public override void SetUp()
{
base.SetUp();
dir = NewDirectory();
IndexWriterConfig iwc = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer
(Random()));
iwc.SetMergePolicy(NewLogMergePolicy());
var iw = new RandomIndexWriter(Random(), dir, iwc);
var doc = new Document
{
NewStringField("id", "1", Field.Store.YES),
NewTextField("body", "some contents and more contents", Field.Store.NO),
new NumericDocValuesField("popularity", 5)
};
iw.AddDocument(doc);
doc = new Document
{
NewStringField("id", "2", Field.Store.YES),
NewTextField("body", "another document with different contents", Field.Store
.NO),
new NumericDocValuesField("popularity", 20)
};
iw.AddDocument(doc);
doc = new Document
{
NewStringField("id", "3", Field.Store.YES),
NewTextField("body", "crappy contents", Field.Store.NO),
new NumericDocValuesField("popularity", 2)
};
iw.AddDocument(doc);
iw.ForceMerge(1);
reader = iw.Reader;
iw.Dispose();
}
public void TestFieldNotPresent()
{
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);
int num = AtLeast(3);
int skip = Random().Next(num);
var terms = new List<Term>();
for (int i = 0; i < num; i++)
{
terms.Add(new Term("field" + i, "content1"));
Document doc = new Document();
if (skip == i)
{
continue;
}
doc.Add(NewStringField("field" + i, "content1", Field.Store.YES));
w.AddDocument(doc);
}
w.ForceMerge(1);
IndexReader reader = w.Reader;
w.Dispose();
assertEquals(1, reader.Leaves.size());
AtomicReaderContext context = reader.Leaves.First();
TermsFilter tf = new TermsFilter(terms);
FixedBitSet bits = (FixedBitSet)tf.GetDocIdSet(context, context.AtomicReader.LiveDocs);
assertEquals("Must be num fields - 1 since we skip only one field", num - 1, bits.Cardinality());
reader.Dispose();
dir.Dispose();
}
public virtual void Test()
{
Directory dir = NewDirectory();
IndexWriterConfig conf = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random()));
conf.SetCodec(new Lucene46Codec());
RandomIndexWriter riw = new RandomIndexWriter(Random(), dir, conf);
Document doc = new Document();
// these fields should sometimes get term vectors, etc
Field idField = NewStringField("id", "", Field.Store.NO);
Field bodyField = NewTextField("body", "", Field.Store.NO);
Field dvField = new NumericDocValuesField("dv", 5);
doc.Add(idField);
doc.Add(bodyField);
doc.Add(dvField);
for (int i = 0; i < 100; i++)
{
idField.StringValue = Convert.ToString(i);
bodyField.StringValue = TestUtil.RandomUnicodeString(Random());
riw.AddDocument(doc);
if (Random().Next(7) == 0)
{
riw.Commit();
}
// TODO: we should make a new format with a clean header...
// if (Random().nextInt(20) == 0) {
// riw.DeleteDocuments(new Term("id", Integer.toString(i)));
// }
}
riw.Dispose();
CheckHeaders(dir);
dir.Dispose();
}
public override void SetUp()
{
base.SetUp();
dir = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, new MockAnalyzer(Random(), MockTokenizer.WHITESPACE, true), Similarity, TimeZone);
for (int i = 900; i < 1112; i++)
{
Document doc = new Document();
string num = Regex.Replace(Regex.Replace(English.IntToEnglish(i), "[-]", " "), "[,]", "");
doc.Add(NewTextField("numbers", num, Field.Store.NO));
writer.AddDocument(doc);
}
{
Document doc = new Document();
doc.Add(NewTextField("numbers", "thou hast sand betwixt thy toes", Field.Store.NO));
writer.AddDocument(doc);
}
{
Document doc = new Document();
doc.Add(NewTextField("numbers", "hundredeight eightyeight yeight", Field.Store.NO));
writer.AddDocument(doc);
}
{
Document doc = new Document();
doc.Add(NewTextField("numbers", "tres y cinco", Field.Store.NO));
writer.AddDocument(doc);
}
writer.Commit();
writer.Dispose();
}
public void BeforeClass()
{
Dir = NewDirectory();
Sdir1 = NewDirectory();
Sdir2 = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), Dir, new MockAnalyzer(Random()), Similarity, TimeZone);
RandomIndexWriter swriter1 = new RandomIndexWriter(Random(), Sdir1, new MockAnalyzer(Random()), Similarity, TimeZone);
RandomIndexWriter swriter2 = new RandomIndexWriter(Random(), Sdir2, new MockAnalyzer(Random()), Similarity, TimeZone);
for (int i = 0; i < 10; i++)
{
Document doc = new Document();
doc.Add(NewStringField("data", Convert.ToString(i), Field.Store.NO));
writer.AddDocument(doc);
((i % 2 == 0) ? swriter1 : swriter2).AddDocument(doc);
}
writer.ForceMerge(1);
swriter1.ForceMerge(1);
swriter2.ForceMerge(1);
writer.Dispose();
swriter1.Dispose();
swriter2.Dispose();
Reader = DirectoryReader.Open(Dir);
Searcher = NewSearcher(Reader);
MultiReader = new MultiReader(new IndexReader[] { DirectoryReader.Open(Sdir1), DirectoryReader.Open(Sdir2) }, true);
MultiSearcher = NewSearcher(MultiReader);
MultiReaderDupls = new MultiReader(new IndexReader[] { DirectoryReader.Open(Sdir1), DirectoryReader.Open(Dir) }, true);
MultiSearcherDupls = NewSearcher(MultiReaderDupls);
}
public override void SetUp()
{
base.SetUp();
// we generate aweful regexps: good for testing.
// but for preflex codec, the test can be very slow, so use less iterations.
NumIterations = Codec.Default.Name.Equals("Lucene3x") ? 10 * RANDOM_MULTIPLIER : AtLeast(50);
Dir = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), Dir, (IndexWriterConfig)NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random(), MockTokenizer.KEYWORD, false)).SetMaxBufferedDocs(TestUtil.NextInt(Random(), 50, 1000)));
Document doc = new Document();
Field field = NewStringField("field", "", Field.Store.YES);
doc.Add(field);
Terms = new SortedSet<BytesRef>();
int num = AtLeast(200);
for (int i = 0; i < num; i++)
{
string s = TestUtil.RandomUnicodeString(Random());
field.StringValue = s;
Terms.Add(new BytesRef(s));
writer.AddDocument(doc);
}
TermsAutomaton = BasicAutomata.MakeStringUnion(Terms);
Reader = writer.Reader;
Searcher = NewSearcher(Reader);
writer.Dispose();
}
public override void SetUp()
{
base.SetUp();
Dir = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), Dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())).SetMaxBufferedDocs(TestUtil.NextInt(Random(), 50, 1000)));
Document doc = new Document();
FieldType customType = new FieldType(TextField.TYPE_STORED);
customType.OmitNorms = true;
Field field = NewField("field", "", customType);
doc.Add(field);
NumberFormatInfo df = new NumberFormatInfo();
df.NumberDecimalDigits = 0;
//NumberFormat df = new DecimalFormat("000", new DecimalFormatSymbols(Locale.ROOT));
for (int i = 0; i < 1000; i++)
{
field.StringValue = i.ToString(df);
writer.AddDocument(doc);
}
Reader = writer.Reader;
writer.Dispose();
Searcher = NewSearcher(Reader);
}
public virtual void TestBasic()
{
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);
Document doc = new Document();
FieldType ft = new FieldType(TextField.TYPE_NOT_STORED);
ft.IndexOptions = FieldInfo.IndexOptions.DOCS_AND_FREQS;
Field f = NewField("foo", "this is a test test", ft);
doc.Add(f);
for (int i = 0; i < 100; i++)
{
w.AddDocument(doc);
}
IndexReader reader = w.Reader;
w.Dispose();
Assert.IsNull(MultiFields.GetTermPositionsEnum(reader, null, "foo", new BytesRef("test")));
DocsEnum de = TestUtil.Docs(Random(), reader, "foo", new BytesRef("test"), null, null, DocsEnum.FLAG_FREQS);
while (de.NextDoc() != DocIdSetIterator.NO_MORE_DOCS)
{
Assert.AreEqual(2, de.Freq());
}
reader.Dispose();
dir.Dispose();
}
public void BeforeClass()
{
Directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), Directory, Similarity, TimeZone);
Document doc = new Document();
Field field = NewStringField(FIELD, "meaninglessnames", Field.Store.NO);
doc.Add(field);
for (int i = 0; i < 5137; ++i)
{
writer.AddDocument(doc);
}
field.StringValue = "tangfulin";
writer.AddDocument(doc);
field.StringValue = "meaninglessnames";
for (int i = 5138; i < 11377; ++i)
{
writer.AddDocument(doc);
}
field.StringValue = "tangfulin";
writer.AddDocument(doc);
Reader = writer.Reader;
Searcher = NewSearcher(Reader);
writer.Dispose();
}
public override void SetUp()
{
base.SetUp();
dir = NewDirectory();
var iw = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);
int numDocs = TestUtil.NextInt(Random(), 2049, 4000);
for (int i = 0; i < numDocs; i++)
{
var document = new Document
{
NewTextField("english", English.IntToEnglish(i), Field.Store.NO),
NewTextField("oddeven", (i%2 == 0) ? "even" : "odd", Field.Store.NO
),
NewStringField("byte", string.Empty + (unchecked((byte) Random().Next
())), Field.Store.NO),
NewStringField("short", string.Empty + ((short) Random().Next()), Field.Store
.NO),
new IntField("int", Random().Next(), Field.Store.NO),
new LongField("long", Random().NextLong(), Field.Store.NO),
new FloatField("float", Random().NextFloat(), Field.Store.NO),
new DoubleField("double", Random().NextDouble(), Field.Store.NO),
new NumericDocValuesField("intdocvalues", Random().Next()),
new FloatDocValuesField("floatdocvalues", Random().NextFloat())
};
iw.AddDocument(document);
}
reader = iw.Reader;
iw.Dispose();
searcher = NewSearcher(reader);
}
public override void SetUp()
{
base.SetUp();
dir = NewDirectory();
var iw = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);
var doc = new Document
{
NewStringField("id", "1", Field.Store.YES),
NewTextField("body", "some contents and more contents", Field.Store.NO),
new NumericDocValuesField("popularity", 5)
};
iw.AddDocument(doc);
doc = new Document
{
NewStringField("id", "2", Field.Store.YES),
NewTextField("body", "another document with different contents", Field.Store
.NO),
new NumericDocValuesField("popularity", 20)
};
iw.AddDocument(doc);
doc = new Document
{
NewStringField("id", "3", Field.Store.YES),
NewTextField("body", "crappy contents", Field.Store.NO),
new NumericDocValuesField("popularity", 2)
};
iw.AddDocument(doc);
reader = iw.Reader;
searcher = new IndexSearcher(reader);
iw.Dispose();
}
public virtual void TestPrefixQuery_Mem()
{
Directory directory = NewDirectory();
string[] categories = new string[] { "/Computers", "/Computers/Mac", "/Computers/Windows" };
RandomIndexWriter writer = new RandomIndexWriter(Random(), directory, Similarity, TimeZone);
for (int i = 0; i < categories.Length; i++)
{
Document doc = new Document();
doc.Add(NewStringField("category", categories[i], Field.Store.YES));
writer.AddDocument(doc);
}
IndexReader reader = writer.Reader;
PrefixQuery query = new PrefixQuery(new Term("category", "/Computers"));
IndexSearcher searcher = NewSearcher(reader);
ScoreDoc[] hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length, "All documents in /Computers category and below");
query = new PrefixQuery(new Term("category", "/Computers/Mac"));
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(1, hits.Length, "One in /Computers/Mac");
query = new PrefixQuery(new Term("category", ""));
Terms terms = MultiFields.GetTerms(searcher.IndexReader, "category");
Assert.IsFalse(query.GetTermsEnum(terms) is PrefixTermsEnum);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length, "everything");
writer.Dispose();
reader.Dispose();
directory.Dispose();
}
public void TestReverse()
{
Directory dir = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);
Document doc = new Document();
doc.Add(NewStringField("value", "foo", Field.Store.NO));
doc.Add(NewStringField("value", "bar", Field.Store.NO));
doc.Add(NewStringField("id", "1", Field.Store.YES));
writer.AddDocument(doc);
doc = new Document();
doc.Add(NewStringField("value", "baz", Field.Store.NO));
doc.Add(NewStringField("id", "2", Field.Store.YES));
writer.AddDocument(doc);
IndexReader ir = writer.Reader;
writer.Dispose();
IndexSearcher searcher = NewSearcher(ir);
Sort sort = new Sort(new SortedSetSortField("value", true));
TopDocs td = searcher.Search(new MatchAllDocsQuery(), 10, sort);
assertEquals(2, td.TotalHits);
// 'bar' comes before 'baz'
assertEquals("2", searcher.Doc(td.ScoreDocs[0].Doc).Get("id"));
assertEquals("1", searcher.Doc(td.ScoreDocs[1].Doc).Get("id"));
ir.Dispose();
dir.Dispose();
}
public virtual void TestMethod()
{
Directory directory = NewDirectory();
string[] values = new string[] { "1", "2", "3", "4" };
RandomIndexWriter writer = new RandomIndexWriter(Random(), directory, Similarity, TimeZone);
for (int i = 0; i < values.Length; i++)
{
Document doc = new Document();
doc.Add(NewStringField(FIELD, values[i], Field.Store.YES));
writer.AddDocument(doc);
}
IndexReader ir = writer.Reader;
writer.Dispose();
BooleanQuery booleanQuery1 = new BooleanQuery();
booleanQuery1.Add(new TermQuery(new Term(FIELD, "1")), BooleanClause.Occur.SHOULD);
booleanQuery1.Add(new TermQuery(new Term(FIELD, "2")), BooleanClause.Occur.SHOULD);
BooleanQuery query = new BooleanQuery();
query.Add(booleanQuery1, BooleanClause.Occur.MUST);
query.Add(new TermQuery(new Term(FIELD, "9")), BooleanClause.Occur.MUST_NOT);
IndexSearcher indexSearcher = NewSearcher(ir);
ScoreDoc[] hits = indexSearcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(2, hits.Length, "Number of matched documents");
ir.Dispose();
directory.Dispose();
}
public void TestInternalLevenshteinDistance()
{
DirectSpellChecker spellchecker = new DirectSpellChecker();
Directory dir = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir,
new MockAnalyzer(Random(), MockTokenizer.KEYWORD, true), Similarity, TimeZone);
string[] termsToAdd = { "metanoia", "metanoian", "metanoiai", "metanoias", "metanoið‘" };
for (int i = 0; i < termsToAdd.Length; i++)
{
Document doc = new Document();
doc.Add(NewTextField("repentance", termsToAdd[i], Field.Store.NO));
writer.AddDocument(doc);
}
IndexReader ir = writer.Reader;
string misspelled = "metanoix";
SuggestWord[] similar = spellchecker.SuggestSimilar(new Term("repentance", misspelled), 4, ir, SuggestMode.SUGGEST_WHEN_NOT_IN_INDEX);
assertTrue(similar.Length == 4);
IStringDistance sd = spellchecker.Distance;
assertTrue(sd is LuceneLevenshteinDistance);
foreach (SuggestWord word in similar)
{
assertTrue(word.Score == sd.GetDistance(word.String, misspelled));
assertTrue(word.Score == sd.GetDistance(misspelled, word.String)); // LUCNENET TODO: Perhaps change this to word.ToString()?
