func (s *TruncateTokenFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
for _, token := range input {
wordLen := utf8.RuneCount(token.Term)
if wordLen > s.length {
token.Term = analysis.TruncateRunes(token.Term, wordLen-s.length)
}
}
return input
}
func (s *PossessiveFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
for _, token := range input {
runes := bytes.Runes(token.Term)
if len(runes) >= 2 {
secondToLastRune := runes[len(runes)-2]
lastRune := runes[len(runes)-1]
if (secondToLastRune == rightSingleQuotationMark ||
secondToLastRune == apostrophe ||
secondToLastRune == fullWidthApostrophe) &&
(lastRune == 's' || lastRune == 'S') {
token.Term = analysis.TruncateRunes(token.Term, 2)
}
}
}
return input
}
func stem(input []byte) []byte {
inputLen := utf8.RuneCount(input)
// 5
if inputLen > 6 &&
(bytes.HasSuffix(input, []byte("ाएंगी")) ||
bytes.HasSuffix(input, []byte("ाएंगे")) ||
bytes.HasSuffix(input, []byte("ाऊंगी")) ||
bytes.HasSuffix(input, []byte("ाऊंगा")) ||
bytes.HasSuffix(input, []byte("ाइयाँ")) ||
bytes.HasSuffix(input, []byte("ाइयों")) ||
bytes.HasSuffix(input, []byte("ाइयां"))) {
return analysis.TruncateRunes(input, 5)
}
// 4
if inputLen > 5 &&
(bytes.HasSuffix(input, []byte("ाएगी")) ||
bytes.HasSuffix(input, []byte("ाएगा")) ||
bytes.HasSuffix(input, []byte("ाओगी")) ||
bytes.HasSuffix(input, []byte("ाओगे")) ||
bytes.HasSuffix(input, []byte("एंगी")) ||
bytes.HasSuffix(input, []byte("ेंगी")) ||
bytes.HasSuffix(input, []byte("एंगे")) ||
bytes.HasSuffix(input, []byte("ेंगे")) ||
bytes.HasSuffix(input, []byte("ूंगी")) ||
bytes.HasSuffix(input, []byte("ूंगा")) ||
bytes.HasSuffix(input, []byte("ातीं")) ||
bytes.HasSuffix(input, []byte("नाओं")) ||
bytes.HasSuffix(input, []byte("नाएं")) ||
bytes.HasSuffix(input, []byte("ताओं")) ||
bytes.HasSuffix(input, []byte("ताएं")) ||
bytes.HasSuffix(input, []byte("ियाँ")) ||
bytes.HasSuffix(input, []byte("ियों")) ||
bytes.HasSuffix(input, []byte("ियां"))) {
return analysis.TruncateRunes(input, 4)
}
// 3
if inputLen > 4 &&
(bytes.HasSuffix(input, []byte("ाकर")) ||
bytes.HasSuffix(input, []byte("ाइए")) ||
bytes.HasSuffix(input, []byte("ाईं")) ||
bytes.HasSuffix(input, []byte("ाया")) ||
bytes.HasSuffix(input, []byte("ेगी")) ||
bytes.HasSuffix(input, []byte("ेगा")) ||
bytes.HasSuffix(input, []byte("ोगी")) ||
bytes.HasSuffix(input, []byte("ोगे")) ||
bytes.HasSuffix(input, []byte("ाने")) ||
bytes.HasSuffix(input, []byte("ाना")) ||
bytes.HasSuffix(input, []byte("ाते")) ||
bytes.HasSuffix(input, []byte("ाती")) ||
bytes.HasSuffix(input, []byte("ाता")) ||
bytes.HasSuffix(input, []byte("तीं")) ||
bytes.HasSuffix(input, []byte("ाओं")) ||
bytes.HasSuffix(input, []byte("ाएं")) ||
bytes.HasSuffix(input, []byte("ुओं")) ||
bytes.HasSuffix(input, []byte("ुएं")) ||
bytes.HasSuffix(input, []byte("ुआं"))) {
return analysis.TruncateRunes(input, 3)
}
// 2
if inputLen > 3 &&
(bytes.HasSuffix(input, []byte("कर")) ||
bytes.HasSuffix(input, []byte("ाओ")) ||
bytes.HasSuffix(input, []byte("िए")) ||
bytes.HasSuffix(input, []byte("ाई")) ||
bytes.HasSuffix(input, []byte("ाए")) ||
bytes.HasSuffix(input, []byte("ने")) ||
bytes.HasSuffix(input, []byte("नी")) ||
bytes.HasSuffix(input, []byte("ना")) ||
bytes.HasSuffix(input, []byte("ते")) ||
bytes.HasSuffix(input, []byte("ीं")) ||
bytes.HasSuffix(input, []byte("ती")) ||
bytes.HasSuffix(input, []byte("ता")) ||
bytes.HasSuffix(input, []byte("ाँ")) ||
bytes.HasSuffix(input, []byte("ां")) ||
bytes.HasSuffix(input, []byte("ों")) ||
bytes.HasSuffix(input, []byte("ें"))) {
return analysis.TruncateRunes(input, 2)
}
// 1
if inputLen > 2 &&
(bytes.HasSuffix(input, []byte("ो")) ||
bytes.HasSuffix(input, []byte("े")) ||
bytes.HasSuffix(input, []byte("ू")) ||
bytes.HasSuffix(input, []byte("ु")) ||
bytes.HasSuffix(input, []byte("ी")) ||
bytes.HasSuffix(input, []byte("ि")) ||
bytes.HasSuffix(input, []byte("ा"))) {
return analysis.TruncateRunes(input, 1)
}
return input
}