Search in sources :

Example 71 with Document

use of org.apache.lucene.document.Document in project elasticsearch by elastic.

the class IndexSearcherWrapperTests method testNoWrap.

public void testNoWrap() throws IOException {
    Directory dir = newDirectory();
    IndexWriterConfig iwc = newIndexWriterConfig();
    IndexWriter writer = new IndexWriter(dir, iwc);
    Document doc = new Document();
    doc.add(new StringField("id", "1", random().nextBoolean() ? Field.Store.YES : Field.Store.NO));
    doc.add(new TextField("field", "doc", random().nextBoolean() ? Field.Store.YES : Field.Store.NO));
    writer.addDocument(doc);
    DirectoryReader open = ElasticsearchDirectoryReader.wrap(DirectoryReader.open(writer), new ShardId("foo", "_na_", 1));
    IndexSearcher searcher = new IndexSearcher(open);
    assertEquals(1, searcher.search(new TermQuery(new Term("field", "doc")), 1).totalHits);
    searcher.setSimilarity(iwc.getSimilarity());
    IndexSearcherWrapper wrapper = new IndexSearcherWrapper();
    try (Engine.Searcher engineSearcher = new Engine.Searcher("foo", searcher)) {
        final Engine.Searcher wrap = wrapper.wrap(engineSearcher);
        assertSame(wrap, engineSearcher);
    }
    IOUtils.close(open, writer, dir);
}
Also used : IndexSearcher(org.apache.lucene.search.IndexSearcher) TermQuery(org.apache.lucene.search.TermQuery) DirectoryReader(org.apache.lucene.index.DirectoryReader) ElasticsearchDirectoryReader(org.elasticsearch.common.lucene.index.ElasticsearchDirectoryReader) FilterDirectoryReader(org.apache.lucene.index.FilterDirectoryReader) IndexSearcher(org.apache.lucene.search.IndexSearcher) Term(org.apache.lucene.index.Term) Document(org.apache.lucene.document.Document) IndexWriter(org.apache.lucene.index.IndexWriter) StringField(org.apache.lucene.document.StringField) TextField(org.apache.lucene.document.TextField) Engine(org.elasticsearch.index.engine.Engine) Directory(org.apache.lucene.store.Directory) IndexWriterConfig(org.apache.lucene.index.IndexWriterConfig)

Example 72 with Document

use of org.apache.lucene.document.Document in project elasticsearch by elastic.

the class ShardUtilsTests method testExtractShardId.

public void testExtractShardId() throws IOException {
    BaseDirectoryWrapper dir = newDirectory();
    IndexWriter writer = new IndexWriter(dir, newIndexWriterConfig());
    writer.commit();
    ShardId id = new ShardId("foo", "_na_", random().nextInt());
    try (DirectoryReader reader = DirectoryReader.open(writer)) {
        ElasticsearchDirectoryReader wrap = ElasticsearchDirectoryReader.wrap(reader, id);
        assertEquals(id, ShardUtils.extractShardId(wrap));
    }
    final int numDocs = 1 + random().nextInt(5);
    for (int i = 0; i < numDocs; i++) {
        Document d = new Document();
        d.add(newField("name", "foobar", StringField.TYPE_STORED));
        writer.addDocument(d);
        if (random().nextBoolean()) {
            writer.commit();
        }
    }
    try (DirectoryReader reader = DirectoryReader.open(writer)) {
        ElasticsearchDirectoryReader wrap = ElasticsearchDirectoryReader.wrap(reader, id);
        assertEquals(id, ShardUtils.extractShardId(wrap));
        CompositeReaderContext context = wrap.getContext();
        for (LeafReaderContext leaf : context.leaves()) {
            assertEquals(id, ShardUtils.extractShardId(leaf.reader()));
        }
    }
    IOUtils.close(writer, dir);
}
Also used : ElasticsearchDirectoryReader(org.elasticsearch.common.lucene.index.ElasticsearchDirectoryReader) CompositeReaderContext(org.apache.lucene.index.CompositeReaderContext) IndexWriter(org.apache.lucene.index.IndexWriter) DirectoryReader(org.apache.lucene.index.DirectoryReader) ElasticsearchDirectoryReader(org.elasticsearch.common.lucene.index.ElasticsearchDirectoryReader) BaseDirectoryWrapper(org.apache.lucene.store.BaseDirectoryWrapper) LeafReaderContext(org.apache.lucene.index.LeafReaderContext) Document(org.apache.lucene.document.Document)

