parent
9a777aaa80
commit
c19270108c
@ -0,0 +1,91 @@
|
||||
// Copyright (c) 2014 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 edge_ngram_filter
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"unicode/utf8"
|
||||
|
||||
"github.com/couchbaselabs/bleve/analysis"
|
||||
)
|
||||
|
||||
type Side bool
|
||||
|
||||
const BACK Side = true
|
||||
const FRONT Side = false
|
||||
|
||||
type EdgeNgramFilter struct {
|
||||
back Side
|
||||
minLength int
|
||||
maxLength int
|
||||
}
|
||||
|
||||
func NewEdgeNgramFilter(side Side, minLength, maxLength int) *EdgeNgramFilter {
|
||||
return &EdgeNgramFilter{
|
||||
back: side,
|
||||
minLength: minLength,
|
||||
maxLength: maxLength,
|
||||
}
|
||||
}
|
||||
|
||||
func (s *EdgeNgramFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
|
||||
rv := make(analysis.TokenStream, 0)
|
||||
|
||||
for _, token := range input {
|
||||
runeCount := utf8.RuneCount(token.Term)
|
||||
runes := bytes.Runes(token.Term)
|
||||
if s.back {
|
||||
i := runeCount
|
||||
// index of the starting rune for this token
|
||||
for ngramSize := s.minLength; ngramSize <= s.maxLength; ngramSize++ {
|
||||
// build an ngram of this size starting at i
|
||||
if i-ngramSize > 0 {
|
||||
ngramTerm := buildTermFromRunes(runes[i-ngramSize : i])
|
||||
token := analysis.Token{
|
||||
Position: token.Position,
|
||||
Start: token.Start,
|
||||
End: token.End,
|
||||
Type: token.Type,
|
||||
Term: ngramTerm,
|
||||
}
|
||||
rv = append(rv, &token)
|
||||
}
|
||||
}
|
||||
} else {
|
||||
i := 0
|
||||
// index of the starting rune for this token
|
||||
for ngramSize := s.minLength; ngramSize <= s.maxLength; ngramSize++ {
|
||||
// build an ngram of this size starting at i
|
||||
if i+ngramSize <= runeCount {
|
||||
ngramTerm := buildTermFromRunes(runes[i : i+ngramSize])
|
||||
token := analysis.Token{
|
||||
Position: token.Position,
|
||||
Start: token.Start,
|
||||
End: token.End,
|
||||
Type: token.Type,
|
||||
Term: ngramTerm,
|
||||
}
|
||||
rv = append(rv, &token)
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return rv
|
||||
}
|
||||
|
||||
func buildTermFromRunes(runes []rune) []byte {
|
||||
rv := make([]byte, 0, len(runes)*4)
|
||||
for _, r := range runes {
|
||||
runeBytes := make([]byte, utf8.RuneLen(r))
|
||||
utf8.EncodeRune(runeBytes, r)
|
||||
rv = append(rv, runeBytes...)
|
||||
}
|
||||
return rv
|
||||
}
|
@ -0,0 +1,141 @@
|
||||
// Copyright (c) 2014 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 edge_ngram_filter
|
||||
|
||||
import (
|
||||
"reflect"
|
||||
"testing"
|
||||
|
||||
"github.com/couchbaselabs/bleve/analysis"
|
||||
)
|
||||
|
||||
func TestEdgeNgramFilter(t *testing.T) {
|
||||
|
||||
tests := []struct {
|
||||
side Side
|
||||
min int
|
||||
max int
|
||||
input analysis.TokenStream
|
||||
output analysis.TokenStream
|
||||
}{
|
||||
{
|
||||
side: FRONT,
|
||||
min: 1,
|
||||
max: 1,
|
||||
input: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("abcde"),
|
||||
},
|
||||
},
|
||||
output: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("a"),
|
||||
},
|
||||
},
|
||||
},
|
||||
{
|
||||
side: BACK,
|
||||
min: 1,
|
||||
max: 1,
|
||||
input: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("abcde"),
|
||||
},
|
||||
},
|
||||
output: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("e"),
|
||||
},
|
||||
},
|
||||
},
|
||||
{
|
||||
side: FRONT,
|
||||
min: 1,
|
||||
max: 3,
|
||||
input: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("abcde"),
|
||||
},
|
||||
},
|
||||
output: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("a"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("ab"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("abc"),
|
||||
},
|
||||
},
|
||||
},
|
||||
{
|
||||
side: BACK,
|
||||
min: 1,
|
||||
max: 3,
|
||||
input: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("abcde"),
|
||||
},
|
||||
},
|
||||
output: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("e"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("de"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("cde"),
|
||||
},
|
||||
},
|
||||
},
|
||||
{
|
||||
side: FRONT,
|
||||
min: 1,
|
||||
max: 3,
|
||||
input: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("abcde"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("vwxyz"),
|
||||
},
|
||||
},
|
||||
output: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("a"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("ab"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("abc"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("v"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("vw"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("vwx"),
|
||||
},
|
||||
},
|
||||
},
|
||||
}
|
||||
|
||||
for _, test := range tests {
