Computes facets counts, assuming the default encoding into DocValues was used. @lucene.experimental
Inheritance: IntTaxonomyFacets
Example #1
0
        public virtual void TestBasic()
        {
            Store.Directory dir     = NewDirectory();
            Store.Directory taxoDir = NewDirectory();

            // Writes facet ords to a separate directory from the
            // main index:
            var taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode_e.CREATE);

            FacetsConfig config = new FacetsConfig();

            config.SetHierarchical("Publish Date", true);

            RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);

            Document doc = new Document();

            doc.Add(new FacetField("Author", "Bob"));
            doc.Add(new FacetField("Publish Date", "2010", "10", "15"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            doc = new Document();
            doc.Add(new FacetField("Author", "Lisa"));
            doc.Add(new FacetField("Publish Date", "2010", "10", "20"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            doc = new Document();
            doc.Add(new FacetField("Author", "Lisa"));
            doc.Add(new FacetField("Publish Date", "2012", "1", "1"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            doc = new Document();
            doc.Add(new FacetField("Author", "Susan"));
            doc.Add(new FacetField("Publish Date", "2012", "1", "7"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            doc = new Document();
            doc.Add(new FacetField("Author", "Frank"));
            doc.Add(new FacetField("Publish Date", "1999", "5", "5"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            // NRT open
            IndexSearcher searcher = NewSearcher(writer.Reader);

            // NRT open
            var taxoReader = new DirectoryTaxonomyReader(taxoWriter);

            // Aggregate the facet counts:
            FacetsCollector c = new FacetsCollector();

            // MatchAllDocsQuery is for "browsing" (counts facets
            // for all non-deleted docs in the index); normally
            // you'd use a "normal" query, and use MultiCollector to
            // wrap collecting the "normal" hits and also facets:
            searcher.Search(new MatchAllDocsQuery(), c);

            Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, c);

            // Retrieve & verify results:
            Assert.AreEqual("dim=Publish Date path=[] value=5 childCount=3\n  2010 (2)\n  2012 (2)\n  1999 (1)\n", facets.GetTopChildren(10, "Publish Date").ToString());
            Assert.AreEqual("dim=Author path=[] value=5 childCount=4\n  Lisa (2)\n  Bob (1)\n  Susan (1)\n  Frank (1)\n", facets.GetTopChildren(10, "Author").ToString());

            // Now user drills down on Publish Date/2010:
            DrillDownQuery q2 = new DrillDownQuery(config);

            q2.Add("Publish Date", "2010");
            c = new FacetsCollector();
            searcher.Search(q2, c);
            facets = new FastTaxonomyFacetCounts(taxoReader, config, c);
            Assert.AreEqual("dim=Author path=[] value=2 childCount=2\n  Bob (1)\n  Lisa (1)\n", facets.GetTopChildren(10, "Author").ToString());

            Assert.AreEqual(1, facets.GetSpecificValue("Author", "Lisa"));

            Assert.Null(facets.GetTopChildren(10, "Non exitent dim"));

            // Smoke test PrintTaxonomyStats:
            string result;

            using (ByteArrayOutputStream bos = new ByteArrayOutputStream())
            {
                using (StreamWriter w = new StreamWriter(bos, Encoding.UTF8, 2048, true)
                {
                    AutoFlush = true
                })
                {
                    PrintTaxonomyStats.PrintStats(taxoReader, w, true);
                }
                result = bos.ToString();
            }
            Assert.True(result.IndexOf("/Author: 4 immediate children; 5 total categories", StringComparison.Ordinal) != -1);
            Assert.True(result.IndexOf("/Publish Date: 3 immediate children; 12 total categories", StringComparison.Ordinal) != -1);
            // Make sure at least a few nodes of the tree came out:
            Assert.True(result.IndexOf("  /1999", StringComparison.Ordinal) != -1);
            Assert.True(result.IndexOf("  /2012", StringComparison.Ordinal) != -1);
            Assert.True(result.IndexOf("      /20", StringComparison.Ordinal) != -1);

            IOUtils.Close(writer, taxoWriter, searcher.IndexReader, taxoReader, taxoDir, dir);
        }
Example #2
0
        public virtual void TestWrongIndexFieldName()
        {
            Store.Directory dir     = NewDirectory();
            Store.Directory taxoDir = NewDirectory();

            // Writes facet ords to a separate directory from the
            // main index:
            DirectoryTaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode_e.CREATE);

