gibheer
/
bleve
0
0
Fork 0
Code

add analyzers for several languages 

Having pure Go snowball stemmers allows us to add support for
many languages into the core of bleve.  Specifically we just
added: Russian, Danish, Finnish, Hungarian, Dutch, Norwegian,
Romanian, Swedish, Turkish
This commit is contained in:
Marty Schoch 2018-01-10 16:00:29 -05:00
parent e68b70aa82
commit 09a61a7a38
40 changed files with 3199 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

56
analysis/lang/da/analyzer_da.go Normal file
View File

@ -0,0 +1,56 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package da
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/analysis/token/lowercase"
"github.com/blevesearch/bleve/analysis/tokenizer/unicode"
"github.com/blevesearch/bleve/registry"
)
const AnalyzerName = "da"
func AnalyzerConstructor(config map[string]interface{}, cache *registry.Cache) (*analysis.Analyzer, error) {
unicodeTokenizer, err := cache.TokenizerNamed(unicode.Name)
if err != nil {
return nil, err
}
toLowerFilter, err := cache.TokenFilterNamed(lowercase.Name)
if err != nil {
return nil, err
}
stopDaFilter, err := cache.TokenFilterNamed(StopName)
if err != nil {
return nil, err
}
stemmerDaFilter, err := cache.TokenFilterNamed(SnowballStemmerName)
if err != nil {
return nil, err
}
rv := analysis.Analyzer{
Tokenizer: unicodeTokenizer,
TokenFilters: []analysis.TokenFilter{
toLowerFilter,
stopDaFilter,
stemmerDaFilter,
},
}
return &rv, nil
}
func init() {
registry.RegisterAnalyzer(AnalyzerName, AnalyzerConstructor)
}

71
analysis/lang/da/analyzer_da_test.go Normal file
View File

@ -0,0 +1,71 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package da
import (
"reflect"
"testing"
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
func TestDanishAnalyzer(t *testing.T) {
tests := []struct {
input []byte
output analysis.TokenStream
}{
// stemming
{
input: []byte("undersøg"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("undersøg"),
Position: 1,
Start: 0,
End: 9,
},
},
},
{
input: []byte("undersøgelse"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("undersøg"),
Position: 1,
Start: 0,
End: 13,
},
},
},
// stop word
{
input: []byte("på"),
output: analysis.TokenStream{},
},
}
cache := registry.NewCache()
analyzer, err := cache.AnalyzerNamed(AnalyzerName)
if err != nil {
t.Fatal(err)
}
for _, test := range tests {
actual := analyzer.Analyze(test.input)
if !reflect.DeepEqual(actual, test.output) {
t.Errorf("expected %v, got %v", test.output, actual)
}
}
}

49
analysis/lang/da/stemmer_da.go Normal file
View File

@ -0,0 +1,49 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package da
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/snowballstem"
"github.com/blevesearch/snowballstem/danish"
)
const SnowballStemmerName = "stemmer_da_snowball"
type DanishStemmerFilter struct {
}
func NewDanishStemmerFilter() *DanishStemmerFilter {
return &DanishStemmerFilter{}
}
func (s *DanishStemmerFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
for _, token := range input {
env := snowballstem.NewEnv(string(token.Term))
danish.Stem(env)
token.Term = []byte(env.Current())
}
return input
}
func DanishStemmerFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
return NewDanishStemmerFilter(), nil
}
func init() {
registry.RegisterTokenFilter(SnowballStemmerName, DanishStemmerFilterConstructor)
}

33
analysis/lang/da/stop_filter_da.go Normal file
View File

@ -0,0 +1,33 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package da
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/analysis/token/stop"
"github.com/blevesearch/bleve/registry"
)
func StopTokenFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
tokenMap, err := cache.TokenMapNamed(StopName)
if err != nil {
return nil, err
}
return stop.NewStopTokensFilter(tokenMap), nil
}
func init() {
registry.RegisterTokenFilter(StopName, StopTokenFilterConstructor)
}

134
analysis/lang/da/stop_words_da.go Normal file
View File

@ -0,0 +1,134 @@
package da
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
const StopName = "stop_da"
// this content was obtained from:
// lucene-4.7.2/analysis/common/src/resources/org/apache/lucene/analysis/snowball/
// ` was changed to ' to allow for literal string
var DanishStopWords = []byte(` | From svn.tartarus.org/snowball/trunk/website/algorithms/danish/stop.txt
| This file is distributed under the BSD License.
| See http://snowball.tartarus.org/license.php
| Also see http://www.opensource.org/licenses/bsd-license.html
| - Encoding was converted to UTF-8.
| - This notice was added.
|
| NOTE: To use this file with StopFilterFactory, you must specify format="snowball"
| A Danish stop word list. Comments begin with vertical bar. Each stop
| word is at the start of a line.
| This is a ranked list (commonest to rarest) of stopwords derived from
| a large text sample.
og | and
i | in
jeg | I
det | that (dem. pronoun)/it (pers. pronoun)
at | that (in front of a sentence)/to (with infinitive)
en | a/an
den | it (pers. pronoun)/that (dem. pronoun)
til | to/at/for/until/against/by/of/into, more
er | present tense of "to be"
som | who, as
på | on/upon/in/on/at/to/after/of/with/for, on
de | they
med | with/by/in, along
han | he
af | of/by/from/off/for/in/with/on, off
for | at/for/to/from/by/of/ago, in front/before, because
ikke | not
der | who/which, there/those
var | past tense of "to be"
mig | me/myself
sig | oneself/himself/herself/itself/themselves
men | but
et | a/an/one, one (number), someone/somebody/one
har | present tense of "to have"
om | round/about/for/in/a, about/around/down, if
vi | we
min | my
havde | past tense of "to have"
ham | him
hun | she
nu | now
over | over/above/across/by/beyond/past/on/about, over/past
da | then, when/as/since
fra | from/off/since, off, since
du | you
ud | out
sin | his/her/its/one's
dem | them
os | us/ourselves
op | up
man | you/one
hans | his
hvor | where
eller | or
hvad | what
skal | must/shall etc.
selv | myself/youself/herself/ourselves etc., even
her | here
alle | all/everyone/everybody etc.
vil | will (verb)
blev | past tense of "to stay/to remain/to get/to become"
kunne | could
ind | in
når | when
være | present tense of "to be"
dog | however/yet/after all
noget | something
ville | would
jo | you know/you see (adv), yes
deres | their/theirs
efter | after/behind/according to/for/by/from, later/afterwards
ned | down
skulle | should
denne | this
end | than
dette | this
mit | my/mine
også | also
under | under/beneath/below/during, below/underneath
have | have
dig | you
anden | other
hende | her
mine | my
alt | everything
meget | much/very, plenty of
sit | his, her, its, one's
sine | his, her, its, one's
vor | our
mod | against
disse | these
hvis | if
din | your/yours
nogle | some
hos | by/at
blive | be/become
mange | many
ad | by/through
bliver | present tense of "to be/to become"
hendes | her/hers
været | be
thi | for (conj)
jer | you
sådan | such, like this/like that
`)
func TokenMapConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenMap, error) {
rv := analysis.NewTokenMap()
err := rv.LoadBytes(DanishStopWords)
return rv, err
}
func init() {
registry.RegisterTokenMap(StopName, TokenMapConstructor)
}

