/** User runs a query and aggregates facets by summing their association values. */
private List<FacetResult> sumAssociations() throws IOException {
DirectoryReader indexReader = DirectoryReader.open(indexDir);
IndexSearcher searcher = new IndexSearcher(indexReader);
TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
FacetsCollector fc = new FacetsCollector();
// MatchAllDocsQuery is for "browsing" (counts facets
// for all non-deleted docs in the index); normally
// you'd use a "normal" query:
FacetsCollector.search(searcher, new MatchAllDocsQuery(), 10, fc);
Facets tags = new TaxonomyFacetSumIntAssociations("$tags", taxoReader, config, fc);
Facets genre = new TaxonomyFacetSumFloatAssociations("$genre", taxoReader, config, fc);
// Retrieve results
List<FacetResult> results = new ArrayList<>();
results.add(tags.getTopChildren(10, "tags"));
results.add(genre.getTopChildren(10, "genre"));
indexReader.close();
taxoReader.close();
return results;
}
private List<FacetResult> facetsOnly() throws IOException {
FacetsCollector fc = new FacetsCollector();
// MatchAllDocsQuery is for "browsing" (counts facets
// for all non-deleted docs in the index); normally
// you'd use a "normal" query:
TopDocs docs = FacetsCollector.search(indexSearcher, new MatchAllDocsQuery(), 10, fc);
// indexSearcher.search(new MatchAllDocsQuery(), null /*Filter */, fc);
// Retrieve results
List<FacetResult> results = new ArrayList<FacetResult>();
System.out.println("results size: " + results.size());
// Count both "Publish Date" and "Author" dimensions
Facets facets = new FastTaxonomyFacetCounts(taxoReader, taxoConfig, fc);
results.add(facets.getTopChildren(20, "objects"));
// results.add(facets.getTopChildren(10, "Publish Date"));
List<FacetResult> result = facets.getAllDims(10);
// indexReader.close();
// taxoReader.close();
return result;
}
private TopDocsName facetsWithSearch(String query) throws IOException, ParseException {
FacetsCollector fc = new FacetsCollector();
ComplexPhraseQueryParser queryParser =
new ComplexPhraseQueryParser("content", new StandardAnalyzer());
Query luceneQuery = queryParser.parse(query);
TopDocs docs = FacetsCollector.search(indexSearcher, luceneQuery, 400, fc);
List<TopDocName> topDocNames = new ArrayList<TopDocName>();
for (ScoreDoc scoreDoc : docs.scoreDocs) {
String filename = indexSearcher.doc(scoreDoc.doc).get("filename");
float score = scoreDoc.score;
TopDocName docname = new TopDocName(score, filename.substring(0, filename.length()));
topDocNames.add(docname);
}
List<FacetResult> facetResults = new ArrayList<FacetResult>();
// Count both "Publish Date" and "Author" dimensions
Facets facets = new FastTaxonomyFacetCounts(taxoReader, taxoConfig, fc);
facetResults.add(facets.getTopChildren(11, "person"));
facetResults.add(facets.getTopChildren(20, "objects"));
TopDocsName results = new TopDocsName(1.0f, topDocNames, facetResults);
return results;
}
/**
* TODO: Facets are based on DocValue fields and not on separate taxonomy index, this will be
* slower but will take much less memory. Provide option to support hierarchical facets support
* and add taxonomy index based faceting.
