public virtual void Test()
{
Assert.IsTrue(dir != null);
}
// LUCENENET NOTE: This was not added because it causes naming collisions with
// member variables "private readonly Random random;". Capitlizing it would collide
// with the Random property. Better (and safer) just to convert all of these
// from "random()" to "Random" going forward.
//internal static Random random()
//{
// return Random;
//}
internal static void assertTrue(bool condition)
{
Assert.IsTrue(condition);
}
internal static void assertFalse(string message, bool condition)
{
Assert.IsFalse(condition, message);
}
internal static void assertSame(Object expected, Object actual)
{
Assert.AreSame(expected, actual);
}
internal static void fail()
{
Assert.Fail();
}
internal static void assertNotSame(string message, object unexpected, object actual)
{
Assert.AreNotSame(unexpected, actual, message);
}
internal static void assertNull(string msg, object o)
{
Assert.Null(o, msg);
}
internal static void assertEquals(string message, byte expected, byte actual)
{
Assert.AreEqual(expected, actual, message);
}
internal static void assertEquals(double d1, double d2, double delta)
{
Assert.AreEqual(d1, d2, delta);
}
public virtual void TestCharFilter11()
{
CharFilter cs = new CharFilter1(new CharFilter1(new StringReader("")));
Assert.AreEqual(2, cs.CorrectOffset(0), "corrected offset is invalid");
}
internal static void assertEquals(byte expected, byte actual)
{
Assert.AreEqual(expected, actual);
}
public virtual void TestSimple()
{
Directory dir = NewDirectory();
RandomIndexWriter iw = new RandomIndexWriter(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, dir);
Document doc = new Document();
Field field = NewTextField("foo", "", Field.Store.NO);
doc.Add(field);
Field dvField = new SingleDocValuesField("foo_boost", 0.0F);
doc.Add(dvField);
Field field2 = NewTextField("bar", "", Field.Store.NO);
doc.Add(field2);
field.SetStringValue("quick brown fox");
field2.SetStringValue("quick brown fox");
dvField.SetSingleValue(2f); // boost x2
iw.AddDocument(doc);
field.SetStringValue("jumps over lazy brown dog");
field2.SetStringValue("jumps over lazy brown dog");
dvField.SetSingleValue(4f); // boost x4
iw.AddDocument(doc);
IndexReader ir = iw.GetReader();
iw.Dispose();
// no boosting
IndexSearcher searcher1 = NewSearcher(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
ir, false);
Similarity @base = searcher1.Similarity;
// boosting
IndexSearcher searcher2 = NewSearcher(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
ir, false);
searcher2.Similarity = new PerFieldSimilarityWrapperAnonymousClass(this, field, @base);
// in this case, we searched on field "foo". first document should have 2x the score.
TermQuery tq = new TermQuery(new Term("foo", "quick"));
QueryUtils.Check(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, tq, searcher1);
QueryUtils.Check(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, tq, searcher2);
TopDocs noboost = searcher1.Search(tq, 10);
TopDocs boost = searcher2.Search(tq, 10);
Assert.AreEqual(1, noboost.TotalHits);
Assert.AreEqual(1, boost.TotalHits);
//System.out.println(searcher2.Explain(tq, boost.ScoreDocs[0].Doc));
Assert.AreEqual(boost.ScoreDocs[0].Score, noboost.ScoreDocs[0].Score * 2f, SCORE_EPSILON);
// this query matches only the second document, which should have 4x the score.
tq = new TermQuery(new Term("foo", "jumps"));
QueryUtils.Check(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, tq, searcher1);
QueryUtils.Check(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, tq, searcher2);
noboost = searcher1.Search(tq, 10);
boost = searcher2.Search(tq, 10);
Assert.AreEqual(1, noboost.TotalHits);
Assert.AreEqual(1, boost.TotalHits);
Assert.AreEqual(boost.ScoreDocs[0].Score, noboost.ScoreDocs[0].Score * 4f, SCORE_EPSILON);
// search on on field bar just for kicks, nothing should happen, since we setup
// our sim provider to only use foo_boost for field foo.