57
analysis/lang/fi/analyzer_fi.go Normal file
View File

@ -0,0 +1,57 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package fi
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/bleve/analysis/token/lowercase"
"github.com/blevesearch/bleve/analysis/tokenizer/unicode"
)
const AnalyzerName = "fi"
func AnalyzerConstructor(config map[string]interface{}, cache *registry.Cache) (*analysis.Analyzer, error) {
unicodeTokenizer, err := cache.TokenizerNamed(unicode.Name)
if err != nil {
return nil, err
}
toLowerFilter, err := cache.TokenFilterNamed(lowercase.Name)
if err != nil {
return nil, err
}
stopFiFilter, err := cache.TokenFilterNamed(StopName)
if err != nil {
return nil, err
}
stemmerFiFilter, err := cache.TokenFilterNamed(SnowballStemmerName)
if err != nil {
return nil, err
}
rv := analysis.Analyzer{
Tokenizer: unicodeTokenizer,
TokenFilters: []analysis.TokenFilter{
toLowerFilter,
stopFiFilter,
stemmerFiFilter,
},
}
return &rv, nil
}
func init() {
registry.RegisterAnalyzer(AnalyzerName, AnalyzerConstructor)
}

70
analysis/lang/fi/analyzer_fi_test.go Normal file
View File

@ -0,0 +1,70 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package fi
import (
"reflect"
"testing"
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
func TestFinishAnalyzer(t *testing.T) {
tests := []struct {
input []byte
output analysis.TokenStream
}{
// stemming
{
input: []byte("edeltäjiinsä"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("edeltäj"),
},
},
},
{
input: []byte("edeltäjistään"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("edeltäj"),
},
},
},
// stop word
{
input: []byte("olla"),
output: analysis.TokenStream{},
},
}
cache := registry.NewCache()
analyzer, err := cache.AnalyzerNamed(AnalyzerName)
if err != nil {
t.Fatal(err)
}
for _, test := range tests {
actual := analyzer.Analyze(test.input)
if len(actual) != len(test.output) {
t.Fatalf("expected length: %d, got %d", len(test.output), len(actual))
}
for i, tok := range actual {
if !reflect.DeepEqual(tok.Term, test.output[i].Term) {
t.Errorf("expected term %s (% x) got %s (% x)", test.output[i].Term, test.output[i].Term, tok.Term, tok.Term)
}
}
}
}

49
analysis/lang/fi/stemmer_fi.go Normal file
View File

@ -0,0 +1,49 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package fi
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/snowballstem"
"github.com/blevesearch/snowballstem/finnish"
)
const SnowballStemmerName = "stemmer_fi_snowball"
type FinnishStemmerFilter struct {
}
func NewFinnishStemmerFilter() *FinnishStemmerFilter {
return &FinnishStemmerFilter{}
}
func (s *FinnishStemmerFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
for _, token := range input {
env := snowballstem.NewEnv(string(token.Term))
finnish.Stem(env)
token.Term = []byte(env.Current())
}
return input
}
func FinnishStemmerFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
return NewFinnishStemmerFilter(), nil
}
func init() {
registry.RegisterTokenFilter(SnowballStemmerName, FinnishStemmerFilterConstructor)
}

33
analysis/lang/fi/stop_filter_fi.go Normal file
View File

@ -0,0 +1,33 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package fi
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/analysis/token/stop"
"github.com/blevesearch/bleve/registry"
)
func StopTokenFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
tokenMap, err := cache.TokenMapNamed(StopName)
if err != nil {
return nil, err
}
return stop.NewStopTokensFilter(tokenMap), nil
}
func init() {
registry.RegisterTokenFilter(StopName, StopTokenFilterConstructor)
}

121
analysis/lang/fi/stop_words_fi.go Normal file
View File

@ -0,0 +1,121 @@
package fi
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
const StopName = "stop_fi"
// this content was obtained from:
// lucene-4.7.2/analysis/common/src/resources/org/apache/lucene/analysis/snowball/
// ` was changed to ' to allow for literal string
var FinnishStopWords = []byte(` | From svn.tartarus.org/snowball/trunk/website/algorithms/finnish/stop.txt
| This file is distributed under the BSD License.
| See http://snowball.tartarus.org/license.php
| Also see http://www.opensource.org/licenses/bsd-license.html
| - Encoding was converted to UTF-8.
| - This notice was added.
|
| NOTE: To use this file with StopFilterFactory, you must specify format="snowball"
| forms of BE
olla
olen
olet
on
olemme
olette
ovat
ole | negative form
oli
olisi
olisit
olisin
olisimme
olisitte
olisivat
olit
olin
olimme
olitte
olivat
ollut
olleet
en | negation
et
ei
emme
ette
eivät
|Nom Gen Acc Part Iness Elat Illat Adess Ablat Allat Ess Trans
minä minun minut minua minussa minusta minuun minulla minulta minulle | I
sinä sinun sinut sinua sinussa sinusta sinuun sinulla sinulta sinulle | you
hän hänen hänet häntä hänessä hänestä häneen hänellä häneltä hänelle | he she
me meidän meidät meitä meissä meistä meihin meillä meiltä meille | we
te teidän teidät teitä teissä teistä teihin teillä teiltä teille | you
he heidän heidät heitä heissä heistä heihin heillä heiltä heille | they
tämä tämän tätä tässä tästä tähän tallä tältä tälle tänä täksi | this
tuo tuon tuotä tuossa tuosta tuohon tuolla tuolta tuolle tuona tuoksi | that
se sen sitä siinä siitä siihen sillä siltä sille sinä siksi | it
nämä näiden näitä näissä näistä näihin näillä näiltä näille näinä näiksi | these
nuo noiden noita noissa noista noihin noilla noilta noille noina noiksi | those
ne niiden niitä niissä niistä niihin niillä niiltä niille niinä niiksi | they
kuka kenen kenet ketä kenessä kenestä keneen kenellä keneltä kenelle kenenä keneksi| who
ketkä keiden ketkä keitä keissä keistä keihin keillä keiltä keille keinä keiksi | (pl)
mikä minkä minkä mitä missä mistä mihin millä miltä mille minä miksi | which what
mitkä | (pl)
joka jonka jota jossa josta johon jolla jolta jolle jona joksi | who which
jotka joiden joita joissa joista joihin joilla joilta joille joina joiksi | (pl)
| conjunctions
että | that
ja | and
jos | if
koska | because
kuin | than
mutta | but
niin | so
sekä | and
sillä | for
tai | or
vaan | but
vai | or
vaikka | although
| prepositions
kanssa | with
mukaan | according to
noin | about
poikki | across
yli | over, across
| other
kun | when
niin | so
nyt | now
itse | self
`)
func TokenMapConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenMap, error) {
rv := analysis.NewTokenMap()
err := rv.LoadBytes(FinnishStopWords)
return rv, err
}
func init() {
registry.RegisterTokenMap(StopName, TokenMapConstructor)
}

