Examples with SortedSetDocValuesFacetField org.apache.lucene.facet.sortedset.SortedSetDocValuesFacetField used on opensource projects

Example 1 with SortedSetDocValuesFacetField

use of org.apache.lucene.facet.sortedset.SortedSetDocValuesFacetField in project lucene-solr by apache.

the class FacetsConfig method processSSDVFacetFields.

private void processSSDVFacetFields(Map<String, List<SortedSetDocValuesFacetField>> byField, Document doc) throws IOException {
    //System.out.println("process SSDV: " + byField);
    for (Map.Entry<String, List<SortedSetDocValuesFacetField>> ent : byField.entrySet()) {
        String indexFieldName = ent.getKey();
        for (SortedSetDocValuesFacetField facetField : ent.getValue()) {
            FacetLabel cp = new FacetLabel(facetField.dim, facetField.label);
            String fullPath = pathToString(cp.components, cp.length);
            //System.out.println("add " + fullPath);
            // For facet counts:
            doc.add(new SortedSetDocValuesField(indexFieldName, new BytesRef(fullPath)));
            // For drill-down:
            doc.add(new StringField(indexFieldName, fullPath, Field.Store.NO));
            doc.add(new StringField(indexFieldName, facetField.dim, Field.Store.NO));
        }
    }
}

Example 2 with SortedSetDocValuesFacetField

use of org.apache.lucene.facet.sortedset.SortedSetDocValuesFacetField in project lucene-solr by apache.

the class TestDrillSideways method testRandom.

