/// <summary>
/// Mainly remove the definite article </summary>
/// <param name="s"> input buffer </param>
/// <param name="len"> length of input buffer </param>
/// <returns> new stemmed length </returns>
private int RemoveArticle(char[] s, int len)
{
if (len > 6 && StemmerUtil.EndsWith(s, len, "ият"))
{
return(len - 3);
}
if (len > 5)
{
if (StemmerUtil.EndsWith(s, len, "ът") || StemmerUtil.EndsWith(s, len, "то") || StemmerUtil.EndsWith(s, len, "те") || StemmerUtil.EndsWith(s, len, "та") || StemmerUtil.EndsWith(s, len, "ия"))
{
return(len - 2);
}
}
if (len > 4 && StemmerUtil.EndsWith(s, len, "ят"))
{
return(len - 2);
}
return(len);
}
private int Step2(char[] s, int len)
{
if (len > 5)
{
if (StemmerUtil.EndsWith(s, len, "lla") || StemmerUtil.EndsWith(s, len, "tse") || StemmerUtil.EndsWith(s, len, "sti"))
{
return(len - 3);
}
if (StemmerUtil.EndsWith(s, len, "ni"))
{
return(len - 2);
}
if (StemmerUtil.EndsWith(s, len, "aa"))
{
return(len - 1); // aa -> a
}
}
return(len);
}
private int Rule16(char[] s, int len)
{
bool removed = false;
if (len > 4 && StemmerUtil.EndsWith(s, len, "ησου"))
{
len -= 4;
removed = true;
}
else if (len > 3 && (StemmerUtil.EndsWith(s, len, "ησε") || StemmerUtil.EndsWith(s, len, "ησα")))
{
len -= 3;
removed = true;
}
if (removed && exc16.Contains(s, 0, len))
{
len += 2; // add back -ησ
}
return(len);
}
private int Rule15(char[] s, int len)
{
bool removed = false;
if (len > 4 && StemmerUtil.EndsWith(s, len, "αγεσ"))
{
len -= 4;
removed = true;
}
else if (len > 3 && (StemmerUtil.EndsWith(s, len, "αγα") || StemmerUtil.EndsWith(s, len, "αγε")))
{
len -= 3;
removed = true;
}
if (removed)
{
bool cond1 = exc15a.Contains(s, 0, len) ||
StemmerUtil.EndsWith(s, len, "οφ") ||
StemmerUtil.EndsWith(s, len, "πελ") ||
StemmerUtil.EndsWith(s, len, "χορτ") ||
StemmerUtil.EndsWith(s, len, "λλ") ||
StemmerUtil.EndsWith(s, len, "σφ") ||
StemmerUtil.EndsWith(s, len, "ρπ") ||
StemmerUtil.EndsWith(s, len, "φρ") ||
StemmerUtil.EndsWith(s, len, "πρ") ||
StemmerUtil.EndsWith(s, len, "λοχ") ||
StemmerUtil.EndsWith(s, len, "σμην");
bool cond2 = exc15b.Contains(s, 0, len) || StemmerUtil.EndsWith(s, len, "κολλ");
if (cond1 && !cond2)
{
len += 2; // add back -αγ
}
}
return(len);
}
#pragma warning restore 612, 618
private int Rule12(char[] s, int len)
{
if (len > 5 && StemmerUtil.EndsWith(s, len, "ιεστε"))
{
len -= 5;
if (exc12a.Contains(s, 0, len))
{
len += 4; // add back -ιεστ
}
}
if (len > 4 && StemmerUtil.EndsWith(s, len, "εστε"))
{
len -= 4;
if (exc12b.Contains(s, 0, len))
{
len += 3; // add back -εστ
}
}
return(len);
}
private int Rule1(char[] s, int len)
{
if (len > 4 && (StemmerUtil.EndsWith(s, len, "αδεσ") ||
StemmerUtil.EndsWith(s, len, "αδων")))
{
len -= 4;
if (!(StemmerUtil.EndsWith(s, len, "οκ") ||
StemmerUtil.EndsWith(s, len, "μαμ") ||
StemmerUtil.EndsWith(s, len, "μαν") ||
StemmerUtil.EndsWith(s, len, "μπαμπ") ||
StemmerUtil.EndsWith(s, len, "πατερ") ||
StemmerUtil.EndsWith(s, len, "γιαγι") ||
StemmerUtil.EndsWith(s, len, "νταντ") ||
StemmerUtil.EndsWith(s, len, "κυρ") ||
StemmerUtil.EndsWith(s, len, "θει") ||
StemmerUtil.EndsWith(s, len, "πεθερ")))
{
len += 2; // add back -αδ
}
}
return(len);
}
private int Rule13(char[] s, int len)
{
if (len > 6 && StemmerUtil.EndsWith(s, len, "ηθηκεσ"))
{
len -= 6;
}
else if (len > 5 && (StemmerUtil.EndsWith(s, len, "ηθηκα") || StemmerUtil.EndsWith(s, len, "ηθηκε")))
{
len -= 5;
}
bool removed = false;
if (len > 4 && StemmerUtil.EndsWith(s, len, "ηκεσ"))
{
len -= 4;
removed = true;
}
else if (len > 3 && (StemmerUtil.EndsWith(s, len, "ηκα") || StemmerUtil.EndsWith(s, len, "ηκε")))
{
len -= 3;
removed = true;
}
if (removed && (exc13.Contains(s, 0, len) ||
StemmerUtil.EndsWith(s, len, "σκωλ") ||
StemmerUtil.EndsWith(s, len, "σκουλ") ||
StemmerUtil.EndsWith(s, len, "ναρθ") ||
StemmerUtil.EndsWith(s, len, "σφ") ||
StemmerUtil.EndsWith(s, len, "οθ") ||
StemmerUtil.EndsWith(s, len, "πιθ")))
{
len += 2; // add back the -ηκ
}
return(len);
}
private int Rule19(char[] s, int len)
{
bool removed = false;
if (len > 6 && (StemmerUtil.EndsWith(s, len, "ησουμε") || StemmerUtil.EndsWith(s, len, "ηθουμε")))
{
len -= 6;
removed = true;
}
else if (len > 4 && StemmerUtil.EndsWith(s, len, "ουμε"))
{
len -= 4;
removed = true;
}
if (removed && exc19.Contains(s, 0, len))
{
len += 3;
s[len - 3] = 'ο';
s[len - 2] = 'υ';
s[len - 1] = 'μ';
}
return(len);
}
private int Rule14(char[] s, int len)
{
bool removed = false;
if (len > 5 && StemmerUtil.EndsWith(s, len, "ουσεσ"))
{
len -= 5;
removed = true;
}
else if (len > 4 && (StemmerUtil.EndsWith(s, len, "ουσα") ||
StemmerUtil.EndsWith(s, len, "ουσε")))
{
len -= 4;
removed = true;
}
if (removed && (exc14.Contains(s, 0, len) ||
EndsWithVowel(s, len) ||
StemmerUtil.EndsWith(s, len, "ποδαρ") ||