tq = new TermQuery(new Term("bar", "quick"));
QueryUtils.Check(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, tq, searcher1);
QueryUtils.Check(
#if FEATURE_INSTANCE_TESTDATA_INITIALIZATION
this,
#endif
Random, tq, searcher2);
noboost = searcher1.Search(tq, 10);
boost = searcher2.Search(tq, 10);
Assert.AreEqual(1, noboost.TotalHits);
Assert.AreEqual(1, boost.TotalHits);
Assert.AreEqual(boost.ScoreDocs[0].Score, noboost.ScoreDocs[0].Score, SCORE_EPSILON);
ir.Dispose();
dir.Dispose();
}
public virtual void TestPhrasePrefix()
{
Directory indexStore = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random, indexStore);
Add("blueberry pie", writer);
Add("blueberry strudel", writer);
Add("blueberry pizza", writer);
Add("blueberry chewing gum", writer);
Add("bluebird pizza", writer);
Add("bluebird foobar pizza", writer);
Add("piccadilly circus", writer);
IndexReader reader = writer.GetReader();
IndexSearcher searcher = NewSearcher(reader);
// search for "blueberry pi*":
MultiPhraseQuery query1 = new MultiPhraseQuery();
// search for "strawberry pi*":
MultiPhraseQuery query2 = new MultiPhraseQuery();
query1.Add(new Term("body", "blueberry"));
query2.Add(new Term("body", "strawberry"));
LinkedList <Term> termsWithPrefix = new LinkedList <Term>();
// this TermEnum gives "piccadilly", "pie" and "pizza".
string prefix = "pi";
TermsEnum te = MultiFields.GetFields(reader).GetTerms("body").GetEnumerator();
te.SeekCeil(new BytesRef(prefix));
do
{
string s = te.Term.Utf8ToString();
if (s.StartsWith(prefix, StringComparison.Ordinal))
{
termsWithPrefix.AddLast(new Term("body", s));
}
else
{
break;
}
} while (te.MoveNext());
query1.Add(termsWithPrefix.ToArray(/*new Term[0]*/));
Assert.AreEqual("body:\"blueberry (piccadilly pie pizza)\"", query1.ToString());
query2.Add(termsWithPrefix.ToArray(/*new Term[0]*/));
Assert.AreEqual("body:\"strawberry (piccadilly pie pizza)\"", query2.ToString());
ScoreDoc[] result;
result = searcher.Search(query1, null, 1000).ScoreDocs;
Assert.AreEqual(2, result.Length);
result = searcher.Search(query2, null, 1000).ScoreDocs;
Assert.AreEqual(0, result.Length);
// search for "blue* pizza":
MultiPhraseQuery query3 = new MultiPhraseQuery();
termsWithPrefix.Clear();
prefix = "blue";
te.SeekCeil(new BytesRef(prefix));
do
{
if (te.Term.Utf8ToString().StartsWith(prefix, StringComparison.Ordinal))
{
termsWithPrefix.AddLast(new Term("body", te.Term.Utf8ToString()));
}
} while (te.MoveNext());
query3.Add(termsWithPrefix.ToArray(/*new Term[0]*/));
query3.Add(new Term("body", "pizza"));
result = searcher.Search(query3, null, 1000).ScoreDocs;
Assert.AreEqual(2, result.Length); // blueberry pizza, bluebird pizza
Assert.AreEqual("body:\"(blueberry bluebird) pizza\"", query3.ToString());
// test slop:
query3.Slop = 1;
result = searcher.Search(query3, null, 1000).ScoreDocs;
// just make sure no exc:
searcher.Explain(query3, 0);
Assert.AreEqual(3, result.Length); // blueberry pizza, bluebird pizza, bluebird
// foobar pizza
MultiPhraseQuery query4 = new MultiPhraseQuery();
try
{
query4.Add(new Term("field1", "foo"));
query4.Add(new Term("field2", "foobar"));
Assert.Fail();
}
catch (Exception e) when(e.IsIllegalArgumentException())
{
// okay, all terms must belong to the same field
}
writer.Dispose();
reader.Dispose();
indexStore.Dispose();
}
public virtual void TestZeroPosIncr()
{
Directory dir = new RAMDirectory();
Token[] tokens = new Token[3];
tokens[0] = new Token();
tokens[0].Append("a");
tokens[0].PositionIncrement = 1;
tokens[1] = new Token();
tokens[1].Append("b");
tokens[1].PositionIncrement = 0;
tokens[2] = new Token();
tokens[2].Append("c");
tokens[2].PositionIncrement = 0;
RandomIndexWriter writer = new RandomIndexWriter(Random, dir);
Document doc = new Document();
doc.Add(new TextField("field", new CannedTokenStream(tokens)));
writer.AddDocument(doc);
doc = new Document();
doc.Add(new TextField("field", new CannedTokenStream(tokens)));
writer.AddDocument(doc);
IndexReader r = writer.GetReader();
writer.Dispose();
IndexSearcher s = NewSearcher(r);
MultiPhraseQuery mpq = new MultiPhraseQuery();
//mpq.setSlop(1);
