/// <summary>
/// Returns a <see cref="StemmerOverrideMap"/> to be used with the <see cref="StemmerOverrideFilter"/> </summary>
/// <returns> a <see cref="StemmerOverrideMap"/> to be used with the <see cref="StemmerOverrideFilter"/> </returns>
/// <exception cref="IOException"> if an <see cref="IOException"/> occurs; </exception>
public virtual StemmerOverrideMap Build()
{
ByteSequenceOutputs outputs = ByteSequenceOutputs.Singleton;
Builder <BytesRef> builder = new Builder <BytesRef>(FST.INPUT_TYPE.BYTE4, outputs);
int[] sort = hash.Sort(BytesRef.UTF8SortedAsUnicodeComparer);
Int32sRef intsSpare = new Int32sRef();
int size = hash.Count;
for (int i = 0; i < size; i++)
{
int id = sort[i];
BytesRef bytesRef = hash.Get(id, spare);
UnicodeUtil.UTF8toUTF32(bytesRef, intsSpare);
builder.Add(intsSpare, new BytesRef(outputValues[id]));
}
return(new StemmerOverrideMap(builder.Finish(), ignoreCase));
}
/// <summary>
/// Returns an <seealso cref="StemmerOverrideMap"/> to be used with the <seealso cref="StemmerOverrideFilter"/> </summary>
/// <returns> an <seealso cref="StemmerOverrideMap"/> to be used with the <seealso cref="StemmerOverrideFilter"/> </returns>
/// <exception cref="IOException"> if an <seealso cref="IOException"/> occurs; </exception>
//JAVA TO C# CONVERTER WARNING: Method 'throws' clauses are not available in .NET:
//ORIGINAL LINE: public StemmerOverrideMap build() throws java.io.IOException
public virtual StemmerOverrideMap build()
{
ByteSequenceOutputs outputs = ByteSequenceOutputs.Singleton;
org.apache.lucene.util.fst.Builder <BytesRef> builder = new org.apache.lucene.util.fst.Builder <BytesRef>(FST.INPUT_TYPE.BYTE4, outputs);
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final int[] sort = hash.sort(org.apache.lucene.util.BytesRef.getUTF8SortedAsUnicodeComparator());
int[] sort = hash.sort(BytesRef.UTF8SortedAsUnicodeComparator);
IntsRef intsSpare = new IntsRef();
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final int size = hash.size();
int size = hash.size();
for (int i = 0; i < size; i++)
{
int id = sort[i];
BytesRef bytesRef = hash.get(id, spare);
UnicodeUtil.UTF8toUTF32(bytesRef, intsSpare);
builder.add(intsSpare, new BytesRef(outputValues[id]));
}
return(new StemmerOverrideMap(builder.finish(), ignoreCase));
}
/// <summary>
/// Builds an <see cref="SynonymMap"/> and returns it.
/// </summary>
public virtual SynonymMap Build()
{
ByteSequenceOutputs outputs = ByteSequenceOutputs.Singleton;
// TODO: are we using the best sharing options?
var builder = new Builder <BytesRef>(FST.INPUT_TYPE.BYTE4, outputs);
BytesRef scratch = new BytesRef(64);
ByteArrayDataOutput scratchOutput = new ByteArrayDataOutput();
HashSet <int?> dedupSet;
if (dedup)
{
dedupSet = new HashSet <int?>();
}
else
{
dedupSet = null;
}
var spare = new byte[5];
ICollection <CharsRef> keys = workingSet.Keys;
CharsRef[] sortedKeys = keys.ToArray();
#pragma warning disable 612, 618
System.Array.Sort(sortedKeys, CharsRef.UTF16SortedAsUTF8Comparer);
#pragma warning restore 612, 618
Int32sRef scratchIntsRef = new Int32sRef();
//System.out.println("fmap.build");
for (int keyIdx = 0; keyIdx < sortedKeys.Length; keyIdx++)
{
CharsRef input = sortedKeys[keyIdx];
MapEntry output = workingSet[input];
int numEntries = output.ords.Count;
// output size, assume the worst case
int estimatedSize = 5 + numEntries * 5; // numEntries + one ord for each entry
scratch.Grow(estimatedSize);
scratchOutput.Reset(scratch.Bytes, scratch.Offset, scratch.Bytes.Length);
Debug.Assert(scratch.Offset == 0);
// now write our output data:
int count = 0;
for (int i = 0; i < numEntries; i++)
{
if (dedupSet != null)
{
// box once
int?ent = output.ords[i];
if (dedupSet.Contains(ent))
{
continue;
}
dedupSet.Add(ent);
}
scratchOutput.WriteVInt32(output.ords[i]);
count++;
}
int pos = scratchOutput.Position;
scratchOutput.WriteVInt32(count << 1 | (output.includeOrig ? 0 : 1));
int pos2 = scratchOutput.Position;
int vIntLen = pos2 - pos;
// Move the count + includeOrig to the front of the byte[]:
Array.Copy(scratch.Bytes, pos, spare, 0, vIntLen);
Array.Copy(scratch.Bytes, 0, scratch.Bytes, vIntLen, pos);
Array.Copy(spare, 0, scratch.Bytes, 0, vIntLen);
if (dedupSet != null)
{
dedupSet.Clear();
}
scratch.Length = scratchOutput.Position - scratch.Offset;
//System.out.println(" add input=" + input + " output=" + scratch + " offset=" + scratch.offset + " length=" + scratch.length + " count=" + count);
builder.Add(Lucene.Net.Util.Fst.Util.ToUTF32(input.ToString(), scratchIntsRef), BytesRef.DeepCopyOf(scratch));
}
FST <BytesRef> fst = builder.Finish();
return(new SynonymMap(fst, words, maxHorizontalContext));
}
/// <summary>
/// Builds an <seealso cref="SynonymMap"/> and returns it.