57
analysis/lang/hu/analyzer_hu.go Normal file
View File

@ -0,0 +1,57 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package hu
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/bleve/analysis/token/lowercase"
"github.com/blevesearch/bleve/analysis/tokenizer/unicode"
)
const AnalyzerName = "hu"
func AnalyzerConstructor(config map[string]interface{}, cache *registry.Cache) (*analysis.Analyzer, error) {
unicodeTokenizer, err := cache.TokenizerNamed(unicode.Name)
if err != nil {
return nil, err
}
toLowerFilter, err := cache.TokenFilterNamed(lowercase.Name)
if err != nil {
return nil, err
}
stopHuFilter, err := cache.TokenFilterNamed(StopName)
if err != nil {
return nil, err
}
stemmerHuFilter, err := cache.TokenFilterNamed(SnowballStemmerName)
if err != nil {
return nil, err
}
rv := analysis.Analyzer{
Tokenizer: unicodeTokenizer,
TokenFilters: []analysis.TokenFilter{
toLowerFilter,
stopHuFilter,
stemmerHuFilter,
},
}
return &rv, nil
}
func init() {
registry.RegisterAnalyzer(AnalyzerName, AnalyzerConstructor)
}

70
analysis/lang/hu/analyzer_hu_test.go Normal file
View File

@ -0,0 +1,70 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package hu
import (
"reflect"
"testing"
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
func TestHungarianAnalyzer(t *testing.T) {
tests := []struct {
input []byte
output analysis.TokenStream
}{
// stemming
{
input: []byte("babakocsi"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("babakocs"),
},
},
},
{
input: []byte("babakocsijáért"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("babakocs"),
},
},
},
// stop word
{
input: []byte("által"),
output: analysis.TokenStream{},
},
}
cache := registry.NewCache()
analyzer, err := cache.AnalyzerNamed(AnalyzerName)
if err != nil {
t.Fatal(err)
}
for _, test := range tests {
actual := analyzer.Analyze(test.input)
if len(actual) != len(test.output) {
t.Fatalf("expected length: %d, got %d", len(test.output), len(actual))
}
for i, tok := range actual {
if !reflect.DeepEqual(tok.Term, test.output[i].Term) {
t.Errorf("expected term %s (% x) got %s (% x)", test.output[i].Term, test.output[i].Term, tok.Term, tok.Term)
}
}
}
}

49
analysis/lang/hu/stemmer_hu.go Normal file
View File

@ -0,0 +1,49 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package hu
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/snowballstem"
"github.com/blevesearch/snowballstem/hungarian"
)
const SnowballStemmerName = "stemmer_hu_snowball"
type HungarianStemmerFilter struct {
}
func NewHungarianStemmerFilter() *HungarianStemmerFilter {
return &HungarianStemmerFilter{}
}
func (s *HungarianStemmerFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
for _, token := range input {
env := snowballstem.NewEnv(string(token.Term))
hungarian.Stem(env)
token.Term = []byte(env.Current())
}
return input
}
func HungarianStemmerFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
return NewHungarianStemmerFilter(), nil
}
func init() {
registry.RegisterTokenFilter(SnowballStemmerName, HungarianStemmerFilterConstructor)
}

33
analysis/lang/hu/stop_filter_hu.go Normal file
View File

@ -0,0 +1,33 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package hu
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/analysis/token/stop"
"github.com/blevesearch/bleve/registry"
)
func StopTokenFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
tokenMap, err := cache.TokenMapNamed(StopName)
if err != nil {
return nil, err
}
return stop.NewStopTokensFilter(tokenMap), nil
}
func init() {
registry.RegisterTokenFilter(StopName, StopTokenFilterConstructor)
}

235
analysis/lang/hu/stop_words_hu.go Normal file
View File

@ -0,0 +1,235 @@
package hu
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
const StopName = "stop_hu"
// this content was obtained from:
// lucene-4.7.2/analysis/common/src/resources/org/apache/lucene/analysis/snowball/
// ` was changed to ' to allow for literal string
var HungarianStopWords = []byte(` | From svn.tartarus.org/snowball/trunk/website/algorithms/hungarian/stop.txt
| This file is distributed under the BSD License.
| See http://snowball.tartarus.org/license.php
| Also see http://www.opensource.org/licenses/bsd-license.html
| - Encoding was converted to UTF-8.
| - This notice was added.
|
| NOTE: To use this file with StopFilterFactory, you must specify format="snowball"
| Hungarian stop word list
| prepared by Anna Tordai
a
ahogy
ahol
aki
akik
akkor
alatt
által
általában
amely
amelyek
amelyekben
amelyeket
amelyet
amelynek
ami
amit
amolyan
amíg
amikor
át
abban
ahhoz
annak
arra
arról
az
azok
azon
azt
azzal
azért
aztán
azután
azonban
bár
be
belül
benne
cikk
cikkek
cikkeket
csak
de
e
eddig
egész
egy
egyes
egyetlen
egyéb
egyik
egyre
ekkor
el
elég
ellen
elő
először
előtt
első
én
éppen
ebben
ehhez
emilyen
ennek
erre
ez
ezt
ezek
ezen
ezzel
ezért
és
fel
felé
hanem
hiszen
hogy
hogyan
igen
így
illetve
ill.
ill
ilyen
ilyenkor
ison
ismét
itt
jó
jól
jobban
kell
kellett
keresztül
keressünk
ki
kívül
között
közül
legalább
lehet
lehetett
legyen
lenne
lenni
lesz
lett
maga
magát
majd
majd
már
más
másik
meg
még
mellett
mert
mely
melyek
mi
mit
míg
miért
milyen
mikor
minden
mindent
mindenki
mindig
mint
mintha
mivel
most
nagy
nagyobb
nagyon
ne
néha
nekem
neki
nem
néhány
nélkül
nincs
olyan
ott
össze
ő
ők
őket
pedig
persze
rá
s
saját
sem
semmi
sok
sokat
sokkal
számára
szemben
szerint
szinte
talán
tehát
teljes
tovább
továbbá
több
úgy
ugyanis
új
újabb
újra
után
utána
utolsó
vagy
vagyis
valaki
valami
valamint
való
vagyok
van
vannak
volt
voltam
voltak
voltunk
vissza
vele
viszont
volna
`)
func TokenMapConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenMap, error) {
rv := analysis.NewTokenMap()
err := rv.LoadBytes(HungarianStopWords)
return rv, err
}
func init() {
registry.RegisterTokenMap(StopName, TokenMapConstructor)
}