StemmerUtil.EndsWith(s, len, "βλεπ") ||
StemmerUtil.EndsWith(s, len, "πανταχ") ||
StemmerUtil.EndsWith(s, len, "φρυδ") ||
StemmerUtil.EndsWith(s, len, "μαντιλ") ||
StemmerUtil.EndsWith(s, len, "μαλλ") ||
StemmerUtil.EndsWith(s, len, "κυματ") ||
StemmerUtil.EndsWith(s, len, "λαχ") ||
StemmerUtil.EndsWith(s, len, "ληγ") ||
StemmerUtil.EndsWith(s, len, "φαγ") ||
StemmerUtil.EndsWith(s, len, "ομ") ||
StemmerUtil.EndsWith(s, len, "πρωτ")))
{
len += 3; // add back -ουσ
}
return(len);
}
private int Norm2(char[] s, int len)
{
if (len > 8)
{
if (s[len - 1] == 'e' || s[len - 1] == 'o' || s[len - 1] == 'u')
{
len--;
}
}
if (len > 4)
{
if (s[len - 1] == 'i')
{
len--;
}
if (len > 4)
{
char ch = s[0];
for (int i = 1; i < len; i++)
{
if (s[i] == ch && (ch == 'k' || ch == 'p' || ch == 't'))
{
len = StemmerUtil.Delete(s, i--, len);
}
else
{
ch = s[i];
}
}
}
}
return(len);
}
/// <returns> new valid length of the string after applying the entire step. </returns>
public virtual int Apply(char[] s, int len)
{
if (len < m_min)
{
return(len);
}
if (m_suffixes != null)
{
bool found = false;
for (int i = 0; i < m_suffixes.Length; i++)
{
if (StemmerUtil.EndsWith(s, len, m_suffixes[i]))
{
found = true;
break;
}
}
if (!found)
{
return(len);
}
}
for (int i = 0; i < m_rules.Length; i++)
{
if (m_rules[i].Matches(s, len))
{
return(m_rules[i].Replace(s, len));
}
}
return(len);
}
private int RemoveCase(char[] s, int len)
{
if (len > 6 && (StemmerUtil.EndsWith(s, len, "иями") || StemmerUtil.EndsWith(s, len, "оями")))
{
return(len - 4);
}
if (len > 5 && (StemmerUtil.EndsWith(s, len, "иям") || StemmerUtil.EndsWith(s, len, "иях") || StemmerUtil.EndsWith(s, len, "оях") || StemmerUtil.EndsWith(s, len, "ями") || StemmerUtil.EndsWith(s, len, "оям") || StemmerUtil.EndsWith(s, len, "оьв") || StemmerUtil.EndsWith(s, len, "ами") || StemmerUtil.EndsWith(s, len, "его") || StemmerUtil.EndsWith(s, len, "ему") || StemmerUtil.EndsWith(s, len, "ери") || StemmerUtil.EndsWith(s, len, "ими") || StemmerUtil.EndsWith(s, len, "ого") || StemmerUtil.EndsWith(s, len, "ому") || StemmerUtil.EndsWith(s, len, "ыми") || StemmerUtil.EndsWith(s, len, "оев")))
{
return(len - 3);
}
if (len > 4 && (StemmerUtil.EndsWith(s, len, "ая") || StemmerUtil.EndsWith(s, len, "яя") || StemmerUtil.EndsWith(s, len, "ях") || StemmerUtil.EndsWith(s, len, "юю") || StemmerUtil.EndsWith(s, len, "ах") || StemmerUtil.EndsWith(s, len, "ею") || StemmerUtil.EndsWith(s, len, "их") || StemmerUtil.EndsWith(s, len, "ия") || StemmerUtil.EndsWith(s, len, "ию") || StemmerUtil.EndsWith(s, len, "ьв") || StemmerUtil.EndsWith(s, len, "ою") || StemmerUtil.EndsWith(s, len, "ую") || StemmerUtil.EndsWith(s, len, "ям") || StemmerUtil.EndsWith(s, len, "ых") || StemmerUtil.EndsWith(s, len, "ея") || StemmerUtil.EndsWith(s, len, "ам") || StemmerUtil.EndsWith(s, len, "ем") || StemmerUtil.EndsWith(s, len, "ей") || StemmerUtil.EndsWith(s, len, "ём") || StemmerUtil.EndsWith(s, len, "ев") || StemmerUtil.EndsWith(s, len, "ий") || StemmerUtil.EndsWith(s, len, "им") || StemmerUtil.EndsWith(s, len, "ое") || StemmerUtil.EndsWith(s, len, "ой") || StemmerUtil.EndsWith(s, len, "ом") || StemmerUtil.EndsWith(s, len, "ов") || StemmerUtil.EndsWith(s, len, "ые") || StemmerUtil.EndsWith(s, len, "ый") || StemmerUtil.EndsWith(s, len, "ым") || StemmerUtil.EndsWith(s, len, "ми")))
{
return(len - 2);
}
if (len > 3)
{
switch (s[len - 1])
{
case 'а':
case 'е':
case 'и':
case 'о':
case 'у':
case 'й':
case 'ы':
case 'я':
case 'ь':
return(len - 1);
}
}
return(len);
}
public virtual int stem(char[] s, int len)
{
// Remove posessive -s (bilens -> bilen) and continue checking
if (len > 4 && s[len - 1] == 's')
{
len--;
}
// Remove common endings, single-pass
if (len > 7 && ((StemmerUtil.EndsWith(s, len, "heter") && useBokmaal) || (StemmerUtil.EndsWith(s, len, "heten") && useBokmaal) || (StemmerUtil.EndsWith(s, len, "heita") && useNynorsk))) // general ending (hemmeleg-heita -> hemmeleg) - general ending (hemmelig-heten -> hemmelig) - general ending (hemmelig-heter -> hemmelig)
{
return(len - 5);
}
// Remove Nynorsk common endings, single-pass
if (len > 8 && useNynorsk && (StemmerUtil.EndsWith(s, len, "heiter") || StemmerUtil.EndsWith(s, len, "leiken") || StemmerUtil.EndsWith(s, len, "leikar"))) // general ending (trygg-leikar -> trygg) - general ending (trygg-leiken -> trygg) - general ending (hemmeleg-heiter -> hemmeleg)
{
return(len - 6);
}
if (len > 5 && (StemmerUtil.EndsWith(s, len, "dom") || (StemmerUtil.EndsWith(s, len, "het") && useBokmaal))) // general ending (hemmelig-het -> hemmelig) - general ending (kristen-dom -> kristen)
{
return(len - 3);
}