Example 73 with Document

use of org.apache.lucene.document.Document in project elasticsearch by elastic.

the class NestedSortingTests method testDuel.

public void testDuel() throws Exception {
    final int numDocs = scaledRandomIntBetween(100, 1000);
    for (int i = 0; i < numDocs; ++i) {
        final int numChildren = randomInt(2);
        List<Document> docs = new ArrayList<>(numChildren + 1);
        for (int j = 0; j < numChildren; ++j) {
            Document doc = new Document();
            doc.add(new StringField("f", TestUtil.randomSimpleString(random(), 2), Field.Store.NO));
            doc.add(new StringField("__type", "child", Field.Store.NO));
            docs.add(doc);
        }
        if (randomBoolean()) {
            docs.add(new Document());
        }
        Document parent = new Document();
        parent.add(new StringField("__type", "parent", Field.Store.NO));
        docs.add(parent);
        writer.addDocuments(docs);
        if (rarely()) {
            // we need to have a bit more segments than what RandomIndexWriter would do by default
            DirectoryReader.open(writer).close();
        }
    }
    writer.commit();
    MultiValueMode sortMode = randomFrom(Arrays.asList(MultiValueMode.MIN, MultiValueMode.MAX));
    DirectoryReader reader = DirectoryReader.open(writer);
    reader = ElasticsearchDirectoryReader.wrap(reader, new ShardId(indexService.index(), 0));
    IndexSearcher searcher = new IndexSearcher(reader);
    PagedBytesIndexFieldData indexFieldData1 = getForField("f");
    IndexFieldData<?> indexFieldData2 = NoOrdinalsStringFieldDataTests.hideOrdinals(indexFieldData1);
    final String missingValue = randomBoolean() ? null : TestUtil.randomSimpleString(random(), 2);
    final int n = randomIntBetween(1, numDocs + 2);
    final boolean reverse = randomBoolean();
    final TopDocs topDocs1 = getTopDocs(searcher, indexFieldData1, missingValue, sortMode, n, reverse);
    final TopDocs topDocs2 = getTopDocs(searcher, indexFieldData2, missingValue, sortMode, n, reverse);
    for (int i = 0; i < topDocs1.scoreDocs.length; ++i) {
        final FieldDoc fieldDoc1 = (FieldDoc) topDocs1.scoreDocs[i];
        final FieldDoc fieldDoc2 = (FieldDoc) topDocs2.scoreDocs[i];
        assertEquals(fieldDoc1.doc, fieldDoc2.doc);
        assertArrayEquals(fieldDoc1.fields, fieldDoc2.fields);
    }
    searcher.getIndexReader().close();
}
Also used : IndexSearcher(org.apache.lucene.search.IndexSearcher) PagedBytesIndexFieldData(org.elasticsearch.index.fielddata.plain.PagedBytesIndexFieldData) FieldDoc(org.apache.lucene.search.FieldDoc) ElasticsearchDirectoryReader(org.elasticsearch.common.lucene.index.ElasticsearchDirectoryReader) DirectoryReader(org.apache.lucene.index.DirectoryReader) ArrayList(java.util.ArrayList) Document(org.apache.lucene.document.Document) MultiValueMode(org.elasticsearch.search.MultiValueMode) ShardId(org.elasticsearch.index.shard.ShardId) TopDocs(org.apache.lucene.search.TopDocs) StringField(org.apache.lucene.document.StringField)

Example 74 with Document

use of org.apache.lucene.document.Document in project elasticsearch by elastic.

the class IndicesQueryCacheTests method testStatsOnEviction.