57
analysis/lang/nl/analyzer_nl.go Normal file
View File

@ -0,0 +1,57 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package nl
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/bleve/analysis/token/lowercase"
"github.com/blevesearch/bleve/analysis/tokenizer/unicode"
)
const AnalyzerName = "nl"
func AnalyzerConstructor(config map[string]interface{}, cache *registry.Cache) (*analysis.Analyzer, error) {
unicodeTokenizer, err := cache.TokenizerNamed(unicode.Name)
if err != nil {
return nil, err
}
toLowerFilter, err := cache.TokenFilterNamed(lowercase.Name)
if err != nil {
return nil, err
}
stopNlFilter, err := cache.TokenFilterNamed(StopName)
if err != nil {
return nil, err
}
stemmerNlFilter, err := cache.TokenFilterNamed(SnowballStemmerName)
if err != nil {
return nil, err
}
rv := analysis.Analyzer{
Tokenizer: unicodeTokenizer,
TokenFilters: []analysis.TokenFilter{
toLowerFilter,
stopNlFilter,
stemmerNlFilter,
},
}
return &rv, nil
}
func init() {
registry.RegisterAnalyzer(AnalyzerName, AnalyzerConstructor)
}

70
analysis/lang/nl/analyzer_nl_test.go Normal file
View File

@ -0,0 +1,70 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package nl
import (
"reflect"
"testing"
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
func TestDutchAnalyzer(t *testing.T) {
tests := []struct {
input []byte
output analysis.TokenStream
}{
// stemming
{
input: []byte("lichamelijk"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("licham"),
},
},
},
{
input: []byte("lichamelijke"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("licham"),
},
},
},
// stop word
{
input: []byte("van"),
output: analysis.TokenStream{},
},
}
cache := registry.NewCache()
analyzer, err := cache.AnalyzerNamed(AnalyzerName)
if err != nil {
t.Fatal(err)
}
for _, test := range tests {
actual := analyzer.Analyze(test.input)
if len(actual) != len(test.output) {
t.Fatalf("expected length: %d, got %d", len(test.output), len(actual))
}
for i, tok := range actual {
if !reflect.DeepEqual(tok.Term, test.output[i].Term) {
t.Errorf("expected term %s (% x) got %s (% x)", test.output[i].Term, test.output[i].Term, tok.Term, tok.Term)
}
}
}
}

49
analysis/lang/nl/stemmer_nl.go Normal file
View File

@ -0,0 +1,49 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package nl
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/snowballstem"
"github.com/blevesearch/snowballstem/dutch"
)
const SnowballStemmerName = "stemmer_nl_snowball"
type DutchStemmerFilter struct {
}
func NewDutchStemmerFilter() *DutchStemmerFilter {
return &DutchStemmerFilter{}
}
func (s *DutchStemmerFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
for _, token := range input {
env := snowballstem.NewEnv(string(token.Term))
dutch.Stem(env)
token.Term = []byte(env.Current())
}
return input
}
func DutchStemmerFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
return NewDutchStemmerFilter(), nil
}
func init() {
registry.RegisterTokenFilter(SnowballStemmerName, DutchStemmerFilterConstructor)
}

33
analysis/lang/nl/stop_filter_nl.go Normal file
View File

@ -0,0 +1,33 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package nl
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/analysis/token/stop"
"github.com/blevesearch/bleve/registry"
)
func StopTokenFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
tokenMap, err := cache.TokenMapNamed(StopName)
if err != nil {
return nil, err
}
return stop.NewStopTokensFilter(tokenMap), nil
}
func init() {
registry.RegisterTokenFilter(StopName, StopTokenFilterConstructor)
}

143
analysis/lang/nl/stop_words_nl.go Normal file
View File

@ -0,0 +1,143 @@
package nl
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
const StopName = "stop_nl"
// this content was obtained from:
// lucene-4.7.2/analysis/common/src/resources/org/apache/lucene/analysis/snowball/
// ` was changed to ' to allow for literal string
var DutchStopWords = []byte(` | From svn.tartarus.org/snowball/trunk/website/algorithms/dutch/stop.txt
| This file is distributed under the BSD License.
| See http://snowball.tartarus.org/license.php
| Also see http://www.opensource.org/licenses/bsd-license.html
| - Encoding was converted to UTF-8.
| - This notice was added.
|
| NOTE: To use this file with StopFilterFactory, you must specify format="snowball"
| A Dutch stop word list. Comments begin with vertical bar. Each stop
| word is at the start of a line.
| This is a ranked list (commonest to rarest) of stopwords derived from
| a large sample of Dutch text.
| Dutch stop words frequently exhibit homonym clashes. These are indicated
| clearly below.
de | the
en | and
van | of, from
ik | I, the ego
te | (1) chez, at etc, (2) to, (3) too
dat | that, which
die | that, those, who, which
in | in, inside
een | a, an, one
hij | he
het | the, it
niet | not, nothing, naught
zijn | (1) to be, being, (2) his, one's, its
is | is
was | (1) was, past tense of all persons sing. of 'zijn' (to be) (2) wax, (3) the washing, (4) rise of river
op | on, upon, at, in, up, used up
aan | on, upon, to (as dative)
met | with, by
als | like, such as, when
voor | (1) before, in front of, (2) furrow
had | had, past tense all persons sing. of 'hebben' (have)
er | there
maar | but, only
om | round, about, for etc
hem | him
dan | then
zou | should/would, past tense all persons sing. of 'zullen'
of | or, whether, if
wat | what, something, anything
mijn | possessive and noun 'mine'
men | people, 'one'
dit | this
zo | so, thus, in this way
door | through by
over | over, across
ze | she, her, they, them
zich | oneself
bij | (1) a bee, (2) by, near, at
ook | also, too
tot | till, until
je | you
mij | me
uit | out of, from
der | Old Dutch form of 'van der' still found in surnames
daar | (1) there, (2) because
haar | (1) her, their, them, (2) hair
naar | (1) unpleasant, unwell etc, (2) towards, (3) as
heb | present first person sing. of 'to have'
hoe | how, why
heeft | present third person sing. of 'to have'
hebben | 'to have' and various parts thereof
deze | this
u | you
want | (1) for, (2) mitten, (3) rigging
nog | yet, still
zal | 'shall', first and third person sing. of verb 'zullen' (will)
me | me
zij | she, they
nu | now
ge | 'thou', still used in Belgium and south Netherlands
geen | none
omdat | because
iets | something, somewhat
worden | to become, grow, get
toch | yet, still
al | all, every, each
waren | (1) 'were' (2) to wander, (3) wares, (3)
veel | much, many
meer | (1) more, (2) lake
doen | to do, to make
toen | then, when
moet | noun 'spot/mote' and present form of 'to must'
ben | (1) am, (2) 'are' in interrogative second person singular of 'to be'
zonder | without
kan | noun 'can' and present form of 'to be able'
hun | their, them
dus | so, consequently
alles | all, everything, anything
onder | under, beneath
ja | yes, of course
eens | once, one day
hier | here
wie | who
werd | imperfect third person sing. of 'become'
altijd | always
doch | yet, but etc
wordt | present third person sing. of 'become'
wezen | (1) to be, (2) 'been' as in 'been fishing', (3) orphans
kunnen | to be able
ons | us/our
zelf | self
tegen | against, towards, at
na | after, near
reeds | already
wil | (1) present tense of 'want', (2) 'will', noun, (3) fender
kon | could; past tense of 'to be able'
niets | nothing
uw | your
iemand | somebody
geweest | been; past participle of 'be'
andere | other
`)
func TokenMapConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenMap, error) {
rv := analysis.NewTokenMap()
err := rv.LoadBytes(DutchStopWords)
return rv, err
}
func init() {
registry.RegisterTokenMap(StopName, TokenMapConstructor)
}