// NOTE: not great that if we do the else clause here we
// get different scores! MultiPhraseQuery counts that
// phrase as occurring twice per doc (it should be 1, I
// think?). this is because MultipleTermPositions is able to
// return the same position more than once (0, in this
// case):
if (true)
{
mpq.Add(new Term[] { new Term("field", "b"), new Term("field", "c") }, 0);
mpq.Add(new Term[] { new Term("field", "a") }, 0);
}
else
{
#pragma warning disable 162
mpq.Add(new Term[] { new Term("field", "a") }, 0);
mpq.Add(new Term[] { new Term("field", "b"), new Term("field", "c") }, 0);
#pragma warning restore 162
}
TopDocs hits = s.Search(mpq, 2);
Assert.AreEqual(2, hits.TotalHits);
Assert.AreEqual(hits.ScoreDocs[0].Score, hits.ScoreDocs[1].Score, 1e-5);
/*
* for(int hit=0;hit<hits.TotalHits;hit++) {
* ScoreDoc sd = hits.ScoreDocs[hit];
* System.out.println(" hit doc=" + sd.Doc + " score=" + sd.Score);
* }
*/
r.Dispose();
dir.Dispose();
}
internal static void assertEquals <TKey, TValue>(string message, IDictionary <TKey, TValue> expected, IDictionary <TKey, TValue> actual, bool aggressive = true)
{
Assert.AreEqual(expected, actual, aggressive, message);
}
internal static void assertEquals(string msg, float d1, float d2, float delta)
{
Assert.AreEqual(d1, d2, delta, msg);
}
internal static void assertNotSame(object unexpected, object actual)
{
Assert.AreNotSame(unexpected, actual);
}
internal static void assertEquals(float d1, float d2, float delta)
{
Assert.AreEqual(d1, d2, delta);
}
internal static void assertNull(object o)
{
Assert.Null(o);
}
internal static void assertEquals(string msg, double d1, double d2, double delta)
{
Assert.AreEqual(d1, d2, delta, msg);
}
internal static void assertArrayEquals <T>(T[] a1, T[] a2)
{
Assert.AreEqual(a1, a2);
}
internal static void assertEquals <T>(ISet <T> expected, ISet <T> actual, bool aggressive = true)
{
Assert.AreEqual(expected, actual, aggressive);
}
internal static void assertSame(string message, Object expected, Object actual)
{
Assert.AreSame(expected, actual, message);
}
internal static void assertEquals <T>(string message, IList <T> expected, IList <T> actual, bool aggressive = true)
{
Assert.AreEqual(expected, actual, aggressive, message);
}
internal static void fail(string message)
{
Assert.Fail(message);
}
internal static void assertEquals <T>(T[] expected, T[] actual)
{
Assert.AreEqual(expected, actual);
}
internal static void assertFalse(bool condition)
{
Assert.IsFalse(condition);
}
internal static void assertEquals <T>(string message, T[] expected, T[] actual)
{
Assert.AreEqual(expected, actual, message);
}
internal static void assertEquals(object expected, object actual)
{
Assert.AreEqual(expected, actual);
}
public virtual void TestSkipTo(int indexDivisor)
{
Directory dir = NewDirectory();
IndexWriter writer = new IndexWriter(dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random)).SetMergePolicy(NewLogMergePolicy()));
Term ta = new Term("content", "aaa");
for (int i = 0; i < 10; i++)
{
AddDoc(writer, "aaa aaa aaa aaa");
}
Term tb = new Term("content", "bbb");
for (int i = 0; i < 16; i++)
{
AddDoc(writer, "bbb bbb bbb bbb");
}
Term tc = new Term("content", "ccc");
for (int i = 0; i < 50; i++)
{
AddDoc(writer, "ccc ccc ccc ccc");
}
// assure that we deal with a single segment
writer.ForceMerge(1);
writer.Dispose();
IndexReader reader = DirectoryReader.Open(dir, indexDivisor);
DocsEnum tdocs = TestUtil.Docs(Random, reader, ta.Field, new BytesRef(ta.Text()), MultiFields.GetLiveDocs(reader), null, DocsFlags.FREQS);
// without optimization (assumption skipInterval == 16)
// with next
Assert.IsTrue(tdocs.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(0, tdocs.DocID);
Assert.AreEqual(4, tdocs.Freq);
Assert.IsTrue(tdocs.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(1, tdocs.DocID);
Assert.AreEqual(4, tdocs.Freq);
Assert.IsTrue(tdocs.Advance(2) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(2, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(4) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(4, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(9) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(9, tdocs.DocID);