/// </summary>
//JAVA TO C# CONVERTER WARNING: Method 'throws' clauses are not available in .NET:
//ORIGINAL LINE: public SynonymMap build() throws java.io.IOException
public virtual SynonymMap build()
{
ByteSequenceOutputs outputs = ByteSequenceOutputs.Singleton;
// TODO: are we using the best sharing options?
org.apache.lucene.util.fst.Builder <BytesRef> builder = new org.apache.lucene.util.fst.Builder <BytesRef>(FST.INPUT_TYPE.BYTE4, outputs);
BytesRef scratch = new BytesRef(64);
ByteArrayDataOutput scratchOutput = new ByteArrayDataOutput();
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final java.util.Set<Integer> dedupSet;
HashSet <int?> dedupSet;
if (dedup)
{
dedupSet = new HashSet <>();
}
else
{
dedupSet = null;
}
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final byte[] spare = new byte[5];
sbyte[] spare = new sbyte[5];
Dictionary <CharsRef, MapEntry> .KeyCollection keys = workingSet.Keys;
CharsRef[] sortedKeys = keys.toArray(new CharsRef[keys.size()]);
Arrays.sort(sortedKeys, CharsRef.UTF16SortedAsUTF8Comparator);
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final org.apache.lucene.util.IntsRef scratchIntsRef = new org.apache.lucene.util.IntsRef();
IntsRef scratchIntsRef = new IntsRef();
//System.out.println("fmap.build");
for (int keyIdx = 0; keyIdx < sortedKeys.Length; keyIdx++)
{
CharsRef input = sortedKeys[keyIdx];
MapEntry output = workingSet[input];
int numEntries = output.ords.Count;
// output size, assume the worst case
int estimatedSize = 5 + numEntries * 5; // numEntries + one ord for each entry
scratch.grow(estimatedSize);
scratchOutput.reset(scratch.bytes, scratch.offset, scratch.bytes.length);
Debug.Assert(scratch.offset == 0);
// now write our output data:
int count = 0;
for (int i = 0; i < numEntries; i++)
{
if (dedupSet != null)
{
// box once
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final Integer ent = output.ords.get(i);
int?ent = output.ords[i];
if (dedupSet.Contains(ent))
{
continue;
}
dedupSet.Add(ent);
}
scratchOutput.writeVInt(output.ords[i]);
count++;
}
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final int pos = scratchOutput.getPosition();
int pos = scratchOutput.Position;
scratchOutput.writeVInt(count << 1 | (output.includeOrig ? 0 : 1));
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final int pos2 = scratchOutput.getPosition();
int pos2 = scratchOutput.Position;
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final int vIntLen = pos2-pos;
int vIntLen = pos2 - pos;
// Move the count + includeOrig to the front of the byte[]:
Array.Copy(scratch.bytes, pos, spare, 0, vIntLen);
Array.Copy(scratch.bytes, 0, scratch.bytes, vIntLen, pos);
Array.Copy(spare, 0, scratch.bytes, 0, vIntLen);
if (dedupSet != null)
{
dedupSet.Clear();
}
scratch.length = scratchOutput.Position - scratch.offset;
//System.out.println(" add input=" + input + " output=" + scratch + " offset=" + scratch.offset + " length=" + scratch.length + " count=" + count);
builder.add(Util.toUTF32(input, scratchIntsRef), BytesRef.deepCopyOf(scratch));
}
FST <BytesRef> fst = builder.finish();
return(new SynonymMap(fst, words, maxHorizontalContext));
}