if (len > 6 && useNynorsk && (StemmerUtil.EndsWith(s, len, "heit") || StemmerUtil.EndsWith(s, len, "semd") || StemmerUtil.EndsWith(s, len, "leik"))) // general ending (trygg-leik -> trygg) - general ending (verk-semd -> verk) - general ending (hemmeleg-heit -> hemmeleg)
{
return(len - 4);
}
if (len > 7 && (StemmerUtil.EndsWith(s, len, "elser") || StemmerUtil.EndsWith(s, len, "elsen"))) // general ending (føl-elsen -> føl) - general ending (føl-elser -> føl)
{
return(len - 5);
}
if (len > 6 && ((StemmerUtil.EndsWith(s, len, "ende") && useBokmaal) || (StemmerUtil.EndsWith(s, len, "ande") && useNynorsk) || StemmerUtil.EndsWith(s, len, "else") || (StemmerUtil.EndsWith(s, len, "este") && useBokmaal) || (StemmerUtil.EndsWith(s, len, "aste") && useNynorsk) || (StemmerUtil.EndsWith(s, len, "eren") && useBokmaal) || (StemmerUtil.EndsWith(s, len, "aren") && useNynorsk))) // masc - masc - adj (fin-aste -> fin) - adj (fin-este -> fin) - general ending (føl-else -> føl) - (sov-ande -> sov) - (sov-ende -> sov)
{
return(len - 4);
}
if (len > 5 && ((StemmerUtil.EndsWith(s, len, "ere") && useBokmaal) || (StemmerUtil.EndsWith(s, len, "are") && useNynorsk) || (StemmerUtil.EndsWith(s, len, "est") && useBokmaal) || (StemmerUtil.EndsWith(s, len, "ast") && useNynorsk) || StemmerUtil.EndsWith(s, len, "ene") || (StemmerUtil.EndsWith(s, len, "ane") && useNynorsk))) // masc pl definite (gut-ane) - masc/fem/neutr pl definite (hus-ene) - adj (fin-ast -> fin) - adj (fin-est -> fin) - adj (fin-are -> fin) - adj (fin-ere -> fin)
{
return(len - 3);
}
if (len > 4 && (StemmerUtil.EndsWith(s, len, "er") || StemmerUtil.EndsWith(s, len, "en") || StemmerUtil.EndsWith(s, len, "et") || (StemmerUtil.EndsWith(s, len, "ar") && useNynorsk) || (StemmerUtil.EndsWith(s, len, "st") && useBokmaal) || StemmerUtil.EndsWith(s, len, "te"))) // adj (billig-st -> billig) - masc pl indefinite - neutr definite - masc/fem definite - masc/fem indefinite
{
return(len - 2);
}
if (len > 3)
{
switch (s[len - 1])
{
case 'a': // fem definite
case 'e': // to get correct stem for nouns ending in -e (kake -> kak, kaker -> kak)
case 'n':
return(len - 1);
}
}
return(len);
}
/// <summary>
/// Stem an input buffer of Sorani text.
/// </summary>
/// <param name="s"> input buffer </param>
/// <param name="len"> length of input buffer </param>
/// <returns> length of input buffer after normalization </returns>
public virtual int Stem(char[] s, int len)
{
// postposition
if (len > 5 && StemmerUtil.EndsWith(s, len, "دا"))
{
len -= 2;
}
else if (len > 4 && StemmerUtil.EndsWith(s, len, "نا"))
{
len--;
}
else if (len > 6 && StemmerUtil.EndsWith(s, len, "ەوە"))
{
len -= 3;
}
// possessive pronoun
if (len > 6 && (StemmerUtil.EndsWith(s, len, "مان") || StemmerUtil.EndsWith(s, len, "یان") || StemmerUtil.EndsWith(s, len, "تان")))
{
len -= 3;
}
// indefinite singular ezafe
if (len > 6 && StemmerUtil.EndsWith(s, len, "ێکی"))
{
return(len - 3);
}
else if (len > 7 && StemmerUtil.EndsWith(s, len, "یەکی"))
{
return(len - 4);
}
// indefinite singular
if (len > 5 && StemmerUtil.EndsWith(s, len, "ێک"))
{
return(len - 2);
}
else if (len > 6 && StemmerUtil.EndsWith(s, len, "یەک"))
{
return(len - 3);
}
// definite singular
else if (len > 6 && StemmerUtil.EndsWith(s, len, "ەکە"))
{
return(len - 3);
}
else if (len > 5 && StemmerUtil.EndsWith(s, len, "کە"))
{
return(len - 2);
}
// definite plural
else if (len > 7 && StemmerUtil.EndsWith(s, len, "ەکان"))
{
return(len - 4);
}
else if (len > 6 && StemmerUtil.EndsWith(s, len, "کان"))
{
return(len - 3);
}
// indefinite plural ezafe
else if (len > 7 && StemmerUtil.EndsWith(s, len, "یانی"))
{
return(len - 4);
}
else if (len > 6 && StemmerUtil.EndsWith(s, len, "انی"))
{
return(len - 3);
}
// indefinite plural
else if (len > 6 && StemmerUtil.EndsWith(s, len, "یان"))
{
return(len - 3);
}
else if (len > 5 && StemmerUtil.EndsWith(s, len, "ان"))
{
return(len - 2);
}
// demonstrative plural
else if (len > 7 && StemmerUtil.EndsWith(s, len, "یانە"))
{
return(len - 4);
}
else if (len > 6 && StemmerUtil.EndsWith(s, len, "انە"))
{
return(len - 3);
}
// demonstrative singular
else if (len > 5 && (StemmerUtil.EndsWith(s, len, "ایە") || StemmerUtil.EndsWith(s, len, "ەیە")))
{
return(len - 2);
}
else if (len > 4 && StemmerUtil.EndsWith(s, len, "ە"))
{
return(len - 1);
}
// absolute singular ezafe
else if (len > 4 && StemmerUtil.EndsWith(s, len, "ی"))
{
return(len - 1);
}
return(len);
}
/// <returns> true if the word matches this rule. </returns>
public virtual bool Matches(char[] s, int len)
{
return(len - m_suffix.Length >= m_min && StemmerUtil.EndsWith(s, len, m_suffix));
}
private int RemovePossessive(char[] s, int len)
{
if (len > 6)
{
if (!IsVowel(s[len - 5]) && (StemmerUtil.EndsWith(s, len, "atok") || StemmerUtil.EndsWith(s, len, "otok") || StemmerUtil.EndsWith(s, len, "etek")))
{
return(len - 4);
