public override Synset[] GetSynsets(string lemma, string partOfSpeech)
{
List <Synset> synsets = new List <Synset>();
IndexWord indexWord = GetIndexWord(lemma, partOfSpeech);
if (indexWord != null)
{
foreach (int synsetOffset in indexWord.SynsetOffsets)
{
Synset synset = CreateSynset(partOfSpeech, synsetOffset);
synsets.Add(synset);
}
}
return(synsets.ToArray());
}
public override Synset[] GetSynsets(string lemma)
{
var synsets = new List <Synset>();
foreach (string partOfSpeech in _dataFileDictionary.Keys)
{
IndexWord indexWord = GetIndexWord(lemma, partOfSpeech);
if (indexWord != null)
{
foreach (int synsetOffset in indexWord.SynsetOffsets)
{
Synset synset = CreateSynset(partOfSpeech, synsetOffset);
synsets.Add(synset);
}
}
}
return(synsets.ToArray());
}
public override Synset GetSynset(string lemma, string partOfSpeech, int senseNumber)
{
if (senseNumber < 1)
{
throw new ArgumentOutOfRangeException("senseNumber", senseNumber, "cannot be less than 1");
}
IndexWord indexWord = GetIndexWord(lemma, partOfSpeech);
if (indexWord != null)
{
if (senseNumber > (indexWord.SynsetOffsets.Length + 1))
{
return(null);
}
Synset synset = CreateSynset(partOfSpeech, indexWord.SynsetOffsets[senseNumber - 1]);
return(synset);
}
return(null);
}
protected internal override Synset CreateSynset(string partOfSpeech, int synsetOffset)
{
StreamReader dataFile = mDataFileDictionary[partOfSpeech].DataFile;
dataFile.DiscardBufferedData();
dataFile.BaseStream.Seek(synsetOffset, SeekOrigin.Begin);
string record = dataFile.ReadLine();
Tokenizer tokenizer = new Tokenizer(record);
int offset = int.Parse(tokenizer.NextToken());
string lexicographerFile = mLexicographerFiles[int.Parse(tokenizer.NextToken())];
string synsetType = tokenizer.NextToken();
int wordCount = int.Parse(tokenizer.NextToken(), System.Globalization.NumberStyles.HexNumber);
string[] words = new string[wordCount];
for (int iCurrentWord = 0; iCurrentWord < wordCount; iCurrentWord++)
{
words[iCurrentWord] = tokenizer.NextToken().Replace("_", " ");
int uniqueID = int.Parse(tokenizer.NextToken(), System.Globalization.NumberStyles.HexNumber);
}
int relationCount = int.Parse(tokenizer.NextToken());
Relation[] relations = new Relation[relationCount];
for (int currentRelation = 0; currentRelation < relationCount; currentRelation++)
{
string relationTypeKey = tokenizer.NextToken();
// if (fpos.name=="adj" && sstype==AdjSynSetType.DontKnow)
// {
// if (ptrs[j].ptp.mnemonic=="ANTPTR")
// sstype = AdjSynSetType.DirectAnt;
// else if (ptrs[j].ptp.mnemonic=="PERTPTR")
// sstype = AdjSynSetType.Pertainym;
// }
int targetSynsetOffset = int.Parse(tokenizer.NextToken());
string targetPartOfSpeech = tokenizer.NextToken();
switch (targetPartOfSpeech)
{
case "n":
targetPartOfSpeech = "noun";
break;
case "v":
targetPartOfSpeech = "verb";
break;
case "a":
case "s":
targetPartOfSpeech = "adjective";
break;
case "r":
targetPartOfSpeech = "adverb";
break;
}
int sourceTarget = int.Parse(tokenizer.NextToken(), System.Globalization.NumberStyles.HexNumber);
if (sourceTarget == 0)
{
relations[currentRelation] = new Relation(this, (RelationType)mRelationTypeDictionary[relationTypeKey], targetSynsetOffset, targetPartOfSpeech);
}
else
{
int sourceWord = sourceTarget >> 8;
int targetWord = sourceTarget & 0xff;
relations[currentRelation] = new Relation(this, (RelationType)mRelationTypeDictionary[relationTypeKey], targetSynsetOffset, targetPartOfSpeech, sourceWord, targetWord);
}
}
string frameData = tokenizer.NextToken();
if (frameData != "|")
{
int frameCount = int.Parse(frameData);
for (int currentFrame = 0; currentFrame < frameCount; currentFrame++)
{
frameData = tokenizer.NextToken(); // +
int frameNumber = int.Parse(tokenizer.NextToken());
int wordID = int.Parse(tokenizer.NextToken(), System.Globalization.NumberStyles.HexNumber);
}
