public virtual string GetSenseKey(string lemma, string partOfSpeech, int sense)
{
IndexWord indexWord = mEngine.GetIndexWord(lemma, "noun");
if (indexWord == null)
{
return(null);
}
return(indexWord.SynsetOffsets[sense].ToString(System.Globalization.CultureInfo.InvariantCulture));
}
public virtual int GetSenseCount(string lemma, string pos)
{
IndexWord indexWord = mEngine.GetIndexWord(lemma, "noun");
if (indexWord == null)
{
return(0);
}
return(indexWord.SenseCount);
}
public virtual int getNumSenses(string lemma, string pos)
{
//try
//{
IndexWord indexWord = mEngine.GetIndexWord(lemma, "noun");
//IndexWord indexWord = dict.getIndexWord(POS.NOUN, lemma);
if (indexWord == null)
{
return(0);
}
//return indexWord.getSenseCount();
return(indexWord.SenseCount);
//}
//catch (JWNLException e)
//{
// return 0;
//}
//return 0;
}
public virtual string getSenseKey(string lemma, string pos, int sense)
{
//try
//{
IndexWord indexWord = mEngine.GetIndexWord(lemma, "noun");
//IndexWord indexWord = dict.getIndexWord(POS.NOUN, lemma);
if (indexWord == null)
{
return(null);
}
//return System.Convert.ToString(indexWord.getSynsetOffsets()[sense]);
return(indexWord.SynsetOffsets[sense].ToString(System.Globalization.CultureInfo.InvariantCulture));
//}
//catch (JWNLException e)
//{
// e.printStackTrace();
// return null;
//}
//return null;
}
public static IEnumerable WordBreaker3(SqlString text, SqlInt16 minlen, SqlInt16 repmax, SqlBoolean hashtaguser, SqlInt16 maxlen)
{
var words = new List <IndexWord>();
string _text = (string)text;
int count = _text.Length;
const string whitespace = " \t\n\r";
const string punctuation = ",.!?;:\"-/+(){}[]\\`?_^~<>";
short word_idx = 0; //word count
int ws = 0; // word start
int wc = 0; // word letter count
char ll = '\0'; // last letter processed
uint lc = 0; // same letter count
bool isnoisy = false;
bool isurl = false;
byte[] bsource = new byte[_text.Length * sizeof(char)];
System.Buffer.BlockCopy(_text.ToCharArray(), 0, bsource, 0, bsource.Length);
int i = 0;
char c;
byte bu = 0x00;
byte bl = 0x00;
bool isWhitespace = false;
bool isPunctuation = false;
if (count > 2)
{
// Check unicode signature, skip if any
if (bsource[0] == 0xFF || bsource[0] == 0xFE)
{
i = 1;
}
}
else
{
count = 0;
yield break;
}
// Read input letter by letter and put together into the word buffer
for (; i <= count; i++)
{
if (i == count)
{
c = '\0';
}
else
{
c = _text[i];
bu = bsource[2 * i];
bl = bsource[2 * i + 1];
isWhitespace = whitespace.IndexOf(c) != -1;
isPunctuation = punctuation.IndexOf(c) != -1;
}
if (isWhitespace || isPunctuation || c == '\0')
{
if (wc >= (int)minlen && !isnoisy && !isurl)
{
if (maxlen == 0 || wc <= (int)maxlen)
{
IndexWord tmp = new IndexWord();
tmp.Word = _text.Substring(ws, wc);
tmp.Pos = (short)ws;
tmp.ID = word_idx++;
yield return(tmp);
}
}
wc = 0;
isnoisy = false;
if (isWhitespace)
{
isurl = false;
}
if (c == '\0')
{
break;
}
}
else if (i < count - 4 &&
(_text[i] == 'h' || _text[i] == 'H') &&
(_text[i + 1] == 't' || _text[i + 1] == 'T') &&
(_text[i + 2] == 't' || _text[i + 2] == 'T') &&
(_text[i + 3] == 'p' || _text[i + 3] == 'P'))
{
isurl = true;
}
else
{
if (c == ll || c == ll + 0x20 || c == ll - 0x20) // do not distinguish upper and lower case, might cause some false negatives, but only a few
{
lc++;
}
else
{
ll = c;
lc = 1;
}
isnoisy |= lc > (int)repmax; // letter repeated more than trice
if (wc == 0 && hashtaguser)
{
if (c == '#' || c == '@')
{
goto nosymbols;
}
}
isnoisy |= (bl > 0x05); // Not latin greek or cyrillic
isnoisy |= (bl == 0x00 && bu < 0x41); // Symbols
isnoisy |= (bl == 0x00 && 0x7B <= bu && bu < 0xC0); // More symbols
isnoisy |= (bl == 0x03 && bu < 0x80); // More symbols
nosymbols:
// append to the word
if (wc == 0)
{
ws = i;
}
wc++;
}
}
yield break;
}