public void testRandom() throws Exception {
    while (aChance == 0.0) {
        aChance = random().nextDouble();
    }
    while (bChance == 0.0) {
        bChance = random().nextDouble();
    }
    while (cChance == 0.0) {
        cChance = random().nextDouble();
    }
    //aChance = .01;
    //bChance = 0.5;
    //cChance = 1.0;
    double sum = aChance + bChance + cChance;
    aChance /= sum;
    bChance /= sum;
    cChance /= sum;
    int numDims = TestUtil.nextInt(random(), 2, 5);
    //int numDims = 3;
    int numDocs = atLeast(3000);
    //int numDocs = 20;
    if (VERBOSE) {
        System.out.println("numDims=" + numDims + " numDocs=" + numDocs + " aChance=" + aChance + " bChance=" + bChance + " cChance=" + cChance);
    }
    String[][] dimValues = new String[numDims][];
    int valueCount = 2;
    for (int dim = 0; dim < numDims; dim++) {
        Set<String> values = new HashSet<>();
        while (values.size() < valueCount) {
            String s = TestUtil.randomRealisticUnicodeString(random());
            //String s = _TestUtil.randomString(random());
            if (s.length() > 0) {
                values.add(s);
            }
        }
        dimValues[dim] = values.toArray(new String[values.size()]);
        valueCount *= 2;
    }
    List<Doc> docs = new ArrayList<>();
    for (int i = 0; i < numDocs; i++) {
        Doc doc = new Doc();
        doc.id = "" + i;
        doc.contentToken = randomContentToken(false);
        doc.dims = new int[numDims];
        doc.dims2 = new int[numDims];
        for (int dim = 0; dim < numDims; dim++) {
            if (random().nextInt(5) == 3) {
                // This doc is missing this dim:
                doc.dims[dim] = -1;
            } else if (dimValues[dim].length <= 4) {
                int dimUpto = 0;
                doc.dims[dim] = dimValues[dim].length - 1;
                while (dimUpto < dimValues[dim].length) {
                    if (random().nextBoolean()) {
                        doc.dims[dim] = dimUpto;
                        break;
                    }
                    dimUpto++;
                }
            } else {
                doc.dims[dim] = random().nextInt(dimValues[dim].length);
            }
            if (random().nextInt(5) == 3) {
                // 2nd value:
                doc.dims2[dim] = random().nextInt(dimValues[dim].length);
            } else {
                doc.dims2[dim] = -1;
            }
        }
        docs.add(doc);
    }
    Directory d = newDirectory();
    Directory td = newDirectory();
    IndexWriterConfig iwc = newIndexWriterConfig(new MockAnalyzer(random()));
    iwc.setInfoStream(InfoStream.NO_OUTPUT);
    RandomIndexWriter w = new RandomIndexWriter(random(), d, iwc);
    DirectoryTaxonomyWriter tw = new DirectoryTaxonomyWriter(td, IndexWriterConfig.OpenMode.CREATE);
    FacetsConfig config = new FacetsConfig();
    for (int i = 0; i < numDims; i++) {
        config.setMultiValued("dim" + i, true);
    }
    boolean doUseDV = random().nextBoolean();
    for (Doc rawDoc : docs) {
        Document doc = new Document();
        doc.add(newStringField("id", rawDoc.id, Field.Store.YES));
        doc.add(new SortedDocValuesField("id", new BytesRef(rawDoc.id)));
        doc.add(newStringField("content", rawDoc.contentToken, Field.Store.NO));
        if (VERBOSE) {
            System.out.println("  doc id=" + rawDoc.id + " token=" + rawDoc.contentToken);
        }
        for (int dim = 0; dim < numDims; dim++) {
            int dimValue = rawDoc.dims[dim];
            if (dimValue != -1) {
                if (doUseDV) {
                    doc.add(new SortedSetDocValuesFacetField("dim" + dim, dimValues[dim][dimValue]));
                } else {
                    doc.add(new FacetField("dim" + dim, dimValues[dim][dimValue]));
                }
                doc.add(new StringField("dim" + dim, dimValues[dim][dimValue], Field.Store.YES));
                if (VERBOSE) {
                    System.out.println("    dim" + dim + "=" + new BytesRef(dimValues[dim][dimValue]));
                }
            }
            int dimValue2 = rawDoc.dims2[dim];
            if (dimValue2 != -1) {
                if (doUseDV) {
                    doc.add(new SortedSetDocValuesFacetField("dim" + dim, dimValues[dim][dimValue2]));
                } else {
                    doc.add(new FacetField("dim" + dim, dimValues[dim][dimValue2]));
                }
                doc.add(new StringField("dim" + dim, dimValues[dim][dimValue2], Field.Store.YES));
                if (VERBOSE) {
                    System.out.println("      dim" + dim + "=" + new BytesRef(dimValues[dim][dimValue2]));
                }
            }
        }
        w.addDocument(config.build(tw, doc));
    }
    if (random().nextBoolean()) {
        // Randomly delete a few docs:
        int numDel = TestUtil.nextInt(random(), 1, (int) (numDocs * 0.05));
        if (VERBOSE) {
            System.out.println("delete " + numDel);
        }
        int delCount = 0;
        while (delCount < numDel) {
            Doc doc = docs.get(random().nextInt(docs.size()));
            if (!doc.deleted) {
                if (VERBOSE) {
                    System.out.println("  delete id=" + doc.id);
                }
                doc.deleted = true;
                w.deleteDocuments(new Term("id", doc.id));
                delCount++;
            }
        }
    }
    if (random().nextBoolean()) {
        if (VERBOSE) {
            System.out.println("TEST: forceMerge(1)...");
        }
        w.forceMerge(1);
    }
    IndexReader r = w.getReader();
    final SortedSetDocValuesReaderState sortedSetDVState;
    IndexSearcher s = newSearcher(r);
    if (doUseDV) {
        sortedSetDVState = new DefaultSortedSetDocValuesReaderState(s.getIndexReader());
    } else {
        sortedSetDVState = null;
    }
    if (VERBOSE) {
        System.out.println("r.numDocs() = " + r.numDocs());
    }
    // NRT open
    TaxonomyReader tr = new DirectoryTaxonomyReader(tw);
    int numIters = atLeast(10);
    for (int iter = 0; iter < numIters; iter++) {
        String contentToken = random().nextInt(30) == 17 ? null : randomContentToken(true);
        int numDrillDown = TestUtil.nextInt(random(), 1, Math.min(4, numDims));
        if (VERBOSE) {
            System.out.println("\nTEST: iter=" + iter + " baseQuery=" + contentToken + " numDrillDown=" + numDrillDown + " useSortedSetDV=" + doUseDV);
        }
        String[][] drillDowns = new String[numDims][];
        int count = 0;
        boolean anyMultiValuedDrillDowns = false;
        while (count < numDrillDown) {
            int dim = random().nextInt(numDims);
            if (drillDowns[dim] == null) {
                if (random().nextBoolean()) {
                    // Drill down on one value:
                    drillDowns[dim] = new String[] { dimValues[dim][random().nextInt(dimValues[dim].length)] };
                } else {
                    int orCount = TestUtil.nextInt(random(), 1, Math.min(5, dimValues[dim].length));
                    drillDowns[dim] = new String[orCount];
                    anyMultiValuedDrillDowns |= orCount > 1;
                    for (int i = 0; i < orCount; i++) {
                        while (true) {
                            String value = dimValues[dim][random().nextInt(dimValues[dim].length)];
                            for (int j = 0; j < i; j++) {
                                if (value.equals(drillDowns[dim][j])) {
                                    value = null;
                                    break;
                                }
                            }
                            if (value != null) {
                                drillDowns[dim][i] = value;
                                break;
                            }
                        }
                    }
                }
                if (VERBOSE) {
                    BytesRef[] values = new BytesRef[drillDowns[dim].length];
                    for (int i = 0; i < values.length; i++) {
                        values[i] = new BytesRef(drillDowns[dim][i]);
                    }
                    System.out.println("  dim" + dim + "=" + Arrays.toString(values));
                }
                count++;
            }
        }
        Query baseQuery;
        if (contentToken == null) {
            baseQuery = new MatchAllDocsQuery();
        } else {
            baseQuery = new TermQuery(new Term("content", contentToken));
        }
        DrillDownQuery ddq = new DrillDownQuery(config, baseQuery);
        for (int dim = 0; dim < numDims; dim++) {
            if (drillDowns[dim] != null) {
                for (String value : drillDowns[dim]) {
                    ddq.add("dim" + dim, value);
                }
            }
        }
        Query filter;
        if (random().nextInt(7) == 6) {
            if (VERBOSE) {
                System.out.println("  only-even filter");
            }
            filter = new Query() {