}
if (StemmerUtil.EndsWith(s, len, "itek") || StemmerUtil.EndsWith(s, len, "itok"))
{
return(len - 4);
}
}
if (len > 5)
{
if (!IsVowel(s[len - 4]) && (StemmerUtil.EndsWith(s, len, "unk") || StemmerUtil.EndsWith(s, len, "tok") || StemmerUtil.EndsWith(s, len, "tek")))
{
return(len - 3);
}
if (IsVowel(s[len - 4]) && StemmerUtil.EndsWith(s, len, "juk"))
{
return(len - 3);
}
if (StemmerUtil.EndsWith(s, len, "ink"))
{
return(len - 3);
}
}
if (len > 4)
{
if (!IsVowel(s[len - 3]) && (StemmerUtil.EndsWith(s, len, "am") || StemmerUtil.EndsWith(s, len, "em") || StemmerUtil.EndsWith(s, len, "om") || StemmerUtil.EndsWith(s, len, "ad") || StemmerUtil.EndsWith(s, len, "ed") || StemmerUtil.EndsWith(s, len, "od") || StemmerUtil.EndsWith(s, len, "uk")))
{
return(len - 2);
}
if (IsVowel(s[len - 3]) && (StemmerUtil.EndsWith(s, len, "nk") || StemmerUtil.EndsWith(s, len, "ja") || StemmerUtil.EndsWith(s, len, "je")))
{
return(len - 2);
}
if (StemmerUtil.EndsWith(s, len, "im") || StemmerUtil.EndsWith(s, len, "id") || StemmerUtil.EndsWith(s, len, "ik"))
{
return(len - 2);
}
}
if (len > 3)
{
switch (s[len - 1])
{
case 'a':
case 'e':
if (!IsVowel(s[len - 2]))
{
return(len - 1);
}
break;
case 'm':
case 'd':
if (IsVowel(s[len - 2]))
{
return(len - 1);
}
break;
case 'i':
return(len - 1);
}
}
return(len);
}
private int RemoveCase(char[] s, int len)
{
if (len > 6 && StemmerUtil.EndsWith(s, len, "kent"))
{
return(len - 4);
}
if (len > 5)
{
if (StemmerUtil.EndsWith(s, len, "nak") || StemmerUtil.EndsWith(s, len, "nek") || StemmerUtil.EndsWith(s, len, "val") || StemmerUtil.EndsWith(s, len, "vel") || StemmerUtil.EndsWith(s, len, "ert") || StemmerUtil.EndsWith(s, len, "rol") || StemmerUtil.EndsWith(s, len, "ban") || StemmerUtil.EndsWith(s, len, "ben") || StemmerUtil.EndsWith(s, len, "bol") || StemmerUtil.EndsWith(s, len, "nal") || StemmerUtil.EndsWith(s, len, "nel") || StemmerUtil.EndsWith(s, len, "hoz") || StemmerUtil.EndsWith(s, len, "hez") || StemmerUtil.EndsWith(s, len, "tol"))
{
return(len - 3);
}
if (StemmerUtil.EndsWith(s, len, "al") || StemmerUtil.EndsWith(s, len, "el"))
{
if (!IsVowel(s[len - 3]) && s[len - 3] == s[len - 4])
{
return(len - 3);
}
}
}
if (len > 4)
{
if (StemmerUtil.EndsWith(s, len, "at") || StemmerUtil.EndsWith(s, len, "et") || StemmerUtil.EndsWith(s, len, "ot") || StemmerUtil.EndsWith(s, len, "va") || StemmerUtil.EndsWith(s, len, "ve") || StemmerUtil.EndsWith(s, len, "ra") || StemmerUtil.EndsWith(s, len, "re") || StemmerUtil.EndsWith(s, len, "ba") || StemmerUtil.EndsWith(s, len, "be") || StemmerUtil.EndsWith(s, len, "ul") || StemmerUtil.EndsWith(s, len, "ig"))
{
return(len - 2);
}
if ((StemmerUtil.EndsWith(s, len, "on") || StemmerUtil.EndsWith(s, len, "en")) && !IsVowel(s[len - 3]))
{
return(len - 2);
}
switch (s[len - 1])
{
case 't':
case 'n':
return(len - 1);
case 'a':
case 'e':
if (s[len - 2] == s[len - 3] && !IsVowel(s[len - 2]))
{
return(len - 2);
}
break;
}
}
return(len);
}
private int Step3(char[] s, int len)
{
if (len > 8)
{
if (StemmerUtil.EndsWith(s, len, "nnen"))
{
s[len - 4] = 's';
return(len - 3);
}
if (StemmerUtil.EndsWith(s, len, "ntena"))
{
s[len - 5] = 's';
return(len - 4);
}
if (StemmerUtil.EndsWith(s, len, "tten"))
{
return(len - 4);
}
if (StemmerUtil.EndsWith(s, len, "eiden"))
{
return(len - 5);
}
}
if (len > 6)
{
if (StemmerUtil.EndsWith(s, len, "neen") || StemmerUtil.EndsWith(s, len, "niin") || StemmerUtil.EndsWith(s, len, "seen") || StemmerUtil.EndsWith(s, len, "teen") || StemmerUtil.EndsWith(s, len, "inen"))
{
return(len - 4);
}
if (s[len - 3] == 'h' && IsVowel(s[len - 2]) && s[len - 1] == 'n')
{
return(len - 3);
}
if (StemmerUtil.EndsWith(s, len, "den"))
{
s[len - 3] = 's';
return(len - 2);
}
if (StemmerUtil.EndsWith(s, len, "ksen"))
{
s[len - 4] = 's';
return(len - 3);
}
if (StemmerUtil.EndsWith(s, len, "ssa") || StemmerUtil.EndsWith(s, len, "sta") || StemmerUtil.EndsWith(s, len, "lla") || StemmerUtil.EndsWith(s, len, "lta") || StemmerUtil.EndsWith(s, len, "tta") || StemmerUtil.EndsWith(s, len, "ksi") || StemmerUtil.EndsWith(s, len, "lle"))
{
return(len - 3);
}
}
if (len > 5)
{
if (StemmerUtil.EndsWith(s, len, "na") || StemmerUtil.EndsWith(s, len, "ne"))
{
return(len - 2);
}
if (StemmerUtil.EndsWith(s, len, "nei"))
{
return(len - 3);
}
}
if (len > 4)
{
if (StemmerUtil.EndsWith(s, len, "ja") || StemmerUtil.EndsWith(s, len, "ta"))
{
return(len - 2);
}
if (s[len - 1] == 'a')
{
return(len - 1);
}
if (s[len - 1] == 'n' && IsVowel(s[len - 2]))
{
return(len - 2);
}
if (s[len - 1] == 'n')
{
return(len - 1);
}
}
return(len);
}
private int RemoveCase(char[] s, int len)
{
if (len > 7 && StemmerUtil.EndsWith(s, len, "atech"))
{
return(len - 5);
}
if (len > 6 && (StemmerUtil.EndsWith(s, len, "ětem") || StemmerUtil.EndsWith(s, len, "etem") || StemmerUtil.EndsWith(s, len, "atům")))