57
analysis/lang/no/analyzer_no.go Normal file
View File

@ -0,0 +1,57 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package no
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/bleve/analysis/token/lowercase"
"github.com/blevesearch/bleve/analysis/tokenizer/unicode"
)
const AnalyzerName = "no"
func AnalyzerConstructor(config map[string]interface{}, cache *registry.Cache) (*analysis.Analyzer, error) {
unicodeTokenizer, err := cache.TokenizerNamed(unicode.Name)
if err != nil {
return nil, err
}
toLowerFilter, err := cache.TokenFilterNamed(lowercase.Name)
if err != nil {
return nil, err
}
stopNoFilter, err := cache.TokenFilterNamed(StopName)
if err != nil {
return nil, err
}
stemmerNoFilter, err := cache.TokenFilterNamed(SnowballStemmerName)
if err != nil {
return nil, err
}
rv := analysis.Analyzer{
Tokenizer: unicodeTokenizer,
TokenFilters: []analysis.TokenFilter{
toLowerFilter,
stopNoFilter,
stemmerNoFilter,
},
}
return &rv, nil
}
func init() {
registry.RegisterAnalyzer(AnalyzerName, AnalyzerConstructor)
}

70
analysis/lang/no/analyzer_no_test.go Normal file
View File

@ -0,0 +1,70 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package no
import (
"reflect"
"testing"
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
func TestNorwegianAnalyzer(t *testing.T) {
tests := []struct {
input []byte
output analysis.TokenStream
}{
// stemming
{
input: []byte("havnedistriktene"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("havnedistrikt"),
},
},
},
{
input: []byte("havnedistrikter"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("havnedistrikt"),
},
},
},
// stop word
{
input: []byte("det"),
output: analysis.TokenStream{},
},
}
cache := registry.NewCache()
analyzer, err := cache.AnalyzerNamed(AnalyzerName)
if err != nil {
t.Fatal(err)
}
for _, test := range tests {
actual := analyzer.Analyze(test.input)
if len(actual) != len(test.output) {
t.Fatalf("expected length: %d, got %d", len(test.output), len(actual))
}
for i, tok := range actual {
if !reflect.DeepEqual(tok.Term, test.output[i].Term) {
t.Errorf("expected term %s (% x) got %s (% x)", test.output[i].Term, test.output[i].Term, tok.Term, tok.Term)
}
}
}
}

49
analysis/lang/no/stemmer_no.go Normal file
View File

@ -0,0 +1,49 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package no
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/snowballstem"
"github.com/blevesearch/snowballstem/norwegian"
)
const SnowballStemmerName = "stemmer_no_snowball"
type NorwegianStemmerFilter struct {
}
func NewNorwegianStemmerFilter() *NorwegianStemmerFilter {
return &NorwegianStemmerFilter{}
}
func (s *NorwegianStemmerFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
for _, token := range input {
env := snowballstem.NewEnv(string(token.Term))
norwegian.Stem(env)
token.Term = []byte(env.Current())
}
return input
}
func NorwegianStemmerFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
return NewNorwegianStemmerFilter(), nil
}
func init() {
registry.RegisterTokenFilter(SnowballStemmerName, NorwegianStemmerFilterConstructor)
}

33
analysis/lang/no/stop_filter_no.go Normal file
View File

@ -0,0 +1,33 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package no
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/analysis/token/stop"
"github.com/blevesearch/bleve/registry"
)
func StopTokenFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
tokenMap, err := cache.TokenMapNamed(StopName)
if err != nil {
return nil, err
}
return stop.NewStopTokensFilter(tokenMap), nil
}
func init() {
registry.RegisterTokenFilter(StopName, StopTokenFilterConstructor)
}

218
analysis/lang/no/stop_words_no.go Normal file
View File

@ -0,0 +1,218 @@
package no
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
const StopName = "stop_no"
// this content was obtained from:
// lucene-4.7.2/analysis/common/src/resources/org/apache/lucene/analysis/snowball/
// ` was changed to ' to allow for literal string
var NorwegianStopWords = []byte(` | From svn.tartarus.org/snowball/trunk/website/algorithms/norwegian/stop.txt
| This file is distributed under the BSD License.
| See http://snowball.tartarus.org/license.php
| Also see http://www.opensource.org/licenses/bsd-license.html
| - Encoding was converted to UTF-8.
| - This notice was added.
|
| NOTE: To use this file with StopFilterFactory, you must specify format="snowball"
| A Norwegian stop word list. Comments begin with vertical bar. Each stop
| word is at the start of a line.
| This stop word list is for the dominant bokmål dialect. Words unique
| to nynorsk are marked *.
| Revised by Jan Bruusgaard <Jan.Bruusgaard@ssb.no>, Jan 2005
og | and
i | in
jeg | I
det | it/this/that
at | to (w. inf.)
en | a/an
et | a/an
den | it/this/that
til | to
er | is/am/are
som | who/that
på | on
de | they / you(formal)
med | with
han | he
av | of
ikke | not
ikkje | not *
der | there
så | so
var | was/were
meg | me
seg | you
men | but
ett | one
har | have
om | about
vi | we
min | my
mitt | my
ha | have
hadde | had
hun | she
nå | now
over | over
da | when/as
ved | by/know
fra | from
du | you
ut | out
sin | your
dem | them
oss | us
opp | up
man | you/one
kan | can
hans | his
hvor | where
eller | or
hva | what
skal | shall/must
selv | self (reflective)
sjøl | self (reflective)
her | here
alle | all
vil | will
bli | become
ble | became
blei | became *
blitt | have become
kunne | could
inn | in
når | when
være | be
kom | come
noen | some
noe | some
ville | would
dere | you
som | who/which/that
deres | their/theirs
kun | only/just
ja | yes
etter | after
ned | down
skulle | should
denne | this
for | for/because
deg | you
si | hers/his
sine | hers/his
sitt | hers/his
mot | against
å | to
meget | much
hvorfor | why
dette | this
disse | these/those
uten | without
hvordan | how
ingen | none
din | your
ditt | your
blir | become
samme | same
hvilken | which
hvilke | which (plural)
sånn | such a
inni | inside/within
mellom | between
vår | our
hver | each
hvem | who
vors | us/ours
hvis | whose
både | both
bare | only/just
enn | than
fordi | as/because
før | before
mange | many
også | also
slik | just
vært | been
være | to be
båe | both *
begge | both
siden | since
dykk | your *
dykkar | yours *
dei | they *
deira | them *
deires | theirs *
deim | them *
di | your (fem.) *
då | as/when *
eg | I *
ein | a/an *
eit | a/an *
eitt | a/an *
elles | or *
honom | he *
hjå | at *
ho | she *
hoe | she *
henne | her
hennar | her/hers
hennes | hers
hoss | how *
hossen | how *
ikkje | not *
ingi | noone *
inkje | noone *
korleis | how *
korso | how *
kva | what/which *
kvar | where *
kvarhelst | where *
kven | who/whom *
kvi | why *
kvifor | why *
me | we *
medan | while *
mi | my *
mine | my *
mykje | much *
no | now *
nokon | some (masc./neut.) *
noka | some (fem.) *
nokor | some *
noko | some *
nokre | some *
si | his/hers *
sia | since *
sidan | since *
so | so *
somt | some *
somme | some *
um | about*
upp | up *
vere | be *
vore | was *
verte | become *
vort | become *
varte | became *
vart | became *
`)
func TokenMapConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenMap, error) {
rv := analysis.NewTokenMap()
err := rv.LoadBytes(NorwegianStopWords)
return rv, err
}
func init() {
registry.RegisterTokenMap(StopName, TokenMapConstructor)
}