*
* @param user
* @param req
* @return
* @throws ParseException
* @throws IOException
*/
public Map<Long, Integer> getTrend(User user, Request req) throws ParseException, IOException {
user = userService.validate(user);
Facets facets = runFacetedSearch(req, user);
FacetResult result = facets.getTopChildren(10, req.trendField.name());
Map<Long, Integer> countsByRes = new LinkedHashMap<>();
for (LabelAndValue labelAndValue : result.labelValues) {
countsByRes.put(Long.valueOf(labelAndValue.label), labelAndValue.value.intValue());
}
return countsByRes;
}
/** User runs a query and counts facets. */
public FacetResult search() throws IOException {
FacetsCollector fc = new FacetsCollector();
searcher.search(new MatchAllDocsQuery(), fc);
Facets facets =
new DoubleRangeFacetCounts(
"field",
getDistanceValueSource(),
fc,
getBoundingBoxQuery(ORIGIN_LATITUDE, ORIGIN_LONGITUDE, 10.0),
ONE_KM,
TWO_KM,
FIVE_KM,
TEN_KM);
return facets.getTopChildren(10, "field");
}
@Test
public void testAllCounts() throws Exception {
DirectoryReader indexReader = DirectoryReader.open(indexDir);
TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
IndexSearcher searcher = newSearcher(indexReader);
FacetsCollector sfc = new FacetsCollector();
searcher.search(new MatchAllDocsQuery(), sfc);
Facets facets = getTaxonomyFacetCounts(taxoReader, getConfig(), sfc);
FacetResult result = facets.getTopChildren(NUM_CHILDREN_CP_A, CP_A);
assertEquals(-1, result.value.intValue());
int prevValue = Integer.MAX_VALUE;
for (LabelAndValue labelValue : result.labelValues) {
assertEquals(allExpectedCounts.get(CP_A + "/" + labelValue.label), labelValue.value);
assertTrue(
"wrong sort order of sub results: labelValue.value="
+ labelValue.value
+ " prevValue="
+ prevValue,
labelValue.value.intValue() <= prevValue);
prevValue = labelValue.value.intValue();
}
result = facets.getTopChildren(NUM_CHILDREN_CP_B, CP_B);
assertEquals(allExpectedCounts.get(CP_B), result.value);
prevValue = Integer.MAX_VALUE;
for (LabelAndValue labelValue : result.labelValues) {
assertEquals(allExpectedCounts.get(CP_B + "/" + labelValue.label), labelValue.value);
assertTrue(
"wrong sort order of sub results: labelValue.value="
+ labelValue.value
+ " prevValue="
+ prevValue,
labelValue.value.intValue() <= prevValue);
prevValue = labelValue.value.intValue();
}
IOUtils.close(indexReader, taxoReader);
}
/** User drills down on 'tags/solr'. */
private FacetResult drillDown() throws IOException {
DirectoryReader indexReader = DirectoryReader.open(indexDir);
IndexSearcher searcher = new IndexSearcher(indexReader);
TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
// Passing no baseQuery means we drill down on all
// documents ("browse only"):
DrillDownQuery q = new DrillDownQuery(config);
// Now user drills down on Publish Date/2010:
q.add("tags", "solr");
FacetsCollector fc = new FacetsCollector();
FacetsCollector.search(searcher, q, 10, fc);
// Retrieve results
Facets facets = new TaxonomyFacetSumFloatAssociations("$genre", taxoReader, config, fc);
FacetResult result = facets.getTopChildren(10, "genre");
indexReader.close();
taxoReader.close();
return result;
}
private TopDocsName drillDown(Map<String, String> facets_value) throws IOException {
// Passing no baseQuery means we drill down on all
// documents ("browse only"):
DrillDownQuery q = new DrillDownQuery(taxoConfig);
for (Map.Entry<String, String> entry : facets_value.entrySet()) {
q.add(entry.getKey(), entry.getValue());
}
FacetsCollector fc = new FacetsCollector();
TopDocs docs = FacetsCollector.search(indexSearcher, q, 100000, fc);
List<TopDocName> topDocNames = new ArrayList<TopDocName>();
for (ScoreDoc scoreDoc : docs.scoreDocs) {