{
return(len - 4);
}
if (len > 5 && (StemmerUtil.EndsWith(s, len, "ech") || StemmerUtil.EndsWith(s, len, "ich") || StemmerUtil.EndsWith(s, len, "ích") || StemmerUtil.EndsWith(s, len, "ého") || StemmerUtil.EndsWith(s, len, "ěmi") || StemmerUtil.EndsWith(s, len, "emi") || StemmerUtil.EndsWith(s, len, "ému") || StemmerUtil.EndsWith(s, len, "ěte") || StemmerUtil.EndsWith(s, len, "ete") || StemmerUtil.EndsWith(s, len, "ěti") || StemmerUtil.EndsWith(s, len, "eti") || StemmerUtil.EndsWith(s, len, "ího") || StemmerUtil.EndsWith(s, len, "iho") || StemmerUtil.EndsWith(s, len, "ími") || StemmerUtil.EndsWith(s, len, "ímu") || StemmerUtil.EndsWith(s, len, "imu") || StemmerUtil.EndsWith(s, len, "ách") || StemmerUtil.EndsWith(s, len, "ata") || StemmerUtil.EndsWith(s, len, "aty") || StemmerUtil.EndsWith(s, len, "ých") || StemmerUtil.EndsWith(s, len, "ama") || StemmerUtil.EndsWith(s, len, "ami") || StemmerUtil.EndsWith(s, len, "ové") || StemmerUtil.EndsWith(s, len, "ovi") || StemmerUtil.EndsWith(s, len, "ými")))
{
return(len - 3);
}
if (len > 4 && (StemmerUtil.EndsWith(s, len, "em") || StemmerUtil.EndsWith(s, len, "es") || StemmerUtil.EndsWith(s, len, "ém") || StemmerUtil.EndsWith(s, len, "ím") || StemmerUtil.EndsWith(s, len, "ům") || StemmerUtil.EndsWith(s, len, "at") || StemmerUtil.EndsWith(s, len, "ám") || StemmerUtil.EndsWith(s, len, "os") || StemmerUtil.EndsWith(s, len, "us") || StemmerUtil.EndsWith(s, len, "ým") || StemmerUtil.EndsWith(s, len, "mi") || StemmerUtil.EndsWith(s, len, "ou")))
{
return(len - 2);
}
if (len > 3)
{
switch (s[len - 1])
{
case 'a':
case 'e':
case 'i':
case 'o':
case 'u':
case 'ů':
case 'y':
case 'á':
case 'é':
case 'í':
case 'ý':
case 'ě':
return(len - 1);
}
}
return(len);
}
public virtual int Stem(char[] s, int len)
{
if (len > 5 && s[len - 1] == 'x')
{
if (s[len - 3] == 'a' && s[len - 2] == 'u' && s[len - 4] != 'e')
{
s[len - 2] = 'l';
}
len--;
}
if (len > 3 && s[len - 1] == 'x')
{
len--;
}
if (len > 3 && s[len - 1] == 's')
{
len--;
}
if (len > 9 && StemmerUtil.EndsWith(s, len, "issement"))
{
len -= 6;
s[len - 1] = 'r';
return(Norm(s, len));
}
if (len > 8 && StemmerUtil.EndsWith(s, len, "issant"))
{
len -= 4;
s[len - 1] = 'r';
return(Norm(s, len));
}
if (len > 6 && StemmerUtil.EndsWith(s, len, "ement"))
{
len -= 4;
if (len > 3 && StemmerUtil.EndsWith(s, len, "ive"))
{
len--;
s[len - 1] = 'f';
}
return(Norm(s, len));
}
if (len > 11 && StemmerUtil.EndsWith(s, len, "ficatrice"))
{
len -= 5;
s[len - 2] = 'e';
s[len - 1] = 'r';
return(Norm(s, len));
}
if (len > 10 && StemmerUtil.EndsWith(s, len, "ficateur"))
{
len -= 4;
s[len - 2] = 'e';
s[len - 1] = 'r';
return(Norm(s, len));
}
if (len > 9 && StemmerUtil.EndsWith(s, len, "catrice"))
{
len -= 3;
s[len - 4] = 'q';
s[len - 3] = 'u';
s[len - 2] = 'e';
//s[len-1] = 'r' <-- unnecessary, already 'r'.
return(Norm(s, len));
}
if (len > 8 && StemmerUtil.EndsWith(s, len, "cateur"))
{
len -= 2;
s[len - 4] = 'q';
s[len - 3] = 'u';
s[len - 2] = 'e';
s[len - 1] = 'r';
return(Norm(s, len));
}
if (len > 8 && StemmerUtil.EndsWith(s, len, "atrice"))
{
len -= 4;
s[len - 2] = 'e';
s[len - 1] = 'r';
return(Norm(s, len));
}
if (len > 7 && StemmerUtil.EndsWith(s, len, "ateur"))
{
len -= 3;
s[len - 2] = 'e';
s[len - 1] = 'r';
return(Norm(s, len));
}
if (len > 6 && StemmerUtil.EndsWith(s, len, "trice"))
{
len--;
s[len - 3] = 'e';
s[len - 2] = 'u';
s[len - 1] = 'r';
}
if (len > 5 && StemmerUtil.EndsWith(s, len, "ième"))
{
return(Norm(s, len - 4));
}
if (len > 7 && StemmerUtil.EndsWith(s, len, "teuse"))
{
len -= 2;
s[len - 1] = 'r';
return(Norm(s, len));
}
if (len > 6 && StemmerUtil.EndsWith(s, len, "teur"))
{
len--;
s[len - 1] = 'r';
return(Norm(s, len));
}
if (len > 5 && StemmerUtil.EndsWith(s, len, "euse"))
{
return(Norm(s, len - 2));
}
if (len > 8 && StemmerUtil.EndsWith(s, len, "ère"))
{
len--;
s[len - 2] = 'e';
return(Norm(s, len));
}
if (len > 7 && StemmerUtil.EndsWith(s, len, "ive"))
{
len--;
s[len - 1] = 'f';
return(Norm(s, len));
}
if (len > 4 && (StemmerUtil.EndsWith(s, len, "folle") || StemmerUtil.EndsWith(s, len, "molle")))
{
len -= 2;
s[len - 1] = 'u';
return(Norm(s, len));
}
if (len > 9 && StemmerUtil.EndsWith(s, len, "nnelle"))
{
return(Norm(s, len - 5));
}
if (len > 9 && StemmerUtil.EndsWith(s, len, "nnel"))
{
return(Norm(s, len - 3));
}
if (len > 4 && StemmerUtil.EndsWith(s, len, "ète"))
{
len--;
s[len - 2] = 'e';
}
if (len > 8 && StemmerUtil.EndsWith(s, len, "ique"))
{
len -= 4;
}
if (len > 8 && StemmerUtil.EndsWith(s, len, "esse"))
{
return(Norm(s, len - 3));
}
if (len > 7 && StemmerUtil.EndsWith(s, len, "inage"))
{
return(Norm(s, len - 3));
}
if (len > 9 && StemmerUtil.EndsWith(s, len, "isation"))
{
len -= 7;