}
ir.Dispose();
writer.Dispose();
dir.Dispose();
}
public virtual void TestString()
{
Directory dir = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir);
Document doc = new Document();
doc.Add(NewStringField("value", "foo", Field.Store.YES));
writer.AddDocument(doc);
doc = new Document();
doc.Add(NewStringField("value", "bar", Field.Store.YES));
writer.AddDocument(doc);
IndexReader ir = writer.Reader;
writer.Dispose();
IndexSearcher searcher = NewSearcher(ir);
Sort sort = new Sort(new SortField("value", SortField.Type_e.STRING));
TopDocs td = searcher.Search(new MatchAllDocsQuery(), 10, sort);
Assert.AreEqual(2, td.TotalHits);
// 'bar' comes before 'foo'
Assert.AreEqual("bar", searcher.Doc(td.ScoreDocs[0].Doc).Get("value"));
Assert.AreEqual("foo", searcher.Doc(td.ScoreDocs[1].Doc).Get("value"));
ir.Dispose();
dir.Dispose();
}
public virtual void TestRollbackIntegrityWithBufferFlush()
{
Directory dir = NewDirectory();
RandomIndexWriter rw = new RandomIndexWriter(Random(), dir);
for (int i = 0; i < 5; i++)
{
Document doc = new Document();
doc.Add(NewStringField("pk", Convert.ToString(i), Field.Store.YES));
rw.AddDocument(doc);
}
rw.Dispose();
// If buffer size is small enough to cause a flush, errors ensue...
IndexWriter w = new IndexWriter(dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())).SetMaxBufferedDocs(2).SetOpenMode(IndexWriterConfig.OpenMode_e.APPEND));
for (int i = 0; i < 3; i++)
{
Document doc = new Document();
string value = Convert.ToString(i);
doc.Add(NewStringField("pk", value, Field.Store.YES));
doc.Add(NewStringField("text", "foo", Field.Store.YES));
w.UpdateDocument(new Term("pk", value), doc);
}
w.Rollback();
IndexReader r = DirectoryReader.Open(dir);
Assert.AreEqual(5, r.NumDocs, "index should contain same number of docs post rollback");
r.Dispose();
dir.Dispose();
}
public override void SetUp()
{
base.SetUp();
INDEX_SIZE = AtLeast(2000);
Index = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), Index);
RandomGen random = new RandomGen(this, Random());
for (int i = 0; i < INDEX_SIZE; ++i) // don't decrease; if to low the
{
// problem doesn't show up
Document doc = new Document();
if ((i % 5) != 0) // some documents must not have an entry in the first
{
// sort field
doc.Add(NewStringField("publicationDate_", random.LuceneDate, Field.Store.YES));
}
if ((i % 7) == 0) // some documents to match the query (see below)
{
doc.Add(NewTextField("content", "test", Field.Store.YES));
}
// every document has a defined 'mandant' field
doc.Add(NewStringField("mandant", Convert.ToString(i % 3), Field.Store.YES));
writer.AddDocument(doc);
}
Reader = writer.Reader;
writer.Dispose();
Query = new TermQuery(new Term("content", "test"));
}
public override void SetUp()
{
base.SetUp();
Dir = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), Dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())).SetMaxBufferedDocs(TestUtil.NextInt(Random(), 50, 1000)));
Document doc = new Document();
Field field = NewStringField("field", "", Field.Store.NO);
doc.Add(field);
NumberFormatInfo df = new NumberFormatInfo();
df.NumberDecimalDigits = 0;
//NumberFormat df = new DecimalFormat("000", new DecimalFormatSymbols(Locale.ROOT));
for (int i = 0; i < 1000; i++)
{
field.StringValue = i.ToString(df);
writer.AddDocument(doc);
}
Reader = writer.Reader;
Searcher = NewSearcher(Reader);
writer.Dispose();
if (VERBOSE)
{
Console.WriteLine("TEST: setUp searcher=" + Searcher);
}
}
public static void BeforeClass()
{
Directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), Directory, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())).SetMergePolicy(NewLogMergePolicy()));
for (int i = 0; i < DocFields.Length; i++)
{
Document doc = new Document();
doc.Add(NewTextField(field, DocFields[i], Field.Store.NO));
writer.AddDocument(doc);
}
writer.Dispose();
LittleReader = DirectoryReader.Open(Directory);
Searcher = NewSearcher(LittleReader);
// this is intentionally using the baseline sim, because it compares against bigSearcher (which uses a random one)
Searcher.Similarity = new DefaultSimilarity();
// Make big index
Dir2 = new MockDirectoryWrapper(Random(), new RAMDirectory(Directory, IOContext.DEFAULT));
// First multiply small test index:
MulFactor = 1;
int docCount = 0;
if (VERBOSE)
{
Console.WriteLine("\nTEST: now copy index...");
}
do
{
if (VERBOSE)
{
Console.WriteLine("\nTEST: cycle...");
}
Directory copy = new MockDirectoryWrapper(Random(), new RAMDirectory(Dir2, IOContext.DEFAULT));
RandomIndexWriter w = new RandomIndexWriter(Random(), Dir2);
w.AddIndexes(copy);
docCount = w.MaxDoc();
w.Dispose();
MulFactor *= 2;
} while (docCount < 3000);
RandomIndexWriter riw = new RandomIndexWriter(Random(), Dir2, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())).SetMaxBufferedDocs(TestUtil.NextInt(Random(), 50, 1000)));
Document doc_ = new Document();
doc_.Add(NewTextField("field2", "xxx", Field.Store.NO));
for (int i = 0; i < NUM_EXTRA_DOCS / 2; i++)
{
riw.AddDocument(doc_);
}
doc_ = new Document();
doc_.Add(NewTextField("field2", "big bad bug", Field.Store.NO));
for (int i = 0; i < NUM_EXTRA_DOCS / 2; i++)
{
riw.AddDocument(doc_);
}
Reader = riw.Reader;
BigSearcher = NewSearcher(Reader);
riw.Dispose();
}
public virtual void Test()
{
Directory dir = NewDirectory();
RandomIndexWriter riw = new RandomIndexWriter(Random(), dir);
Document doc = new Document();
doc.Add(new TextField("eng", new BugReproTokenStream()));
riw.AddDocument(doc);
riw.Dispose();
dir.Dispose();
}
public virtual void TestBasic()
{
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(Random(), dir, Iwc);
Document doc = new Document();
FieldType ft = new FieldType(TextField.TYPE_NOT_STORED);
ft.IndexOptionsValue = FieldInfo.IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS;
if (Random().NextBoolean())
{
ft.StoreTermVectors = true;
ft.StoreTermVectorPositions = Random().NextBoolean();
ft.StoreTermVectorOffsets = Random().NextBoolean();
}
Token[] tokens = new Token[] { MakeToken("a", 1, 0, 6), MakeToken("b", 1, 8, 9), MakeToken("a", 1, 9, 17), MakeToken("c", 1, 19, 50) };
doc.Add(new Field("content", new CannedTokenStream(tokens), ft));
w.AddDocument(doc);
IndexReader r = w.Reader;
w.Dispose();
DocsAndPositionsEnum dp = MultiFields.GetTermPositionsEnum(r, null, "content", new BytesRef("a"));
Assert.IsNotNull(dp);
Assert.AreEqual(0, dp.NextDoc());
Assert.AreEqual(2, dp.Freq());
Assert.AreEqual(0, dp.NextPosition());
Assert.AreEqual(0, dp.StartOffset());
Assert.AreEqual(6, dp.EndOffset());
Assert.AreEqual(2, dp.NextPosition());
Assert.AreEqual(9, dp.StartOffset());
Assert.AreEqual(17, dp.EndOffset());
Assert.AreEqual(DocIdSetIterator.NO_MORE_DOCS, dp.NextDoc());
dp = MultiFields.GetTermPositionsEnum(r, null, "content", new BytesRef("b"));
Assert.IsNotNull(dp);
Assert.AreEqual(0, dp.NextDoc());
Assert.AreEqual(1, dp.Freq());
Assert.AreEqual(1, dp.NextPosition());
Assert.AreEqual(8, dp.StartOffset());
Assert.AreEqual(9, dp.EndOffset());
Assert.AreEqual(DocIdSetIterator.NO_MORE_DOCS, dp.NextDoc());
dp = MultiFields.GetTermPositionsEnum(r, null, "content", new BytesRef("c"));
Assert.IsNotNull(dp);
Assert.AreEqual(0, dp.NextDoc());
Assert.AreEqual(1, dp.Freq());
Assert.AreEqual(3, dp.NextPosition());
Assert.AreEqual(19, dp.StartOffset());
Assert.AreEqual(50, dp.EndOffset());
Assert.AreEqual(DocIdSetIterator.NO_MORE_DOCS, dp.NextDoc());
r.Dispose();
dir.Dispose();
}
public virtual void TestBefore()
{
// create an index
Directory indexStore = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), indexStore);
long now = DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond;
Document doc = new Document();
// add time that is in the past
doc.Add(NewStringField("datefield", DateTools.TimeToString(now - 1000, DateTools.Resolution.MILLISECOND), Field.Store.YES));
doc.Add(NewTextField("body", "Today is a very sunny day in New York City", Field.Store.YES));
writer.AddDocument(doc);
IndexReader reader = writer.Reader;
writer.Dispose();
IndexSearcher searcher = NewSearcher(reader);
// filter that should preserve matches
// DateFilter df1 = DateFilter.Before("datefield", now);
TermRangeFilter df1 = TermRangeFilter.NewStringRange("datefield", DateTools.TimeToString(now - 2000, DateTools.Resolution.MILLISECOND), DateTools.TimeToString(now, DateTools.Resolution.MILLISECOND), false, true);
// filter that should discard matches
// DateFilter df2 = DateFilter.Before("datefield", now - 999999);
TermRangeFilter df2 = TermRangeFilter.NewStringRange("datefield", DateTools.TimeToString(0, DateTools.Resolution.MILLISECOND), DateTools.TimeToString(now - 2000, DateTools.Resolution.MILLISECOND), true, false);
// search something that doesn't exist with DateFilter
Query query1 = new TermQuery(new Term("body", "NoMatchForthis"));
// search for something that does exists
Query query2 = new TermQuery(new Term("body", "sunny"));
ScoreDoc[] result;
// ensure that queries return expected results without DateFilter first
result = searcher.Search(query1, null, 1000).ScoreDocs;
Assert.AreEqual(0, result.Length);
result = searcher.Search(query2, null, 1000).ScoreDocs;
Assert.AreEqual(1, result.Length);
// run queries with DateFilter
result = searcher.Search(query1, df1, 1000).ScoreDocs;
Assert.AreEqual(0, result.Length);
result = searcher.Search(query1, df2, 1000).ScoreDocs;
Assert.AreEqual(0, result.Length);
result = searcher.Search(query2, df1, 1000).ScoreDocs;
Assert.AreEqual(1, result.Length);
result = searcher.Search(query2, df2, 1000).ScoreDocs;
Assert.AreEqual(0, result.Length);
reader.Dispose();
indexStore.Dispose();
}
public override void SetUp()
{
base.SetUp();
Directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), Directory, Similarity, TimeZone);
Document doc = new Document();
doc.Add(NewTextField(FN, "the quick brown fox jumps over the lazy ??? dog 493432 49344", Field.Store.NO));
writer.AddDocument(doc);
Reader = writer.Reader;
writer.Dispose();
Searcher = NewSearcher(Reader);
}
public override void SetUp()
{
base.SetUp();
directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), directory, new MockAnalyzer(Random(), MockTokenizer.WHITESPACE, false), Similarity, TimeZone);
AddDoc(writer, @"admin guest", @"010", @"20040101", @"Y");
AddDoc(writer, @"guest", @"020", @"20040101", @"Y");
AddDoc(writer, @"guest", @"020", @"20050101", @"Y");
AddDoc(writer, @"admin", @"020", @"20050101", @"Maybe");
AddDoc(writer, @"admin guest", @"030", @"20050101", @"N");
reader = SlowCompositeReaderWrapper.Wrap(writer.Reader);
writer.Dispose();
}
public override void SetUp()
{
base.SetUp();
directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), directory, Similarity, TimeZone);
// Add series of docs with specific information for MoreLikeThis
AddDoc(writer, "lucene");
AddDoc(writer, "lucene release");
reader = writer.Reader;
writer.Dispose();
searcher = NewSearcher(reader);
}
public override void SetUp()
{
base.SetUp();
// create test index
MDirectory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), MDirectory, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random(), MockTokenizer.SIMPLE, true, MockTokenFilter.ENGLISH_STOPSET)).SetMergePolicy(NewLogMergePolicy()).SetSimilarity(new DefaultSimilarity()));
AddDocument(writer, "1", "I think it should work.");
AddDocument(writer, "2", "I think it should work.");
AddDocument(writer, "3", "I think it should work.");
AddDocument(writer, "4", "I think it should work.");
Reader = writer.Reader;
writer.Dispose();
Searcher = NewSearcher(Reader);
Searcher.Similarity = new DefaultSimilarity();
}
public override void SetUp()
{
base.SetUp();
Dir = NewDirectory();
RandomIndexWriter iw = new RandomIndexWriter(Random(), Dir, Similarity, TimeZone);
for (int i = 0; i < 100; i++)
{
Document doc = new Document();
doc.Add(NewStringField("field", Convert.ToString(i), Field.Store.NO));
doc.Add(NewStringField("field2", Convert.ToString(i % 2 == 0), Field.Store.NO));
iw.AddDocument(doc);
}
Reader = iw.Reader;
iw.Dispose();
}
public override void SetUp()
{
base.SetUp();
Directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), Directory, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())).SetMergePolicy(NewLogMergePolicy()));
for (int i = 0; i < DocFields.Length; i++)
{
Document doc = new Document();
doc.Add(NewTextField(field, DocFields[i], Field.Store.YES));
writer.AddDocument(doc);
}
Reader = writer.Reader;
writer.Dispose();
Searcher = NewSearcher(Reader);
}
public override void SetUp()
{
base.SetUp();
dir = NewDirectory();
IndexWriterConfig iwConfig = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random()));
iwConfig.SetMergePolicy(NewLogMergePolicy());
RandomIndexWriter iw = new RandomIndexWriter(Random(), dir, iwConfig);
Document document = new Document();
Field idField = new StringField("id", "", Field.Store.NO);
document.Add(idField);
iw.AddDocument(document);
ir = iw.Reader;
@is = NewSearcher(ir);
iw.Dispose();
}
public override void BeforeClass()
{
base.BeforeClass();
Directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), Directory, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())).SetMergePolicy(NewLogMergePolicy()));
for (int i = 0; i < DocFields.Length; i++)
{
Document doc = new Document();
doc.Add(NewTextField(field, DocFields[i], Field.Store.NO));
writer.AddDocument(doc);
}
writer.Dispose();
LittleReader = DirectoryReader.Open(Directory);
Searcher = NewSearcher(LittleReader);
// this is intentionally using the baseline sim, because it compares against bigSearcher (which uses a random one)
Searcher.Similarity = new DefaultSimilarity();
// Make big index
Dir2 = new MockDirectoryWrapper(Random(), new RAMDirectory(Directory, IOContext.DEFAULT));
// First multiply small test index:
MulFactor = 1;
int docCount = 0;
if (VERBOSE)
{
Console.WriteLine("\nTEST: now copy index...");
}
do
{
if (VERBOSE)
{
Console.WriteLine("\nTEST: cycle...");
}
Directory copy = new MockDirectoryWrapper(Random(), new RAMDirectory(Dir2, IOContext.DEFAULT));
RandomIndexWriter w = new RandomIndexWriter(Random(), Dir2, Similarity, TimeZone);
w.AddIndexes(copy);
docCount = w.MaxDoc;
w.Dispose();
MulFactor *= 2;
} while (docCount < 3000);
RandomIndexWriter riw = new RandomIndexWriter(Random(), Dir2, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())).SetMaxBufferedDocs(TestUtil.NextInt(Random(), 50, 1000)));
Document doc_ = new Document();
doc_.Add(NewTextField("field2", "xxx", Field.Store.NO));
for (int i = 0; i < NUM_EXTRA_DOCS / 2; i++)
{
riw.AddDocument(doc_);
}
doc_ = new Document();
doc_.Add(NewTextField("field2", "big bad bug", Field.Store.NO));
for (int i = 0; i < NUM_EXTRA_DOCS / 2; i++)
{
riw.AddDocument(doc_);
}
Reader = riw.Reader;
BigSearcher = NewSearcher(Reader);
riw.Dispose();
}
public void TestSimpleWithScoring()
{
const string idField = "id";
const string toField = "movieId";
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(Random, dir,
NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random))
.SetMergePolicy(NewLogMergePolicy()));
// 0
Document doc = new Document();
doc.Add(new TextField("description", "A random movie", Field.Store.NO));
doc.Add(new TextField("name", "Movie 1", Field.Store.NO));
doc.Add(new TextField(idField, "1", Field.Store.NO));
w.AddDocument(doc);
// 1
doc = new Document();
doc.Add(new TextField("subtitle", "The first subtitle of this movie", Field.Store.NO));
doc.Add(new TextField(idField, "2", Field.Store.NO));
doc.Add(new TextField(toField, "1", Field.Store.NO));
w.AddDocument(doc);
// 2
doc = new Document();
doc.Add(new TextField("subtitle", "random subtitle; random event movie", Field.Store.NO));
doc.Add(new TextField(idField, "3", Field.Store.NO));
doc.Add(new TextField(toField, "1", Field.Store.NO));
w.AddDocument(doc);
// 3
doc = new Document();
doc.Add(new TextField("description", "A second random movie", Field.Store.NO));
doc.Add(new TextField("name", "Movie 2", Field.Store.NO));
doc.Add(new TextField(idField, "4", Field.Store.NO));
w.AddDocument(doc);
w.Commit();
// 4
doc = new Document();
doc.Add(new TextField("subtitle", "a very random event happened during christmas night", Field.Store.NO));
doc.Add(new TextField(idField, "5", Field.Store.NO));
doc.Add(new TextField(toField, "4", Field.Store.NO));
w.AddDocument(doc);
// 5
doc = new Document();
doc.Add(new TextField("subtitle", "movie end movie test 123 test 123 random", Field.Store.NO));
doc.Add(new TextField(idField, "6", Field.Store.NO));
doc.Add(new TextField(toField, "4", Field.Store.NO));
w.AddDocument(doc);
IndexSearcher indexSearcher = new IndexSearcher(w.GetReader());
w.Dispose();
// Search for movie via subtitle
Query joinQuery = JoinUtil.CreateJoinQuery(toField, false, idField,
new TermQuery(new Term("subtitle", "random")), indexSearcher, ScoreMode.Max);
TopDocs result = indexSearcher.Search(joinQuery, 10);
assertEquals(2, result.TotalHits);
assertEquals(0, result.ScoreDocs[0].Doc);
assertEquals(3, result.ScoreDocs[1].Doc);
// Score mode max.