String filename = indexSearcher.doc(scoreDoc.doc).get("filename");
String videoID = indexSearcher.doc(scoreDoc.doc).get("videoID");
String frameNumber = indexSearcher.doc(scoreDoc.doc).get("frameNumber");
float score = scoreDoc.score;
TopDocName docname = new TopDocName(score, filename, videoID, frameNumber);
topDocNames.add(docname);
}
List<FacetResult> facetResults = new ArrayList<FacetResult>();
// Count both "Publish Date" and "Author" dimensions
Facets facets = new FastTaxonomyFacetCounts(taxoReader, taxoConfig, fc);
facetResults.add(facets.getTopChildren(11, "person"));
facetResults.add(facets.getTopChildren(20, "objects"));
facetResults.add(facets.getTopChildren(200, "background"));
TopDocsName results = new TopDocsName(1.0f, topDocNames, facetResults);
return results;
}
@Test
public void testNoParents() throws Exception {
DirectoryReader indexReader = DirectoryReader.open(indexDir);
TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
IndexSearcher searcher = newSearcher(indexReader);
FacetsCollector sfc = new FacetsCollector();
searcher.search(new MatchAllDocsQuery(), sfc);
Facets facets = getTaxonomyFacetCounts(taxoReader, getConfig(), sfc);
FacetResult result = facets.getTopChildren(NUM_CHILDREN_CP_C, CP_C);
assertEquals(allExpectedCounts.get(CP_C), result.value);
for (LabelAndValue labelValue : result.labelValues) {
assertEquals(allExpectedCounts.get(CP_C + "/" + labelValue.label), labelValue.value);
}
result = facets.getTopChildren(NUM_CHILDREN_CP_D, CP_D);
assertEquals(allExpectedCounts.get(CP_C), result.value);
for (LabelAndValue labelValue : result.labelValues) {
assertEquals(allExpectedCounts.get(CP_D + "/" + labelValue.label), labelValue.value);
}
IOUtils.close(indexReader, taxoReader);
}
@Test
public void testBigNumResults() throws Exception {
DirectoryReader indexReader = DirectoryReader.open(indexDir);
TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
IndexSearcher searcher = newSearcher(indexReader);
FacetsCollector sfc = new FacetsCollector();
searcher.search(new MatchAllDocsQuery(), sfc);
Facets facets = getTaxonomyFacetCounts(taxoReader, getConfig(), sfc);
FacetResult result = facets.getTopChildren(Integer.MAX_VALUE, CP_A);
assertEquals(-1, result.value.intValue());
for (LabelAndValue labelValue : result.labelValues) {
assertEquals(allExpectedCounts.get(CP_A + "/" + labelValue.label), labelValue.value);
}
result = facets.getTopChildren(Integer.MAX_VALUE, CP_B);
assertEquals(allExpectedCounts.get(CP_B), result.value);
for (LabelAndValue labelValue : result.labelValues) {
assertEquals(allExpectedCounts.get(CP_B + "/" + labelValue.label), labelValue.value);
}
IOUtils.close(indexReader, taxoReader);
}
@Test
public void testDifferentNumResults() throws Exception {
// test the collector w/ FacetRequests and different numResults
DirectoryReader indexReader = DirectoryReader.open(indexDir);
TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
IndexSearcher searcher = newSearcher(indexReader);
FacetsCollector sfc = new FacetsCollector();
TermQuery q = new TermQuery(A);
searcher.search(q, sfc);
Facets facets = getTaxonomyFacetCounts(taxoReader, getConfig(), sfc);
FacetResult result = facets.getTopChildren(NUM_CHILDREN_CP_A, CP_A);
assertEquals(-1, result.value.intValue());
for (LabelAndValue labelValue : result.labelValues) {
assertEquals(termExpectedCounts.get(CP_A + "/" + labelValue.label), labelValue.value);
}
result = facets.getTopChildren(NUM_CHILDREN_CP_B, CP_B);
assertEquals(termExpectedCounts.get(CP_B), result.value);
for (LabelAndValue labelValue : result.labelValues) {
assertEquals(termExpectedCounts.get(CP_B + "/" + labelValue.label), labelValue.value);
}
IOUtils.close(indexReader, taxoReader);
}