/// <summary>User runs a query and counts facets.</summary>
private IList <FacetResult> FacetsWithSearch()
{
using (DirectoryReader indexReader = DirectoryReader.Open(indexDir))
using (TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir))
{
IndexSearcher searcher = new IndexSearcher(indexReader);
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);
// Retrieve results
IList <FacetResult> results = new List <FacetResult>();
// Count both "Publish Date" and "Author" dimensions
Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, fc);
results.Add(facets.GetTopChildren(10, "Author"));
results.Add(facets.GetTopChildren(10, "Publish Date"));
return(results);
}// Disposes indexReader and taxoReader
}
/// <summary>User runs a query and counts facets only without collecting the matching documents.</summary>
private IList <FacetResult> FacetsOnly()
{
using DirectoryReader indexReader = DirectoryReader.Open(indexDir);
using TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
IndexSearcher searcher = new IndexSearcher(indexReader);
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:
searcher.Search(new MatchAllDocsQuery(), null /*Filter */, fc);
Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, fc);
// Retrieve results
IList <FacetResult> results = new List <FacetResult>
{
// Count both "Publish Date" and "Author" dimensions
facets.GetTopChildren(10, "Author"),
facets.GetTopChildren(10, "Publish Date")
};
return(results);
}
/// <summary>
/// User drills down on 'Publish Date/2010', and we
/// return facets for 'Author'
/// </summary>
private FacetResult DrillDown()
{
using DirectoryReader indexReader = DirectoryReader.Open(indexDir);
using TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
IndexSearcher searcher = new IndexSearcher(indexReader);
// 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("Publish Date", "2010");
FacetsCollector fc = new FacetsCollector();
FacetsCollector.Search(searcher, q, 10, fc);
// Retrieve results
Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, fc);
FacetResult result = facets.GetTopChildren(10, "Author");
return(result);
}
public virtual void TestRandomSampling()
{
Directory dir = NewDirectory();
Directory taxoDir = NewDirectory();
DirectoryTaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir);
RandomIndexWriter writer = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, dir);
FacetsConfig config = new FacetsConfig();
int numDocs = AtLeast(10000);
for (int i = 0; i < numDocs; i++)
{
Document doc = new Document();
doc.Add(new StringField("EvenOdd", (i % 2 == 0) ? "even" : "odd", Store.NO));
doc.Add(new FacetField("iMod10", Convert.ToString(i % 10, CultureInfo.InvariantCulture)));
writer.AddDocument(config.Build(taxoWriter, doc));
}
Random random = Random;
// NRT open
IndexSearcher searcher = NewSearcher(writer.GetReader());
var taxoReader = new DirectoryTaxonomyReader(taxoWriter);
IOUtils.Dispose(writer, taxoWriter);
// Test empty results
RandomSamplingFacetsCollector collectRandomZeroResults = new RandomSamplingFacetsCollector(numDocs / 10, random.NextInt64());
// There should be no divisions by zero
searcher.Search(new TermQuery(new Term("EvenOdd", "NeverMatches")), collectRandomZeroResults);
// There should be no divisions by zero and no null result
Assert.IsNotNull(collectRandomZeroResults.GetMatchingDocs());
// There should be no results at all
foreach (MatchingDocs doc in collectRandomZeroResults.GetMatchingDocs())
{
Assert.AreEqual(0, doc.TotalHits);
}
// Now start searching and retrieve results.
// Use a query to select half of the documents.
TermQuery query = new TermQuery(new Term("EvenOdd", "even"));
// there will be 5 facet values (0, 2, 4, 6 and 8), as only the even (i %
// 10) are hits.
// there is a REAL small chance that one of the 5 values will be missed when
// sampling.