joinQuery = JoinUtil.CreateJoinQuery(toField, false, idField, new TermQuery(new Term("subtitle", "movie")),
indexSearcher, ScoreMode.Max);
result = indexSearcher.Search(joinQuery, 10);
assertEquals(2, result.TotalHits);
assertEquals(3, result.ScoreDocs[0].Doc);
assertEquals(0, result.ScoreDocs[1].Doc);
// Score mode total
joinQuery = JoinUtil.CreateJoinQuery(toField, false, idField, new TermQuery(new Term("subtitle", "movie")),
indexSearcher, ScoreMode.Total);
result = indexSearcher.Search(joinQuery, 10);
assertEquals(2, result.TotalHits);
assertEquals(0, result.ScoreDocs[0].Doc);
assertEquals(3, result.ScoreDocs[1].Doc);
//Score mode avg
joinQuery = JoinUtil.CreateJoinQuery(toField, false, idField, new TermQuery(new Term("subtitle", "movie")),
indexSearcher, ScoreMode.Avg);
result = indexSearcher.Search(joinQuery, 10);
assertEquals(2, result.TotalHits);
assertEquals(3, result.ScoreDocs[0].Doc);
assertEquals(0, result.ScoreDocs[1].Doc);
indexSearcher.IndexReader.Dispose();
dir.Dispose();
}
public virtual void TestSparseFacets()
{
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);
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);
FacetsConfig config = new FacetsConfig();
Document doc = new Document();
doc.Add(new IntField("num", 10, Field.Store.NO));
doc.Add(new FacetField("a", "foo1"));
writer.AddDocument(config.Build(taxoWriter, doc));
if (Random().NextBoolean())
{
writer.Commit();
}
doc = new Document();
doc.Add(new IntField("num", 20, Field.Store.NO));
doc.Add(new FacetField("a", "foo2"));
doc.Add(new FacetField("b", "bar1"));
writer.AddDocument(config.Build(taxoWriter, doc));
if (Random().NextBoolean())
{
writer.Commit();
}
doc = new Document();
doc.Add(new IntField("num", 30, Field.Store.NO));
doc.Add(new FacetField("a", "foo3"));
doc.Add(new FacetField("b", "bar2"));
doc.Add(new FacetField("c", "baz1"));
writer.AddDocument(config.Build(taxoWriter, doc));
// NRT open
IndexSearcher searcher = NewSearcher(writer.Reader);
writer.Dispose();
// NRT open
var taxoReader = new DirectoryTaxonomyReader(taxoWriter);
taxoWriter.Dispose();
FacetsCollector c = new FacetsCollector();
searcher.Search(new MatchAllDocsQuery(), c);
TaxonomyFacetSumValueSource facets = new TaxonomyFacetSumValueSource(taxoReader, new FacetsConfig(), c, new IntFieldSource("num"));
// Ask for top 10 labels for any dims that have counts:
IList <FacetResult> results = facets.GetAllDims(10);
Assert.AreEqual(3, results.Count);
Assert.AreEqual("dim=a path=[] value=60.0 childCount=3\n foo3 (30.0)\n foo2 (20.0)\n foo1 (10.0)\n", results[0].ToString());
Assert.AreEqual("dim=b path=[] value=50.0 childCount=2\n bar2 (30.0)\n bar1 (20.0)\n", results[1].ToString());
Assert.AreEqual("dim=c path=[] value=30.0 childCount=1\n baz1 (30.0)\n", results[2].ToString());
IOUtils.Close(searcher.IndexReader, taxoReader, dir, taxoDir);
}
public virtual void TestBasic()
{
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);
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);
FacetsConfig config = new FacetsConfig();
// Reused across documents, to add the necessary facet
// fields:
Document doc = new Document();
doc.Add(new IntField("num", 10, Field.Store.NO));
doc.Add(new FacetField("Author", "Bob"));
writer.AddDocument(config.Build(taxoWriter, doc));
doc = new Document();
doc.Add(new IntField("num", 20, Field.Store.NO));
doc.Add(new FacetField("Author", "Lisa"));
writer.AddDocument(config.Build(taxoWriter, doc));
doc = new Document();
doc.Add(new IntField("num", 30, Field.Store.NO));
doc.Add(new FacetField("Author", "Lisa"));
writer.AddDocument(config.Build(taxoWriter, doc));
doc = new Document();
doc.Add(new IntField("num", 40, Field.Store.NO));
doc.Add(new FacetField("Author", "Susan"));
writer.AddDocument(config.Build(taxoWriter, doc));
doc = new Document();
doc.Add(new IntField("num", 45, Field.Store.NO));
doc.Add(new FacetField("Author", "Frank"));
writer.AddDocument(config.Build(taxoWriter, doc));
// NRT open
IndexSearcher searcher = NewSearcher(writer.Reader);
writer.Dispose();
// NRT open
var taxoReader = new DirectoryTaxonomyReader(taxoWriter);
taxoWriter.Dispose();
// 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 one of the
// Facets.search utility methods:
searcher.Search(new MatchAllDocsQuery(), c);
TaxonomyFacetSumValueSource facets = new TaxonomyFacetSumValueSource(taxoReader, new FacetsConfig(), c, new IntFieldSource("num"));
// Retrieve & verify results:
Assert.AreEqual("dim=Author path=[] value=145.0 childCount=4\n Lisa (50.0)\n Frank (45.0)\n Susan (40.0)\n Bob (10.0)\n", facets.GetTopChildren(10, "Author").ToString());
taxoReader.Dispose();
searcher.IndexReader.Dispose();
dir.Dispose();
taxoDir.Dispose();
}
public virtual void TestSimple()
{
Random random = Random();
DocValuesType[] dvTypes = new DocValuesType[] {
DocValuesType.NUMERIC,
DocValuesType.BINARY,
DocValuesType.SORTED,
};
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(
random,
dir,
NewIndexWriterConfig(TEST_VERSION_CURRENT,
new MockAnalyzer(random)).SetMergePolicy(NewLogMergePolicy()));
bool canUseDV = !"Lucene3x".Equals(w.w.Config.Codec.Name, StringComparison.Ordinal);
DocValuesType dvType = canUseDV ? dvTypes[random.nextInt(dvTypes.Length)] : DocValuesType.NONE;
Document doc = new Document();
AddField(doc, groupField, "1", dvType);
AddField(doc, countField, "1", dvType);
doc.Add(new TextField("content", "random text", Field.Store.NO));
doc.Add(new StringField("id", "1", Field.Store.NO));
w.AddDocument(doc);
// 1
doc = new Document();
AddField(doc, groupField, "1", dvType);
AddField(doc, countField, "1", dvType);
doc.Add(new TextField("content", "some more random text blob", Field.Store.NO));
doc.Add(new StringField("id", "2", Field.Store.NO));
w.AddDocument(doc);
// 2
doc = new Document();
AddField(doc, groupField, "1", dvType);
AddField(doc, countField, "2", dvType);
doc.Add(new TextField("content", "some more random textual data", Field.Store.NO));
doc.Add(new StringField("id", "3", Field.Store.NO));
w.AddDocument(doc);
w.Commit(); // To ensure a second segment
// 3
doc = new Document();
AddField(doc, groupField, "2", dvType);
doc.Add(new TextField("content", "some random text", Field.Store.NO));
doc.Add(new StringField("id", "4", Field.Store.NO));
w.AddDocument(doc);
// 4
doc = new Document();
AddField(doc, groupField, "3", dvType);
AddField(doc, countField, "1", dvType);
doc.Add(new TextField("content", "some more random text", Field.Store.NO));
doc.Add(new StringField("id", "5", Field.Store.NO));
w.AddDocument(doc);
// 5
doc = new Document();
AddField(doc, groupField, "3", dvType);
AddField(doc, countField, "1", dvType);
doc.Add(new TextField("content", "random blob", Field.Store.NO));
doc.Add(new StringField("id", "6", Field.Store.NO));
w.AddDocument(doc);
// 6 -- no author field
doc = new Document();
doc.Add(new TextField("content", "random word stuck in alot of other text", Field.Store.YES));
AddField(doc, countField, "1", dvType);
doc.Add(new StringField("id", "6", Field.Store.NO));
w.AddDocument(doc);
IndexSearcher indexSearcher = NewSearcher(w.Reader);
w.Dispose();
var cmp = new ComparerAnonymousHelper1(this);
// === Search for content:random
IAbstractFirstPassGroupingCollector <IComparable> firstCollector = CreateRandomFirstPassCollector(dvType, new Sort(), groupField, 10);
indexSearcher.Search(new TermQuery(new Term("content", "random")), firstCollector);
IAbstractDistinctValuesCollector <AbstractDistinctValuesCollector.IGroupCount <IComparable> > distinctValuesCollector
= CreateDistinctCountCollector(firstCollector, groupField, countField, dvType);
indexSearcher.Search(new TermQuery(new Term("content", "random")), distinctValuesCollector);
//var gcs = distinctValuesCollector.Groups as List<IGroupCount<IComparable>>;
// LUCENENET TODO: Try to work out how to do this without an O(n) operation
var gcs = new List <AbstractDistinctValuesCollector.IGroupCount <IComparable> >(distinctValuesCollector.Groups);
gcs.Sort(cmp);
assertEquals(4, gcs.Count);
CompareNull(gcs[0].GroupValue);
List <IComparable> countValues = new List <IComparable>(gcs[0].UniqueValues);
assertEquals(1, countValues.size());
Compare("1", countValues[0]);
Compare("1", gcs[1].GroupValue);
countValues = new List <IComparable>(gcs[1].UniqueValues);
countValues.Sort(nullComparer);
assertEquals(2, countValues.size());
Compare("1", countValues[0]);
Compare("2", countValues[1]);
Compare("2", gcs[2].GroupValue);
countValues = new List <IComparable>(gcs[2].UniqueValues);
assertEquals(1, countValues.size());
CompareNull(countValues[0]);
Compare("3", gcs[3].GroupValue);
countValues = new List <IComparable>(gcs[3].UniqueValues);
assertEquals(1, countValues.size());
Compare("1", countValues[0]);
// === Search for content:some
firstCollector = CreateRandomFirstPassCollector(dvType, new Sort(), groupField, 10);
indexSearcher.Search(new TermQuery(new Term("content", "some")), firstCollector);
distinctValuesCollector = CreateDistinctCountCollector(firstCollector, groupField, countField, dvType);
indexSearcher.Search(new TermQuery(new Term("content", "some")), distinctValuesCollector);
// LUCENENET TODO: Try to work out how to do this without an O(n) operation
//gcs = distinctValuesCollector.Groups as List<IGroupCount<IComparable>>;
gcs = new List <AbstractDistinctValuesCollector.IGroupCount <IComparable> >(distinctValuesCollector.Groups);
gcs.Sort(cmp);
assertEquals(3, gcs.Count);
Compare("1", gcs[0].GroupValue);
countValues = new List <IComparable>(gcs[0].UniqueValues);
assertEquals(2, countValues.size());
countValues.Sort(nullComparer);
Compare("1", countValues[0]);
Compare("2", countValues[1]);
Compare("2", gcs[1].GroupValue);
countValues = new List <IComparable>(gcs[1].UniqueValues);
assertEquals(1, countValues.size());
CompareNull(countValues[0]);
Compare("3", gcs[2].GroupValue);
countValues = new List <IComparable>(gcs[2].UniqueValues);
assertEquals(1, countValues.size());
Compare("1", countValues[0]);
// === Search for content:blob
firstCollector = CreateRandomFirstPassCollector(dvType, new Sort(), groupField, 10);
indexSearcher.Search(new TermQuery(new Term("content", "blob")), firstCollector);
distinctValuesCollector = CreateDistinctCountCollector(firstCollector, groupField, countField, dvType);
indexSearcher.Search(new TermQuery(new Term("content", "blob")), distinctValuesCollector);
// LUCENENET TODO: Try to work out how to do this without an O(n) operation
//gcs = distinctValuesCollector.Groups as List<IGroupCount<IComparable>>;
gcs = new List <AbstractDistinctValuesCollector.IGroupCount <IComparable> >(distinctValuesCollector.Groups);
gcs.Sort(cmp);
assertEquals(2, gcs.Count);
Compare("1", gcs[0].GroupValue);
countValues = new List <IComparable>(gcs[0].UniqueValues);
// B/c the only one document matched with blob inside the author 1 group
assertEquals(1, countValues.Count);
Compare("1", countValues[0]);
Compare("3", gcs[1].GroupValue);
countValues = new List <IComparable>(gcs[1].UniqueValues);
assertEquals(1, countValues.Count);
Compare("1", countValues[0]);
indexSearcher.IndexReader.Dispose();
dir.Dispose();
}
public override void SetUp()
{
base.SetUp();
AllSortFields = new List <SortField>(Arrays.AsList(new SortField[] { new SortField("byte", SortField.Type_e.BYTE, false), new SortField("short", SortField.Type_e.SHORT, false), new SortField("int", SortField.Type_e.INT, false), new SortField("long", SortField.Type_e.LONG, false), new SortField("float", SortField.Type_e.FLOAT, false), new SortField("double", SortField.Type_e.DOUBLE, false), new SortField("bytes", SortField.Type_e.STRING, false), new SortField("bytesval", SortField.Type_e.STRING_VAL, false), new SortField("byte", SortField.Type_e.BYTE, true), new SortField("short", SortField.Type_e.SHORT, true), new SortField("int", SortField.Type_e.INT, true), new SortField("long", SortField.Type_e.LONG, true), new SortField("float", SortField.Type_e.FLOAT, true), new SortField("double", SortField.Type_e.DOUBLE, true), new SortField("bytes", SortField.Type_e.STRING, true), new SortField("bytesval", SortField.Type_e.STRING_VAL, true), SortField.FIELD_SCORE, SortField.FIELD_DOC }));
