/// <summary>
/// Sort a single partition in-memory. </summary>
private FileInfo SortPartition(int len) // LUCENENET NOTE: made private, since protected is not valid in a sealed class
{
var data = this.buffer;
FileInfo tempFile = FileSupport.CreateTempFile("sort", "partition", DefaultTempDir());
long start = Environment.TickCount;
sortInfo.SortTime += (Environment.TickCount - start);
using (var @out = new ByteSequencesWriter(tempFile))
{
BytesRef spare;
IBytesRefIterator iter = buffer.GetIterator(comparer);
while ((spare = iter.Next()) != null)
{
Debug.Assert(spare.Length <= ushort.MaxValue);
@out.Write(spare);
}
}
// Clean up the buffer for the next partition.
data.Clear();
return(tempFile);
}
private void Check(IBytesRefSorter sorter)
{
for (int i = 0; i < 100; i++)
{
byte[] current = new byte[Random.nextInt(256)];
Random.NextBytes(current);
sorter.Add(new BytesRef(current));
}
// Create two iterators and check that they're aligned with each other.
IBytesRefIterator i1 = sorter.GetIterator();
IBytesRefIterator i2 = sorter.GetIterator();
// Verify sorter contract.
try
{
sorter.Add(new BytesRef(new byte[1]));
fail("expected contract violation.");
}
catch (InvalidOperationException /*e*/)
{
// Expected.
}
BytesRef spare1;
BytesRef spare2;
while ((spare1 = i1.Next()) != null && (spare2 = i2.Next()) != null)
{
assertEquals(spare1, spare2);
}
assertNull(i1.Next());
assertNull(i2.Next());
}
/// <summary>
/// Builds the final automaton from a list of entries.
/// </summary>
private FST <object> BuildAutomaton(IBytesRefSorter sorter)
{
// Build the automaton.
Outputs <object> outputs = NoOutputs.Singleton;
object empty = outputs.NoOutput;
Builder <object> builder = new Builder <object>(FST.INPUT_TYPE.BYTE1, 0, 0, true, true, shareMaxTailLength, outputs, null, false, PackedInt32s.DEFAULT, true, 15);
BytesRef scratch = new BytesRef();
BytesRef entry;
Int32sRef scratchIntsRef = new Int32sRef();
int count = 0;
IBytesRefIterator iter = sorter.GetIterator();
while ((entry = iter.Next()) != null)
{
count++;
if (scratch.CompareTo(entry) != 0)
{
builder.Add(Util.Fst.Util.ToInt32sRef(entry, scratchIntsRef), empty);
scratch.CopyBytes(entry);
}
}
return(count == 0 ? null : builder.Finish());
}
public void TestFieldContents_2()
{
IBytesRefEnumerator it;
try
{
indexReader = DirectoryReader.Open(store);
ld = new LuceneDictionary(indexReader, "contents");
it = ld.GetEntryEnumerator();
// just iterate through words
assertTrue(it.MoveNext());
assertEquals("First element isn't correct", "Jerry", it.Current.Utf8ToString());
assertTrue(it.MoveNext());
assertEquals("Second element isn't correct", "Tom", it.Current.Utf8ToString());
assertFalse("Nonexistent element is really null", it.MoveNext());
}
finally
{
if (indexReader != null)
{
indexReader.Dispose();
}
}
}
public void TestFieldContents_1()
{
try
{
indexReader = DirectoryReader.Open(store);
ld = new LuceneDictionary(indexReader, "contents");
it = ld.GetEntryIterator();
assertNotNull("First element doesn't exist.", spare = it.Next());
assertTrue("First element isn't correct", spare.Utf8ToString().equals("Jerry"));
assertNotNull("Second element doesn't exist.", spare = it.Next());
assertTrue("Second element isn't correct", spare.Utf8ToString().equals("Tom"));
assertNull("More elements than expected", it.Next());
ld = new LuceneDictionary(indexReader, "contents");
it = ld.GetEntryIterator();
