This commit is contained in:
Alexander Kiryukhin 2018-05-10 03:00:52 +03:00
commit 6796db17de
No known key found for this signature in database
GPG key ID: 5579837FDBF65965
6 changed files with 49899 additions and 0 deletions

1
.gitignore vendored Normal file
View file

@ -0,0 +1 @@
.idea

21
LICENSE Normal file
View file

@ -0,0 +1,21 @@
MIT License
Copyright (c) 2018 Alexander Kiryukhin
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

32
README.md Normal file
View file

@ -0,0 +1,32 @@
# Стеммер Портера для русского языка
Стемминг - процесс получения основы слова из любой его формы. Иными словами, отсекает лишние суффиксы и окончания.
Самое очевидное применение - в полнотекстовом поиске, где нужно, чтобы слово находилось, даже если у него другое окончание.
Этот пакет - реализация [стеммера Портера](https://ru.wikipedia.org/wiki/Стемминг#Стеммерортера) для русского языка на Go.
## Использование
`основа := StemmerRu.StemWord("слово")`
Преобразует слово на входе в его основу на выходе
## Пример
```
package main
import (
"fmt"
"github.com/neonxp/StemmerRu"
)
func main() {
fmt.Println(StemmerRu.StemWord("безмолвны") // выведет: безмолвн
fmt.Println(StemmerRu.StemWord("безмолвные") // выведет: безмолвн
fmt.Println(StemmerRu.StemWord("безмолвный") // выведет: безмолвн
fmt.Println(StemmerRu.StemWord("безмолвным") // выведет: безмолвн
fmt.Println(StemmerRu.StemWord("безмолвных") // выведет: безмолвн
}
```