if (SupportsDocValues)
{
AllSortFields.AddRange(Arrays.AsList(new SortField[] { new SortField("intdocvalues", SortField.Type_e.INT, false), new SortField("floatdocvalues", SortField.Type_e.FLOAT, false), new SortField("sortedbytesdocvalues", SortField.Type_e.STRING, false), new SortField("sortedbytesdocvaluesval", SortField.Type_e.STRING_VAL, false), new SortField("straightbytesdocvalues", SortField.Type_e.STRING_VAL, false), new SortField("intdocvalues", SortField.Type_e.INT, true), new SortField("floatdocvalues", SortField.Type_e.FLOAT, true), new SortField("sortedbytesdocvalues", SortField.Type_e.STRING, true), new SortField("sortedbytesdocvaluesval", SortField.Type_e.STRING_VAL, true), new SortField("straightbytesdocvalues", SortField.Type_e.STRING_VAL, true) }));
}
// Also test missing first / last for the "string" sorts:
foreach (string field in new string[] { "bytes", "sortedbytesdocvalues" })
{
for (int rev = 0; rev < 2; rev++)
{
bool reversed = rev == 0;
SortField sf = new SortField(field, SortField.Type_e.STRING, reversed);
sf.MissingValue = SortField.STRING_FIRST;
AllSortFields.Add(sf);
sf = new SortField(field, SortField.Type_e.STRING, reversed);
sf.MissingValue = SortField.STRING_LAST;
AllSortFields.Add(sf);
}
}
int limit = AllSortFields.Count;
for (int i = 0; i < limit; i++)
{
SortField sf = AllSortFields[i];
if (sf.Type == SortField.Type_e.INT)
{
SortField sf2 = new SortField(sf.Field, SortField.Type_e.INT, sf.Reverse);
sf2.MissingValue = Random().Next();
AllSortFields.Add(sf2);
}
else if (sf.Type == SortField.Type_e.LONG)
{
SortField sf2 = new SortField(sf.Field, SortField.Type_e.LONG, sf.Reverse);
sf2.MissingValue = Random().NextLong();
AllSortFields.Add(sf2);
}
else if (sf.Type == SortField.Type_e.FLOAT)
{
SortField sf2 = new SortField(sf.Field, SortField.Type_e.FLOAT, sf.Reverse);
sf2.MissingValue = (float)Random().NextDouble();
AllSortFields.Add(sf2);
}
else if (sf.Type == SortField.Type_e.DOUBLE)
{
SortField sf2 = new SortField(sf.Field, SortField.Type_e.DOUBLE, sf.Reverse);
sf2.MissingValue = Random().NextDouble();
AllSortFields.Add(sf2);
}
}
Dir = NewDirectory();
RandomIndexWriter iw = new RandomIndexWriter(Random(), Dir);
int numDocs = AtLeast(200);
for (int i = 0; i < numDocs; i++)
{
IList <Field> fields = new List <Field>();
fields.Add(NewTextField("english", English.IntToEnglish(i), Field.Store.NO));
fields.Add(NewTextField("oddeven", (i % 2 == 0) ? "even" : "odd", Field.Store.NO));
fields.Add(NewStringField("byte", "" + ((sbyte)Random().Next()), Field.Store.NO));
fields.Add(NewStringField("short", "" + ((short)Random().Next()), Field.Store.NO));
fields.Add(new IntField("int", Random().Next(), Field.Store.NO));
fields.Add(new LongField("long", Random().NextLong(), Field.Store.NO));
fields.Add(new FloatField("float", (float)Random().NextDouble(), Field.Store.NO));
fields.Add(new DoubleField("double", Random().NextDouble(), Field.Store.NO));
fields.Add(NewStringField("bytes", TestUtil.RandomRealisticUnicodeString(Random()), Field.Store.NO));
fields.Add(NewStringField("bytesval", TestUtil.RandomRealisticUnicodeString(Random()), Field.Store.NO));
fields.Add(new DoubleField("double", Random().NextDouble(), Field.Store.NO));
if (SupportsDocValues)
{
fields.Add(new NumericDocValuesField("intdocvalues", Random().Next()));
fields.Add(new FloatDocValuesField("floatdocvalues", (float)Random().NextDouble()));
fields.Add(new SortedDocValuesField("sortedbytesdocvalues", new BytesRef(TestUtil.RandomRealisticUnicodeString(Random()))));
fields.Add(new SortedDocValuesField("sortedbytesdocvaluesval", new BytesRef(TestUtil.RandomRealisticUnicodeString(Random()))));
fields.Add(new BinaryDocValuesField("straightbytesdocvalues", new BytesRef(TestUtil.RandomRealisticUnicodeString(Random()))));
}
Document document = new Document();
document.Add(new StoredField("id", "" + i));
if (VERBOSE)
{
Console.WriteLine(" add doc id=" + i);
}
foreach (Field field in fields)
{
// So we are sometimes missing that field:
if (Random().Next(5) != 4)
{
document.Add(field);
if (VERBOSE)
{
Console.WriteLine(" " + field);
}
}
}
iw.AddDocument(document);
if (Random().Next(50) == 17)
{
iw.Commit();
}
}
Reader = iw.Reader;
iw.Dispose();
Searcher = NewSearcher(Reader);
if (VERBOSE)
{
Console.WriteLine(" searcher=" + Searcher);
}
}
public virtual void TestPostings()
{
Directory dir = NewFSDirectory(CreateTempDir("postings"));
IndexWriterConfig iwc = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random()));
iwc.SetCodec(Codec.ForName("Lucene40"));
RandomIndexWriter iw = new RandomIndexWriter(Random(), dir, iwc);
Document doc = new Document();
// id field
FieldType idType = new FieldType(StringField.TYPE_NOT_STORED);
idType.StoreTermVectors = true;
Field idField = new Field("id", "", idType);
doc.Add(idField);
// title field: short text field
FieldType titleType = new FieldType(TextField.TYPE_NOT_STORED);
titleType.StoreTermVectors = true;
titleType.StoreTermVectorPositions = true;
titleType.StoreTermVectorOffsets = true;
titleType.IndexOptions = IndexOptions();
Field titleField = new Field("title", "", titleType);
doc.Add(titleField);
// body field: long text field
FieldType bodyType = new FieldType(TextField.TYPE_NOT_STORED);
bodyType.StoreTermVectors = true;
bodyType.StoreTermVectorPositions = true;
bodyType.StoreTermVectorOffsets = true;
bodyType.IndexOptions = IndexOptions();
Field bodyField = new Field("body", "", bodyType);
doc.Add(bodyField);
int numDocs = AtLeast(1000);
for (int i = 0; i < numDocs; i++)
{
idField.SetStringValue(Convert.ToString(i));
titleField.SetStringValue(FieldValue(1));
bodyField.SetStringValue(FieldValue(3));
iw.AddDocument(doc);
if (Random().Next(20) == 0)
{
iw.DeleteDocuments(new Term("id", Convert.ToString(i)));
}
}
if (Random().NextBoolean())
{
// delete 1-100% of docs
iw.DeleteDocuments(new Term("title", Terms[Random().Next(Terms.Length)]));
}
iw.Dispose();
dir.Dispose(); // checkindex
}
public virtual void TestDeMorgan()
{
Directory dir1 = NewDirectory();
RandomIndexWriter iw1 = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, dir1);
Document doc1 = new Document();
doc1.Add(NewTextField("field", "foo bar", Field.Store.NO));
iw1.AddDocument(doc1);
IndexReader reader1 = iw1.GetReader();
iw1.Dispose();
Directory dir2 = NewDirectory();
RandomIndexWriter iw2 = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, dir2);
Document doc2 = new Document();
doc2.Add(NewTextField("field", "foo baz", Field.Store.NO));
iw2.AddDocument(doc2);
IndexReader reader2 = iw2.GetReader();
iw2.Dispose();
BooleanQuery query = new BooleanQuery(); // Query: +foo -ba*
query.Add(new TermQuery(new Term("field", "foo")), Occur.MUST);
WildcardQuery wildcardQuery = new WildcardQuery(new Term("field", "ba*"));
wildcardQuery.MultiTermRewriteMethod = (MultiTermQuery.SCORING_BOOLEAN_QUERY_REWRITE);
query.Add(wildcardQuery, Occur.MUST_NOT);
MultiReader multireader = new MultiReader(reader1, reader2);
IndexSearcher searcher = NewSearcher(multireader);
Assert.AreEqual(0, searcher.Search(query, 10).TotalHits);
Task foo = new Task(TestDeMorgan);
TaskScheduler es = TaskScheduler.Default;
searcher = new IndexSearcher(multireader, es);
if (VERBOSE)
{
Console.WriteLine("rewritten form: " + searcher.Rewrite(query));
}
Assert.AreEqual(0, searcher.Search(query, 10).TotalHits);
multireader.Dispose();
reader1.Dispose();
reader2.Dispose();
dir1.Dispose();
dir2.Dispose();
}
public virtual void TestBS2DisjunctionNextVsAdvance()
{
Directory d = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, d);
int numDocs = AtLeast(300);
for (int docUpto = 0; docUpto < numDocs; docUpto++)
{
string contents = "a";
if (Random.Next(20) <= 16)
{
contents += " b";
}
if (Random.Next(20) <= 8)
{
contents += " c";
}
if (Random.Next(20) <= 4)
{
contents += " d";
}
if (Random.Next(20) <= 2)
{
contents += " e";
}
if (Random.Next(20) <= 1)
{
contents += " f";
}
Document doc = new Document();
doc.Add(new TextField("field", contents, Field.Store.NO));
w.AddDocument(doc);
}
w.ForceMerge(1);
IndexReader r = w.GetReader();
IndexSearcher s = NewSearcher(r);
w.Dispose();
for (int iter = 0; iter < 10 * RANDOM_MULTIPLIER; iter++)
{
if (VERBOSE)
{
Console.WriteLine("iter=" + iter);
}
IList <string> terms = new List <string>(Arrays.AsList("a", "b", "c", "d", "e", "f"));
int numTerms = TestUtil.NextInt32(Random, 1, terms.Count);
while (terms.Count > numTerms)
{
terms.RemoveAt(Random.Next(terms.Count));
}
if (VERBOSE)
{
Console.WriteLine(" terms=" + terms);
}
BooleanQuery q = new BooleanQuery();
foreach (string term in terms)
{
q.Add(new BooleanClause(new TermQuery(new Term("field", term)), Occur.SHOULD));
}
Weight weight = s.CreateNormalizedWeight(q);
Scorer scorer = weight.GetScorer(s.m_leafContexts[0], null);
// First pass: just use .NextDoc() to gather all hits
IList <ScoreDoc> hits = new List <ScoreDoc>();
while (scorer.NextDoc() != DocIdSetIterator.NO_MORE_DOCS)
{
hits.Add(new ScoreDoc(scorer.DocID, scorer.GetScore()));
}
if (VERBOSE)
{
Console.WriteLine(" " + hits.Count + " hits");
}
// Now, randomly next/advance through the list and
// verify exact match:
for (int iter2 = 0; iter2 < 10; iter2++)
{
weight = s.CreateNormalizedWeight(q);
scorer = weight.GetScorer(s.m_leafContexts[0], null);
if (VERBOSE)
{
Console.WriteLine(" iter2=" + iter2);
}
int upto = -1;
while (upto < hits.Count)
{
int nextUpto;
int nextDoc;
int left = hits.Count - upto;
if (left == 1 || Random.nextBoolean())
{
// next
nextUpto = 1 + upto;
nextDoc = scorer.NextDoc();
}
else
{
// advance
int inc = TestUtil.NextInt32(Random, 1, left - 1);
nextUpto = inc + upto;
nextDoc = scorer.Advance(hits[nextUpto].Doc);
}
if (nextUpto == hits.Count)
{
Assert.AreEqual(DocIdSetIterator.NO_MORE_DOCS, nextDoc);
}
else
{
ScoreDoc hit = hits[nextUpto];
Assert.AreEqual(hit.Doc, nextDoc);
// Test for precise float equality:
Assert.IsTrue(hit.Score == scorer.GetScore(), "doc " + hit.Doc + " has wrong score: expected=" + hit.Score + " actual=" + scorer.GetScore());
}
upto = nextUpto;
}
}
}
r.Dispose();
d.Dispose();
}
public virtual void TestNullOrSubScorer()
{
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, dir);
Document doc = new Document();
doc.Add(NewTextField("field", "a b c d", Field.Store.NO));
w.AddDocument(doc);
IndexReader r = w.GetReader();
IndexSearcher s = NewSearcher(r);
// this test relies upon coord being the default implementation,
// otherwise scores are different!
s.Similarity = new DefaultSimilarity();
BooleanQuery q = new BooleanQuery();
q.Add(new TermQuery(new Term("field", "a")), Occur.SHOULD);
// LUCENE-2617: make sure that a term not in the index still contributes to the score via coord factor
float score = s.Search(q, 10).MaxScore;
Query subQuery = new TermQuery(new Term("field", "not_in_index"));
subQuery.Boost = 0;
q.Add(subQuery, Occur.SHOULD);
float score2 = s.Search(q, 10).MaxScore;
Assert.AreEqual(score * .5F, score2, 1e-6);
// LUCENE-2617: make sure that a clause not in the index still contributes to the score via coord factor
BooleanQuery qq = (BooleanQuery)q.Clone();
PhraseQuery phrase = new PhraseQuery();
phrase.Add(new Term("field", "not_in_index"));
phrase.Add(new Term("field", "another_not_in_index"));
phrase.Boost = 0;
qq.Add(phrase, Occur.SHOULD);
score2 = s.Search(qq, 10).MaxScore;
Assert.AreEqual(score * (1 / 3F), score2, 1e-6);
// now test BooleanScorer2
subQuery = new TermQuery(new Term("field", "b"));
subQuery.Boost = 0;
q.Add(subQuery, Occur.MUST);
score2 = s.Search(q, 10).MaxScore;
Assert.AreEqual(score * (2 / 3F), score2, 1e-6);
// PhraseQuery w/ no terms added returns a null scorer
PhraseQuery pq = new PhraseQuery();
q.Add(pq, Occur.SHOULD);
Assert.AreEqual(1, s.Search(q, 10).TotalHits);
// A required clause which returns null scorer should return null scorer to
// IndexSearcher.