int counter = 2;
while (it.Next() != null)
{
counter--;
}
assertTrue("Number of words incorrect", counter == 0);
}
finally
{
if (indexReader != null)
{
indexReader.Dispose();
}
}
}
public void TestEmpty()
{
Directory dir = NewDirectory();
IndexWriter writer = new IndexWriter(dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random)));
writer.Commit();
writer.Dispose();
IndexReader ir = DirectoryReader.Open(dir);
IDictionary dictionary = new HighFrequencyDictionary(ir, "bogus", 0.1f);
IBytesRefIterator tf = dictionary.GetEntryIterator();
assertNull(tf.Comparer);
assertNull(tf.Next());
dir.Dispose();
}
public virtual void TestAppend()
{
Random random = Random;
BytesRefArray list = new BytesRefArray(Util.Counter.NewCounter());
IList <string> stringList = new List <string>();
for (int j = 0; j < 2; j++)
{
if (j > 0 && random.NextBoolean())
{
list.Clear();
stringList.Clear();
}
int entries = AtLeast(500);
BytesRef spare = new BytesRef();
int initSize = list.Length;
for (int i = 0; i < entries; i++)
{
string randomRealisticUnicodeString = TestUtil.RandomRealisticUnicodeString(random);
spare.CopyChars(randomRealisticUnicodeString);
Assert.AreEqual(i + initSize, list.Append(spare));
stringList.Add(randomRealisticUnicodeString);
}
for (int i = 0; i < entries; i++)
{
Assert.IsNotNull(list.Get(spare, i));
Assert.AreEqual(stringList[i], spare.Utf8ToString(), "entry " + i + " doesn't match");
}
// check random
for (int i = 0; i < entries; i++)
{
int e = random.Next(entries);
Assert.IsNotNull(list.Get(spare, e));
Assert.AreEqual(stringList[e], spare.Utf8ToString(), "entry " + i + " doesn't match");
}
for (int i = 0; i < 2; i++)
{
IBytesRefIterator iterator = list.GetIterator();
foreach (string @string in stringList)
{
Assert.AreEqual(@string, iterator.Next().Utf8ToString());
}
}
}
}
public void TestFieldNonExistent()
{
try
{
indexReader = DirectoryReader.Open(store);
ld = new LuceneDictionary(indexReader, "nonexistent_field");
it = ld.GetEntryIterator();
assertNull("More elements than expected", spare = it.Next());
}
finally
{
if (indexReader != null)
{
indexReader.Dispose();
}
}
}
public void TestFieldAaa()
{
try
{
indexReader = DirectoryReader.Open(store);
ld = new LuceneDictionary(indexReader, "aaa");
it = ld.GetEntryIterator();
assertNotNull("First element doesn't exist.", spare = it.Next());
assertTrue("First element isn't correct", spare.Utf8ToString().equals("foo"));
assertNull("More elements than expected", it.Next());
}
finally
{
if (indexReader != null)
{
indexReader.Dispose();
}
}
}
public virtual void TestSort()
{
Random random = Random;
BytesRefArray list = new BytesRefArray(Util.Counter.NewCounter());
List <string> stringList = new List <string>();
for (int j = 0; j < 2; j++)
{
if (j > 0 && random.NextBoolean())
{
list.Clear();
stringList.Clear();
}
int entries = AtLeast(500);
BytesRef spare = new BytesRef();
int initSize = list.Length;
for (int i = 0; i < entries; i++)
{
string randomRealisticUnicodeString = TestUtil.RandomRealisticUnicodeString(random);
spare.CopyChars(randomRealisticUnicodeString);
Assert.AreEqual(initSize + i, list.Append(spare));
stringList.Add(randomRealisticUnicodeString);
}
// LUCENENET NOTE: Must sort using ArrayUtil.GetNaturalComparator<T>()
// to ensure culture isn't taken into consideration during the sort,
// which will match the sort order of BytesRef.UTF8SortedAsUTF16Comparer.
