Golang SnowballWord.RemoveLastNRunes方法代码示例

// Remove adjectival endings and return true if one was removed.
//
func removeAdjectivalEnding(word *snowballword.SnowballWord) bool {

	// Remove adjectival endings.  Start by looking for
	// an adjective ending.
	//
	suffix, _ := word.RemoveFirstSuffixIn(word.RVstart,
		"ими", "ыми", "его", "ого", "ему", "ому", "ее", "ие",
		"ые", "ое", "ей", "ий", "ый", "ой", "ем", "им", "ым",
		"ом", "их", "ых", "ую", "юю", "ая", "яя", "ою", "ею",
	)
	if suffix != "" {

		// We found an adjective ending.  Remove optional participle endings.
		//
		newSuffix, newSuffixRunes := word.FirstSuffixIn(word.RVstart, len(word.RS),
			"ивш", "ывш", "ующ",
			"ем", "нн", "вш", "ющ", "щ",
		)
		switch newSuffix {
		case "ем", "нн", "вш", "ющ", "щ":

			// These are "Group 1" participle endings.
			// Group 1 endings must follow а (a) or я (ia) in RV.
			if precededByARinRV(word, len(newSuffixRunes)) == false {
				newSuffix = ""
			}
		}

		if newSuffix != "" {
			word.RemoveLastNRunes(len(newSuffixRunes))
		}
		return true
	}
	return false
}

// Remove verb endings and return true if one was removed.
//
func removeVerbEnding(word *snowballword.SnowballWord) bool {
	suffix, suffixRunes := word.FirstSuffixIn(word.RVstart, len(word.RS),
		"уйте", "ейте", "ыть", "ыло", "ыли", "ыла", "уют", "ует",
		"нно", "йте", "ишь", "ить", "ите", "ило", "или", "ила",
		"ешь", "ете", "ены", "ено", "ена", "ят", "ют", "ыт", "ым",
		"ыл", "ую", "уй", "ть", "ны", "но", "на", "ло", "ли", "ла",
		"ит", "им", "ил", "ет", "ен", "ем", "ей", "ю", "н", "л", "й",
	)
	switch suffix {
	case "ла", "на", "ете", "йте", "ли", "й", "л", "ем", "н",
		"ло", "но", "ет", "ют", "ны", "ть", "ешь", "нно":

		// These are "Group 1" verb endings.
		// Group 1 endings must follow а (a) or я (ia) in RV.
		if precededByARinRV(word, len(suffixRunes)) == false {
			suffix = ""
		}

	}

	if suffix != "" {
		word.RemoveLastNRunes(len(suffixRunes))
		return true
	}
	return false
}

// Step 3 is the removal of residual suffixes.
//
func step3(word *snowballword.SnowballWord) bool {
	suffix, suffixRunes := word.FirstSuffixIfIn(word.RVstart, len(word.RS),
		"os", "a", "o", "á", "í", "ó", "e", "é",
	)

	// No suffix found, nothing to do.
	//
	if suffix == "" {
		return false
	}

	// Remove all these suffixes
	word.RemoveLastNRunes(len(suffixRunes))

	if suffix == "e" || suffix == "é" {

		// If preceded by gu with the u in RV delete the u
		//
		guSuffix, _ := word.FirstSuffix("gu")
		if guSuffix != "" {
			word.RemoveLastNRunes(1)
		}
	}
	return true
}

// Step 0 is to strip off apostrophes and "s".
//
func step0(w *snowballword.SnowballWord) bool {
	suffix, suffixRunes := w.FirstSuffix("'s'", "'s", "'")
	if suffix == "" {
		return false
	}
	w.RemoveLastNRunes(len(suffixRunes))
	return true
}

// Step 5 Undouble non-vowel endings
//
func step5(word *snowballword.SnowballWord) bool {

	suffix, _ := word.FirstSuffix("enn", "onn", "ett", "ell", "eill")
	if suffix != "" {
		word.RemoveLastNRunes(1)
	}
	return false
}

// Step 2a is the removal of verb suffixes beginning y,
// Search for the longest among the following suffixes
// in RV, and if found, delete if preceded by u.
//
func step2a(word *snowballword.SnowballWord) bool {
	suffix, suffixRunes := word.FirstSuffixIn(word.RVstart, len(word.RS), "ya", "ye", "yan", "yen", "yeron", "yendo", "yo", "yó", "yas", "yes", "yais", "yamos")
	if suffix != "" {
		idx := len(word.RS) - len(suffixRunes) - 1
		if idx >= 0 && word.RS[idx] == 117 {
			word.RemoveLastNRunes(len(suffixRunes))
			return true
		}
	}
	return false
}