// Make sure the cache behaves correctly when a segment that is associated
// with an empty cache gets closed. In that particular case, the eviction
// callback is called with a number of evicted entries equal to 0
// see https://github.com/elastic/elasticsearch/issues/15043
public void testStatsOnEviction() throws IOException {
    Directory dir1 = newDirectory();
    IndexWriter w1 = new IndexWriter(dir1, newIndexWriterConfig());
    w1.addDocument(new Document());
    DirectoryReader r1 = DirectoryReader.open(w1);
    w1.close();
    ShardId shard1 = new ShardId("index", "_na_", 0);
    r1 = ElasticsearchDirectoryReader.wrap(r1, shard1);
    IndexSearcher s1 = new IndexSearcher(r1);
    s1.setQueryCachingPolicy(QueryCachingPolicy.ALWAYS_CACHE);
    Directory dir2 = newDirectory();
    IndexWriter w2 = new IndexWriter(dir2, newIndexWriterConfig());
    w2.addDocument(new Document());
    DirectoryReader r2 = DirectoryReader.open(w2);
    w2.close();
    ShardId shard2 = new ShardId("index", "_na_", 1);
    r2 = ElasticsearchDirectoryReader.wrap(r2, shard2);
    IndexSearcher s2 = new IndexSearcher(r2);
    s2.setQueryCachingPolicy(QueryCachingPolicy.ALWAYS_CACHE);
    Settings settings = Settings.builder().put(IndicesQueryCache.INDICES_CACHE_QUERY_COUNT_SETTING.getKey(), 10).put(IndicesQueryCache.INDICES_QUERIES_CACHE_ALL_SEGMENTS_SETTING.getKey(), true).build();
    IndicesQueryCache cache = new IndicesQueryCache(settings);
    s1.setQueryCache(cache);
    s2.setQueryCache(cache);
    assertEquals(1, s1.count(new DummyQuery(0)));
    for (int i = 1; i <= 20; ++i) {
        assertEquals(1, s2.count(new DummyQuery(i)));
    }
    QueryCacheStats stats1 = cache.getStats(shard1);
    assertEquals(0L, stats1.getCacheSize());
    assertEquals(1L, stats1.getCacheCount());
    // this used to fail because we were evicting an empty cache on
    // the segment from r1
    IOUtils.close(r1, dir1);
    cache.onClose(shard1);
    IOUtils.close(r2, dir2);
    cache.onClose(shard2);
    // this triggers some assertions
    cache.close();
}
Also used : ShardId(org.elasticsearch.index.shard.ShardId) IndexSearcher(org.apache.lucene.search.IndexSearcher) IndicesQueryCache(org.elasticsearch.indices.IndicesQueryCache) IndexWriter(org.apache.lucene.index.IndexWriter) DirectoryReader(org.apache.lucene.index.DirectoryReader) ElasticsearchDirectoryReader(org.elasticsearch.common.lucene.index.ElasticsearchDirectoryReader) QueryCacheStats(org.elasticsearch.index.cache.query.QueryCacheStats) Document(org.apache.lucene.document.Document) Settings(org.elasticsearch.common.settings.Settings) Directory(org.apache.lucene.store.Directory)

Example 75 with Document

use of org.apache.lucene.document.Document in project elasticsearch by elastic.

the class IndicesQueryCacheTests method testBasics.