q = new BooleanQuery();
pq = new PhraseQuery();
q.Add(new TermQuery(new Term("field", "a")), Occur.SHOULD);
q.Add(pq, Occur.MUST);
Assert.AreEqual(0, s.Search(q, 10).TotalHits);
DisjunctionMaxQuery dmq = new DisjunctionMaxQuery(1.0f);
dmq.Add(new TermQuery(new Term("field", "a")));
dmq.Add(pq);
Assert.AreEqual(1, s.Search(dmq, 10).TotalHits);
r.Dispose();
w.Dispose();
dir.Dispose();
}
public void TestFuzziness()
{
//every test with SlowFuzzyQuery.defaultMinSimilarity
//is exercising the Automaton, not the brute force linear method
Directory directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random, directory, Similarity, TimeZone);
addDoc("aaaaa", writer);
addDoc("aaaab", writer);
addDoc("aaabb", writer);
addDoc("aabbb", writer);
addDoc("abbbb", writer);
addDoc("bbbbb", writer);
addDoc("ddddd", writer);
IndexReader reader = writer.GetReader();
IndexSearcher searcher = NewSearcher(reader);
writer.Dispose();
SlowFuzzyQuery query = new SlowFuzzyQuery(new Term("field", "aaaaa"), SlowFuzzyQuery.defaultMinSimilarity, 0);
ScoreDoc[] hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(3, hits.Length);
// same with prefix
query = new SlowFuzzyQuery(new Term("field", "aaaaa"), SlowFuzzyQuery.defaultMinSimilarity, 1);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(3, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "aaaaa"), SlowFuzzyQuery.defaultMinSimilarity, 2);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(3, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "aaaaa"), SlowFuzzyQuery.defaultMinSimilarity, 3);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(3, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "aaaaa"), SlowFuzzyQuery.defaultMinSimilarity, 4);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(2, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "aaaaa"), SlowFuzzyQuery.defaultMinSimilarity, 5);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "aaaaa"), SlowFuzzyQuery.defaultMinSimilarity, 6);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
// test scoring
query = new SlowFuzzyQuery(new Term("field", "bbbbb"), SlowFuzzyQuery.defaultMinSimilarity, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals("3 documents should match", 3, hits.Length);
List <String> order = Arrays.AsList("bbbbb", "abbbb", "aabbb");
for (int i = 0; i < hits.Length; i++)
{
string term = searcher.Doc(hits[i].Doc).Get("field");
//System.out.println(hits[i].score);
assertEquals(order[i], term);
}
// test pq size by supplying maxExpansions=2
// This query would normally return 3 documents, because 3 terms match (see above):
query = new SlowFuzzyQuery(new Term("field", "bbbbb"), SlowFuzzyQuery.defaultMinSimilarity, 0, 2);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals("only 2 documents should match", 2, hits.Length);
order = Arrays.AsList("bbbbb", "abbbb");
for (int i = 0; i < hits.Length; i++)
{
string term = searcher.Doc(hits[i].Doc).Get("field");
//System.out.println(hits[i].score);
assertEquals(order[i], term);
}
// not similar enough:
query = new SlowFuzzyQuery(new Term("field", "xxxxx"), SlowFuzzyQuery.defaultMinSimilarity, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "aaccc"), SlowFuzzyQuery.defaultMinSimilarity, 0); // edit distance to "aaaaa" = 3
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
// query identical to a word in the index:
query = new SlowFuzzyQuery(new Term("field", "aaaaa"), SlowFuzzyQuery.defaultMinSimilarity, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(3, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
// default allows for up to two edits:
assertEquals(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
assertEquals(searcher.Doc(hits[2].Doc).Get("field"), ("aaabb"));
// query similar to a word in the index:
query = new SlowFuzzyQuery(new Term("field", "aaaac"), SlowFuzzyQuery.defaultMinSimilarity, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(3, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
assertEquals(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
assertEquals(searcher.Doc(hits[2].Doc).Get("field"), ("aaabb"));
// now with prefix
query = new SlowFuzzyQuery(new Term("field", "aaaac"), SlowFuzzyQuery.defaultMinSimilarity, 1);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(3, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
assertEquals(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
assertEquals(searcher.Doc(hits[2].Doc).Get("field"), ("aaabb"));
query = new SlowFuzzyQuery(new Term("field", "aaaac"), SlowFuzzyQuery.defaultMinSimilarity, 2);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(3, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
assertEquals(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
assertEquals(searcher.Doc(hits[2].Doc).Get("field"), ("aaabb"));
query = new SlowFuzzyQuery(new Term("field", "aaaac"), SlowFuzzyQuery.defaultMinSimilarity, 3);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(3, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
assertEquals(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
assertEquals(searcher.Doc(hits[2].Doc).Get("field"), ("aaabb"));
query = new SlowFuzzyQuery(new Term("field", "aaaac"), SlowFuzzyQuery.defaultMinSimilarity, 4);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(2, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
assertEquals(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
query = new SlowFuzzyQuery(new Term("field", "aaaac"), SlowFuzzyQuery.defaultMinSimilarity, 5);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "ddddX"), SlowFuzzyQuery.defaultMinSimilarity, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("ddddd"));
// now with prefix
query = new SlowFuzzyQuery(new Term("field", "ddddX"), SlowFuzzyQuery.defaultMinSimilarity, 1);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("ddddd"));
query = new SlowFuzzyQuery(new Term("field", "ddddX"), SlowFuzzyQuery.defaultMinSimilarity, 2);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("ddddd"));
query = new SlowFuzzyQuery(new Term("field", "ddddX"), SlowFuzzyQuery.defaultMinSimilarity, 3);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("ddddd"));
query = new SlowFuzzyQuery(new Term("field", "ddddX"), SlowFuzzyQuery.defaultMinSimilarity, 4);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("ddddd"));
query = new SlowFuzzyQuery(new Term("field", "ddddX"), SlowFuzzyQuery.defaultMinSimilarity, 5);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
// different field = no match:
query = new SlowFuzzyQuery(new Term("anotherfield", "ddddX"), SlowFuzzyQuery.defaultMinSimilarity, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
reader.Dispose();
directory.Dispose();
}
public void TestFuzzinessLong()
{
Directory directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random, directory, Similarity, TimeZone);
addDoc("aaaaaaa", writer);
addDoc("segment", writer);
IndexReader reader = writer.GetReader();
IndexSearcher searcher = NewSearcher(reader);
writer.Dispose();
SlowFuzzyQuery query;
// not similar enough:
query = new SlowFuzzyQuery(new Term("field", "xxxxx"), 0.5f, 0);
ScoreDoc[] hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
// edit distance to "aaaaaaa" = 3, this matches because the string is longer than
// in testDefaultFuzziness so a bigger difference is allowed:
query = new SlowFuzzyQuery(new Term("field", "aaaaccc"), 0.5f, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaaaa"));
// now with prefix
query = new SlowFuzzyQuery(new Term("field", "aaaaccc"), 0.5f, 1);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaaaa"));
query = new SlowFuzzyQuery(new Term("field", "aaaaccc"), 0.5f, 4);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
assertEquals(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaaaa"));
query = new SlowFuzzyQuery(new Term("field", "aaaaccc"), 0.5f, 5);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
// no match, more than half of the characters is wrong:
query = new SlowFuzzyQuery(new Term("field", "aaacccc"), 0.5f, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
// now with prefix
query = new SlowFuzzyQuery(new Term("field", "aaacccc"), 0.5f, 2);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
// "student" and "stellent" are indeed similar to "segment" by default:
query = new SlowFuzzyQuery(new Term("field", "student"), 0.5f, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "stellent"), 0.5f, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
// now with prefix
query = new SlowFuzzyQuery(new Term("field", "student"), 0.5f, 1);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "stellent"), 0.5f, 1);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(1, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "student"), 0.5f, 2);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
query = new SlowFuzzyQuery(new Term("field", "stellent"), 0.5f, 2);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
// "student" doesn't match anymore thanks to increased minimum similarity:
query = new SlowFuzzyQuery(new Term("field", "student"), 0.6f, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
assertEquals(0, hits.Length);
try
{
query = new SlowFuzzyQuery(new Term("field", "student"), 1.1f);
fail("Expected IllegalArgumentException");
}
#pragma warning disable 168
catch (ArgumentException e)
#pragma warning restore 168
{
// expecting exception
}
try
{
query = new SlowFuzzyQuery(new Term("field", "student"), -0.1f);
fail("Expected IllegalArgumentException");
}
#pragma warning disable 168
catch (ArgumentException e)
#pragma warning restore 168
{
// expecting exception
}
reader.Dispose();
directory.Dispose();
}
public virtual void Test()
{
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, dir);
long startTime = Environment.TickCount;
// TODO: replace w/ the @nightly test data; make this
// into an optional @nightly stress test
Document doc = new Document();
Field body = NewTextField("body", "", Field.Store.NO);
doc.Add(body);
StringBuilder sb = new StringBuilder();
for (int docCount = 0; docCount < NUM_DOCS; docCount++)
{
int numTerms = Random.Next(10);
for (int termCount = 0; termCount < numTerms; termCount++)
{
sb.Append(Random.NextBoolean() ? "aaa" : "bbb");
sb.Append(' ');
}
body.SetStringValue(sb.ToString());
w.AddDocument(doc);
sb.Remove(0, sb.Length);
}
IndexReader r = w.GetReader();
w.Dispose();
long endTime = Environment.TickCount;
if (Verbose)
{
Console.WriteLine("BUILD took " + (endTime - startTime));
}
IndexSearcher s = NewSearcher(r);
AtomicBoolean failed = new AtomicBoolean();
AtomicInt64 netSearch = new AtomicInt64();
ThreadJob[] threads = new ThreadJob[NUM_SEARCH_THREADS];
for (int threadID = 0; threadID < NUM_SEARCH_THREADS; threadID++)
{
threads[threadID] = new ThreadAnonymousInnerClassHelper(this, s, failed, netSearch);
threads[threadID].IsBackground = (true);
}
foreach (ThreadJob t in threads)
{
t.Start();
}
foreach (ThreadJob t in threads)
{
t.Join();
}
if (Verbose)
{
Console.WriteLine(NUM_SEARCH_THREADS + " threads did " + netSearch + " searches");
}
r.Dispose();
dir.Dispose();
}
public virtual void TestTransitionAPI()
{
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(Random, dir, Similarity, TimeZone);
Documents.Document doc = new Documents.Document();
#pragma warning disable 612, 618
doc.Add(new Field("stored", "abc", Field.Store.YES, Field.Index.NO));
doc.Add(new Field("stored_indexed", "abc xyz", Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.Add(new Field("stored_tokenized", "abc xyz", Field.Store.YES, Field.Index.ANALYZED));
doc.Add(new Field("indexed", "abc xyz", Field.Store.NO, Field.Index.NOT_ANALYZED));
doc.Add(new Field("tokenized", "abc xyz", Field.Store.NO, Field.Index.ANALYZED));
doc.Add(new Field("tokenized_reader", new StringReader("abc xyz")));
doc.Add(new Field("tokenized_tokenstream", w.IndexWriter.Analyzer.GetTokenStream("tokenized_tokenstream", new StringReader("abc xyz"))));
doc.Add(new Field("binary", new byte[10]));
doc.Add(new Field("tv", "abc xyz", Field.Store.NO, Field.Index.ANALYZED, Field.TermVector.YES));
doc.Add(new Field("tv_pos", "abc xyz", Field.Store.NO, Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS));
doc.Add(new Field("tv_off", "abc xyz", Field.Store.NO, Field.Index.ANALYZED, Field.TermVector.WITH_OFFSETS));
doc.Add(new Field("tv_pos_off", "abc xyz", Field.Store.NO, Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS_OFFSETS));
#pragma warning restore 612, 618
w.AddDocument(doc);
IndexReader r = w.GetReader();
w.Dispose();
doc = r.Document(0);
// 4 stored fields
Assert.AreEqual(4, doc.Fields.Count);
Assert.AreEqual("abc", doc.Get("stored"));
Assert.AreEqual("abc xyz", doc.Get("stored_indexed"));
Assert.AreEqual("abc xyz", doc.Get("stored_tokenized"));
BytesRef br = doc.GetBinaryValue("binary");
Assert.IsNotNull(br);
Assert.AreEqual(10, br.Length);
IndexSearcher s = new IndexSearcher(r);
Assert.AreEqual(1, s.Search(new TermQuery(new Term("stored_indexed", "abc xyz")), 1).TotalHits);
Assert.AreEqual(1, s.Search(new TermQuery(new Term("stored_tokenized", "abc")), 1).TotalHits);
Assert.AreEqual(1, s.Search(new TermQuery(new Term("stored_tokenized", "xyz")), 1).TotalHits);
Assert.AreEqual(1, s.Search(new TermQuery(new Term("indexed", "abc xyz")), 1).TotalHits);
Assert.AreEqual(1, s.Search(new TermQuery(new Term("tokenized", "abc")), 1).TotalHits);
Assert.AreEqual(1, s.Search(new TermQuery(new Term("tokenized", "xyz")), 1).TotalHits);
Assert.AreEqual(1, s.Search(new TermQuery(new Term("tokenized_reader", "abc")), 1).TotalHits);
Assert.AreEqual(1, s.Search(new TermQuery(new Term("tokenized_reader", "xyz")), 1).TotalHits);
Assert.AreEqual(1, s.Search(new TermQuery(new Term("tokenized_tokenstream", "abc")), 1).TotalHits);
Assert.AreEqual(1, s.Search(new TermQuery(new Term("tokenized_tokenstream", "xyz")), 1).TotalHits);
foreach (string field in new string[] { "tv", "tv_pos", "tv_off", "tv_pos_off" })
{
Fields tvFields = r.GetTermVectors(0);
Terms tvs = tvFields.GetTerms(field);
Assert.IsNotNull(tvs);
Assert.AreEqual(2, tvs.Count);
TermsEnum tvsEnum = tvs.GetIterator(null);
Assert.AreEqual(new BytesRef("abc"), tvsEnum.Next());
DocsAndPositionsEnum dpEnum = tvsEnum.DocsAndPositions(null, null);
if (field.Equals("tv", StringComparison.Ordinal))
{
Assert.IsNull(dpEnum);
}
else
{
Assert.IsNotNull(dpEnum);
}
Assert.AreEqual(new BytesRef("xyz"), tvsEnum.Next());
Assert.IsNull(tvsEnum.Next());
}
r.Dispose();
dir.Dispose();
}
private IndexContext CreateIndexContext()
{
Random random = Random();
DocValuesType[] dvTypes = new DocValuesType[] {
DocValuesType.BINARY,
DocValuesType.SORTED
};
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(
random,
dir,
NewIndexWriterConfig(TEST_VERSION_CURRENT,
new MockAnalyzer(random)).SetMergePolicy(NewLogMergePolicy())
);
bool canUseDV = !"Lucene3x".Equals(w.w.Config.Codec.Name, StringComparison.Ordinal);
DocValuesType dvType = canUseDV ? dvTypes[random.nextInt(dvTypes.Length)] : DocValuesType.NONE;
int numDocs = 86 + random.nextInt(1087) * RANDOM_MULTIPLIER;
string[] groupValues = new string[numDocs / 5];
string[] countValues = new string[numDocs / 10];
for (int i = 0; i < groupValues.Length; i++)
{
groupValues[i] = GenerateRandomNonEmptyString();
}
for (int i = 0; i < countValues.Length; i++)
{
countValues[i] = GenerateRandomNonEmptyString();
}
List <string> contentStrings = new List <string>();
IDictionary <string, IDictionary <string, ISet <string> > > searchTermToGroupCounts = new HashMap <string, IDictionary <string, ISet <string> > >();
for (int i = 1; i <= numDocs; i++)
{
string groupValue = random.nextInt(23) == 14 ? null : groupValues[random.nextInt(groupValues.Length)];
string countValue = random.nextInt(21) == 13 ? null : countValues[random.nextInt(countValues.Length)];
string content = "random" + random.nextInt(numDocs / 20);
IDictionary <string, ISet <string> > groupToCounts;
if (!searchTermToGroupCounts.TryGetValue(content, out groupToCounts))
{
// Groups sort always DOCID asc...