Assert.IsFalse(tdocs.Advance(10) != DocIdSetIterator.NO_MORE_DOCS);
// without next
tdocs = TestUtil.Docs(Random, reader, ta.Field, new BytesRef(ta.Text()), MultiFields.GetLiveDocs(reader), null, 0);
Assert.IsTrue(tdocs.Advance(0) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(0, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(4) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(4, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(9) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(9, tdocs.DocID);
Assert.IsFalse(tdocs.Advance(10) != DocIdSetIterator.NO_MORE_DOCS);
// exactly skipInterval documents and therefore with optimization
// with next
tdocs = TestUtil.Docs(Random, reader, tb.Field, new BytesRef(tb.Text()), MultiFields.GetLiveDocs(reader), null, DocsFlags.FREQS);
Assert.IsTrue(tdocs.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(10, tdocs.DocID);
Assert.AreEqual(4, tdocs.Freq);
Assert.IsTrue(tdocs.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(11, tdocs.DocID);
Assert.AreEqual(4, tdocs.Freq);
Assert.IsTrue(tdocs.Advance(12) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(12, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(15) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(15, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(24) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(24, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(25) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(25, tdocs.DocID);
Assert.IsFalse(tdocs.Advance(26) != DocIdSetIterator.NO_MORE_DOCS);
// without next
tdocs = TestUtil.Docs(Random, reader, tb.Field, new BytesRef(tb.Text()), MultiFields.GetLiveDocs(reader), null, DocsFlags.FREQS);
Assert.IsTrue(tdocs.Advance(5) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(10, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(15) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(15, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(24) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(24, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(25) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(25, tdocs.DocID);
Assert.IsFalse(tdocs.Advance(26) != DocIdSetIterator.NO_MORE_DOCS);
// much more than skipInterval documents and therefore with optimization
// with next
tdocs = TestUtil.Docs(Random, reader, tc.Field, new BytesRef(tc.Text()), MultiFields.GetLiveDocs(reader), null, DocsFlags.FREQS);
Assert.IsTrue(tdocs.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(26, tdocs.DocID);
Assert.AreEqual(4, tdocs.Freq);
Assert.IsTrue(tdocs.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(27, tdocs.DocID);
Assert.AreEqual(4, tdocs.Freq);
Assert.IsTrue(tdocs.Advance(28) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(28, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(40) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(40, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(57) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(57, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(74) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(74, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(75) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(75, tdocs.DocID);
Assert.IsFalse(tdocs.Advance(76) != DocIdSetIterator.NO_MORE_DOCS);
//without next
tdocs = TestUtil.Docs(Random, reader, tc.Field, new BytesRef(tc.Text()), MultiFields.GetLiveDocs(reader), null, 0);
Assert.IsTrue(tdocs.Advance(5) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(26, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(40) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(40, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(57) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(57, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(74) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(74, tdocs.DocID);
Assert.IsTrue(tdocs.Advance(75) != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(75, tdocs.DocID);
Assert.IsFalse(tdocs.Advance(76) != DocIdSetIterator.NO_MORE_DOCS);
reader.Dispose();
dir.Dispose();
}