if (len > 5 && StemmerUtil.EndsWith(s, len, "ual"))
{
s[len - 2] = 'e';
}
return(Norm(s, len));
}
if (len > 9 && StemmerUtil.EndsWith(s, len, "isateur"))
{
return(Norm(s, len - 7));
}
if (len > 8 && StemmerUtil.EndsWith(s, len, "ation"))
{
return(Norm(s, len - 5));
}
if (len > 8 && StemmerUtil.EndsWith(s, len, "ition"))
{
return(Norm(s, len - 5));
}
return(Norm(s, len));
}
private int Norm(char[] s, int len)
{
if (len > 4)
{
for (int i = 0; i < len; i++)
{
switch (s[i])
{
case 'à':
case 'á':
case 'â':
s[i] = 'a';
break;
case 'ô':
s[i] = 'o';
break;
case 'è':
case 'é':
case 'ê':
s[i] = 'e';
break;
case 'ù':
case 'û':
s[i] = 'u';
break;
case 'î':
s[i] = 'i';
break;
case 'ç':
s[i] = 'c';
break;
}
}
char ch = s[0];
for (int i = 1; i < len; i++)
{
if (s[i] == ch && char.IsLetter(ch))
{
len = StemmerUtil.Delete(s, i--, len);
}
else
{
ch = s[i];
}
}
}
if (len > 4 && StemmerUtil.EndsWith(s, len, "ie"))
{
len -= 2;
}
if (len > 4)
{
if (s[len - 1] == 'r')
{
len--;
}
if (s[len - 1] == 'e')
{
len--;
}
if (s[len - 1] == 'e')
{
len--;
}
if (s[len - 1] == s[len - 2] && char.IsLetter(s[len - 1]))
{
len--;
}
}
return(len);
}
private int Rule21(char[] s, int len)
{
if (len > 9 && StemmerUtil.EndsWith(s, len, "ιοντουσαν"))
{
return(len - 9);
}
if (len > 8 && (StemmerUtil.EndsWith(s, len, "ιομασταν") ||
StemmerUtil.EndsWith(s, len, "ιοσασταν") ||
StemmerUtil.EndsWith(s, len, "ιουμαστε") ||
StemmerUtil.EndsWith(s, len, "οντουσαν")))
{
return(len - 8);
}
if (len > 7 && (StemmerUtil.EndsWith(s, len, "ιεμαστε") ||
StemmerUtil.EndsWith(s, len, "ιεσαστε") ||
StemmerUtil.EndsWith(s, len, "ιομουνα") ||
StemmerUtil.EndsWith(s, len, "ιοσαστε") ||
StemmerUtil.EndsWith(s, len, "ιοσουνα") ||
StemmerUtil.EndsWith(s, len, "ιουνται") ||
StemmerUtil.EndsWith(s, len, "ιουνταν") ||
StemmerUtil.EndsWith(s, len, "ηθηκατε") ||
StemmerUtil.EndsWith(s, len, "ομασταν") ||
StemmerUtil.EndsWith(s, len, "οσασταν") ||
StemmerUtil.EndsWith(s, len, "ουμαστε")))
{
return(len - 7);
}
if (len > 6 && (StemmerUtil.EndsWith(s, len, "ιομουν") ||
StemmerUtil.EndsWith(s, len, "ιονταν") ||
StemmerUtil.EndsWith(s, len, "ιοσουν") ||
StemmerUtil.EndsWith(s, len, "ηθειτε") ||
StemmerUtil.EndsWith(s, len, "ηθηκαν") ||
StemmerUtil.EndsWith(s, len, "ομουνα") ||
StemmerUtil.EndsWith(s, len, "οσαστε") ||
StemmerUtil.EndsWith(s, len, "οσουνα") ||
StemmerUtil.EndsWith(s, len, "ουνται") ||
StemmerUtil.EndsWith(s, len, "ουνταν") ||
StemmerUtil.EndsWith(s, len, "ουσατε")))
{
return(len - 6);
}
if (len > 5 && (StemmerUtil.EndsWith(s, len, "αγατε") ||
StemmerUtil.EndsWith(s, len, "ιεμαι") ||
StemmerUtil.EndsWith(s, len, "ιεται") ||
StemmerUtil.EndsWith(s, len, "ιεσαι") ||
StemmerUtil.EndsWith(s, len, "ιοταν") ||
StemmerUtil.EndsWith(s, len, "ιουμα") ||
StemmerUtil.EndsWith(s, len, "ηθεισ") ||
StemmerUtil.EndsWith(s, len, "ηθουν") ||
StemmerUtil.EndsWith(s, len, "ηκατε") ||
StemmerUtil.EndsWith(s, len, "ησατε") ||
StemmerUtil.EndsWith(s, len, "ησουν") ||
StemmerUtil.EndsWith(s, len, "ομουν") ||
StemmerUtil.EndsWith(s, len, "ονται") ||
StemmerUtil.EndsWith(s, len, "ονταν") ||
StemmerUtil.EndsWith(s, len, "οσουν") ||
StemmerUtil.EndsWith(s, len, "ουμαι") ||
StemmerUtil.EndsWith(s, len, "ουσαν")))
{
return(len - 5);
}
if (len > 4 && (StemmerUtil.EndsWith(s, len, "αγαν") ||
StemmerUtil.EndsWith(s, len, "αμαι") ||
StemmerUtil.EndsWith(s, len, "ασαι") ||
StemmerUtil.EndsWith(s, len, "αται") ||
StemmerUtil.EndsWith(s, len, "ειτε") ||
StemmerUtil.EndsWith(s, len, "εσαι") ||
StemmerUtil.EndsWith(s, len, "εται") ||
StemmerUtil.EndsWith(s, len, "ηδεσ") ||
StemmerUtil.EndsWith(s, len, "ηδων") ||
StemmerUtil.EndsWith(s, len, "ηθει") ||
StemmerUtil.EndsWith(s, len, "ηκαν") ||
StemmerUtil.EndsWith(s, len, "ησαν") ||
StemmerUtil.EndsWith(s, len, "ησει") ||
StemmerUtil.EndsWith(s, len, "ησεσ") ||
StemmerUtil.EndsWith(s, len, "ομαι") ||
StemmerUtil.EndsWith(s, len, "οταν")))
{
return(len - 4);
}
if (len > 3 && (StemmerUtil.EndsWith(s, len, "αει") ||
StemmerUtil.EndsWith(s, len, "εισ") ||
StemmerUtil.EndsWith(s, len, "ηθω") ||
StemmerUtil.EndsWith(s, len, "ησω") ||
StemmerUtil.EndsWith(s, len, "ουν") ||
StemmerUtil.EndsWith(s, len, "ουσ")))
{
return(len - 3);
}
if (len > 2 && (StemmerUtil.EndsWith(s, len, "αν") ||
StemmerUtil.EndsWith(s, len, "ασ") ||
StemmerUtil.EndsWith(s, len, "αω") ||
StemmerUtil.EndsWith(s, len, "ει") ||
StemmerUtil.EndsWith(s, len, "εσ") ||
StemmerUtil.EndsWith(s, len, "ησ") ||
StemmerUtil.EndsWith(s, len, "οι") ||
StemmerUtil.EndsWith(s, len, "οσ") ||
StemmerUtil.EndsWith(s, len, "ου") ||
StemmerUtil.EndsWith(s, len, "υσ") ||
StemmerUtil.EndsWith(s, len, "ων")))
{
return(len - 2);
}
if (len > 1 && EndsWithVowel(s, len))
{
return(len - 1);
}
return(len);
}
/// <summary>
/// Stem an input buffer of Bulgarian text.