// Step 1a is normalization of various special "s"-endings.
//
func step1a(w *snowballword.SnowballWord) bool {

	suffix, suffixRunes := w.FirstSuffix("sses", "ied", "ies", "us", "ss", "s")
	switch suffix {

	case "sses":

		// Replace by ss
		w.ReplaceSuffixRunes(suffixRunes, []rune("ss"), true)
		return true

	case "ies", "ied":

		// Replace by i if preceded by more than one letter,
		// otherwise by ie (so ties -> tie, cries -> cri).

		var repl string
		if len(w.RS) > 4 {
			repl = "i"
		} else {
			repl = "ie"
		}
		w.ReplaceSuffixRunes(suffixRunes, []rune(repl), true)
		return true

	case "us", "ss":

		// Do nothing
		return false

	case "s":

		// Delete if the preceding word part contains a vowel
		// not immediately before the s (so gas and this retain
		// the s, gaps and kiwis lose it)
		//
		for i := 0; i < len(w.RS)-2; i++ {
			if isLowerVowel(w.RS[i]) {
				w.RemoveLastNRunes(len(suffixRunes))
				return true
			}
		}
	}
	return false
}

// Remove perfective gerund endings and return true if one was removed.
//
func removePerfectiveGerundEnding(word *snowballword.SnowballWord) bool {
	suffix, suffixRunes := word.FirstSuffixIn(word.RVstart, len(word.RS),
		"ившись", "ывшись", "вшись", "ивши", "ывши", "вши", "ив", "ыв", "в",
	)
	switch suffix {
	case "в", "вши", "вшись":

		// These are "Group 1" perfective gerund endings.
		// Group 1 endings must follow а (a) or я (ia) in RV.
		if precededByARinRV(word, len(suffixRunes)) == false {
			suffix = ""
		}

	}

	if suffix != "" {
		word.RemoveLastNRunes(len(suffixRunes))
		return true
	}
	return false
}

// Step 2b is the removal of Verb suffixes in RV
// that do not begin with "i".
//
func step2b(word *snowballword.SnowballWord) bool {

	// Search for the longest among the following suffixes in RV.
	//
	suffix, suffixRunes := word.FirstSuffixIn(word.RVstart, len(word.RS),
		"eraIent", "assions", "erions", "assiez", "assent",
		"èrent", "eront", "erons", "eriez", "erait", "erais",
		"asses", "antes", "aIent", "âtes", "âmes", "ions",
		"erez", "eras", "erai", "asse", "ants", "ante", "ées",
		"iez", "era", "ant", "ait", "ais", "és", "ée", "ât",
		"ez", "er", "as", "ai", "é", "a",
	)

	switch suffix {
	case "ions":

		// Delete if in R2
		suffixLen := len(suffixRunes)
		if word.FitsInR2(suffixLen) {
			word.RemoveLastNRunes(suffixLen)
			return true
		}
		return false

	case "é", "ée", "ées", "és", "èrent", "er", "era",
		"erai", "eraIent", "erais", "erait", "eras", "erez",
		"eriez", "erions", "erons", "eront", "ez", "iez":

		// Delete
		word.RemoveLastNRunes(len(suffixRunes))
		return true

	case "âmes", "ât", "âtes", "a", "ai", "aIent",
		"ais", "ait", "ant", "ante", "antes", "ants", "as",
		"asse", "assent", "asses", "assiez", "assions":

		// Delete
		word.RemoveLastNRunes(len(suffixRunes))

		// If preceded by e (unicode code point 101), delete
		//
		idx := len(word.RS) - 1
		if idx >= 0 && word.RS[idx] == 101 && word.FitsInRV(1) {
			word.RemoveLastNRunes(1)
		}
		return true

	}
	return false
}

// Step 4 is the undoubling of double non-vowel endings
// and removal of superlative endings.
//
func step4(word *snowballword.SnowballWord) bool {

	// (1) Undouble "н", or, 2) if the word ends with a SUPERLATIVE ending,
	// (remove it and undouble н n), or 3) if the word ends ь (') (soft sign)
	// remove it.