57
analysis/lang/ro/analyzer_ro.go Normal file
View File

@ -0,0 +1,57 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package ro
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/bleve/analysis/token/lowercase"
"github.com/blevesearch/bleve/analysis/tokenizer/unicode"
)
const AnalyzerName = "ro"
func AnalyzerConstructor(config map[string]interface{}, cache *registry.Cache) (*analysis.Analyzer, error) {
unicodeTokenizer, err := cache.TokenizerNamed(unicode.Name)
if err != nil {
return nil, err
}
toLowerFilter, err := cache.TokenFilterNamed(lowercase.Name)
if err != nil {
return nil, err
}
stopRoFilter, err := cache.TokenFilterNamed(StopName)
if err != nil {
return nil, err
}
stemmerRoFilter, err := cache.TokenFilterNamed(SnowballStemmerName)
if err != nil {
return nil, err
}
rv := analysis.Analyzer{
Tokenizer: unicodeTokenizer,
TokenFilters: []analysis.TokenFilter{
toLowerFilter,
stopRoFilter,
stemmerRoFilter,
},
}
return &rv, nil
}
func init() {
registry.RegisterAnalyzer(AnalyzerName, AnalyzerConstructor)
}

70
analysis/lang/ro/analyzer_ro_test.go Normal file
View File

@ -0,0 +1,70 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package ro
import (
"reflect"
"testing"
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
func TestRomanianAnalyzer(t *testing.T) {
tests := []struct {
input []byte
output analysis.TokenStream
}{
// stemming
{
input: []byte("absenţa"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("absenţ"),
},
},
},
{
input: []byte("absenţi"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("absenţ"),
},
},
},
// stop word
{
input: []byte("îl"),
output: analysis.TokenStream{},
},
}
cache := registry.NewCache()
analyzer, err := cache.AnalyzerNamed(AnalyzerName)
if err != nil {
t.Fatal(err)
}
for _, test := range tests {
actual := analyzer.Analyze(test.input)
if len(actual) != len(test.output) {
t.Fatalf("expected length: %d, got %d", len(test.output), len(actual))
}
for i, tok := range actual {
if !reflect.DeepEqual(tok.Term, test.output[i].Term) {
t.Errorf("expected term %s (% x) got %s (% x)", test.output[i].Term, test.output[i].Term, tok.Term, tok.Term)
}
}
}
}

49
analysis/lang/ro/stemmer_ro.go Normal file
View File

@ -0,0 +1,49 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package ro
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/snowballstem"
"github.com/blevesearch/snowballstem/romanian"
)
const SnowballStemmerName = "stemmer_ro_snowball"
type RomanianStemmerFilter struct {
}
func NewRomanianStemmerFilter() *RomanianStemmerFilter {
return &RomanianStemmerFilter{}
}
func (s *RomanianStemmerFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
for _, token := range input {
env := snowballstem.NewEnv(string(token.Term))
romanian.Stem(env)
token.Term = []byte(env.Current())
}
return input
}
func RomanianStemmerFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
return NewRomanianStemmerFilter(), nil
}
func init() {
registry.RegisterTokenFilter(SnowballStemmerName, RomanianStemmerFilterConstructor)
}

33
analysis/lang/ro/stop_filter_ro.go Normal file
View File

@ -0,0 +1,33 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package ro
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/analysis/token/stop"
"github.com/blevesearch/bleve/registry"
)
func StopTokenFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
tokenMap, err := cache.TokenMapNamed(StopName)
if err != nil {
return nil, err
}
return stop.NewStopTokensFilter(tokenMap), nil
}
func init() {
registry.RegisterTokenFilter(StopName, StopTokenFilterConstructor)
}

257
analysis/lang/ro/stop_words_ro.go Normal file
View File

@ -0,0 +1,257 @@
package ro
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
const StopName = "stop_ro"
// this content was obtained from:
// lucene-4.7.2/analysis/common/src/resources/org/apache/lucene/analysis/
// ` was changed to ' to allow for literal string
var RomanianStopWords = []byte(`# This file was created by Jacques Savoy and is distributed under the BSD license.
# See http://members.unine.ch/jacques.savoy/clef/index.html.
# Also see http://www.opensource.org/licenses/bsd-license.html
acea
aceasta
această
aceea
acei
aceia
acel
acela
acele
acelea
acest
acesta
aceste
acestea
aceşti
aceştia
acolo
acum
ai
aia
aibă
aici
al
ăla
ale
alea
ălea
altceva
altcineva
am
ar
are
aş
aşadar
asemenea
asta
ăsta
astăzi
astea
ăstea
ăştia
asupra
aţi
au
avea
avem
aveţi
azi
bine
bucur
bună
ca
că
căci
când
care
cărei
căror
cărui
cât
câte
câţi
către
câtva
ce
cel
ceva
chiar
cînd
cine
cineva
cît
cîte
cîţi
cîtva
contra
cu
cum
cumva
curând
curînd
da
dă
dacă
dar
datorită
de
deci
deja
deoarece
departe
deşi
din
dinaintea
dintr
dintre
drept
după
ea
ei
el
ele
eram
este
eşti
eu
face
fără
fi
fie
fiecare
fii
fim
fiţi
iar
ieri
îi
îl
îmi
împotriva
în
înainte
înaintea
încât
încît
încotro
între
întrucât
întrucît
îţi
la
lângă
le
li
lîngă
lor
lui
mă
mâine
mea
mei
mele
mereu
meu
mi
mine
mult
multă
mulţi
ne
nicăieri
nici
nimeni
nişte
noastră
noastre
noi
noştri
nostru
nu
ori
oricând
oricare
oricât
orice
oricînd
oricine
oricît
oricum
oriunde
până
pe
pentru
peste
pînă
poate
pot
prea
prima
primul
prin
printr
sa
să
săi
sale
sau
său
se
şi
sînt
sîntem
sînteţi
spre
sub
sunt
suntem
sunteţi
ta
tăi
tale
tău
te
ţi
ţie
tine
toată
toate
tot
toţi
totuşi
tu
un
una
unde
undeva
unei
unele
uneori
unor
vă
vi
voastră
voastre
voi
voştri
vostru
vouă
vreo
vreun
`)
func TokenMapConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenMap, error) {
rv := analysis.NewTokenMap()
err := rv.LoadBytes(RomanianStopWords)
return rv, err
}
func init() {
registry.RegisterTokenMap(StopName, TokenMapConstructor)
}

