protected internal override string[] GetExceptionForms(string lemma, string partOfSpeech)
{
string line = BinarySearch(lemma, mDataFileDictionary[partOfSpeech].ExceptionFile);
if (line != null)
{
List<string> exceptionForms = new List<string>();
Tokenizer tokenizer = new Tokenizer(line);
string skipWord = tokenizer.NextToken();
string word = tokenizer.NextToken();
while (word != null)
{
exceptionForms.Add(word);
word = tokenizer.NextToken();
}
return exceptionForms.ToArray();
}
return mEmpty;
}
private IndexWord CreateIndexWord(string partOfSpeech, string line)
{
Tokenizer tokenizer = new Tokenizer(line);
string word = tokenizer.NextToken().Replace('_', ' ');
string redundantPartOfSpeech = tokenizer.NextToken();
int senseCount = int.Parse(tokenizer.NextToken());
int relationTypeCount = int.Parse(tokenizer.NextToken());
string[] relationTypes = null;
if (relationTypeCount > 0)
{
relationTypes = new string[relationTypeCount];
for (int currentRelationType = 0; currentRelationType < relationTypeCount; currentRelationType++)
{
relationTypes[currentRelationType] = tokenizer.NextToken();
}
}
int redundantSenseCount = int.Parse(tokenizer.NextToken());
int tagSenseCount = int.Parse(tokenizer.NextToken());
int[] synsetOffsets = null;
if (senseCount > 0)
{
synsetOffsets = new int[senseCount];
for (int currentOffset = 0; currentOffset < senseCount; currentOffset++)
{
synsetOffsets[currentOffset] = int.Parse(tokenizer.NextToken());
}
}
return new IndexWord(word, partOfSpeech, relationTypes, synsetOffsets, tagSenseCount);
}
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;
}