            FacetsConfig config = new FacetsConfig();

            config.SetIndexFieldName("a", "$facets2");
            RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);

            Document doc = new Document();

            doc.Add(new FacetField("a", "foo1"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            // NRT open
            IndexSearcher searcher = NewSearcher(writer.Reader);

            // NRT open
            var taxoReader = new DirectoryTaxonomyReader(taxoWriter);

            FacetsCollector c = new FacetsCollector();

            searcher.Search(new MatchAllDocsQuery(), c);

            // Uses default $facets field:
            Facets facets;

            if (Random().NextBoolean())
            {
                facets = new FastTaxonomyFacetCounts(taxoReader, config, c);
            }
            else
            {
                OrdinalsReader ordsReader = new DocValuesOrdinalsReader();
                if (Random().NextBoolean())
                {
                    ordsReader = new CachedOrdinalsReader(ordsReader);
                }
                facets = new TaxonomyFacetCounts(ordsReader, taxoReader, config, c);
            }

            // Ask for top 10 labels for any dims that have counts:
            IList <FacetResult> results = facets.GetAllDims(10);

            Assert.True(results.Count == 0);

            try
            {
                facets.GetSpecificValue("a");
                Fail("should have hit exc");
            }
            catch (System.ArgumentException)
            {
                // expected
            }

            try
            {
                facets.GetTopChildren(10, "a");
                Fail("should have hit exc");
            }
            catch (System.ArgumentException)
            {
                // expected
            }

            IOUtils.Close(writer, taxoWriter, searcher.IndexReader, taxoReader, taxoDir, dir);
        }
        public virtual void TestWrongIndexFieldName()
        {
            Store.Directory dir = NewDirectory();
            Store.Directory taxoDir = NewDirectory();

            // Writes facet ords to a separate directory from the
            // main index:
            DirectoryTaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode_e.CREATE);

            FacetsConfig config = new FacetsConfig();
            config.SetIndexFieldName("a", "$facets2");
            RandomIndexWriter writer = new RandomIndexWriter(Random(), dir);

            Document doc = new Document();
            doc.Add(new FacetField("a", "foo1"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            // NRT open
            IndexSearcher searcher = NewSearcher(writer.Reader);

            // NRT open
            var taxoReader = new DirectoryTaxonomyReader(taxoWriter);

            FacetsCollector c = new FacetsCollector();
            searcher.Search(new MatchAllDocsQuery(), c);

            // Uses default $facets field:
            Facets facets;
            if (Random().NextBoolean())
            {
                facets = new FastTaxonomyFacetCounts(taxoReader, config, c);
            }
            else
            {
                OrdinalsReader ordsReader = new DocValuesOrdinalsReader();
                if (Random().NextBoolean())
                {
                    ordsReader = new CachedOrdinalsReader(ordsReader);
                }
                facets = new TaxonomyFacetCounts(ordsReader, taxoReader, config, c);
            }

            // Ask for top 10 labels for any dims that have counts:
            IList<FacetResult> results = facets.GetAllDims(10);
            Assert.True(results.Count == 0);

            try
            {
                facets.GetSpecificValue("a");
                Fail("should have hit exc");
            }
            catch (System.ArgumentException)
            {
                // expected
            }

            try
            {
                facets.GetTopChildren(10, "a");
                Fail("should have hit exc");
            }
            catch (System.ArgumentException)
            {
                // expected
            }

            IOUtils.Close(writer, taxoWriter, searcher.IndexReader, taxoReader, taxoDir, dir);
        }
        public virtual void TestBasic()
        {
            Store.Directory dir = NewDirectory();
            Store.Directory taxoDir = NewDirectory();

            // Writes facet ords to a separate directory from the
            // main index:
            var taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode_e.CREATE);

            FacetsConfig config = new FacetsConfig();
            config.SetHierarchical("Publish Date", true);

            RandomIndexWriter writer = new RandomIndexWriter(Random(), dir);

            Document doc = new Document();
            doc.Add(new FacetField("Author", "Bob"));
            doc.Add(new FacetField("Publish Date", "2010", "10", "15"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            doc = new Document();
            doc.Add(new FacetField("Author", "Lisa"));
            doc.Add(new FacetField("Publish Date", "2010", "10", "20"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            doc = new Document();
            doc.Add(new FacetField("Author", "Lisa"));
            doc.Add(new FacetField("Publish Date", "2012", "1", "1"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            doc = new Document();
            doc.Add(new FacetField("Author", "Susan"));
            doc.Add(new FacetField("Publish Date", "2012", "1", "7"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            doc = new Document();
            doc.Add(new FacetField("Author", "Frank"));
            doc.Add(new FacetField("Publish Date", "1999", "5", "5"));
            writer.AddDocument(config.Build(taxoWriter, doc));