frameData = tokenizer.NextToken();
}
string gloss = record.Substring(record.IndexOf('|') + 1);
Synset synset = new Synset(synsetOffset, gloss, words, lexicographerFile, relations);
return(synset);
}
//private void getParents(Synset synset, System.Collections.IList parents)
//{
//Pointer[] pointers = synset.getPointers();
//for (int pi = 0, pn = pointers.length; pi < pn; pi++)
//{
// if (pointers[pi].getType() == PointerType.HYPERNYM)
// {
// Synset parent = pointers[pi].getTargetSynset();
// parents.Add(System.Convert.ToString(parent.getOffset()));
// getParents(parent, parents);
// }
//}
//}
private void getParents(Synset currentSynset, List<string> parentOffsets)
{
for (int currentRelation = 0;currentRelation < currentSynset.RelationCount;currentRelation++)
{
Relation relation = currentSynset.GetRelation(currentRelation);
if (relation.SynsetRelationType.Name == "Hypernym")
{
Synset parentSynset = relation.TargetSynset;
parentOffsets.Add(parentSynset.Offset.ToString(System.Globalization.CultureInfo.InvariantCulture));
getParents(parentSynset, parentOffsets);
}
}
}
protected internal override Synset CreateSynset(string partOfSpeech, int synsetOffset)
{
StreamReader dataFile = mDataFileDictionary[partOfSpeech].DataFile;
dataFile.DiscardBufferedData();
dataFile.BaseStream.Seek(synsetOffset, SeekOrigin.Begin);
string record = dataFile.ReadLine();
Tokenizer tokenizer = new Tokenizer(record);
int offset = int.Parse(tokenizer.NextToken());
string lexicographerFile = mLexicographerFiles[int.Parse(tokenizer.NextToken())];
string synsetType = tokenizer.NextToken();
int wordCount = int.Parse(tokenizer.NextToken(), System.Globalization.NumberStyles.HexNumber);
string[] words = new string[wordCount];
for (int iCurrentWord = 0; iCurrentWord < wordCount; iCurrentWord++)
{
words[iCurrentWord] = tokenizer.NextToken().Replace("_", " ");
int uniqueID = int.Parse(tokenizer.NextToken(), System.Globalization.NumberStyles.HexNumber);
}
int relationCount = int.Parse(tokenizer.NextToken());
Relation[] relations = new Relation[relationCount];
for (int currentRelation = 0; currentRelation < relationCount; currentRelation++)
{
string relationTypeKey = tokenizer.NextToken();
// if (fpos.name=="adj" && sstype==AdjSynSetType.DontKnow)
// {
// if (ptrs[j].ptp.mnemonic=="ANTPTR")
// sstype = AdjSynSetType.DirectAnt;
// else if (ptrs[j].ptp.mnemonic=="PERTPTR")
// sstype = AdjSynSetType.Pertainym;
// }
int targetSynsetOffset = int.Parse(tokenizer.NextToken());
string targetPartOfSpeech = tokenizer.NextToken();
switch (targetPartOfSpeech)
{
case "n":
targetPartOfSpeech = "noun";
break;
case "v":
targetPartOfSpeech = "verb";
break;
case "a":
case "s":
targetPartOfSpeech = "adjective";
break;
case "r":
targetPartOfSpeech = "adverb";
break;
}
int sourceTarget = int.Parse(tokenizer.NextToken(), System.Globalization.NumberStyles.HexNumber);
if (sourceTarget == 0)
{
relations[currentRelation] = new Relation(this, (RelationType)mRelationTypeDictionary[relationTypeKey], targetSynsetOffset, targetPartOfSpeech);
}
else
{
int sourceWord = sourceTarget >> 8;
int targetWord = sourceTarget & 0xff;
relations[currentRelation] = new Relation(this, (RelationType)mRelationTypeDictionary[relationTypeKey], targetSynsetOffset, targetPartOfSpeech, sourceWord, targetWord);
}
}
string frameData = tokenizer.NextToken();
if (frameData != "|")
{
int frameCount = int.Parse(frameData);
for (int currentFrame = 0; currentFrame < frameCount; currentFrame++)
{
frameData = tokenizer.NextToken(); // +
int frameNumber = int.Parse(tokenizer.NextToken());
int wordID = int.Parse(tokenizer.NextToken(), System.Globalization.NumberStyles.HexNumber);
}
frameData = tokenizer.NextToken();
}
string gloss = record.Substring(record.IndexOf('|') + 1);
Synset synset = new Synset(synsetOffset, gloss, words, lexicographerFile, relations);
return synset;
}