	// Undouble "н"
	if word.HasSuffixRunes([]rune("нн")) {
		word.RemoveLastNRunes(1)
		return true
	}

	// Remove superlative endings
	suffix, _ := word.RemoveFirstSuffix("ейше", "ейш")
	if suffix != "" {
		// Undouble "н"
		if word.HasSuffixRunes([]rune("нн")) {
			word.RemoveLastNRunes(1)
		}
		return true
	}

	// Remove soft sign
	if rsLen := len(word.RS); rsLen > 0 && word.RS[rsLen-1] == 'ь' {
		word.RemoveLastNRunes(1)
		return true
	}
	return false
}

// Step 1b is the normalization of various "ly" and "ed" sufficies.
//
func step1b(w *snowballword.SnowballWord) bool {

	suffix, suffixRunes := w.FirstSuffix("eedly", "ingly", "edly", "ing", "eed", "ed")

	switch suffix {

	case "":
		// No suffix found
		return false

	case "eed", "eedly":

		// Replace by ee if in R1
		if len(suffixRunes) <= len(w.RS)-w.R1start {
			w.ReplaceSuffixRunes(suffixRunes, []rune("ee"), true)
		}
		return true

	case "ed", "edly", "ing", "ingly":
		hasLowerVowel := false
		for i := 0; i < len(w.RS)-len(suffixRunes); i++ {
			if isLowerVowel(w.RS[i]) {
				hasLowerVowel = true
				break
			}
		}
		if hasLowerVowel {

			// This case requires a two-step transformation and, due
			// to the way we've implemented the `ReplaceSuffix` method
			// here, information about R1 and R2 would be lost between
			// the two.  Therefore, we need to keep track of the
			// original R1 & R2, so that we may set them below, at the
			// end of this case.
			//
			originalR1start := w.R1start
			originalR2start := w.R2start

			// Delete if the preceding word part contains a vowel
			w.RemoveLastNRunes(len(suffixRunes))

			// ...and after the deletion...

			newSuffix, newSuffixRunes := w.FirstSuffix("at", "bl", "iz", "bb", "dd", "ff", "gg", "mm", "nn", "pp", "rr", "tt")
			switch newSuffix {

			case "":

				// If the word is short, add "e"
				if isShortWord(w) {

					// By definition, r1 and r2 are the empty string for
					// short words.
					w.RS = append(w.RS, []rune("e")...)
					w.R1start = len(w.RS)
					w.R2start = len(w.RS)
					return true
				}

			case "at", "bl", "iz":

				// If the word ends "at", "bl" or "iz" add "e"
				w.ReplaceSuffixRunes(newSuffixRunes, []rune(newSuffix+"e"), true)

			case "bb", "dd", "ff", "gg", "mm", "nn", "pp", "rr", "tt":

				// If the word ends with a double remove the last letter.
				// Note that, "double" does not include all possible doubles,
				// just those shown above.
				//
				w.RemoveLastNRunes(1)
			}

			// Because we did a double replacement, we need to fix
			// R1 and R2 manually. This is just becase of how we've
			// implemented the `ReplaceSuffix` method.
			//
			rsLen := len(w.RS)
			if originalR1start < rsLen {
				w.R1start = originalR1start
			} else {
				w.R1start = rsLen
			}
			if originalR2start < rsLen {
				w.R2start = originalR2start
			} else {
				w.R2start = rsLen
			}

			return true
		}

	}

	return false
}

// Step 2 is the stemming of various endings found in
// R1 including "al", "ness", and "li".
//
func step2(w *snowballword.SnowballWord) bool {

	// Possible sufficies for this step, longest first.
	suffix, suffixRunes := w.FirstSuffix(
		"ational", "fulness", "iveness", "ization", "ousness",
		"biliti", "lessli", "tional", "alism", "aliti", "ation",
		"entli", "fulli", "iviti", "ousli", "anci", "abli",
		"alli", "ator", "enci", "izer", "bli", "ogi", "li",
	)

	// If it is not in R1, do nothing
	if suffix == "" || len(suffixRunes) > len(w.RS)-w.R1start {
		return false
	}

	// Handle special cases where we're not just going to
	// replace the suffix with another suffix: there are
	// other things we need to do.
	//
	switch suffix {

	case "li":

		// Delete if preceded by a valid li-ending. Valid li-endings inlude the
		// following charaters: cdeghkmnrt. (Note, the unicode code points for
		// these characters are, respectively, as follows:
		// 99 100 101 103 104 107 109 110 114 116)
		//
		rsLen := len(w.RS)
		if rsLen >= 3 {
			switch w.RS[rsLen-3] {
			case 99, 100, 101, 103, 104, 107, 109, 110, 114, 116:
				w.RemoveLastNRunes(len(suffixRunes))
				return true
			}
		}
		return false

	case "ogi":