searchTermToGroupCounts.Add(content, groupToCounts = new LinkedHashMap <string, ISet <string> >());
contentStrings.Add(content);
}
ISet <string> countsVals;
if (!groupToCounts.TryGetValue(groupValue, out countsVals))
{
groupToCounts.Add(groupValue, countsVals = new HashSet <string>());
}
countsVals.Add(countValue);
Document doc = new Document();
doc.Add(new StringField("id", string.Format(CultureInfo.InvariantCulture, "{0:D9}", i), Field.Store.YES));
if (groupValue != null)
{
AddField(doc, groupField, groupValue, dvType);
}
if (countValue != null)
{
AddField(doc, countField, countValue, dvType);
}
doc.Add(new TextField("content", content, Field.Store.YES));
w.AddDocument(doc);
}
DirectoryReader reader = w.Reader;
if (VERBOSE)
{
for (int docID = 0; docID < reader.MaxDoc; docID++)
{
Document doc = reader.Document(docID);
Console.WriteLine("docID=" + docID + " id=" + doc.Get("id") + " content=" + doc.Get("content") + " author=" + doc.Get("author") + " publisher=" + doc.Get("publisher"));
}
}
w.Dispose();
return(new IndexContext(dir, reader, dvType, searchTermToGroupCounts, contentStrings.ToArray(/*new String[contentStrings.size()]*/)));
}
public void assertFromTestData(int[] codePointTable)
{
if (VERBOSE)
{
Console.WriteLine("TEST: codePointTable=" + codePointTable);
}
Stream stream = GetType().getResourceAsStream("fuzzyTestData.txt");
TextReader reader = new StreamReader(stream, Encoding.UTF8);
int bits = int.Parse(reader.ReadLine(), CultureInfo.InvariantCulture);
int terms = (int)Math.Pow(2, bits);
Store.Directory dir = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random, dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random, MockTokenizer.KEYWORD, false)).SetMergePolicy(NewLogMergePolicy()));
Document doc = new Document();
Field field = NewTextField("field", "", Field.Store.NO);
doc.Add(field);
for (int i = 0; i < terms; i++)
{
field.SetStringValue(MapInt(codePointTable, i));
writer.AddDocument(doc);
}
IndexReader r = writer.GetReader();
IndexSearcher searcher = NewSearcher(r);
if (VERBOSE)
{
Console.WriteLine("TEST: searcher=" + searcher);
}
// even though this uses a boost-only rewrite, this test relies upon queryNorm being the default implementation,
// otherwise scores are different!
searcher.Similarity = (new DefaultSimilarity());
writer.Dispose();
String line;
int lineNum = 0;
while ((line = reader.ReadLine()) != null)
{
lineNum++;
String[] @params = line.Split(',').TrimEnd();
String query = MapInt(codePointTable, int.Parse(@params[0], CultureInfo.InvariantCulture));
int prefix = int.Parse(@params[1], CultureInfo.InvariantCulture);
int pqSize = int.Parse(@params[2], CultureInfo.InvariantCulture);
float minScore = float.Parse(@params[3], CultureInfo.InvariantCulture);
#pragma warning disable 612, 618
SlowFuzzyQuery q = new SlowFuzzyQuery(new Term("field", query), minScore, prefix);
#pragma warning restore 612, 618
q.MultiTermRewriteMethod = new MultiTermQuery.TopTermsBoostOnlyBooleanQueryRewrite(pqSize);
int expectedResults = int.Parse(reader.ReadLine(), CultureInfo.InvariantCulture);
TopDocs docs = searcher.Search(q, expectedResults);
assertEquals(expectedResults, docs.TotalHits);
for (int i = 0; i < expectedResults; i++)
{
String[] scoreDoc = reader.ReadLine().Split(',').TrimEnd();
assertEquals(int.Parse(scoreDoc[0], CultureInfo.InvariantCulture), docs.ScoreDocs[i].Doc);
assertEquals(float.Parse(scoreDoc[1], CultureInfo.InvariantCulture), docs.ScoreDocs[i].Score, epsilon);
}
}
r.Dispose();
dir.Dispose();
}
public virtual void TestWrongIndexFieldName()
{
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.SetIndexFieldName("a", "$facets2");
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);
Document doc = new Document();
doc.Add(new IntField("num", 10, Field.Store.NO));
doc.Add(new FacetField("a", "foo1"));
writer.AddDocument(config.Build(taxoWriter, doc));
// NRT open
IndexSearcher searcher = NewSearcher(writer.Reader);
writer.Dispose();
// NRT open
var taxoReader = new DirectoryTaxonomyReader(taxoWriter);
taxoWriter.Dispose();
FacetsCollector c = new FacetsCollector();
searcher.Search(new MatchAllDocsQuery(), c);
TaxonomyFacetSumValueSource facets = new TaxonomyFacetSumValueSource(taxoReader, config, c, new IntFieldSource("num"));
// 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(searcher.IndexReader, taxoReader, dir, taxoDir);
}
public virtual void Test()
{
Directory dir = NewDirectory();
MockAnalyzer analyzer = new MockAnalyzer(Random);
analyzer.MaxTokenLength = TestUtil.NextInt32(Random, 1, IndexWriter.MAX_TERM_LENGTH);
RandomIndexWriter w = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, dir, analyzer);
LineFileDocs docs = new LineFileDocs(Random, DefaultCodecSupportsDocValues);
int charsToIndex = AtLeast(100000);
int charsIndexed = 0;
//System.out.println("bytesToIndex=" + charsToIndex);
while (charsIndexed < charsToIndex)
{
Document doc = docs.NextDoc();
charsIndexed += doc.Get("body").Length;
w.AddDocument(doc);
//System.out.println(" bytes=" + charsIndexed + " add: " + doc);
}
IndexReader r = w.GetReader();
//System.out.println("numDocs=" + r.NumDocs);
w.Dispose();
IndexSearcher s = NewSearcher(r);
Terms terms = MultiFields.GetFields(r).GetTerms("body");
int termCount = 0;
TermsEnum termsEnum = terms.GetIterator(null);
while (termsEnum.Next() != null)
{
termCount++;
}
Assert.IsTrue(termCount > 0);
// Target ~10 terms to search:
double chance = 10.0 / termCount;
termsEnum = terms.GetIterator(termsEnum);
IDictionary <BytesRef, TopDocs> answers = new Dictionary <BytesRef, TopDocs>();
while (termsEnum.Next() != null)
{
if (Random.NextDouble() <= chance)
{
BytesRef term = BytesRef.DeepCopyOf(termsEnum.Term);
answers[term] = s.Search(new TermQuery(new Term("body", term)), 100);
}
}
if (answers.Count > 0)
{
CountdownEvent startingGun = new CountdownEvent(1);
int numThreads = TestUtil.NextInt32(Random, 2, 5);
ThreadClass[] threads = new ThreadClass[numThreads];
for (int threadID = 0; threadID < numThreads; threadID++)
{
ThreadClass thread = new ThreadAnonymousInnerClassHelper(this, s, answers, startingGun);
threads[threadID] = thread;
thread.Start();
}
startingGun.Signal();
foreach (ThreadClass thread in threads)
{
thread.Join();
}
}
r.Dispose();
dir.Dispose();
}
private void ExecuteRandomJoin(bool multipleValuesPerDocument, int maxIndexIter, int maxSearchIter,
int numberOfDocumentsToIndex)
{
for (int indexIter = 1; indexIter <= maxIndexIter; indexIter++)
{
if (Verbose)
{
Console.WriteLine("indexIter=" + indexIter);
}
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(Random, dir,
NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random, MockTokenizer.KEYWORD, false))
.SetMergePolicy(NewLogMergePolicy()));
bool scoreDocsInOrder = TestJoinUtil.Random.NextBoolean();
IndexIterationContext context = CreateContext(numberOfDocumentsToIndex, w, multipleValuesPerDocument,
scoreDocsInOrder);
IndexReader topLevelReader = w.GetReader();
w.Dispose();
for (int searchIter = 1; searchIter <= maxSearchIter; searchIter++)
{
if (Verbose)
{
Console.WriteLine("searchIter=" + searchIter);
}
IndexSearcher indexSearcher = NewSearcher(topLevelReader);
int r = Random.Next(context.RandomUniqueValues.Length);
bool from = context.RandomFrom[r];
string randomValue = context.RandomUniqueValues[r];
FixedBitSet expectedResult = CreateExpectedResult(randomValue, from, indexSearcher.IndexReader,
context);
Query actualQuery = new TermQuery(new Term("value", randomValue));
if (Verbose)
{
Console.WriteLine("actualQuery=" + actualQuery);
}
var scoreModeLength = Enum.GetNames(typeof(ScoreMode)).Length;
ScoreMode scoreMode = (ScoreMode)Random.Next(scoreModeLength);
if (Verbose)
{
Console.WriteLine("scoreMode=" + scoreMode);
}
Query joinQuery;
if (from)
{
joinQuery = JoinUtil.CreateJoinQuery("from", multipleValuesPerDocument, "to", actualQuery,
indexSearcher, scoreMode);
}
else
{
joinQuery = JoinUtil.CreateJoinQuery("to", multipleValuesPerDocument, "from", actualQuery,
indexSearcher, scoreMode);
}
if (Verbose)
{
Console.WriteLine("joinQuery=" + joinQuery);
}
// Need to know all documents that have matches. TopDocs doesn't give me that and then I'd be also testing TopDocsCollector...
FixedBitSet actualResult = new FixedBitSet(indexSearcher.IndexReader.MaxDoc);
TopScoreDocCollector topScoreDocCollector = TopScoreDocCollector.Create(10, false);
indexSearcher.Search(joinQuery,
new CollectorAnonymousInnerClassHelper2(this, scoreDocsInOrder, context, actualResult,
topScoreDocCollector));
// Asserting bit set...
if (Verbose)
{
Console.WriteLine("expected cardinality:" + expectedResult.Cardinality());
DocIdSetIterator iterator = expectedResult.GetIterator();
for (int doc = iterator.NextDoc();
doc != DocIdSetIterator.NO_MORE_DOCS;
doc = iterator.NextDoc())
{
Console.WriteLine(string.Format("Expected doc[{0}] with id value {1}", doc, indexSearcher.Doc(doc).Get("id")));
}
Console.WriteLine("actual cardinality:" + actualResult.Cardinality());
iterator = actualResult.GetIterator();
for (int doc = iterator.NextDoc();
doc != DocIdSetIterator.NO_MORE_DOCS;
doc = iterator.NextDoc())
{
Console.WriteLine(string.Format("Actual doc[{0}] with id value {1}", doc, indexSearcher.Doc(doc).Get("id")));
}
}
assertEquals(expectedResult, actualResult);
// Asserting TopDocs...
TopDocs expectedTopDocs = CreateExpectedTopDocs(randomValue, from, scoreMode, context);
TopDocs actualTopDocs = topScoreDocCollector.GetTopDocs();
assertEquals(expectedTopDocs.TotalHits, actualTopDocs.TotalHits);
assertEquals(expectedTopDocs.ScoreDocs.Length, actualTopDocs.ScoreDocs.Length);
if (scoreMode == ScoreMode.None)
{
continue;
}
assertEquals(expectedTopDocs.MaxScore, actualTopDocs.MaxScore, 0.0f);
for (int i = 0; i < expectedTopDocs.ScoreDocs.Length; i++)
{
if (Verbose)
{
string.Format("Expected doc: {0} | Actual doc: {1}\n", expectedTopDocs.ScoreDocs[i].Doc, actualTopDocs.ScoreDocs[i].Doc);
string.Format("Expected score: {0} | Actual score: {1}\n", expectedTopDocs.ScoreDocs[i].Score, actualTopDocs.ScoreDocs[i].Score);
}
assertEquals(expectedTopDocs.ScoreDocs[i].Doc, actualTopDocs.ScoreDocs[i].Doc);
assertEquals(expectedTopDocs.ScoreDocs[i].Score, actualTopDocs.ScoreDocs[i].Score, 0.0f);
Explanation explanation = indexSearcher.Explain(joinQuery, expectedTopDocs.ScoreDocs[i].Doc);
assertEquals(expectedTopDocs.ScoreDocs[i].Score, explanation.Value, 0.0f);
}
}
topLevelReader.Dispose();
dir.Dispose();
}
}
public void TestMinShouldMatch()
{
Directory dir = NewDirectory();
MockAnalyzer analyzer = new MockAnalyzer(Random());
RandomIndexWriter w = new RandomIndexWriter(Random(), dir, analyzer, Similarity, TimeZone);
string[] docs = new string[]
{
@"this is the end of the world right", @"is this it or maybe not",
@"this is the end of the universe as we know it",
@"there is the famous restaurant at the end of the universe"
};
for (int i = 0; i < docs.Length; i++)
{
Document doc = new Document();
doc.Add(NewStringField(@"id", @"" + i, Field.Store.YES));
doc.Add(NewTextField(@"field", docs[i], Field.Store.NO));
w.AddDocument(doc);
}
IndexReader r = w.Reader;
IndexSearcher s = NewSearcher(r);
{
CommonTermsQuery query = new CommonTermsQuery(Occur.SHOULD, Occur.SHOULD, Random().NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "is"));
query.Add(new Term("field", "this"));
query.Add(new Term("field", "end"));
query.Add(new Term("field", "world"));
query.Add(new Term("field", "universe"));
query.Add(new Term("field", "right"));
query.LowFreqMinimumNumberShouldMatch = 0.5F;
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 1);
assertEquals(@"0", r.Document(search.ScoreDocs[0].Doc).Get(@"id"));
}
{
CommonTermsQuery query = new CommonTermsQuery(Occur.SHOULD, Occur.SHOULD, Random().NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "is"));
query.Add(new Term("field", "this"));
query.Add(new Term("field", "end"));
query.Add(new Term("field", "world"));
query.Add(new Term("field", "universe"));
query.Add(new Term("field", "right"));
query.LowFreqMinimumNumberShouldMatch = 2F;
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 1);
assertEquals(@"0", r.Document(search.ScoreDocs[0].Doc).Get(@"id"));
}
{
CommonTermsQuery query = new CommonTermsQuery(Occur.SHOULD, Occur.SHOULD, Random().NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "is"));
query.Add(new Term("field", "this"));
query.Add(new Term("field", "end"));
query.Add(new Term("field", "world"));
query.Add(new Term("field", "universe"));
query.Add(new Term("field", "right"));
query.LowFreqMinimumNumberShouldMatch = 0.49F;
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 3);
assertEquals(@"0", r.Document(search.ScoreDocs[0].Doc).Get(@"id"));
assertEquals(@"2", r.Document(search.ScoreDocs[1].Doc).Get(@"id"));
assertEquals(@"3", r.Document(search.ScoreDocs[2].Doc).Get(@"id"));
}
{
CommonTermsQuery query = new CommonTermsQuery(Occur.SHOULD, Occur.SHOULD, Random().NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "is"));
query.Add(new Term("field", "this"));
query.Add(new Term("field", "end"));
query.Add(new Term("field", "world"));
query.Add(new Term("field", "universe"));
query.Add(new Term("field", "right"));
query.LowFreqMinimumNumberShouldMatch = 1F;
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 3);
assertEquals(@"0", r.Document(search.ScoreDocs[0].Doc).Get(@"id"));
assertEquals(@"2", r.Document(search.ScoreDocs[1].Doc).Get(@"id"));
assertEquals(@"3", r.Document(search.ScoreDocs[2].Doc).Get(@"id"));
assertTrue(search.ScoreDocs[1].Score > search.ScoreDocs[2].Score);
}
{
CommonTermsQuery query = new CommonTermsQuery(Occur.SHOULD, Occur.SHOULD, Random().NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "is"));
query.Add(new Term("field", "this"));
query.Add(new Term("field", "end"));
query.Add(new Term("field", "world"));
query.Add(new Term("field", "universe"));
query.Add(new Term("field", "right"));
query.LowFreqMinimumNumberShouldMatch = 1F;
query.HighFreqMinimumNumberShouldMatch = 4F;
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 3);
assertEquals(search.ScoreDocs[1].Score, search.ScoreDocs[2].Score, 0F);
assertEquals(@"0", r.Document(search.ScoreDocs[0].Doc).Get(@"id"));
assertEquals(new HashSet <string>(Arrays.AsList(@"2", @"3")), new HashSet <string>(Arrays.AsList(r.Document(search.ScoreDocs[1].Doc).Get(@"id"), r.Document(search.ScoreDocs[2].Doc).Get(@"id"))));
}
{