142
StemmerRu/stemmer.go Normal file
View file

@ -0,0 +1,142 @@
package StemmerRu
import (
"strings"
)
var (
perfectiveGerund = [][]string{{`в`, `вши`, `вшись`}, {`ив`, `ивши`, `ившись`, `ыв`, `ывши`, `ывшись`}}
adjective = []string{`ее`, `ие`, `ые`, `ое`, `ими`, `ыми`, `ей`, `ий`, `ый`, `ой`, `ем`, `им`, `ым`, `ом`, `его`, `ого`, `ему`, `ому`, `их`, `ых`, `ую`, `юю`, `ая`, `яя`, `ою`, `ею`}
participle = [][]string{{`ем`, `нн`, `вш`, `ющ`, `щ`}, {`ивш`, `ывш`, `ующ`}}
reflexive = []string{`ся`, `сь`}
verb = [][]string{{`ла`, `на`, `ете`, `йте`, `ли`, `й`, `л`, `ем`, `н`, `ло`, `но`, `ет`, `ют`, `ны`, `ть`, `ешь`, `нно`}, {`ила`, `ыла`, `ена`, `ейте`, `уйте`, `ите`, `или`, `ыли`, `ей`, `уй`, `ил`, `ыл`, `им`, `ым`, `ен`, `ило`, `ыло`, `ено`, `ят`, `ует`, `уют`, `ит`, `ыт`, `ены`, `ить`, `ыть`, `ишь`, `ую`, `ю`}}
noun = []string{`а`, `ев`, `ов`, `ие`, `ье`, `е`, `иями`, `ями`, `ами`, `еи`, `ии`, `и`, `ией`, `ей`, `ой`, `ий`, `й`, `иям`, `ям`, `ием`, `ем`, `ам`, `ом`, `о`, `у`, `ах`, `иях`, `ях`, `ы`, `ь`, `ию`, `ью`, `ю`, `ия`, `ья`, `я`}
superlative = []string{`ейш`, `ейше`}
derivational = []string{`ост`, `ость`}
vowels = `аеиоуыэюя`
)
func StemWord(word string) string {
word = strings.Replace(word, `ё`, `е`, -1)
RVpos := getRVPart(word)
if RVpos == -1 {
return word
}
R1pos := getRNPart(word, 0)
R2pos := getRNPart(word, R1pos)
if (R2pos < RVpos) {
R2pos = 0
} else {
R2pos -= RVpos
}
suffix := string([]rune(word)[RVpos:])
prefix := string([]rune(word)[:RVpos])
// Step 1
suffix, isTrimmed := trimSuffix(suffix, perfectiveGerund[1], perfectiveGerund[0])
if !isTrimmed {
suffix, isTrimmed = trimSuffix(suffix, reflexive, nil)
suffix, isTrimmed = trimAdjectival(suffix)
if !isTrimmed {
suffix, isTrimmed = trimSuffix(suffix, verb[1], verb[0])
if !isTrimmed {
suffix, _ = trimSuffix(suffix, noun, nil)
}
}
}
// Step 2
suffix = strings.TrimSuffix(suffix, `и`)
// Step 3
if R2pos < len([]rune(suffix)) {
R2suffix := string([]rune(suffix)[R2pos:])
R2prefix := string([]rune(suffix)[:R2pos])
R2suffix, _ = trimSuffix(R2suffix, derivational, nil)
suffix = R2prefix + R2suffix
}
// Step 4
suffix, isTrimmed = trimNN(suffix)
if !isTrimmed {
suffix, isTrimmed = trimSuffix(suffix, superlative, nil)
if isTrimmed {
suffix, _ = trimNN(suffix)
} else {
suffix = strings.TrimSuffix(suffix, `ь`)
}
}
return prefix + suffix
}
func trimNN(word string) (string, bool) {
if strings.HasSuffix(word, `нн`) {
return strings.TrimSuffix(word, `нн`) + `н`, true
}
return word, false
}
func trimAdjectival(word string) (string, bool) {
isTrimmedParticiple := false
word, isTrimmedAdjective := trimSuffix(word, adjective, nil)
if isTrimmedAdjective {
word, isTrimmedParticiple = trimSuffix(word, participle[1], participle[0])
}
return word, isTrimmedAdjective || isTrimmedParticiple
}
func trimSuffix(word string, suffixes []string, suffixes2 []string) (string, bool) {
for _, suffix := range suffixes {
if strings.HasSuffix(word, suffix) {
return strings.TrimSuffix(word, suffix), true
}
}
if suffixes2 != nil {
for _, suffix := range suffixes2 {
if strings.HasSuffix(word, `а`+suffix) ||
strings.HasSuffix(word, `я`+suffix) {
return strings.TrimSuffix(word, suffix), true
}
}
}
return word, false
}
func isVowel(char rune) bool {
return strings.Contains(vowels, string(char))
}
func getRVPart(word string) int {
chars := []rune(word)
for idx, char := range chars {
if isVowel(char) {
return idx + 1
}
}
return -1
}
func getRNPart(word string, startPos int) int {
chars := []rune(word)[startPos:]
for idx, char := range chars {
if idx+2 < len(chars) {
if isVowel(char) && !isVowel(chars[idx+1]) {
return startPos + idx + 2
}
}
}
return startPos
}

28
StemmerRu/stemmer_test.go Normal file
View file

@ -0,0 +1,28 @@
package StemmerRu
import (
"testing"
"io/ioutil"
"encoding/json"
"path"
)
var testFile = path.Join(`..`, `tests.json`)
func TestStemWord(t *testing.T) {
file, err := ioutil.ReadFile(testFile)
if err != nil {
t.Error("Can't open file", testFile)
}
tests := &map[string]string{}
err = json.Unmarshal(file, tests)
if err != nil {
t.Error("Can't parse json", err)
}
for source, expected := range *tests {
result := StemWord(source);
if expected != result {
t.Errorf(`Expected "%s" (source: %s) but got "%s"`, result, source, result)
}
}
}

49675
tests.json Normal file

File diff suppressed because it is too large Load diff