CollectionUtil.TimSort(stringList);
#pragma warning disable 612, 618
IBytesRefIterator iter = list.GetIterator(BytesRef.UTF8SortedAsUTF16Comparer);
#pragma warning restore 612, 618
int a = 0;
while ((spare = iter.Next()) != null)
{
Assert.AreEqual(stringList[a], spare.Utf8ToString(), "entry " + a + " doesn't match");
a++;
}
Assert.IsNull(iter.Next());
Assert.AreEqual(a, stringList.Count);
}
}
public void TestFieldNonExistent()
{
IBytesRefEnumerator it;
try
{
indexReader = DirectoryReader.Open(store);
ld = new LuceneDictionary(indexReader, "nonexistent_field");
it = ld.GetEntryEnumerator();
assertFalse("More elements than expected", it.MoveNext());
}
finally
{
if (indexReader != null)
{
indexReader.Dispose();
}
}
}
public void TestFieldAaa()
{
IBytesRefEnumerator it;
try
{
indexReader = DirectoryReader.Open(store);
ld = new LuceneDictionary(indexReader, "aaa");
it = ld.GetEntryEnumerator();
assertTrue("First element doesn't exist.", it.MoveNext());
assertTrue("First element isn't correct", it.Current.Utf8ToString().Equals("foo", StringComparison.Ordinal));
assertFalse("More elements than expected", it.MoveNext());
}
finally
{
if (indexReader != null)
{
indexReader.Dispose();
}
}
}
public void TestFieldContents_1()
{
IBytesRefEnumerator it;
try
{
indexReader = DirectoryReader.Open(store);
ld = new LuceneDictionary(indexReader, "contents");
it = ld.GetEntryEnumerator();
assertTrue("First element doesn't exist.", it.MoveNext());
assertTrue("First element isn't correct", it.Current.Utf8ToString().Equals("Jerry", StringComparison.Ordinal));
assertTrue("Second element doesn't exist.", it.MoveNext());
assertTrue("Second element isn't correct", it.Current.Utf8ToString().Equals("Tom", StringComparison.Ordinal));
assertFalse("More elements than expected", it.MoveNext());
ld = new LuceneDictionary(indexReader, "contents");
it = ld.GetEntryEnumerator();
int counter = 2;
while (it.MoveNext())
{
counter--;
}
assertTrue("Number of words incorrect", counter == 0);
}
finally
{
if (indexReader != null)
{
indexReader.Dispose();
}
}
}
/// <summary>
/// Creates a new wrapper, wrapping the specified iterator and
/// specifying a weight value of <c>1</c> for all terms
/// and nullifies associated payloads.
/// </summary>
public InputIteratorWrapper(IBytesRefIterator wrapped)
{
this.wrapped = wrapped;
}
/// <summary>
/// Creates a new wrapper, wrapping the specified iterator and
/// specifying a weight value of <code>1</code> for all terms.
/// </summary>
public TermFreqIteratorWrapper(IBytesRefIterator wrapped)
{
this.wrapped = wrapped;
}
/// <summary>
/// Indexes the data from the given <see cref="IDictionary"/>. </summary>
/// <param name="dict"> Dictionary to index </param>
/// <param name="config"> <see cref="IndexWriterConfig"/> to use </param>
/// <param name="fullMerge"> whether or not the spellcheck index should be fully merged </param>
/// <exception cref="ObjectDisposedException"> if the <see cref="SpellChecker"/> is already disposed </exception>
/// <exception cref="System.IO.IOException"> If there is a low-level I/O error. </exception>
public void IndexDictionary(IDictionary dict, IndexWriterConfig config, bool fullMerge)
{
lock (modifyCurrentIndexLock)
{
EnsureOpen();
Directory dir = this.spellIndex;
using (var writer = new IndexWriter(dir, config))
{
IndexSearcher indexSearcher = ObtainSearcher();
IList <TermsEnum> termsEnums = new List <TermsEnum>();
IndexReader reader = searcher.IndexReader;
if (reader.MaxDoc > 0)
{
foreach (AtomicReaderContext ctx in reader.Leaves)
{
Terms terms = ctx.AtomicReader.GetTerms(F_WORD);
if (terms != null)
{
termsEnums.Add(terms.GetIterator(null));
}
}
}
bool isEmpty = termsEnums.Count == 0;
try
{
IBytesRefIterator iter = dict.GetEntryIterator();
BytesRef currentTerm;
while ((currentTerm = iter.Next()) != null)
{
string word = currentTerm.Utf8ToString();
int len = word.Length;
if (len < 3)
{
continue; // too short we bail but "too long" is fine...
}
if (!isEmpty)
{
foreach (TermsEnum te in termsEnums)
{
if (te.SeekExact(currentTerm))
{
goto termsContinue;
}
}
}
// ok index the word
var doc = CreateDocument(word, GetMin(len), GetMax(len));
writer.AddDocument(doc);
termsContinue:
;
}
}
finally
{
ReleaseSearcher(indexSearcher);
}
if (fullMerge)
{
writer.ForceMerge(1);
}
}
// TODO: this isn't that great, maybe in the future SpellChecker should take
// IWC in its ctor / keep its writer open?
// also re-open the spell index to see our own changes when the next suggestion
// is fetched:
SwapSearcher(dir);
}
}