CommonTermsQuery query = new CommonTermsQuery(Occur.SHOULD, Occur.SHOULD, Random().NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "is"));
query.Add(new Term("field", "this"));
query.Add(new Term("field", "the"));
query.LowFreqMinimumNumberShouldMatch = 1F;
query.HighFreqMinimumNumberShouldMatch = 2F;
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 4);
}
{
CommonTermsQuery query = new CommonTermsQuery(Occur.MUST, Occur.SHOULD, Random().NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "is"));
query.Add(new Term("field", "this"));
query.Add(new Term("field", "the"));
query.LowFreqMinimumNumberShouldMatch = 1F;
query.HighFreqMinimumNumberShouldMatch = 2F;
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 2);
assertEquals(new HashSet <string>(Arrays.AsList(@"0", @"2")), new HashSet <string>(Arrays.AsList(r.Document(search.ScoreDocs[0].Doc).Get(@"id"), r.Document(search.ScoreDocs[1].Doc).Get(@"id"))));
}
r.Dispose();
w.Dispose();
dir.Dispose();
}
public void TestSimple()
{
const string idField = "id";
const string toField = "productId";
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(Random, dir,
NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random))
.SetMergePolicy(NewLogMergePolicy()));
// 0
Document doc = new Document();
doc.Add(new TextField("description", "random text", Field.Store.NO));
doc.Add(new TextField("name", "name1", Field.Store.NO));
doc.Add(new TextField(idField, "1", Field.Store.NO));
w.AddDocument(doc);
// 1
doc = new Document();
doc.Add(new TextField("price", "10.0", Field.Store.NO));
doc.Add(new TextField(idField, "2", Field.Store.NO));
doc.Add(new TextField(toField, "1", Field.Store.NO));
w.AddDocument(doc);
// 2
doc = new Document();
doc.Add(new TextField("price", "20.0", Field.Store.NO));
doc.Add(new TextField(idField, "3", Field.Store.NO));
doc.Add(new TextField(toField, "1", Field.Store.NO));
w.AddDocument(doc);
// 3
doc = new Document();
doc.Add(new TextField("description", "more random text", Field.Store.NO));
doc.Add(new TextField("name", "name2", Field.Store.NO));
doc.Add(new TextField(idField, "4", Field.Store.NO));
w.AddDocument(doc);
w.Commit();
// 4
doc = new Document();
doc.Add(new TextField("price", "10.0", Field.Store.NO));
doc.Add(new TextField(idField, "5", Field.Store.NO));
doc.Add(new TextField(toField, "4", Field.Store.NO));
w.AddDocument(doc);
// 5
doc = new Document();
doc.Add(new TextField("price", "20.0", Field.Store.NO));
doc.Add(new TextField(idField, "6", Field.Store.NO));
doc.Add(new TextField(toField, "4", Field.Store.NO));
w.AddDocument(doc);
IndexSearcher indexSearcher = new IndexSearcher(w.GetReader());
w.Dispose();
// Search for product
Query joinQuery = JoinUtil.CreateJoinQuery(idField, false, toField, new TermQuery(new Term("name", "name2")),
indexSearcher, ScoreMode.None);
TopDocs result = indexSearcher.Search(joinQuery, 10);
assertEquals(2, result.TotalHits);
assertEquals(4, result.ScoreDocs[0].Doc);
assertEquals(5, result.ScoreDocs[1].Doc);
joinQuery = JoinUtil.CreateJoinQuery(idField, false, toField, new TermQuery(new Term("name", "name1")),
indexSearcher, ScoreMode.None);
result = indexSearcher.Search(joinQuery, 10);
assertEquals(2, result.TotalHits);
assertEquals(1, result.ScoreDocs[0].Doc);
assertEquals(2, result.ScoreDocs[1].Doc);
// Search for offer
joinQuery = JoinUtil.CreateJoinQuery(toField, false, idField, new TermQuery(new Term("id", "5")),
indexSearcher, ScoreMode.None);
result = indexSearcher.Search(joinQuery, 10);
assertEquals(1, result.TotalHits);
assertEquals(3, result.ScoreDocs[0].Doc);
indexSearcher.IndexReader.Dispose();
dir.Dispose();
}
public virtual void TestRandomPhrases()
{
Directory dir = NewDirectory();
Analyzer analyzer = new MockAnalyzer(Random());
RandomIndexWriter w = new RandomIndexWriter(Random(), dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, analyzer).SetMergePolicy(NewLogMergePolicy()));
IList <IList <string> > docs = new List <IList <string> >();
Documents.Document d = new Documents.Document();
Field f = NewTextField("f", "", Field.Store.NO);
d.Add(f);
Random r = Random();
int NUM_DOCS = AtLeast(10);
for (int i = 0; i < NUM_DOCS; i++)
{
// must be > 4096 so it spans multiple chunks
int termCount = TestUtil.NextInt(Random(), 4097, 8200);
IList <string> doc = new List <string>();
StringBuilder sb = new StringBuilder();
while (doc.Count < termCount)
{
if (r.Next(5) == 1 || docs.Count == 0)
{
// make new non-empty-string term
string term;
while (true)
{
term = TestUtil.RandomUnicodeString(r);
if (term.Length > 0)
{
break;
}
}
IOException priorException = null;
TokenStream ts = analyzer.GetTokenStream("ignore", new StringReader(term));
try
{
ICharTermAttribute termAttr = ts.AddAttribute <ICharTermAttribute>();
ts.Reset();
while (ts.IncrementToken())
{
string text = termAttr.ToString();
doc.Add(text);
sb.Append(text).Append(' ');
}
ts.End();
}
catch (IOException e)
{
priorException = e;
}
finally
{
IOUtils.CloseWhileHandlingException(priorException, ts);
}
}
else
{
// pick existing sub-phrase
IList <string> lastDoc = docs[r.Next(docs.Count)];
int len = TestUtil.NextInt(r, 1, 10);
int start = r.Next(lastDoc.Count - len);
for (int k = start; k < start + len; k++)
{
string t = lastDoc[k];
doc.Add(t);
sb.Append(t).Append(' ');
}
}
}
docs.Add(doc);
f.SetStringValue(sb.ToString());
w.AddDocument(d);
}
IndexReader reader = w.Reader;
IndexSearcher s = NewSearcher(reader);
w.Dispose();
// now search
int num = AtLeast(10);
for (int i = 0; i < num; i++)
{
int docID = r.Next(docs.Count);
IList <string> doc = docs[docID];
int numTerm = TestUtil.NextInt(r, 2, 20);
int start = r.Next(doc.Count - numTerm);
PhraseQuery pq = new PhraseQuery();
StringBuilder sb = new StringBuilder();
for (int t = start; t < start + numTerm; t++)
{
pq.Add(new Term("f", doc[t]));
sb.Append(doc[t]).Append(' ');
}
TopDocs hits = s.Search(pq, NUM_DOCS);
bool found = false;
for (int j = 0; j < hits.ScoreDocs.Length; j++)
{
if (hits.ScoreDocs[j].Doc == docID)
{
found = true;
break;
}
}
Assert.IsTrue(found, "phrase '" + sb + "' not found; start=" + start);
}
reader.Dispose();
dir.Dispose();
}
public virtual void TestPhraseQueryInConjunctionScorer()
{
Directory directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), directory, Similarity, TimeZone);
Documents.Document doc = new Documents.Document();
doc.Add(NewTextField("source", "marketing info", Field.Store.YES));
writer.AddDocument(doc);
doc = new Documents.Document();
doc.Add(NewTextField("contents", "foobar", Field.Store.YES));
doc.Add(NewTextField("source", "marketing info", Field.Store.YES));
writer.AddDocument(doc);
IndexReader reader = writer.Reader;
writer.Dispose();
IndexSearcher searcher = NewSearcher(reader);
PhraseQuery phraseQuery = new PhraseQuery();
phraseQuery.Add(new Term("source", "marketing"));
phraseQuery.Add(new Term("source", "info"));
ScoreDoc[] hits = searcher.Search(phraseQuery, null, 1000).ScoreDocs;
Assert.AreEqual(2, hits.Length);
QueryUtils.Check(Random(), phraseQuery, searcher, Similarity);
TermQuery termQuery = new TermQuery(new Term("contents", "foobar"));
BooleanQuery booleanQuery = new BooleanQuery();
booleanQuery.Add(termQuery, Occur.MUST);
booleanQuery.Add(phraseQuery, Occur.MUST);
hits = searcher.Search(booleanQuery, null, 1000).ScoreDocs;
Assert.AreEqual(1, hits.Length);
QueryUtils.Check(Random(), termQuery, searcher, Similarity);
reader.Dispose();
writer = new RandomIndexWriter(Random(), directory, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())).SetOpenMode(OpenMode.CREATE));
doc = new Documents.Document();
doc.Add(NewTextField("contents", "map entry woo", Field.Store.YES));
writer.AddDocument(doc);
doc = new Documents.Document();
doc.Add(NewTextField("contents", "woo map entry", Field.Store.YES));
writer.AddDocument(doc);
doc = new Documents.Document();
doc.Add(NewTextField("contents", "map foobarword entry woo", Field.Store.YES));
writer.AddDocument(doc);
reader = writer.Reader;
writer.Dispose();
searcher = NewSearcher(reader);
termQuery = new TermQuery(new Term("contents", "woo"));
phraseQuery = new PhraseQuery();
phraseQuery.Add(new Term("contents", "map"));
phraseQuery.Add(new Term("contents", "entry"));
hits = searcher.Search(termQuery, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length);
hits = searcher.Search(phraseQuery, null, 1000).ScoreDocs;
Assert.AreEqual(2, hits.Length);
booleanQuery = new BooleanQuery();
booleanQuery.Add(termQuery, Occur.MUST);
booleanQuery.Add(phraseQuery, Occur.MUST);
hits = searcher.Search(booleanQuery, null, 1000).ScoreDocs;
Assert.AreEqual(2, hits.Length);
booleanQuery = new BooleanQuery();
booleanQuery.Add(phraseQuery, Occur.MUST);
booleanQuery.Add(termQuery, Occur.MUST);
hits = searcher.Search(booleanQuery, null, 1000).ScoreDocs;
Assert.AreEqual(2, hits.Length);
QueryUtils.Check(Random(), booleanQuery, searcher, Similarity);
reader.Dispose();
directory.Dispose();
}
public virtual void Test10kPulsed()
{
// we always run this test with pulsing codec.
Codec cp = TestUtil.AlwaysPostingsFormat(new Pulsing41PostingsFormat(1));
DirectoryInfo f = CreateTempDir("10kpulsed");
BaseDirectoryWrapper dir = NewFSDirectory(f);
dir.CheckIndexOnDispose = false; // we do this ourselves explicitly
RandomIndexWriter iw = new RandomIndexWriter(Random, dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random)).SetCodec(cp));
Document document = new Document();
FieldType ft = new FieldType(TextField.TYPE_STORED);
switch (TestUtil.NextInt32(Random, 0, 2))
{
case 0:
ft.IndexOptions = IndexOptions.DOCS_ONLY;
break;
case 1:
ft.IndexOptions = IndexOptions.DOCS_AND_FREQS;
break;
default:
ft.IndexOptions = IndexOptions.DOCS_AND_FREQS_AND_POSITIONS;
break;
}
Field field = NewField("field", "", ft);
document.Add(field);
//NumberFormat df = new DecimalFormat("00000", new DecimalFormatSymbols(Locale.ROOT));
for (int i = 0; i < 10050; i++)
{
//field.StringValue = df.format(i);
field.SetStringValue(i.ToString("00000", CultureInfo.InvariantCulture));
iw.AddDocument(document);
}
IndexReader ir = iw.GetReader();
iw.Dispose();
TermsEnum te = MultiFields.GetTerms(ir, "field").GetIterator(null);
DocsEnum de = null;
for (int i = 0; i < 10050; i++)
{
//string expected = df.format(i);
string expected = i.ToString("00000", CultureInfo.InvariantCulture);
assertEquals(expected, te.Next().Utf8ToString());
de = TestUtil.Docs(Random, te, null, de, DocsFlags.NONE);
assertTrue(de.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
assertEquals(DocIdSetIterator.NO_MORE_DOCS, de.NextDoc());
}
ir.Dispose();
TestUtil.CheckIndex(dir);
dir.Dispose();
}
public void TestInsideBooleanQuery()
{
const string idField = "id";
const string toField = "productId";
Directory dir = NewDirectory();
RandomIndexWriter w = new RandomIndexWriter(Random, dir,
NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random))
.SetMergePolicy(NewLogMergePolicy()));
// 0
Document doc = new Document();
doc.Add(new TextField("description", "random text", Field.Store.NO));
doc.Add(new TextField("name", "name1", Field.Store.NO));
doc.Add(new TextField(idField, "7", Field.Store.NO));
w.AddDocument(doc);
// 1
doc = new Document();
doc.Add(new TextField("price", "10.0", Field.Store.NO));
doc.Add(new TextField(idField, "2", Field.Store.NO));
doc.Add(new TextField(toField, "7", Field.Store.NO));
w.AddDocument(doc);
// 2
doc = new Document();
doc.Add(new TextField("price", "20.0", Field.Store.NO));
doc.Add(new TextField(idField, "3", Field.Store.NO));
doc.Add(new TextField(toField, "7", Field.Store.NO));
w.AddDocument(doc);
// 3
doc = new Document();
doc.Add(new TextField("description", "more random text", Field.Store.NO));
doc.Add(new TextField("name", "name2", Field.Store.NO));
doc.Add(new TextField(idField, "0", Field.Store.NO));
w.AddDocument(doc);
w.Commit();
// 4
doc = new Document();
doc.Add(new TextField("price", "10.0", Field.Store.NO));
doc.Add(new TextField(idField, "5", Field.Store.NO));
doc.Add(new TextField(toField, "0", Field.Store.NO));
w.AddDocument(doc);
// 5
doc = new Document();
doc.Add(new TextField("price", "20.0", Field.Store.NO));
doc.Add(new TextField(idField, "6", Field.Store.NO));
doc.Add(new TextField(toField, "0", Field.Store.NO));
w.AddDocument(doc);
w.ForceMerge(1);
IndexSearcher indexSearcher = new IndexSearcher(w.GetReader());
w.Dispose();
// Search for product
Query joinQuery = JoinUtil.CreateJoinQuery(idField, false, toField,
new TermQuery(new Term("description", "random")), indexSearcher, ScoreMode.Avg);
BooleanQuery bq = new BooleanQuery();
bq.Add(joinQuery, Occur.SHOULD);
bq.Add(new TermQuery(new Term("id", "3")), Occur.SHOULD);
indexSearcher.Search(bq, new CollectorAnonymousInnerClassHelper(this));
indexSearcher.IndexReader.Dispose();
dir.Dispose();
}
public virtual void TestFuzziness()
{
Directory directory = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, directory);
AddDoc("aaaaa", writer);
AddDoc("aaaab", writer);
AddDoc("aaabb", writer);
AddDoc("aabbb", writer);
AddDoc("abbbb", writer);
AddDoc("bbbbb", writer);
AddDoc("ddddd", writer);
IndexReader reader = writer.GetReader();
IndexSearcher searcher = NewSearcher(reader);
writer.Dispose();
FuzzyQuery query = new FuzzyQuery(new Term("field", "aaaaa"), FuzzyQuery.DefaultMaxEdits, 0);
ScoreDoc[] hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length);
// same with prefix
query = new FuzzyQuery(new Term("field", "aaaaa"), FuzzyQuery.DefaultMaxEdits, 1);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length);
query = new FuzzyQuery(new Term("field", "aaaaa"), FuzzyQuery.DefaultMaxEdits, 2);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length);
query = new FuzzyQuery(new Term("field", "aaaaa"), FuzzyQuery.DefaultMaxEdits, 3);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length);
query = new FuzzyQuery(new Term("field", "aaaaa"), FuzzyQuery.DefaultMaxEdits, 4);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(2, hits.Length);
query = new FuzzyQuery(new Term("field", "aaaaa"), FuzzyQuery.DefaultMaxEdits, 5);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(1, hits.Length);
query = new FuzzyQuery(new Term("field", "aaaaa"), FuzzyQuery.DefaultMaxEdits, 6);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(1, hits.Length);
// test scoring
query = new FuzzyQuery(new Term("field", "bbbbb"), FuzzyQuery.DefaultMaxEdits, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length, "3 documents should match");
IList <string> order = Arrays.AsList("bbbbb", "abbbb", "aabbb");
