2ee7289bc8
this started initially to relocate highlighting into a self contained package, which would then also use the registry however, it turned into a much larger refactor in order to avoid cyclic imports now facets, searchers, scorers and collectors are also broken out into subpackages of search
184 lines
4.2 KiB
Go
184 lines
4.2 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 searchers
|
|
|
|
import (
|
|
"math"
|
|
|
|
"github.com/blevesearch/bleve/index"
|
|
"github.com/blevesearch/bleve/search"
|
|
)
|
|
|
|
type PhraseSearcher struct {
|
|
initialized bool
|
|
index index.Index
|
|
mustSearcher *ConjunctionSearcher
|
|
queryNorm float64
|
|
currMust *search.DocumentMatch
|
|
slop int
|
|
terms []string
|
|
}
|
|
|
|
func NewPhraseSearcher(index index.Index, mustSearcher *ConjunctionSearcher, terms []string) (*PhraseSearcher, error) {
|
|
|
|
// build our searcher
|
|
rv := PhraseSearcher{
|
|
index: index,
|
|
mustSearcher: mustSearcher,
|
|
terms: terms,
|
|
}
|
|
rv.computeQueryNorm()
|
|
return &rv, nil
|
|
}
|
|
|
|
func (s *PhraseSearcher) computeQueryNorm() {
|
|
// first calculate sum of squared weights
|
|
sumOfSquaredWeights := 0.0
|
|
if s.mustSearcher != nil {
|
|
sumOfSquaredWeights += s.mustSearcher.Weight()
|
|
}
|
|
|
|
// now compute query norm from this
|
|
s.queryNorm = 1.0 / math.Sqrt(sumOfSquaredWeights)
|
|
// finally tell all the downsteam searchers the norm
|
|
if s.mustSearcher != nil {
|
|
s.mustSearcher.SetQueryNorm(s.queryNorm)
|
|
}
|
|
}
|
|
|
|
func (s *PhraseSearcher) initSearchers() error {
|
|
var err error
|
|
// get all searchers pointing at their first match
|
|
if s.mustSearcher != nil {
|
|
s.currMust, err = s.mustSearcher.Next()
|
|
if err != nil {
|
|
return err
|
|
}
|
|
}
|
|
|
|
s.initialized = true
|
|
return nil
|
|
}
|
|
|
|
func (s *PhraseSearcher) advanceNextMust() error {
|
|
var err error
|
|
|
|
if s.mustSearcher != nil {
|
|
s.currMust, err = s.mustSearcher.Next()
|
|
if err != nil {
|
|
return err
|
|
}
|
|
}
|
|
|
|
return nil
|
|
}
|
|
|
|
func (s *PhraseSearcher) Weight() float64 {
|
|
var rv float64
|
|
rv += s.mustSearcher.Weight()
|
|
|
|
return rv
|
|
}
|
|
|
|
func (s *PhraseSearcher) SetQueryNorm(qnorm float64) {
|
|
s.mustSearcher.SetQueryNorm(qnorm)
|
|
}
|
|
|
|
func (s *PhraseSearcher) Next() (*search.DocumentMatch, error) {
|
|
if !s.initialized {
|
|
err := s.initSearchers()
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
}
|
|
|
|
var rv *search.DocumentMatch
|
|
for s.currMust != nil {
|
|
rvftlm := make(search.FieldTermLocationMap, 0)
|
|
freq := 0
|
|
firstTerm := s.terms[0]
|
|
for field, termLocMap := range s.currMust.Locations {
|
|
rvtlm := make(search.TermLocationMap, 0)
|
|
locations, ok := termLocMap[firstTerm]
|
|
if ok {
|
|
OUTER:
|
|
for _, location := range locations {
|
|
crvtlm := make(search.TermLocationMap, 0)
|
|
INNER:
|
|
for i := 0; i < len(s.mustSearcher.searchers); i++ {
|
|
nextTerm := s.terms[i]
|
|
if nextTerm != "" {
|
|
// look through all this terms locations
|
|
// to try and find the correct offsets
|
|
nextLocations, ok := termLocMap[nextTerm]
|
|
if ok {
|
|
for _, nextLocation := range nextLocations {
|
|
if nextLocation.Pos == location.Pos+float64(i) {
|
|
// found a location match for this term
|
|
crvtlm.AddLocation(nextTerm, nextLocation)
|
|
continue INNER
|
|
}
|
|
}
|
|
// if we got here we didnt find location match for this term
|
|
continue OUTER
|
|
} else {
|
|
continue OUTER
|
|
}
|
|
}
|
|
}
|
|
// if we got here all the terms matched
|
|
freq += 1
|
|
search.MergeTermLocationMaps(rvtlm, crvtlm)
|
|
rvftlm[field] = rvtlm
|
|
}
|
|
}
|
|
}
|
|
|
|
if freq > 0 {
|
|
// return match
|
|
rv = s.currMust
|
|
rv.Locations = rvftlm
|
|
s.advanceNextMust()
|
|
return rv, nil
|
|
}
|
|
|
|
s.advanceNextMust()
|
|
}
|
|
|
|
return nil, nil
|
|
}
|
|
|
|
func (s *PhraseSearcher) Advance(ID string) (*search.DocumentMatch, error) {
|
|
if !s.initialized {
|
|
err := s.initSearchers()
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
}
|
|
s.mustSearcher.Advance(ID)
|
|
return s.Next()
|
|
}
|
|
|
|
func (s *PhraseSearcher) Count() uint64 {
|
|
// for now return a worst case
|
|
var sum uint64 = 0
|
|
sum += s.mustSearcher.Count()
|
|
return sum
|
|
}
|
|
|
|
func (s *PhraseSearcher) Close() {
|
|
if s.mustSearcher != nil {
|
|
s.mustSearcher.Close()
|
|
}
|
|
}
|
|
|
|
func (s *PhraseSearcher) Min() int {
|
|
return 0
|
|
}
|