2968d3538a
removed analyzers (these are now built as needed through config) removed html chacter filter (now built as needed through config) added missing license header changed constructor signature of filters that cannot return errors filter constructors that can have errors, now have Must variant which panics change cdl2 tokenizer into filter (should only see lower-case input) new top level index api, closes #5 refactored index tests to not rely directly on analyzers moved query objects to top-level new top level search api, closes #12 top score collector allows skipping results index mapping supports _all by default, closes #3 and closes #6 index mapping supports disabled sections, closes #7 new http sub package with reusable http.Handler's, closes #22
169 lines
4.0 KiB
Go
169 lines
4.0 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 search
|
|
|
|
import (
|
|
"math"
|
|
|
|
"github.com/couchbaselabs/bleve/index"
|
|
)
|
|
|
|
type PhraseSearcher struct {
|
|
index index.Index
|
|
mustSearcher *TermConjunctionSearcher
|
|
queryNorm float64
|
|
currMust *DocumentMatch
|
|
slop int
|
|
terms []string
|
|
}
|
|
|
|
func NewPhraseSearcher(index index.Index, mustSearcher *TermConjunctionSearcher, terms []string) (*PhraseSearcher, error) {
|
|
|
|
// build our searcher
|
|
rv := PhraseSearcher{
|
|
index: index,
|
|
mustSearcher: mustSearcher,
|
|
terms: terms,
|
|
}
|
|
rv.computeQueryNorm()
|
|
err := rv.initSearchers()
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
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
|
|
}
|
|
}
|
|
|
|
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() (*DocumentMatch, error) {
|
|
var rv *DocumentMatch
|
|
for s.currMust != nil {
|
|
rvftlm := make(FieldTermLocationMap, 0)
|
|
freq := 0
|
|
firstTerm := s.terms[0]
|
|
for field, termLocMap := range s.currMust.Locations {
|
|
rvtlm := make(TermLocationMap, 0)
|
|
locations, ok := termLocMap[firstTerm]
|
|
if ok {
|
|
OUTER:
|
|
for _, location := range locations {
|
|
crvtlm := make(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
|
|
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) (*DocumentMatch, error) {
|
|
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()
|
|
}
|
|
}
|