57
analysis/lang/sv/analyzer_sv.go Normal file
View File

@ -0,0 +1,57 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package sv
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/bleve/analysis/token/lowercase"
"github.com/blevesearch/bleve/analysis/tokenizer/unicode"
)
const AnalyzerName = "sv"
func AnalyzerConstructor(config map[string]interface{}, cache *registry.Cache) (*analysis.Analyzer, error) {
unicodeTokenizer, err := cache.TokenizerNamed(unicode.Name)
if err != nil {
return nil, err
}
toLowerFilter, err := cache.TokenFilterNamed(lowercase.Name)
if err != nil {
return nil, err
}
stopSvFilter, err := cache.TokenFilterNamed(StopName)
if err != nil {
return nil, err
}
stemmerSvFilter, err := cache.TokenFilterNamed(SnowballStemmerName)
if err != nil {
return nil, err
}
rv := analysis.Analyzer{
Tokenizer: unicodeTokenizer,
TokenFilters: []analysis.TokenFilter{
toLowerFilter,
stopSvFilter,
stemmerSvFilter,
},
}
return &rv, nil
}
func init() {
registry.RegisterAnalyzer(AnalyzerName, AnalyzerConstructor)
}

70
analysis/lang/sv/analyzer_sv_test.go Normal file
View File

@ -0,0 +1,70 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package sv
import (
"reflect"
"testing"
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
func TestSwedishAnalyzer(t *testing.T) {
tests := []struct {
input []byte
output analysis.TokenStream
}{
// stemming
{
input: []byte("jaktkarlarne"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("jaktkarl"),
},
},
},
{
input: []byte("jaktkarlens"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("jaktkarl"),
},
},
},
// stop word
{
input: []byte("och"),
output: analysis.TokenStream{},
},
}
cache := registry.NewCache()
analyzer, err := cache.AnalyzerNamed(AnalyzerName)
if err != nil {
t.Fatal(err)
}
for _, test := range tests {
actual := analyzer.Analyze(test.input)
if len(actual) != len(test.output) {
t.Fatalf("expected length: %d, got %d", len(test.output), len(actual))
}
for i, tok := range actual {
if !reflect.DeepEqual(tok.Term, test.output[i].Term) {
t.Errorf("expected term %s (% x) got %s (% x)", test.output[i].Term, test.output[i].Term, tok.Term, tok.Term)
}
}
}
}

49
analysis/lang/sv/stemmer_sv.go Normal file
View File

@ -0,0 +1,49 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package sv
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/snowballstem"
"github.com/blevesearch/snowballstem/swedish"
)
const SnowballStemmerName = "stemmer_sv_snowball"
type SwedishStemmerFilter struct {
}
func NewSwedishStemmerFilter() *SwedishStemmerFilter {
return &SwedishStemmerFilter{}
}
func (s *SwedishStemmerFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
for _, token := range input {
env := snowballstem.NewEnv(string(token.Term))
swedish.Stem(env)
token.Term = []byte(env.Current())
}
return input
}
func SwedishStemmerFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
return NewSwedishStemmerFilter(), nil
}
func init() {
registry.RegisterTokenFilter(SnowballStemmerName, SwedishStemmerFilterConstructor)
}

33
analysis/lang/sv/stop_filter_sv.go Normal file
View File

@ -0,0 +1,33 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package sv
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/analysis/token/stop"
"github.com/blevesearch/bleve/registry"
)
func StopTokenFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
tokenMap, err := cache.TokenMapNamed(StopName)
if err != nil {
return nil, err
}
return stop.NewStopTokensFilter(tokenMap), nil
}
func init() {
registry.RegisterTokenFilter(StopName, StopTokenFilterConstructor)
}

157
analysis/lang/sv/stop_words_sv.go Normal file
View File

@ -0,0 +1,157 @@
package sv
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
const StopName = "stop_sv"
// this content was obtained from:
// lucene-4.7.2/analysis/common/src/resources/org/apache/lucene/analysis/snowball/
// ` was changed to ' to allow for literal string
var SwedishStopWords = []byte(` | From svn.tartarus.org/snowball/trunk/website/algorithms/swedish/stop.txt
| This file is distributed under the BSD License.
| See http://snowball.tartarus.org/license.php
| Also see http://www.opensource.org/licenses/bsd-license.html
| - Encoding was converted to UTF-8.
| - This notice was added.
|
| NOTE: To use this file with StopFilterFactory, you must specify format="snowball"
| A Swedish stop word list. Comments begin with vertical bar. Each stop
| word is at the start of a line.
| This is a ranked list (commonest to rarest) of stopwords derived from
| a large text sample.
| Swedish stop words occasionally exhibit homonym clashes. For example
| så = so, but also seed. These are indicated clearly below.
och | and
det | it, this/that
att | to (with infinitive)
i | in, at
en | a
jag | I
hon | she
som | who, that
han | he
på | on
den | it, this/that
med | with
var | where, each
sig | him(self) etc
för | for
så | so (also: seed)
till | to
är | is
men | but
ett | a
om | if; around, about
hade | had
de | they, these/those
av | of
icke | not, no
mig | me
du | you
henne | her
då | then, when
sin | his
nu | now
har | have
inte | inte någon = no one
hans | his
honom | him
skulle | 'sake'
hennes | her
där | there
min | my
man | one (pronoun)
ej | nor
vid | at, by, on (also: vast)
kunde | could
något | some etc
från | from, off
ut | out
när | when
efter | after, behind
upp | up
vi | we
dem | them
vara | be
vad | what
över | over
än | than
dig | you
kan | can
sina | his
här | here
ha | have
mot | towards
alla | all
under | under (also: wonder)
någon | some etc
eller | or (else)
allt | all
mycket | much
sedan | since
ju | why
denna | this/that
själv | myself, yourself etc
detta | this/that
åt | to
utan | without
varit | was
hur | how
ingen | no
mitt | my
ni | you
bli | to be, become
blev | from bli
oss | us
din | thy
dessa | these/those
några | some etc
deras | their
blir | from bli
mina | my
samma | (the) same
vilken | who, that
er | you, your
sådan | such a
vår | our
blivit | from bli
dess | its
inom | within
mellan | between
sådant | such a
varför | why
varje | each
vilka | who, that
ditt | thy
vem | who
vilket | who, that
sitta | his
sådana | such a
vart | each
dina | thy
vars | whose
vårt | our
våra | our
ert | your
era | your
vilkas | whose
`)
func TokenMapConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenMap, error) {
rv := analysis.NewTokenMap()
err := rv.LoadBytes(SwedishStopWords)
return rv, err
}
func init() {
registry.RegisterTokenMap(StopName, TokenMapConstructor)
}