|
||||
edgeNgramFilter := NewEdgeNgramFilter(test.side, test.min, test.max)
|
||||
actual := edgeNgramFilter.Filter(test.input)
|
||||
if !reflect.DeepEqual(actual, test.output) {
|
||||
t.Errorf("expected %s, got %s", test.output, actual)
|
||||
}
|
||||
}
|
||||
}
|
66
analysis/token_filters/ngram_filter/ngram_filter.go
Normal file
66
analysis/token_filters/ngram_filter/ngram_filter.go
Normal file
@ -0,0 +1,66 @@
|
||||
// Copyright (c) 2014 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 ngram_filter
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"unicode/utf8"
|
||||
|
||||
"github.com/couchbaselabs/bleve/analysis"
|
||||
)
|
||||
|
||||
type NgramFilter struct {
|
||||
minLength int
|
||||
maxLength int
|
||||
}
|
||||
|
||||
func NewNgramFilter(minLength, maxLength int) *NgramFilter {
|
||||
return &NgramFilter{
|
||||
minLength: minLength,
|
||||
maxLength: maxLength,
|
||||
}
|
||||
}
|
||||
|
||||
func (s *NgramFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
|
||||
rv := make(analysis.TokenStream, 0)
|
||||
|
||||
for _, token := range input {
|
||||
runeCount := utf8.RuneCount(token.Term)
|
||||
runes := bytes.Runes(token.Term)
|
||||
for i := 0; i < runeCount; i++ {
|
||||
// index of the starting rune for this token
|
||||
for ngramSize := s.minLength; ngramSize <= s.maxLength; ngramSize++ {
|
||||
// build an ngram of this size starting at i
|
||||
if i+ngramSize <= runeCount {
|
||||
ngramTerm := buildTermFromRunes(runes[i : i+ngramSize])
|
||||
token := analysis.Token{
|
||||
Position: token.Position,
|
||||
Start: token.Start,
|
||||
End: token.End,
|
||||
Type: token.Type,
|
||||
Term: ngramTerm,
|
||||
}
|
||||
rv = append(rv, &token)
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return rv
|
||||
}
|
||||
|
||||
func buildTermFromRunes(runes []rune) []byte {
|
||||
rv := make([]byte, 0, len(runes)*4)
|
||||
for _, r := range runes {
|
||||
runeBytes := make([]byte, utf8.RuneLen(r))
|
||||
utf8.EncodeRune(runeBytes, r)
|
||||
rv = append(rv, runeBytes...)
|
||||
}
|
||||
return rv
|
||||
}
|
131
analysis/token_filters/ngram_filter/ngram_filter_test.go
Normal file
131
analysis/token_filters/ngram_filter/ngram_filter_test.go
Normal file
@ -0,0 +1,131 @@
|
||||
// Copyright (c) 2014 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 ngram_filter
|
||||
|
||||
import (
|
||||
"reflect"
|
||||
"testing"
|
||||
|
||||
"github.com/couchbaselabs/bleve/analysis"
|
||||
)
|
||||
|
||||
func TestNgramFilter(t *testing.T) {
|
||||
|
||||
tests := []struct {
|
||||
min int
|
||||
max int
|
||||
input analysis.TokenStream
|
||||
output analysis.TokenStream
|
||||
}{
|
||||
{
|
||||
min: 1,
|
||||
max: 1,
|
||||
input: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("abcde"),
|
||||
},
|
||||
},
|
||||
output: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("a"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("b"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("c"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("d"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("e"),
|
||||
},
|
||||
},
|
||||
},
|
||||
{
|
||||
min: 2,
|
||||
max: 2,
|
||||
input: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("abcde"),
|
||||
},
|
||||
},
|
||||
output: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("ab"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("bc"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("cd"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("de"),
|
||||
},
|
||||
},
|
||||
},
|
||||
{
|
||||
min: 1,
|
||||
max: 3,
|
||||
input: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("abcde"),
|
||||
},
|
||||
},
|
||||
output: analysis.TokenStream{
|
||||
&analysis.Token{
|
||||
Term: []byte("a"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("ab"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("abc"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("b"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("bc"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("bcd"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("c"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("cd"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("cde"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("d"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("de"),
|
||||
},
|
||||
&analysis.Token{
|
||||
Term: []byte("e"),
|
||||
},
|
||||
},
|
||||
},
|
||||
}
|
||||
|
||||
for _, test := range tests {
|
||||
ngramFilter := NewNgramFilter(test.min, test.max)
|
||||
actual := ngramFilter.Filter(test.input)
|
||||
if !reflect.DeepEqual(actual, test.output) {
|
||||
t.Errorf("expected %s, got %s", test.output, actual)
|
||||
}
|
||||
}
|
||||
}
|
Loading…
Reference in New Issue
Block a user