                @Override
                public Weight createWeight(IndexSearcher searcher, boolean needsScores, float boost) throws IOException {
                    return new ConstantScoreWeight(this, boost) {

                        @Override
                        public Scorer scorer(LeafReaderContext context) throws IOException {
                            DocIdSetIterator approximation = DocIdSetIterator.all(context.reader().maxDoc());
                            return new ConstantScoreScorer(this, score(), new TwoPhaseIterator(approximation) {

                                @Override
                                public boolean matches() throws IOException {
                                    int docID = approximation.docID();
                                    return (Integer.parseInt(context.reader().document(docID).get("id")) & 1) == 0;
                                }

                                @Override
                                public float matchCost() {
                                    return 1000f;
                                }
                            });
                        }
                    };
                }

                @Override
                public String toString(String field) {
                    return "drillSidewaysTestFilter";
                }

                @Override
                public boolean equals(Object o) {
                    return o == this;
                }

                @Override
                public int hashCode() {
                    return System.identityHashCode(this);
                }
            };
        } else {
            filter = null;
        }
        // Verify docs are always collected in order.  If we
        // had an AssertingScorer it could catch it when
        // Weight.scoresDocsOutOfOrder lies!:
        getNewDrillSideways(s, config, tr).search(ddq, new SimpleCollector() {

            int lastDocID;

            @Override
            public void collect(int doc) {
                assert doc > lastDocID;
                lastDocID = doc;
            }

            @Override
            protected void doSetNextReader(LeafReaderContext context) throws IOException {
                lastDocID = -1;
            }