// but is that 0.8 (chance not to take a value) ^ 2000 * 5 (any can be
// missing) ~ 10^-193
// so that is probably not going to happen.
int maxNumChildren = 5;
RandomSamplingFacetsCollector random100Percent = new RandomSamplingFacetsCollector(numDocs, random.NextInt64()); // no sampling
RandomSamplingFacetsCollector random10Percent = new RandomSamplingFacetsCollector(numDocs / 10, random.NextInt64()); // 10 % of total docs, 20% of the hits
FacetsCollector fc = new FacetsCollector();
searcher.Search(query, MultiCollector.Wrap(fc, random100Percent, random10Percent));
FastTaxonomyFacetCounts random10FacetCounts = new FastTaxonomyFacetCounts(taxoReader, config, random10Percent);
FastTaxonomyFacetCounts random100FacetCounts = new FastTaxonomyFacetCounts(taxoReader, config, random100Percent);
FastTaxonomyFacetCounts exactFacetCounts = new FastTaxonomyFacetCounts(taxoReader, config, fc);
FacetResult random10Result = random10Percent.AmortizeFacetCounts(random10FacetCounts.GetTopChildren(10, "iMod10"), config, searcher);
FacetResult random100Result = random100FacetCounts.GetTopChildren(10, "iMod10");
FacetResult exactResult = exactFacetCounts.GetTopChildren(10, "iMod10");
Assert.AreEqual(random100Result, exactResult);
// we should have five children, but there is a small chance we have less.
// (see above).
Assert.IsTrue(random10Result.ChildCount <= maxNumChildren);
// there should be one child at least.
Assert.IsTrue(random10Result.ChildCount >= 1);
// now calculate some statistics to determine if the sampled result is 'ok'.
// because random sampling is used, the results will vary each time.
int sum = 0;
foreach (LabelAndValue lav in random10Result.LabelValues)
{
sum += (int)lav.Value;
}
float mu = (float)sum / (float)maxNumChildren;
float variance = 0;
foreach (LabelAndValue lav in random10Result.LabelValues)
{
variance += (float)Math.Pow((mu - (int)lav.Value), 2);
}
variance = variance / maxNumChildren;
float sigma = (float)Math.Sqrt(variance);
// we query only half the documents and have 5 categories. The average
// number of docs in a category will thus be the total divided by 5*2
float targetMu = numDocs / (5.0f * 2.0f);
// the average should be in the range and the standard deviation should not
// be too great
Assert.IsTrue(sigma < 200);
Assert.IsTrue(targetMu - 3 * sigma < mu && mu < targetMu + 3 * sigma);
IOUtils.Dispose(searcher.IndexReader, taxoReader, dir, taxoDir);
}
public async Task <SearchResult <T> > SearchAsync(SearchQuery queryDefinition, CancellationToken cancellationToken = default)
{
using (await writerLock.ReaderLockAsync(cancellationToken))
{
var result = new SearchResult <T>();
List <T> hits = new List <T>();
using (var writer = getWriter())
{
Query query = new MatchAllDocsQuery();
// Term queries
if (queryDefinition.TermQueries.Any())
{
var phraseQuery = new MultiPhraseQuery();
foreach (var termQuery in queryDefinition.TermQueries)
{
phraseQuery.Add(
termQuery.value
.Split(" ".ToCharArray(), StringSplitOptions.RemoveEmptyEntries)
.Select(phrase => new Term(termQuery.field, phrase.ToLower()))
.ToArray()
);
}
query = phraseQuery;
}
var reader = writer.DocsWriter.GetReader(applyAllDeletes: true);
var searcher = new IndexSearcher(reader);
var luceneResult = searcher.Search(query, queryDefinition.Limit);
foreach (var doc in luceneResult.ScoreDocs)
{
var foundDoc = searcher.Doc(doc.Doc);
hits.Add(await inflateDocument(foundDoc));
}
result.TotalHits = luceneResult.TotalHits;
result.Hits = hits;
// Facets
if (queryDefinition.Facets.Any())
{
FacetsConfig facetsConfig = new FacetsConfig();
FacetsCollector fc = new FacetsCollector();
FacetsCollector.Search(searcher, query, queryDefinition.FacetMax, fc);
using (var taxonomyReader = new DirectoryTaxonomyReader(FSDirectory.Open(Path.Combine(options.IndexPath, indexType, "taxonomy"))))
{
var facets = new FastTaxonomyFacetCounts(taxonomyReader, facetsConfig, fc);
foreach (var facet in queryDefinition.Facets)
{
var facetGroup = new FacetGroup {
Field = facet
};
facetGroup.Facets =
facets.GetTopChildren(queryDefinition.FacetMax, facet).LabelValues
.Select(x => new Facet {
Key = x.Label, Count = (long)x.Value