63
analysis/lang/tr/analyzer_tr.go Normal file
View File

@ -0,0 +1,63 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package tr
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/bleve/analysis/token/apostrophe"
"github.com/blevesearch/bleve/analysis/token/lowercase"
"github.com/blevesearch/bleve/analysis/tokenizer/unicode"
)
const AnalyzerName = "tr"
func AnalyzerConstructor(config map[string]interface{}, cache *registry.Cache) (*analysis.Analyzer, error) {
unicodeTokenizer, err := cache.TokenizerNamed(unicode.Name)
if err != nil {
return nil, err
}
aposFilter, err := cache.TokenFilterNamed(apostrophe.Name)
if err != nil {
return nil, err
}
toLowerFilter, err := cache.TokenFilterNamed(lowercase.Name)
if err != nil {
return nil, err
}
stopTrFilter, err := cache.TokenFilterNamed(StopName)
if err != nil {
return nil, err
}
stemmerTrFilter, err := cache.TokenFilterNamed(SnowballStemmerName)
if err != nil {
return nil, err
}
rv := analysis.Analyzer{
Tokenizer: unicodeTokenizer,
TokenFilters: []analysis.TokenFilter{
aposFilter,
toLowerFilter,
stopTrFilter,
stemmerTrFilter,
},
}
return &rv, nil
}
func init() {
registry.RegisterAnalyzer(AnalyzerName, AnalyzerConstructor)
}

90
analysis/lang/tr/analyzer_tr_test.go Normal file
View File

@ -0,0 +1,90 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package tr
import (
"reflect"
"testing"
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
func TestTurkishAnalyzer(t *testing.T) {
tests := []struct {
input []byte
output analysis.TokenStream
}{
// stemming
{
input: []byte("ağacı"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("ağaç"),
},
},
},
{
input: []byte("ağaç"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("ağaç"),
},
},
},
// stop word
{
input: []byte("dolayı"),
output: analysis.TokenStream{},
},
// apostrophes
{
input: []byte("Kıbrıs'ta"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("kıbrıs"),
},
},
},
{
input: []byte("Van Gölü'ne"),
output: analysis.TokenStream{
&analysis.Token{
Term: []byte("van"),
},
&analysis.Token{
Term: []byte("göl"),
},
},
},
}
cache := registry.NewCache()
analyzer, err := cache.AnalyzerNamed(AnalyzerName)
if err != nil {
t.Fatal(err)
}
for _, test := range tests {
actual := analyzer.Analyze(test.input)
if len(actual) != len(test.output) {
t.Fatalf("expected length: %d, got %d", len(test.output), len(actual))
}
for i, tok := range actual {
if !reflect.DeepEqual(tok.Term, test.output[i].Term) {
t.Errorf("expected term %s (% x) got %s (% x)", test.output[i].Term, test.output[i].Term, tok.Term, tok.Term)
}
}
}
}

49
analysis/lang/tr/stemmer_tr.go Normal file
View File

@ -0,0 +1,49 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package tr
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
"github.com/blevesearch/snowballstem"
"github.com/blevesearch/snowballstem/turkish"
)
const SnowballStemmerName = "stemmer_tr_snowball"
type TurkishStemmerFilter struct {
}
func NewTurkishStemmerFilter() *TurkishStemmerFilter {
return &TurkishStemmerFilter{}
}
func (s *TurkishStemmerFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
for _, token := range input {
env := snowballstem.NewEnv(string(token.Term))
turkish.Stem(env)
token.Term = []byte(env.Current())
}
return input
}
func TurkishStemmerFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
return NewTurkishStemmerFilter(), nil
}
func init() {
registry.RegisterTokenFilter(SnowballStemmerName, TurkishStemmerFilterConstructor)
}

33
analysis/lang/tr/stop_filter_tr.go Normal file
View File

@ -0,0 +1,33 @@
// Copyright (c) 2018 Couchbase, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package tr
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/analysis/token/stop"
"github.com/blevesearch/bleve/registry"
)
func StopTokenFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
tokenMap, err := cache.TokenMapNamed(StopName)
if err != nil {
return nil, err
}
return stop.NewStopTokensFilter(tokenMap), nil
}
func init() {
registry.RegisterTokenFilter(StopName, StopTokenFilterConstructor)
}

236
analysis/lang/tr/stop_words_tr.go Normal file
View File

@ -0,0 +1,236 @@
package tr
import (
"github.com/blevesearch/bleve/analysis"
"github.com/blevesearch/bleve/registry"
)
const StopName = "stop_tr"
// this content was obtained from:
// lucene-4.7.2/analysis/common/src/resources/org/apache/lucene/analysis/snowball/
// ` was changed to ' to allow for literal string
var TurkishStopWords = []byte(`# Turkish stopwords from LUCENE-559
# merged with the list from "Information Retrieval on Turkish Texts"
# (http://www.users.muohio.edu/canf/papers/JASIST2008offPrint.pdf)
acaba
altmış
altı
ama
ancak
arada
aslında
ayrıca
bana
bazı
belki
ben
benden
beni
benim
beri
beş
bile
bin
bir
birçok
biri
birkaç
birkez
birşey
birşeyi
biz
bize
bizden
bizi
bizim
böyle
böylece
bu
buna
bunda
bundan
bunlar
bunları
bunların
bunu
bunun
burada
çok
çünkü
da
daha
dahi
de
defa
değil
diğer
diye
doksan
dokuz
dolayı
dolayısıyla
dört
edecek
eden
ederek
edilecek
ediliyor
edilmesi
ediyor
eğer
elli
en
etmesi
etti
ettiği
ettiğini
gibi
göre
halen
hangi
hatta
hem
henüz
hep
hepsi
her
herhangi
herkesin
hiç
hiçbir
için
iki
ile
ilgili
ise
işte
itibaren
itibariyle
kadar
karşın
katrilyon
kendi
kendilerine
kendini
kendisi
kendisine
kendisini
kez
ki
kim
kimden
kime
kimi
kimse
kırk
milyar
milyon
mu
mü
mı
nasıl
ne
neden
nedenle
nerde
nerede
nereye
niye
niçin
o
olan
olarak
oldu
olduğu
olduğunu
olduklarını
olmadı
olmadığı
olmak
olması
olmayan
olmaz
olsa
olsun
olup
olur
olursa
oluyor
on
ona
ondan
onlar
onlardan
onları
onların
onu
onun
otuz
oysa
öyle
pek
rağmen
sadece
sanki
sekiz
seksen
sen
senden
seni
senin
siz
sizden
sizi
sizin
şey
şeyden
şeyi
şeyler
şöyle
şu
şuna
şunda
şundan
şunları
şunu
tarafından
trilyon
tüm
üç
üzere
var
vardı
ve
veya
ya
yani
yapacak
yapılan
yapılması
yapıyor
yapmak
yaptı
yaptığı
yaptığını
yaptıkları
yedi
yerine
yetmiş
yine
yirmi
yoksa
yüz
zaten
`)
func TokenMapConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenMap, error) {
rv := analysis.NewTokenMap()
err := rv.LoadBytes(TurkishStopWords)
return rv, err
}
func init() {
registry.RegisterTokenMap(StopName, TokenMapConstructor)
}