/// </summary>
/// <param name="s"> input buffer </param>
/// <param name="len"> length of input buffer </param>
/// <returns> length of input buffer after normalization </returns>
public virtual int Stem(char[] s, int len)
{
if (len < 4) // do not stem
{
return(len);
}
if (len > 5 && StemmerUtil.EndsWith(s, len, "ища"))
{
return(len - 3);
}
len = RemoveArticle(s, len);
len = RemovePlural(s, len);
if (len > 3)
{
if (StemmerUtil.EndsWith(s, len, "я"))
{
len--;
}
if (StemmerUtil.EndsWith(s, len, "а") || StemmerUtil.EndsWith(s, len, "о") || StemmerUtil.EndsWith(s, len, "е"))
{
len--;
}
}
// the rule to rewrite ен -> н is duplicated in the paper.
// in the perl implementation referenced by the paper, this is fixed.
// (it is fixed here as well)
if (len > 4 && StemmerUtil.EndsWith(s, len, "ен"))
{
s[len - 2] = 'н'; // replace with н
len--;
}
if (len > 5 && s[len - 2] == 'ъ')
{
s[len - 2] = s[len - 1]; // replace ъN with N
len--;
}
return(len);
}
private int RemovePlural(char[] s, int len)
{
if (len > 6)
{
if (StemmerUtil.EndsWith(s, len, "овци"))
{
return(len - 3); // replace with о
}
if (StemmerUtil.EndsWith(s, len, "ове"))
{
return(len - 3);
}
if (StemmerUtil.EndsWith(s, len, "еве"))
{
s[len - 3] = 'й'; // replace with й
return(len - 2);
}
}
if (len > 5)
{
if (StemmerUtil.EndsWith(s, len, "ища"))
{
return(len - 3);
}
if (StemmerUtil.EndsWith(s, len, "та"))
{
return(len - 2);
}
if (StemmerUtil.EndsWith(s, len, "ци"))
{
s[len - 2] = 'к'; // replace with к
return(len - 1);
}
if (StemmerUtil.EndsWith(s, len, "зи"))
{
s[len - 2] = 'г'; // replace with г
return(len - 1);
}
if (s[len - 3] == 'е' && s[len - 1] == 'и')
{
s[len - 3] = 'я'; // replace е with я, remove и
return(len - 1);
}
}
if (len > 4)
{
if (StemmerUtil.EndsWith(s, len, "си"))
{
s[len - 2] = 'х'; // replace with х
return(len - 1);
}
if (StemmerUtil.EndsWith(s, len, "и"))
{
return(len - 1);
}
}
return(len);
}
public virtual int Stem(char[] buffer, int len)
{
// 5
if ((len > 6) && (StemmerUtil.EndsWith(buffer, len, "ाएंगी") || StemmerUtil.EndsWith(buffer, len, "ाएंगे") || StemmerUtil.EndsWith(buffer, len, "ाऊंगी") || StemmerUtil.EndsWith(buffer, len, "ाऊंगा") || StemmerUtil.EndsWith(buffer, len, "ाइयाँ") || StemmerUtil.EndsWith(buffer, len, "ाइयों") || StemmerUtil.EndsWith(buffer, len, "ाइयां")))
{
return(len - 5);
}
// 4
if ((len > 5) && (StemmerUtil.EndsWith(buffer, len, "ाएगी") || StemmerUtil.EndsWith(buffer, len, "ाएगा") || StemmerUtil.EndsWith(buffer, len, "ाओगी") || StemmerUtil.EndsWith(buffer, len, "ाओगे") || StemmerUtil.EndsWith(buffer, len, "एंगी") || StemmerUtil.EndsWith(buffer, len, "ेंगी") || StemmerUtil.EndsWith(buffer, len, "एंगे") || StemmerUtil.EndsWith(buffer, len, "ेंगे") || StemmerUtil.EndsWith(buffer, len, "ूंगी") || StemmerUtil.EndsWith(buffer, len, "ूंगा") || StemmerUtil.EndsWith(buffer, len, "ातीं") || StemmerUtil.EndsWith(buffer, len, "नाओं") || StemmerUtil.EndsWith(buffer, len, "नाएं") || StemmerUtil.EndsWith(buffer, len, "ताओं") || StemmerUtil.EndsWith(buffer, len, "ताएं") || StemmerUtil.EndsWith(buffer, len, "ियाँ") || StemmerUtil.EndsWith(buffer, len, "ियों") || StemmerUtil.EndsWith(buffer, len, "ियां")))
{
return(len - 4);
}
// 3
if ((len > 4) && (StemmerUtil.EndsWith(buffer, len, "ाकर") || StemmerUtil.EndsWith(buffer, len, "ाइए") || StemmerUtil.EndsWith(buffer, len, "ाईं") || StemmerUtil.EndsWith(buffer, len, "ाया") || StemmerUtil.EndsWith(buffer, len, "ेगी") || StemmerUtil.EndsWith(buffer, len, "ेगा") || StemmerUtil.EndsWith(buffer, len, "ोगी") || StemmerUtil.EndsWith(buffer, len, "ोगे") || StemmerUtil.EndsWith(buffer, len, "ाने") || StemmerUtil.EndsWith(buffer, len, "ाना") || StemmerUtil.EndsWith(buffer, len, "ाते") || StemmerUtil.EndsWith(buffer, len, "ाती") || StemmerUtil.EndsWith(buffer, len, "ाता") || StemmerUtil.EndsWith(buffer, len, "तीं") || StemmerUtil.EndsWith(buffer, len, "ाओं") || StemmerUtil.EndsWith(buffer, len, "ाएं") || StemmerUtil.EndsWith(buffer, len, "ुओं") || StemmerUtil.EndsWith(buffer, len, "ुएं") || StemmerUtil.EndsWith(buffer, len, "ुआं")))
{
return(len - 3);
}
// 2
if ((len > 3) && (StemmerUtil.EndsWith(buffer, len, "कर") || StemmerUtil.EndsWith(buffer, len, "ाओ") || StemmerUtil.EndsWith(buffer, len, "िए") || StemmerUtil.EndsWith(buffer, len, "ाई") || StemmerUtil.EndsWith(buffer, len, "ाए") || StemmerUtil.EndsWith(buffer, len, "ने") || StemmerUtil.EndsWith(buffer, len, "नी") || StemmerUtil.EndsWith(buffer, len, "ना") || StemmerUtil.EndsWith(buffer, len, "ते") || StemmerUtil.EndsWith(buffer, len, "ीं") || StemmerUtil.EndsWith(buffer, len, "ती") || StemmerUtil.EndsWith(buffer, len, "ता") || StemmerUtil.EndsWith(buffer, len, "ाँ") || StemmerUtil.EndsWith(buffer, len, "ां") || StemmerUtil.EndsWith(buffer, len, "ों") || StemmerUtil.EndsWith(buffer, len, "ें")))