            @Override
            public boolean needsScores() {
                return false;
            }
        });
        // subScorers are on the same docID:
        if (!anyMultiValuedDrillDowns) {
            // Can only do this test when there are no OR'd
            // drill-down values, because in that case it's
            // easily possible for one of the DD terms to be on
            // a future docID:
            getNewDrillSidewaysScoreSubdocsAtOnce(s, config, tr).search(ddq, new AssertingSubDocsAtOnceCollector());
        }
        TestFacetResult expected = slowDrillSidewaysSearch(s, docs, contentToken, drillDowns, dimValues, filter);
        Sort sort = new Sort(new SortField("id", SortField.Type.STRING));
        DrillSideways ds;
        if (doUseDV) {
            ds = getNewDrillSideways(s, config, sortedSetDVState);
        } else {
            ds = getNewDrillSidewaysBuildFacetsResult(s, config, tr);
        }
        // Retrieve all facets:
        DrillSidewaysResult actual = ds.search(ddq, filter, null, numDocs, sort, true, true);
        TopDocs hits = s.search(baseQuery, numDocs);
        Map<String, Float> scores = new HashMap<>();
        for (ScoreDoc sd : hits.scoreDocs) {
            scores.put(s.doc(sd.doc).get("id"), sd.score);
        }
        if (VERBOSE) {
            System.out.println("  verify all facets");
        }
        verifyEquals(dimValues, s, expected, actual, scores, doUseDV);
        // Make sure drill down doesn't change score:
        Query q = ddq;
        if (filter != null) {
            q = new BooleanQuery.Builder().add(q, Occur.MUST).add(filter, Occur.FILTER).build();
        }
        TopDocs ddqHits = s.search(q, numDocs);
        assertEquals(expected.hits.size(), ddqHits.totalHits);
        for (int i = 0; i < expected.hits.size(); i++) {
            // Score should be IDENTICAL:
            assertEquals(scores.get(expected.hits.get(i).id), ddqHits.scoreDocs[i].score, 0.0f);
        }
    }
    w.close();
    IOUtils.close(r, tr, tw, d, td);
}

Example 3 with SortedSetDocValuesFacetField

use of org.apache.lucene.facet.sortedset.SortedSetDocValuesFacetField in project lucene-solr by apache.

the class TestFacetLabel method testEmptyNullComponents.

@Test
public void testEmptyNullComponents() throws Exception {
    // LUCENE-4724: CategoryPath should not allow empty or null components
    String[][] components_tests = new String[][] { // empty in the beginning
    new String[] { "", "test" }, // empty in the end
    new String[] { "test", "" }, // empty in the middle
    new String[] { "test", "", "foo" }, // null at the beginning
    new String[] { null, "test" }, // null in the end
    new String[] { "test", null }, // null in the middle
    new String[] { "test", null, "foo" } };
    // empty or null components should not be allowed.
    for (String[] components : components_tests) {
        expectThrows(IllegalArgumentException.class, () -> {
            new FacetLabel(components);
        });
        expectThrows(IllegalArgumentException.class, () -> {
            new FacetField("dim", components);
        });
        expectThrows(IllegalArgumentException.class, () -> {
            new AssociationFacetField(new BytesRef(), "dim", components);
        });
        expectThrows(IllegalArgumentException.class, () -> {
            new IntAssociationFacetField(17, "dim", components);
        });
        expectThrows(IllegalArgumentException.class, () -> {
            new FloatAssociationFacetField(17.0f, "dim", components);
        });
    }
    expectThrows(IllegalArgumentException.class, () -> {
        new FacetField(null, new String[] { "abc" });
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new FacetField("", new String[] { "abc" });
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new IntAssociationFacetField(17, null, new String[] { "abc" });
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new IntAssociationFacetField(17, "", new String[] { "abc" });
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new FloatAssociationFacetField(17.0f, null, new String[] { "abc" });
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new FloatAssociationFacetField(17.0f, "", new String[] { "abc" });
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new AssociationFacetField(new BytesRef(), null, new String[] { "abc" });
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new AssociationFacetField(new BytesRef(), "", new String[] { "abc" });
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new SortedSetDocValuesFacetField(null, "abc");
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new SortedSetDocValuesFacetField("", "abc");
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new SortedSetDocValuesFacetField("dim", null);
    });
    expectThrows(IllegalArgumentException.class, () -> {
        new SortedSetDocValuesFacetField("dim", "");
    });
}