		// Replace by og if preceded by l.
		// (Note, the unicode code point for l is 108)
		//
		rsLen := len(w.RS)
		if rsLen >= 4 && w.RS[rsLen-4] == 108 {
			w.ReplaceSuffixRunes(suffixRunes, []rune("og"), true)
		}
		return true
	}

	// Handle a suffix that was found, which is going
	// to be replaced with a different suffix.
	//
	var repl string
	switch suffix {
	case "tional":
		repl = "tion"
	case "enci":
		repl = "ence"
	case "anci":
		repl = "ance"
	case "abli":
		repl = "able"
	case "entli":
		repl = "ent"
	case "izer", "ization":
		repl = "ize"
	case "ational", "ation", "ator":
		repl = "ate"
	case "alism", "aliti", "alli":
		repl = "al"
	case "fulness":
		repl = "ful"
	case "ousli", "ousness":
		repl = "ous"
	case "iveness", "iviti":
		repl = "ive"
	case "biliti", "bli":
		repl = "ble"
	case "fulli":
		repl = "ful"
	case "lessli":
		repl = "less"
	}
	w.ReplaceSuffixRunes(suffixRunes, []rune(repl), true)
	return true

}

// Step 1 is the removal of standard suffixes
//
func step1(word *snowballword.SnowballWord) bool {

	// Possible suffixes, longest first
	suffix, suffixRunes := word.FirstSuffix(
		"amientos", "imientos", "aciones", "amiento", "imiento",
		"uciones", "logías", "idades", "encias", "ancias", "amente",
		"adores", "adoras", "ución", "mente", "logía", "istas",
		"ismos", "ibles", "encia", "anzas", "antes", "ancia",
		"adora", "ación", "ables", "osos", "osas", "ivos", "ivas",
		"ista", "ismo", "idad", "icos", "icas", "ible", "anza",
		"ante", "ador", "able", "oso", "osa", "ivo", "iva",
		"ico", "ica",
	)

	isInR1 := (word.R1start <= len(word.RS)-len(suffixRunes))
	isInR2 := (word.R2start <= len(word.RS)-len(suffixRunes))

	// Deal with special cases first.  All of these will
	// return if they are hit.
	//
	switch suffix {
	case "":

		// Nothing to do
		return false

	case "amente":

		if isInR1 {
			// Delete if in R1
			word.RemoveLastNRunes(len(suffixRunes))

			// if preceded by iv, delete if in R2 (and if further preceded by at,
			// delete if in R2), otherwise,
			// if preceded by os, ic or ad, delete if in R2
			newSuffix, _ := word.RemoveFirstSuffixIfIn(word.R2start, "iv", "os", "ic", "ad")
			if newSuffix == "iv" {
				word.RemoveFirstSuffixIfIn(word.R2start, "at")
			}
			return true
		}
		return false
	}

	// All the following cases require the found suffix
	// to be in R2.
	if isInR2 == false {
		return false
	}

	// Compound replacement cases.  All these cases return
	// if they are hit.
	//
	compoundReplacement := func(otherSuffixes ...string) bool {
		word.RemoveLastNRunes(len(suffixRunes))
		word.RemoveFirstSuffixIfIn(word.R2start, otherSuffixes...)
		return true
	}

	switch suffix {
	case "adora", "ador", "ación", "adoras", "adores", "aciones", "ante", "antes", "ancia", "ancias":
		return compoundReplacement("ic")
	case "mente":
		return compoundReplacement("ante", "able", "ible")
	case "idad", "idades":
		return compoundReplacement("abil", "ic", "iv")
	case "iva", "ivo", "ivas", "ivos":
		return compoundReplacement("at")
	}

	// Simple replacement & deletion cases are all that remain.
	//
	simpleReplacement := func(repl string) bool {
		word.ReplaceSuffixRunes(suffixRunes, []rune(repl), true)
		return true
	}
	switch suffix {
	case "logía", "logías":
		return simpleReplacement("log")
	case "ución", "uciones":
		return simpleReplacement("u")
	case "encia", "encias":
		return simpleReplacement("ente")
	case "anza", "anzas", "ico", "ica", "icos", "icas",
		"ismo", "ismos", "able", "ables", "ible", "ibles",
		"ista", "istas", "oso", "osa", "osos", "osas",
		"amiento", "amientos", "imiento", "imientos":
		word.RemoveLastNRunes(len(suffixRunes))
		return true
	}

	log.Panicln("Unhandled suffix:", suffix)
	return false
}