for (int i = 0; i < hits.Length; i++)
{
string term = searcher.Doc(hits[i].Doc).Get("field");
//System.out.println(hits[i].Score);
Assert.AreEqual(order[i], term);
}
// test pq size by supplying maxExpansions=2
// this query would normally return 3 documents, because 3 terms match (see above):
query = new FuzzyQuery(new Term("field", "bbbbb"), FuzzyQuery.DefaultMaxEdits, 0, 2, false);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(2, hits.Length, "only 2 documents should match");
order = Arrays.AsList("bbbbb", "abbbb");
for (int i = 0; i < hits.Length; i++)
{
string term = searcher.Doc(hits[i].Doc).Get("field");
//System.out.println(hits[i].Score);
Assert.AreEqual(order[i], term);
}
// not similar enough:
query = new FuzzyQuery(new Term("field", "xxxxx"), FuzzyQuery.DefaultMaxEdits, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(0, hits.Length);
query = new FuzzyQuery(new Term("field", "aaccc"), FuzzyQuery.DefaultMaxEdits, 0); // edit distance to "aaaaa" = 3
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(0, hits.Length);
// query identical to a word in the index:
query = new FuzzyQuery(new Term("field", "aaaaa"), FuzzyQuery.DefaultMaxEdits, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
// default allows for up to two edits:
Assert.AreEqual(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
Assert.AreEqual(searcher.Doc(hits[2].Doc).Get("field"), ("aaabb"));
// query similar to a word in the index:
query = new FuzzyQuery(new Term("field", "aaaac"), FuzzyQuery.DefaultMaxEdits, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
Assert.AreEqual(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
Assert.AreEqual(searcher.Doc(hits[2].Doc).Get("field"), ("aaabb"));
// now with prefix
query = new FuzzyQuery(new Term("field", "aaaac"), FuzzyQuery.DefaultMaxEdits, 1);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
Assert.AreEqual(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
Assert.AreEqual(searcher.Doc(hits[2].Doc).Get("field"), ("aaabb"));
query = new FuzzyQuery(new Term("field", "aaaac"), FuzzyQuery.DefaultMaxEdits, 2);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
Assert.AreEqual(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
Assert.AreEqual(searcher.Doc(hits[2].Doc).Get("field"), ("aaabb"));
query = new FuzzyQuery(new Term("field", "aaaac"), FuzzyQuery.DefaultMaxEdits, 3);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(3, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
Assert.AreEqual(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
Assert.AreEqual(searcher.Doc(hits[2].Doc).Get("field"), ("aaabb"));
query = new FuzzyQuery(new Term("field", "aaaac"), FuzzyQuery.DefaultMaxEdits, 4);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(2, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("aaaaa"));
Assert.AreEqual(searcher.Doc(hits[1].Doc).Get("field"), ("aaaab"));
query = new FuzzyQuery(new Term("field", "aaaac"), FuzzyQuery.DefaultMaxEdits, 5);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(0, hits.Length);
query = new FuzzyQuery(new Term("field", "ddddX"), FuzzyQuery.DefaultMaxEdits, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(1, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("ddddd"));
// now with prefix
query = new FuzzyQuery(new Term("field", "ddddX"), FuzzyQuery.DefaultMaxEdits, 1);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(1, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("ddddd"));
query = new FuzzyQuery(new Term("field", "ddddX"), FuzzyQuery.DefaultMaxEdits, 2);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(1, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("ddddd"));
query = new FuzzyQuery(new Term("field", "ddddX"), FuzzyQuery.DefaultMaxEdits, 3);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(1, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("ddddd"));
query = new FuzzyQuery(new Term("field", "ddddX"), FuzzyQuery.DefaultMaxEdits, 4);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(1, hits.Length);
Assert.AreEqual(searcher.Doc(hits[0].Doc).Get("field"), ("ddddd"));
query = new FuzzyQuery(new Term("field", "ddddX"), FuzzyQuery.DefaultMaxEdits, 5);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(0, hits.Length);
// different field = no match:
query = new FuzzyQuery(new Term("anotherfield", "ddddX"), FuzzyQuery.DefaultMaxEdits, 0);
hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(0, hits.Length);
reader.Dispose();
directory.Dispose();
}
public void TestBasics()
{
Directory dir = NewDirectory();
MockAnalyzer analyzer = new MockAnalyzer(Random);
RandomIndexWriter w = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, dir, analyzer);
var docs = new string[]
{
@"this is the end of the world right", @"is this it or maybe not",
@"this is the end of the universe as we know it",
@"there is the famous restaurant at the end of the universe"
};
for (int i = 0; i < docs.Length; i++)
{
Document doc = new Document();
doc.Add(NewStringField(@"id", @"" + i, Field.Store.YES));
doc.Add(NewTextField(@"field", docs[i], Field.Store.NO));
w.AddDocument(doc);
}
IndexReader r = w.GetReader();
IndexSearcher s = NewSearcher(r);
{
CommonTermsQuery query = new CommonTermsQuery(Occur.SHOULD, Occur.SHOULD, Random.NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "is"));
query.Add(new Term("field", "this"));
query.Add(new Term("field", "end"));
query.Add(new Term("field", "world"));
query.Add(new Term("field", "universe"));
query.Add(new Term("field", "right"));
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 3);
assertEquals(@"0", r.Document(search.ScoreDocs[0].Doc).Get(@"id"));
assertEquals(@"2", r.Document(search.ScoreDocs[1].Doc).Get(@"id"));
assertEquals(@"3", r.Document(search.ScoreDocs[2].Doc).Get(@"id"));
}
{
CommonTermsQuery query = new CommonTermsQuery(Occur.SHOULD, Occur.SHOULD, Random.NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "is"));
query.Add(new Term("field", "this"));
query.Add(new Term("field", "end"));
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 2);
assertEquals(@"0", r.Document(search.ScoreDocs[0].Doc).Get(@"id"));
assertEquals(@"2", r.Document(search.ScoreDocs[1].Doc).Get(@"id"));
}
{
CommonTermsQuery query = new CommonTermsQuery(Occur.SHOULD, Occur.MUST, Random.NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "is"));
query.Add(new Term("field", "this"));
query.Add(new Term("field", "end"));
query.Add(new Term("field", "world"));
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 1);
assertEquals(@"0", r.Document(search.ScoreDocs[0].Doc).Get(@"id"));
}
{
CommonTermsQuery query = new CommonTermsQuery(Occur.SHOULD, Occur.MUST, Random.NextBoolean() ? 2F : 0.5F);
query.Add(new Term("field", "restaurant"));
query.Add(new Term("field", "universe"));
TopDocs search = s.Search(query, 10);
assertEquals(search.TotalHits, 1);
assertEquals(@"3", r.Document(search.ScoreDocs[0].Doc).Get(@"id"));
}
r.Dispose();
w.Dispose();
dir.Dispose();
}
private void DoTest(DocValuesType type)
{
Directory d = NewDirectory();
IndexWriterConfig iwConfig = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random));
int nDocs = AtLeast(50);
Field id = new NumericDocValuesField("id", 0);
Field f;
switch (type)
{
case DocValuesType.BINARY:
f = new BinaryDocValuesField("dv", new BytesRef());
break;
case DocValuesType.SORTED:
f = new SortedDocValuesField("dv", new BytesRef());
break;
case DocValuesType.NUMERIC:
f = new NumericDocValuesField("dv", 0);
break;
default:
throw new InvalidOperationException();
}
Document document = new Document();
document.Add(id);
document.Add(f);
object[] vals = new object[nDocs];
RandomIndexWriter iw = new RandomIndexWriter(Random, d, iwConfig);
for (int i = 0; i < nDocs; ++i)
{
id.SetInt64Value(i);
switch (type)
{
case DocValuesType.SORTED:
case DocValuesType.BINARY:
do
{
vals[i] = TestUtil.RandomSimpleString(Random, 20);
} while (((string)vals[i]).Length == 0);
f.SetBytesValue(new BytesRef((string)vals[i]));
break;
case DocValuesType.NUMERIC:
int bitsPerValue = RandomInts.RandomInt32Between(Random, 1, 31); // keep it an int
vals[i] = (long)Random.Next((int)PackedInt32s.MaxValue(bitsPerValue));
f.SetInt64Value((long)vals[i]);
break;
}
iw.AddDocument(document);
if (Random.NextBoolean() && i % 10 == 9)
{
iw.Commit();
}
}
iw.Dispose();
DirectoryReader rd = DirectoryReader.Open(d);
foreach (AtomicReaderContext leave in rd.Leaves)
{
FunctionValues ids = (new Int64FieldSource("id")).GetValues(null, leave);
ValueSource vs;
switch (type)
{
case DocValuesType.BINARY:
case DocValuesType.SORTED:
vs = new BytesRefFieldSource("dv");
break;
case DocValuesType.NUMERIC:
vs = new Int64FieldSource("dv");
break;
default:
throw new InvalidOperationException();
}
FunctionValues values = vs.GetValues(null, leave);
BytesRef bytes = new BytesRef();
for (int i = 0; i < leave.AtomicReader.MaxDoc; ++i)
{
assertTrue(values.Exists(i));
if (vs is BytesRefFieldSource)
{
assertTrue(values.ObjectVal(i) is string);
}
else if (vs is Int64FieldSource)
{
assertTrue(values.ObjectVal(i) is long?);
assertTrue(values.BytesVal(i, bytes));
}
else
{
throw new InvalidOperationException();
}
object expected = vals[ids.Int32Val(i)];
switch (type)
{
case DocValuesType.SORTED:
values.OrdVal(i); // no exception
assertTrue(values.NumOrd >= 1);
goto case DocValuesType.BINARY;
case DocValuesType.BINARY:
assertEquals(expected, values.ObjectVal(i));
assertEquals(expected, values.StrVal(i));
assertEquals(expected, values.ObjectVal(i));
assertEquals(expected, values.StrVal(i));
assertTrue(values.BytesVal(i, bytes));
assertEquals(new BytesRef((string)expected), bytes);
break;
case DocValuesType.NUMERIC:
assertEquals(Convert.ToInt64(expected, CultureInfo.InvariantCulture), values.Int64Val(i));
break;
}
}
}
rd.Dispose();
d.Dispose();
}
public virtual void TestNRTAndCommit()
{
Directory dir = NewDirectory();
NRTCachingDirectory cachedDir = new NRTCachingDirectory(dir, 2.0, 25.0);
MockAnalyzer analyzer = new MockAnalyzer(Random);
analyzer.MaxTokenLength = TestUtil.NextInt32(Random, 1, IndexWriter.MAX_TERM_LENGTH);
IndexWriterConfig conf = NewIndexWriterConfig(TEST_VERSION_CURRENT, analyzer);
RandomIndexWriter w = new RandomIndexWriter(Random, cachedDir, conf);
LineFileDocs docs = new LineFileDocs(Random, DefaultCodecSupportsDocValues);
int numDocs = TestUtil.NextInt32(Random, 100, 400);
if (Verbose)
{
Console.WriteLine("TEST: numDocs=" + numDocs);
}
IList <BytesRef> ids = new List <BytesRef>();
DirectoryReader r = null;
for (int docCount = 0; docCount < numDocs; docCount++)
{
Document doc = docs.NextDoc();
ids.Add(new BytesRef(doc.Get("docid")));
w.AddDocument(doc);
if (Random.Next(20) == 17)
{
if (r == null)
{
r = DirectoryReader.Open(w.IndexWriter, false);
}
else
{
DirectoryReader r2 = DirectoryReader.OpenIfChanged(r);
if (r2 != null)
{
r.Dispose();
r = r2;
}
}
Assert.AreEqual(1 + docCount, r.NumDocs);
IndexSearcher s = NewSearcher(r);
// Just make sure search can run; we can't assert
// totHits since it could be 0
TopDocs hits = s.Search(new TermQuery(new Term("body", "the")), 10);
// System.out.println("tot hits " + hits.totalHits);
}
}
if (r != null)
{
r.Dispose();
}
// Close should force cache to clear since all files are sync'd
w.Dispose();
string[] cachedFiles = cachedDir.ListCachedFiles();
foreach (string file in cachedFiles)
{
Console.WriteLine("FAIL: cached file " + file + " remains after sync");
}
Assert.AreEqual(0, cachedFiles.Length);
r = DirectoryReader.Open(dir);
foreach (BytesRef id in ids)
{
Assert.AreEqual(1, r.DocFreq(new Term("docid", id)));
}
r.Dispose();
cachedDir.Dispose();
docs.Dispose();
}
public void TestRandomIndex()
{
Directory dir = NewDirectory();
MockAnalyzer analyzer = new MockAnalyzer(Random);
analyzer.MaxTokenLength = TestUtil.NextInt32(Random, 1, IndexWriter.MAX_TERM_LENGTH);
RandomIndexWriter w = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, dir, analyzer);
CreateRandomIndex(AtLeast(50), w, Random.NextInt64());
DirectoryReader reader = w.GetReader();
AtomicReader wrapper = SlowCompositeReaderWrapper.Wrap(reader);
string field = @"body";
Terms terms = wrapper.GetTerms(field);
var lowFreqQueue = new AnonymousPriorityQueue(this, 5);
Util.PriorityQueue <TermAndFreq> highFreqQueue = new AnonymousPriorityQueue1(this, 5);
try
{
TermsEnum iterator = terms.GetEnumerator();
while (iterator.MoveNext())
{
if (highFreqQueue.Count < 5)
{
highFreqQueue.Add(new TermAndFreq(BytesRef.DeepCopyOf(iterator.Term), iterator.DocFreq));
lowFreqQueue.Add(new TermAndFreq(BytesRef.DeepCopyOf(iterator.Term), iterator.DocFreq));
}
else
{
if (highFreqQueue.Top.freq < iterator.DocFreq)
{
highFreqQueue.Top.freq = iterator.DocFreq;
highFreqQueue.Top.term = BytesRef.DeepCopyOf(iterator.Term);
highFreqQueue.UpdateTop();
}
if (lowFreqQueue.Top.freq > iterator.DocFreq)
{
lowFreqQueue.Top.freq = iterator.DocFreq;
lowFreqQueue.Top.term = BytesRef.DeepCopyOf(iterator.Term);
lowFreqQueue.UpdateTop();
}
}
}
int lowFreq = lowFreqQueue.Top.freq;
int highFreq = highFreqQueue.Top.freq;
AssumeTrue(@"unlucky index", highFreq - 1 > lowFreq);
List <TermAndFreq> highTerms = QueueToList(highFreqQueue);
List <TermAndFreq> lowTerms = QueueToList(lowFreqQueue);
IndexSearcher searcher = NewSearcher(reader);
Occur lowFreqOccur = RandomOccur(Random);
BooleanQuery verifyQuery = new BooleanQuery();
CommonTermsQuery cq = new CommonTermsQuery(RandomOccur(Random), lowFreqOccur, highFreq - 1, Random.NextBoolean());
foreach (TermAndFreq termAndFreq in lowTerms)
{
cq.Add(new Term(field, termAndFreq.term));
verifyQuery.Add(new BooleanClause(new TermQuery(new Term(field, termAndFreq.term)), lowFreqOccur));
}
foreach (TermAndFreq termAndFreq in highTerms)
{
cq.Add(new Term(field, termAndFreq.term));
}
TopDocs cqSearch = searcher.Search(cq, reader.MaxDoc);
TopDocs verifySearch = searcher.Search(verifyQuery, reader.MaxDoc);
assertEquals(verifySearch.TotalHits, cqSearch.TotalHits);
var hits = new JCG.HashSet <int>();
foreach (ScoreDoc doc in verifySearch.ScoreDocs)
{
hits.Add(doc.Doc);
}
foreach (ScoreDoc doc in cqSearch.ScoreDocs)
{
assertTrue(hits.Remove(doc.Doc));
}
assertTrue(hits.Count == 0);
w.ForceMerge(1);
DirectoryReader reader2 = w.GetReader();
QueryUtils.Check(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, cq, NewSearcher(reader2));
reader2.Dispose();
}
finally
{
reader.Dispose();
wrapper.Dispose();
w.Dispose();
dir.Dispose();
}
}