{
return(len - 2);
}
// 1
if ((len > 2) && (StemmerUtil.EndsWith(buffer, len, "ो") || StemmerUtil.EndsWith(buffer, len, "े") || StemmerUtil.EndsWith(buffer, len, "ू") || StemmerUtil.EndsWith(buffer, len, "ु") || StemmerUtil.EndsWith(buffer, len, "ी") || StemmerUtil.EndsWith(buffer, len, "ि") || StemmerUtil.EndsWith(buffer, len, "ा")))
{
return(len - 1);
}
return(len);
}
private int RemoveParticle(char[] text, int length)
{
if (StemmerUtil.EndsWith(text, length, "kah") || StemmerUtil.EndsWith(text, length, "lah") || StemmerUtil.EndsWith(text, length, "pun"))
{
numSyllables--;
return(length - 3);
}
return(length);
}
private int RemoveFirstOrderPrefix(char[] text, int length)
{
if (StemmerUtil.StartsWith(text, length, "meng"))
{
flags |= REMOVED_MENG;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 4));
}
if (StemmerUtil.StartsWith(text, length, "meny") && length > 4 && IsVowel(text[4]))
{
flags |= REMOVED_MENG;
text[3] = 's';
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 3));
}
if (StemmerUtil.StartsWith(text, length, "men"))
{
flags |= REMOVED_MENG;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 3));
}
if (StemmerUtil.StartsWith(text, length, "mem"))
{
flags |= REMOVED_MENG;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 3));
}
if (StemmerUtil.StartsWith(text, length, "me"))
{
flags |= REMOVED_MENG;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 2));
}
if (StemmerUtil.StartsWith(text, length, "peng"))
{
flags |= REMOVED_PENG;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 4));
}
if (StemmerUtil.StartsWith(text, length, "peny") && length > 4 && IsVowel(text[4]))
{
flags |= REMOVED_PENG;
text[3] = 's';
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 3));
}
if (StemmerUtil.StartsWith(text, length, "peny"))
{
flags |= REMOVED_PENG;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 4));
}
if (StemmerUtil.StartsWith(text, length, "pen") && length > 3 && IsVowel(text[3]))
{
flags |= REMOVED_PENG;
text[2] = 't';
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 2));
}
if (StemmerUtil.StartsWith(text, length, "pen"))
{
flags |= REMOVED_PENG;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 3));
}
if (StemmerUtil.StartsWith(text, length, "pem"))
{
flags |= REMOVED_PENG;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 3));
}
if (StemmerUtil.StartsWith(text, length, "di"))
{
flags |= REMOVED_DI;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 2));
}
if (StemmerUtil.StartsWith(text, length, "ter"))
{
flags |= REMOVED_TER;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 3));
}
if (StemmerUtil.StartsWith(text, length, "ke"))
{
flags |= REMOVED_KE;
numSyllables--;
return(StemmerUtil.DeleteN(text, 0, length, 2));
}
return(length);
}
private int RemovePossessives(char[] s, int len)
{
if (len > 5 && (StemmerUtil.EndsWith(s, len, "ov") || StemmerUtil.EndsWith(s, len, "in") || StemmerUtil.EndsWith(s, len, "ův")))
{
return(len - 2);
}
return(len);
}
public virtual int Stem(char[] s, int len)
{
// Remove genitiv s
if (len > 4 && s[len - 1] == 's')
{
len--;
}
if (len > 5 && (StemmerUtil.EndsWith(s, len, "ene") || (StemmerUtil.EndsWith(s, len, "ane") && useNynorsk))) // masc pl definite (gut-ane) - masc/fem/neutr pl definite (hus-ene)
{
return(len - 3);
}
if (len > 4 && (StemmerUtil.EndsWith(s, len, "er") || StemmerUtil.EndsWith(s, len, "en") || StemmerUtil.EndsWith(s, len, "et") || (StemmerUtil.EndsWith(s, len, "ar") && useNynorsk))) // masc pl indefinite - neutr definite - masc/fem definite - masc/fem indefinite
{
return(len - 2);
}
if (len > 3)
{
switch (s[len - 1])
{
case 'a': // fem definite
case 'e': // to get correct stem for nouns ending in -e (kake -> kak, kaker -> kak)
return(len - 1);
}
}
return(len);
}
public override bool IncrementToken()
{
if (input.IncrementToken())
{
int state = N;
char[] buffer = termAtt.Buffer();
int length = termAtt.Length;
for (int i = 0; i < length; i++)
{
char c = buffer[i];
switch (c)
{
case 'a':
case 'o':
state = U;
break;
case 'u':
state = (state == N) ? U : V;
break;
case 'e':
if (state == U)
{
length = StemmerUtil.Delete(buffer, i--, length);
}
state = V;
break;
case 'i':
case 'q':
case 'y':
state = V;
break;
case 'ä':
buffer[i] = 'a';
state = V;
break;
case 'ö':
buffer[i] = 'o';
state = V;
break;
case 'ü':
buffer[i] = 'u';
state = V;
break;
case 'ß':
buffer[i++] = 's';
buffer = termAtt.ResizeBuffer(1 + length);
if (i < length)
{
Array.Copy(buffer, i, buffer, i + 1, (length - i));
}
buffer[i] = 's';
length++;
state = N;
break;
default:
state = N;
break;
}
}
termAtt.Length = length;
return(true);
}
else
{
return(false);
}
}
