setWithinGroupSort(rb.getGroupingSpec().getSortWithinGroup());
if (rb.getGroupingSpec().getFields().length > 0) {
grouping.addFieldCommand(rb.getGroupingSpec().getFields()[0], req);
} else if (rb.getGroupingSpec().getFunctions().length > 0) {
grouping.addFunctionCommand(rb.getGroupingSpec().getFunctions()[0], req);
} else {
return base;
}
AbstractAllGroupHeadsCollector allGroupHeadsCollector =
grouping.getCommands().get(0).createAllGroupCollector();
searcher.search(base.getTopFilter(), allGroupHeadsCollector);
return new BitDocSet(allGroupHeadsCollector.retrieveGroupHeads(searcher.maxDoc()));
} else {
return base;
}
}
@Override
public void finishStage(ResponseBuilder rb) {
SolrParams params = rb.req.getParams();
if (!params.getBool(COMPONENT_NAME, false)
|| !params.getBool(ClusteringParams.USE_SEARCH_RESULTS, false)) {
return;
}
if (rb.stage == ResponseBuilder.STAGE_GET_FIELDS) {
SearchClusteringEngine engine = getSearchClusteringEngine(rb);
if (engine != null) {
SolrDocumentList solrDocList = (SolrDocumentList) rb.rsp.getValues().get("response");
// TODO: Currently, docIds is set to null in distributed environment.
// This causes CarrotParams.PRODUCE_SUMMARY doesn't work.
// To work CarrotParams.PRODUCE_SUMMARY under distributed mode, we can choose either one of:
// (a) In each shard, ClusteringComponent produces summary and finishStage()
// merges these summaries.
// (b) Adding doHighlighting(SolrDocumentList, ...) method to SolrHighlighter and
// making SolrHighlighter uses "external text" rather than stored values to produce
// snippets.
Map<SolrDocument, Integer> docIds = null;
Object clusters = engine.cluster(rb.getQuery(), solrDocList, docIds, rb.req);
rb.rsp.add("clusters", clusters);
} else {
String name = getClusteringEngineName(rb);
log.warn("No engine for: " + name);
}
}
}
@Override
public void process(ResponseBuilder rb) throws IOException {
SolrParams params = rb.req.getParams();
if (!params.getBool(COMPONENT_NAME, false)) {
return;
}
String name = getClusteringEngineName(rb);
boolean useResults = params.getBool(ClusteringParams.USE_SEARCH_RESULTS, false);
if (useResults == true) {
SearchClusteringEngine engine = getSearchClusteringEngine(rb);
if (engine != null) {
DocListAndSet results = rb.getResults();
Map<SolrDocument, Integer> docIds = Maps.newHashMapWithExpectedSize(results.docList.size());
SolrDocumentList solrDocList =
SolrPluginUtils.docListToSolrDocumentList(
results.docList, rb.req.getSearcher(), engine.getFieldsToLoad(rb.req), docIds);
Object clusters = engine.cluster(rb.getQuery(), solrDocList, docIds, rb.req);
rb.rsp.add("clusters", clusters);
} else {
log.warn("No engine for: " + name);
}
}
boolean useCollection = params.getBool(ClusteringParams.USE_COLLECTION, false);
if (useCollection == true) {
DocumentClusteringEngine engine = documentClusteringEngines.get(name);
if (engine != null) {
boolean useDocSet = params.getBool(ClusteringParams.USE_DOC_SET, false);
NamedList<?> nl = null;
// TODO: This likely needs to be made into a background task that runs in an executor
if (useDocSet == true) {
nl = engine.cluster(rb.getResults().docSet, params);
} else {
nl = engine.cluster(params);
}
rb.rsp.add("clusters", nl);
} else {
log.warn("No engine for " + name);
}
}
}
public void process(ResponseBuilder rb) throws IOException {
SolrParams params = rb.req.getParams();
if (!params.getBool(COMPONENT_NAME, false)) {
return;
}
String name = params.get(ClusteringParams.ENGINE_NAME, ClusteringEngine.DEFAULT_ENGINE_NAME);
boolean useResults = params.getBool(ClusteringParams.USE_SEARCH_RESULTS, false);
if (useResults == true) {
SearchClusteringEngine engine = searchClusteringEngines.get(name);
if (engine != null) {
DocListAndSet results = rb.getResults();
Object clusters = engine.cluster(rb.getQuery(), results.docList, rb.req);
rb.rsp.add("clusters", clusters);
} else {
log.warn("No engine for: " + name);
}
}
boolean useCollection = params.getBool(ClusteringParams.USE_COLLECTION, false);
if (useCollection == true) {
DocumentClusteringEngine engine = documentClusteringEngines.get(name);
if (engine != null) {
boolean useDocSet = params.getBool(ClusteringParams.USE_DOC_SET, false);
NamedList nl = null;
// TODO: This likely needs to be made into a background task that runs in an executor
if (useDocSet == true) {
nl = engine.cluster(rb.getResults().docSet, params);
} else {
nl = engine.cluster(params);
}
rb.rsp.add("clusters", nl);
} else {
log.warn("No engine for " + name);
}
}
}