})
.ToArray();
result.FacetGroups.Add(facetGroup);
}
}
}
}
return(result);
}
}
public virtual void TestRandomSampling()
{
Directory dir = NewDirectory();
Directory taxoDir = NewDirectory();
DirectoryTaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir);
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);
FacetsConfig config = new FacetsConfig();
int numDocs = AtLeast(10000);
for (int i = 0; i < numDocs; i++)
{
Document doc = new Document();
doc.Add(new StringField("EvenOdd", (i % 2 == 0) ? "even" : "odd", Store.NO));
doc.Add(new FacetField("iMod10", Convert.ToString(i % 10)));
writer.AddDocument(config.Build(taxoWriter, doc));
}
Random random = Random();
// NRT open
IndexSearcher searcher = NewSearcher(writer.Reader);
var taxoReader = new DirectoryTaxonomyReader(taxoWriter);
IOUtils.Close(writer, taxoWriter);
// Test empty results
RandomSamplingFacetsCollector collectRandomZeroResults = new RandomSamplingFacetsCollector(numDocs / 10, random.NextLong());
// There should be no divisions by zero
searcher.Search(new TermQuery(new Term("EvenOdd", "NeverMatches")), collectRandomZeroResults);
// There should be no divisions by zero and no null result
Assert.NotNull(collectRandomZeroResults.GetMatchingDocs());
// There should be no results at all
foreach (MatchingDocs doc in collectRandomZeroResults.GetMatchingDocs())
{
Assert.AreEqual(0, doc.TotalHits);
}
// Now start searching and retrieve results.
// Use a query to select half of the documents.
TermQuery query = new TermQuery(new Term("EvenOdd", "even"));
// there will be 5 facet values (0, 2, 4, 6 and 8), as only the even (i %
// 10) are hits.
// there is a REAL small chance that one of the 5 values will be missed when
// sampling.
// but is that 0.8 (chance not to take a value) ^ 2000 * 5 (any can be
// missing) ~ 10^-193
// so that is probably not going to happen.
int maxNumChildren = 5;
RandomSamplingFacetsCollector random100Percent = new RandomSamplingFacetsCollector(numDocs, random.NextLong()); // no sampling
RandomSamplingFacetsCollector random10Percent = new RandomSamplingFacetsCollector(numDocs / 10, random.NextLong()); // 10 % of total docs, 20% of the hits
FacetsCollector fc = new FacetsCollector();
searcher.Search(query, MultiCollector.Wrap(fc, random100Percent, random10Percent));
FastTaxonomyFacetCounts random10FacetCounts = new FastTaxonomyFacetCounts(taxoReader, config, random10Percent);
FastTaxonomyFacetCounts random100FacetCounts = new FastTaxonomyFacetCounts(taxoReader, config, random100Percent);
FastTaxonomyFacetCounts exactFacetCounts = new FastTaxonomyFacetCounts(taxoReader, config, fc);
FacetResult random10Result = random10Percent.AmortizeFacetCounts(random10FacetCounts.GetTopChildren(10, "iMod10"), config, searcher);
FacetResult random100Result = random100FacetCounts.GetTopChildren(10, "iMod10");
FacetResult exactResult = exactFacetCounts.GetTopChildren(10, "iMod10");
Assert.AreEqual(random100Result, exactResult);
// we should have five children, but there is a small chance we have less.
// (see above).
Assert.True(random10Result.ChildCount <= maxNumChildren);
// there should be one child at least.
Assert.True(random10Result.ChildCount >= 1);
// now calculate some statistics to determine if the sampled result is 'ok'.
// because random sampling is used, the results will vary each time.
int sum = 0;
foreach (LabelAndValue lav in random10Result.LabelValues)
{
sum += (int)lav.Value;
}
float mu = (float)sum / (float)maxNumChildren;
float variance = 0;
foreach (LabelAndValue lav in random10Result.LabelValues)
{
variance += (float)Math.Pow((mu - (int)lav.Value), 2);
}
variance = variance / maxNumChildren;
float sigma = (float)Math.Sqrt(variance);
// we query only half the documents and have 5 categories. The average
// number of docs in a category will thus be the total divided by 5*2
float targetMu = numDocs / (5.0f * 2.0f);
// the average should be in the range and the standard deviation should not
// be too great
Assert.True(sigma < 200);
Assert.True(targetMu - 3 * sigma < mu && mu < targetMu + 3 * sigma);
IOUtils.Close(searcher.IndexReader, taxoReader, dir, taxoDir);
}
