1dc466a800
often the result stream was the same length, so can reuse the existing token stream also, in cases where a new stream was required, set capacity to the length of the input stream. most output stream are at least as long as the input, so this may avoid some subsequent resizing
134 lines
3.2 KiB
Go
134 lines
3.2 KiB
Go
// 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 hi
|
|
|
|
import (
|
|
"bytes"
|
|
|
|
"github.com/blevesearch/bleve/analysis"
|
|
"github.com/blevesearch/bleve/registry"
|
|
)
|
|
|
|
const NormalizeName = "normalize_hi"
|
|
|
|
type HindiNormalizeFilter struct {
|
|
}
|
|
|
|
func NewHindiNormalizeFilter() *HindiNormalizeFilter {
|
|
return &HindiNormalizeFilter{}
|
|
}
|
|
|
|
func (s *HindiNormalizeFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
|
|
for _, token := range input {
|
|
term := normalize(token.Term)
|
|
token.Term = term
|
|
}
|
|
return input
|
|
}
|
|
|
|
func normalize(input []byte) []byte {
|
|
runes := bytes.Runes(input)
|
|
for i := 0; i < len(runes); i++ {
|
|
switch runes[i] {
|
|
// dead n -> bindu
|
|
case '\u0928':
|
|
if i+1 < len(runes) && runes[i+1] == '\u094D' {
|
|
runes[i] = '\u0902'
|
|
runes = analysis.DeleteRune(runes, i+1)
|
|
}
|
|
// candrabindu -> bindu
|
|
case '\u0901':
|
|
runes[i] = '\u0902'
|
|
// nukta deletions
|
|
case '\u093C':
|
|
runes = analysis.DeleteRune(runes, i)
|
|
i--
|
|
case '\u0929':
|
|
runes[i] = '\u0928'
|
|
case '\u0931':
|
|
runes[i] = '\u0930'
|
|
case '\u0934':
|
|
runes[i] = '\u0933'
|
|
case '\u0958':
|
|
runes[i] = '\u0915'
|
|
case '\u0959':
|
|
runes[i] = '\u0916'
|
|
case '\u095A':
|
|
runes[i] = '\u0917'
|
|
case '\u095B':
|
|
runes[i] = '\u091C'
|
|
case '\u095C':
|
|
runes[i] = '\u0921'
|
|
case '\u095D':
|
|
runes[i] = '\u0922'
|
|
case '\u095E':
|
|
runes[i] = '\u092B'
|
|
case '\u095F':
|
|
runes[i] = '\u092F'
|
|
// zwj/zwnj -> delete
|
|
case '\u200D', '\u200C':
|
|
runes = analysis.DeleteRune(runes, i)
|
|
i--
|
|
// virama -> delete
|
|
case '\u094D':
|
|
runes = analysis.DeleteRune(runes, i)
|
|
i--
|
|
// chandra/short -> replace
|
|
case '\u0945', '\u0946':
|
|
runes[i] = '\u0947'
|
|
case '\u0949', '\u094A':
|
|
runes[i] = '\u094B'
|
|
case '\u090D', '\u090E':
|
|
runes[i] = '\u090F'
|
|
case '\u0911', '\u0912':
|
|
runes[i] = '\u0913'
|
|
case '\u0972':
|
|
runes[i] = '\u0905'
|
|
// long -> short ind. vowels
|
|
case '\u0906':
|
|
runes[i] = '\u0905'
|
|
case '\u0908':
|
|
runes[i] = '\u0907'
|
|
case '\u090A':
|
|
runes[i] = '\u0909'
|
|
case '\u0960':
|
|
runes[i] = '\u090B'
|
|
case '\u0961':
|
|
runes[i] = '\u090C'
|
|
case '\u0910':
|
|
runes[i] = '\u090F'
|
|
case '\u0914':
|
|
runes[i] = '\u0913'
|
|
// long -> short dep. vowels
|
|
case '\u0940':
|
|
runes[i] = '\u093F'
|
|
case '\u0942':
|
|
runes[i] = '\u0941'
|
|
case '\u0944':
|
|
runes[i] = '\u0943'
|
|
case '\u0963':
|
|
runes[i] = '\u0962'
|
|
case '\u0948':
|
|
runes[i] = '\u0947'
|
|
case '\u094C':
|
|
runes[i] = '\u094B'
|
|
}
|
|
}
|
|
return analysis.BuildTermFromRunes(runes)
|
|
}
|
|
|
|
func NormalizerFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
|
|
return NewHindiNormalizeFilter(), nil
|
|
}
|
|
|
|
func init() {
|
|
registry.RegisterTokenFilter(NormalizeName, NormalizerFilterConstructor)
|
|
}
|