bugfixes more search results and dump restore options

vendor/github.com/blevesearch/bleve/analysis/byte_array_converters/json/json_byte_array_converter.go

@@ -1,42 +0,0 @@
-//  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 json_byte_array_converter
-
-import (
-	"encoding/json"
-
-	"github.com/blevesearch/bleve/analysis"
-	"github.com/blevesearch/bleve/registry"
-)
-
-const Name = "json"
-
-type JSONByteArrayConverter struct{}
-
-func NewJSONByteArrayConverter() *JSONByteArrayConverter {
-	return &JSONByteArrayConverter{}
-}
-
-func (c *JSONByteArrayConverter) Convert(in []byte) (interface{}, error) {
-	var rv map[string]interface{}
-	err := json.Unmarshal(in, &rv)
-	if err != nil {
-		return nil, err
-	}
-	return rv, nil
-}
-
-func Constructor(config map[string]interface{}, cache *registry.Cache) (analysis.ByteArrayConverter, error) {
-	return NewJSONByteArrayConverter(), nil
-}
-
-func init() {
-	registry.RegisterByteArrayConverter(Name, Constructor)
-}

vendor/github.com/blevesearch/bleve/analysis/language/en/possessive_filter_en.go

@@ -10,7 +10,7 @@
 package en
 
 import (
-	"bytes"
+	"unicode/utf8"
 
 	"github.com/blevesearch/bleve/analysis"
 	"github.com/blevesearch/bleve/registry"
@@ -40,15 +40,13 @@ func NewPossessiveFilter() *PossessiveFilter {
 
 func (s *PossessiveFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
 	for _, token := range input {
-		runes := bytes.Runes(token.Term)
-		if len(runes) >= 2 {
-			secondToLastRune := runes[len(runes)-2]
-			lastRune := runes[len(runes)-1]
-			if (secondToLastRune == rightSingleQuotationMark ||
-				secondToLastRune == apostrophe ||
-				secondToLastRune == fullWidthApostrophe) &&
-				(lastRune == 's' || lastRune == 'S') {
-				token.Term = analysis.TruncateRunes(token.Term, 2)
+		lastRune, lastRuneSize := utf8.DecodeLastRune(token.Term)
+		if lastRune == 's' || lastRune == 'S' {
+			nextLastRune, nextLastRuneSize := utf8.DecodeLastRune(token.Term[:len(token.Term)-lastRuneSize])
+			if nextLastRune == rightSingleQuotationMark ||
+				nextLastRune == apostrophe ||
+				nextLastRune == fullWidthApostrophe {
+				token.Term = token.Term[:len(token.Term)-lastRuneSize-nextLastRuneSize]
 			}
 		}
 	}

vendor/github.com/blevesearch/bleve/analysis/token_filters/porter/porter.go

@@ -10,6 +10,8 @@
 package porter
 
 import (
+	"bytes"
+
 	"github.com/blevesearch/bleve/analysis"
 	"github.com/blevesearch/bleve/registry"
 
@@ -29,8 +31,9 @@ func (s *PorterStemmer) Filter(input analysis.TokenStream) analysis.TokenStream
 	for _, token := range input {
 		// if it is not a protected keyword, stem it
 		if !token.KeyWord {
-			stemmed := porterstemmer.StemString(string(token.Term))
-			token.Term = []byte(stemmed)
+			termRunes := bytes.Runes(token.Term)
+			stemmedRunes := porterstemmer.StemWithoutLowerCasing(termRunes)
+			token.Term = analysis.BuildTermFromRunes(stemmedRunes)
 		}
 	}
 	return input

vendor/github.com/blevesearch/bleve/analysis/token_filters/stop_tokens_filter/stop_tokens_filter.go

@@ -36,16 +36,16 @@ func NewStopTokensFilter(stopTokens analysis.TokenMap) *StopTokensFilter {
 }
 
 func (f *StopTokensFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
-	rv := make(analysis.TokenStream, 0, len(input))
-
+	j := 0
 	for _, token := range input {
 		_, isStopToken := f.stopTokens[string(token.Term)]
 		if !isStopToken {
-			rv = append(rv, token)
+			input[j] = token
+			j++
 		}
 	}
 
-	return rv
+	return input[:j]
 }
 
 func StopTokensFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {

vendor/github.com/blevesearch/bleve/analysis/util.go

@@ -34,14 +34,32 @@ func InsertRune(in []rune, pos int, r rune) []rune {
 	return rv
 }
 
-func BuildTermFromRunes(runes []rune) []byte {
-	rv := make([]byte, 0, len(runes)*4)
+// BuildTermFromRunesOptimistic will build a term from the provided runes
+// AND optimistically attempt to encode into the provided buffer
+// if at any point it appears the buffer is too small, a new buffer is
+// allocated and that is used instead
+// this should be used in cases where frequently the new term is the same
+// length or shorter than the original term (in number of bytes)
+func BuildTermFromRunesOptimistic(buf []byte, runes []rune) []byte {
+	rv := buf
+	used := 0
 	for _, r := range runes {
-		runeBytes := make([]byte, utf8.RuneLen(r))
-		utf8.EncodeRune(runeBytes, r)
-		rv = append(rv, runeBytes...)
+		nextLen := utf8.RuneLen(r)
+		if used+nextLen > len(rv) {
+			// alloc new buf
+			buf = make([]byte, len(runes)*utf8.UTFMax)
+			// copy work we've already done
+			copy(buf, rv[:used])
+			rv = buf
+		}
+		written := utf8.EncodeRune(rv[used:], r)
+		used += written
 	}
-	return rv
+	return rv[:used]
+}
+
+func BuildTermFromRunes(runes []rune) []byte {
+	return BuildTermFromRunesOptimistic(make([]byte, len(runes)*utf8.UTFMax), runes)
 }
 
 func TruncateRunes(input []byte, num int) []byte {