Example 4 with SortedSetDocValuesFacetField

use of org.apache.lucene.facet.sortedset.SortedSetDocValuesFacetField in project lucene-solr by apache.

the class SimpleSortedSetFacetsExample method index.

/** Build the example index. */
private void index() throws IOException {
    IndexWriter indexWriter = new IndexWriter(indexDir, new IndexWriterConfig(new WhitespaceAnalyzer()).setOpenMode(OpenMode.CREATE));
    Document doc = new Document();
    doc.add(new SortedSetDocValuesFacetField("Author", "Bob"));
    doc.add(new SortedSetDocValuesFacetField("Publish Year", "2010"));
    indexWriter.addDocument(config.build(doc));
    doc = new Document();
    doc.add(new SortedSetDocValuesFacetField("Author", "Lisa"));
    doc.add(new SortedSetDocValuesFacetField("Publish Year", "2010"));
    indexWriter.addDocument(config.build(doc));
    doc = new Document();
    doc.add(new SortedSetDocValuesFacetField("Author", "Lisa"));
    doc.add(new SortedSetDocValuesFacetField("Publish Year", "2012"));
    indexWriter.addDocument(config.build(doc));
    doc = new Document();
    doc.add(new SortedSetDocValuesFacetField("Author", "Susan"));
    doc.add(new SortedSetDocValuesFacetField("Publish Year", "2012"));
    indexWriter.addDocument(config.build(doc));
    doc = new Document();
    doc.add(new SortedSetDocValuesFacetField("Author", "Frank"));
    doc.add(new SortedSetDocValuesFacetField("Publish Year", "1999"));
    indexWriter.addDocument(config.build(doc));
    indexWriter.close();
}

Example 5 with SortedSetDocValuesFacetField

use of org.apache.lucene.facet.sortedset.SortedSetDocValuesFacetField in project jackrabbit-oak by apache.

the class LuceneDocumentMaker method addFacetFields.

private boolean addFacetFields(List<Field> fields, PropertyState property, String pname, PropertyDefinition pd) {
    String facetFieldName = FieldNames.createFacetFieldName(pname);
    getFacetsConfig().setIndexFieldName(pname, facetFieldName);
    int tag = property.getType().tag();
    int idxDefinedTag = pd.getType();
    // Try converting type to the defined type in the index definition
    if (tag != idxDefinedTag) {
        log.debug("[{}] Facet property defined with type {} differs from property {} with type {} in " + "path {}", getIndexName(), Type.fromTag(idxDefinedTag, false), property.toString(), Type.fromTag(tag, false), path);
        tag = idxDefinedTag;
    }
    boolean fieldAdded = false;
    try {
        if (tag == Type.STRINGS.tag() && property.isArray()) {
            getFacetsConfig().setMultiValued(pname, true);
            Iterable<String> values = property.getValue(Type.STRINGS);
            for (String value : values) {
                if (value != null && value.length() > 0) {
                    fields.add(new SortedSetDocValuesFacetField(pname, value));
                }
            }
            fieldAdded = true;
        } else if (tag == Type.STRING.tag()) {
            String value = property.getValue(Type.STRING);
            if (value.length() > 0) {
                fields.add(new SortedSetDocValuesFacetField(pname, value));
                fieldAdded = true;
            }
        }
    } catch (Throwable e) {
        log.warn("[{}] Ignoring facet property. Could not convert property {} of type {} to type {} for path {}", getIndexName(), pname, Type.fromTag(property.getType().tag(), false), Type.fromTag(tag, false), path, e);
    }
    return fieldAdded;
}