            // NRT open
            IndexSearcher searcher = NewSearcher(writer.Reader);

            // NRT open
            var taxoReader = new DirectoryTaxonomyReader(taxoWriter);

            // Aggregate the facet counts:
            FacetsCollector c = new FacetsCollector();

            // MatchAllDocsQuery is for "browsing" (counts facets
            // for all non-deleted docs in the index); normally
            // you'd use a "normal" query, and use MultiCollector to
            // wrap collecting the "normal" hits and also facets:
            searcher.Search(new MatchAllDocsQuery(), c);

            Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, c);

            // Retrieve & verify results:
            Assert.AreEqual("dim=Publish Date path=[] value=5 childCount=3\n  2010 (2)\n  2012 (2)\n  1999 (1)\n", facets.GetTopChildren(10, "Publish Date").ToString());
            Assert.AreEqual("dim=Author path=[] value=5 childCount=4\n  Lisa (2)\n  Bob (1)\n  Susan (1)\n  Frank (1)\n", facets.GetTopChildren(10, "Author").ToString());

            // Now user drills down on Publish Date/2010:
            DrillDownQuery q2 = new DrillDownQuery(config);
            q2.Add("Publish Date", "2010");
            c = new FacetsCollector();
            searcher.Search(q2, c);
            facets = new FastTaxonomyFacetCounts(taxoReader, config, c);
            Assert.AreEqual("dim=Author path=[] value=2 childCount=2\n  Bob (1)\n  Lisa (1)\n", facets.GetTopChildren(10, "Author").ToString());

            Assert.AreEqual(1, facets.GetSpecificValue("Author", "Lisa"));

            Assert.Null(facets.GetTopChildren(10, "Non exitent dim"));

            // Smoke test PrintTaxonomyStats:
            ByteArrayOutputStream bos = new ByteArrayOutputStream();
            PrintTaxonomyStats.PrintStats(taxoReader, Console.Out, true);
            string result = bos.ToString();
            Assert.True(result.IndexOf("/Author: 4 immediate children; 5 total categories", StringComparison.Ordinal) != -1);
            Assert.True(result.IndexOf("/Publish Date: 3 immediate children; 12 total categories", StringComparison.Ordinal) != -1);
            // Make sure at least a few nodes of the tree came out:
            Assert.True(result.IndexOf("  /1999", StringComparison.Ordinal) != -1);
            Assert.True(result.IndexOf("  /2012", StringComparison.Ordinal) != -1);
            Assert.True(result.IndexOf("      /20", StringComparison.Ordinal) != -1);

            IOUtils.Close(writer, taxoWriter, searcher.IndexReader, taxoReader, taxoDir, dir);
        }
Example #5
0
        /// <summary>
        /// Subclass can override to customize per-dim Facets
        ///  impl. 
        /// </summary>
        protected virtual Facets BuildFacetsResult(FacetsCollector drillDowns, FacetsCollector[] drillSideways, string[] drillSidewaysDims)
        {

            Facets drillDownFacets;
            var drillSidewaysFacets = new Dictionary<string, Facets>();

            if (taxoReader != null)
            {
                drillDownFacets = new FastTaxonomyFacetCounts(taxoReader, config, drillDowns);
                if (drillSideways != null)
                {
                    for (int i = 0; i < drillSideways.Length; i++)
                    {
                        drillSidewaysFacets[drillSidewaysDims[i]] = new FastTaxonomyFacetCounts(taxoReader, config, drillSideways[i]);
                    }
                }
            }
            else
            {
                drillDownFacets = new SortedSetDocValuesFacetCounts(state, drillDowns);
                if (drillSideways != null)
                {
                    for (int i = 0; i < drillSideways.Length; i++)
                    {
                        drillSidewaysFacets[drillSidewaysDims[i]] = new SortedSetDocValuesFacetCounts(state, drillSideways[i]);
                    }
                }
            }

            if (drillSidewaysFacets.Count == 0)
            {
                return drillDownFacets;
            }
            else
            {
                return new MultiFacets(drillSidewaysFacets, drillDownFacets);
            }
        }