public void testBasics() throws IOException {
    Directory dir = newDirectory();
    IndexWriter w = new IndexWriter(dir, newIndexWriterConfig());
    w.addDocument(new Document());
    DirectoryReader r = DirectoryReader.open(w);
    w.close();
    ShardId shard = new ShardId("index", "_na_", 0);
    r = ElasticsearchDirectoryReader.wrap(r, shard);
    IndexSearcher s = new IndexSearcher(r);
    s.setQueryCachingPolicy(QueryCachingPolicy.ALWAYS_CACHE);
    Settings settings = Settings.builder().put(IndicesQueryCache.INDICES_CACHE_QUERY_COUNT_SETTING.getKey(), 10).put(IndicesQueryCache.INDICES_QUERIES_CACHE_ALL_SEGMENTS_SETTING.getKey(), true).build();
    IndicesQueryCache cache = new IndicesQueryCache(settings);
    s.setQueryCache(cache);
    QueryCacheStats stats = cache.getStats(shard);
    assertEquals(0L, stats.getCacheSize());
    assertEquals(0L, stats.getCacheCount());
    assertEquals(0L, stats.getHitCount());
    assertEquals(0L, stats.getMissCount());
    assertEquals(1, s.count(new DummyQuery(0)));
    stats = cache.getStats(shard);
    assertEquals(1L, stats.getCacheSize());
    assertEquals(1L, stats.getCacheCount());
    assertEquals(0L, stats.getHitCount());
    assertEquals(1L, stats.getMissCount());
    for (int i = 1; i < 20; ++i) {
        assertEquals(1, s.count(new DummyQuery(i)));
    }
    stats = cache.getStats(shard);
    assertEquals(10L, stats.getCacheSize());
    assertEquals(20L, stats.getCacheCount());
    assertEquals(0L, stats.getHitCount());
    assertEquals(20L, stats.getMissCount());
    s.count(new DummyQuery(10));
    stats = cache.getStats(shard);
    assertEquals(10L, stats.getCacheSize());
    assertEquals(20L, stats.getCacheCount());
    assertEquals(1L, stats.getHitCount());
    assertEquals(20L, stats.getMissCount());
    IOUtils.close(r, dir);
    // got emptied, but no changes to other metrics
    stats = cache.getStats(shard);
    assertEquals(0L, stats.getCacheSize());
    assertEquals(20L, stats.getCacheCount());
    assertEquals(1L, stats.getHitCount());
    assertEquals(20L, stats.getMissCount());
    cache.onClose(shard);
    // forgot everything
    stats = cache.getStats(shard);
    assertEquals(0L, stats.getCacheSize());
    assertEquals(0L, stats.getCacheCount());
    assertEquals(0L, stats.getHitCount());
    assertEquals(0L, stats.getMissCount());
    // this triggers some assertions
    cache.close();
}
Also used : ShardId(org.elasticsearch.index.shard.ShardId) IndexSearcher(org.apache.lucene.search.IndexSearcher) IndicesQueryCache(org.elasticsearch.indices.IndicesQueryCache) IndexWriter(org.apache.lucene.index.IndexWriter) DirectoryReader(org.apache.lucene.index.DirectoryReader) ElasticsearchDirectoryReader(org.elasticsearch.common.lucene.index.ElasticsearchDirectoryReader) QueryCacheStats(org.elasticsearch.index.cache.query.QueryCacheStats) Document(org.apache.lucene.document.Document) Settings(org.elasticsearch.common.settings.Settings) Directory(org.apache.lucene.store.Directory)

Aggregations

Document (org.apache.lucene.document.Document)2344 Directory (org.apache.lucene.store.Directory)1374 MockAnalyzer (org.apache.lucene.analysis.MockAnalyzer)798 RandomIndexWriter (org.apache.lucene.index.RandomIndexWriter)752 IndexReader (org.apache.lucene.index.IndexReader)598 Field (org.apache.lucene.document.Field)480 IndexSearcher (org.apache.lucene.search.IndexSearcher)470 Term (org.apache.lucene.index.Term)456 BytesRef (org.apache.lucene.util.BytesRef)415 StringField (org.apache.lucene.document.StringField)403 TextField (org.apache.lucene.document.TextField)389 NumericDocValuesField (org.apache.lucene.document.NumericDocValuesField)325 IndexWriter (org.apache.lucene.index.IndexWriter)312 IndexWriterConfig (org.apache.lucene.index.IndexWriterConfig)278 TopDocs (org.apache.lucene.search.TopDocs)270 TermQuery (org.apache.lucene.search.TermQuery)237 FieldType (org.apache.lucene.document.FieldType)231 DirectoryReader (org.apache.lucene.index.DirectoryReader)226 Test (org.junit.Test)222 RAMDirectory (org.apache.lucene.store.RAMDirectory)211