1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339
|
// 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 upsidedown
import (
"bytes"
"sort"
"sync/atomic"
"github.com/blevesearch/bleve/index"
"github.com/blevesearch/bleve/index/store"
)
type UpsideDownCouchTermFieldReader struct {
count uint64
indexReader *IndexReader
iterator store.KVIterator
term []byte
tfrNext *TermFrequencyRow
tfrPrealloc TermFrequencyRow
keyBuf []byte
field uint16
includeTermVectors bool
}
func newUpsideDownCouchTermFieldReader(indexReader *IndexReader, term []byte, field uint16, includeFreq, includeNorm, includeTermVectors bool) (*UpsideDownCouchTermFieldReader, error) {
bufNeeded := termFrequencyRowKeySize(term, nil)
if bufNeeded < dictionaryRowKeySize(term) {
bufNeeded = dictionaryRowKeySize(term)
}
buf := make([]byte, bufNeeded)
bufUsed := dictionaryRowKeyTo(buf, field, term)
val, err := indexReader.kvreader.Get(buf[:bufUsed])
if err != nil {
return nil, err
}
if val == nil {
atomic.AddUint64(&indexReader.index.stats.termSearchersStarted, uint64(1))
rv := &UpsideDownCouchTermFieldReader{
count: 0,
term: term,
field: field,
includeTermVectors: includeTermVectors,
}
rv.tfrNext = &rv.tfrPrealloc
return rv, nil
}
count, err := dictionaryRowParseV(val)
if err != nil {
return nil, err
}
bufUsed = termFrequencyRowKeyTo(buf, field, term, nil)
it := indexReader.kvreader.PrefixIterator(buf[:bufUsed])
atomic.AddUint64(&indexReader.index.stats.termSearchersStarted, uint64(1))
return &UpsideDownCouchTermFieldReader{
indexReader: indexReader,
iterator: it,
count: count,
term: term,
field: field,
includeTermVectors: includeTermVectors,
}, nil
}
func (r *UpsideDownCouchTermFieldReader) Count() uint64 {
return r.count
}
func (r *UpsideDownCouchTermFieldReader) Next(preAlloced *index.TermFieldDoc) (*index.TermFieldDoc, error) {
if r.iterator != nil {
// We treat tfrNext also like an initialization flag, which
// tells us whether we need to invoke the underlying
// iterator.Next(). The first time, don't call iterator.Next().
if r.tfrNext != nil {
r.iterator.Next()
} else {
r.tfrNext = &r.tfrPrealloc
}
key, val, valid := r.iterator.Current()
if valid {
tfr := r.tfrNext
err := tfr.parseKDoc(key, r.term)
if err != nil {
return nil, err
}
err = tfr.parseV(val, r.includeTermVectors)
if err != nil {
return nil, err
}
rv := preAlloced
if rv == nil {
rv = &index.TermFieldDoc{}
}
rv.ID = append(rv.ID, tfr.doc...)
rv.Freq = tfr.freq
rv.Norm = float64(tfr.norm)
if tfr.vectors != nil {
rv.Vectors = r.indexReader.index.termFieldVectorsFromTermVectors(tfr.vectors)
}
return rv, nil
}
}
return nil, nil
}
func (r *UpsideDownCouchTermFieldReader) Advance(docID index.IndexInternalID, preAlloced *index.TermFieldDoc) (rv *index.TermFieldDoc, err error) {
if r.iterator != nil {
if r.tfrNext == nil {
r.tfrNext = &TermFrequencyRow{}
}
tfr := InitTermFrequencyRow(r.tfrNext, r.term, r.field, docID, 0, 0)
r.keyBuf, err = tfr.KeyAppendTo(r.keyBuf[:0])
if err != nil {
return nil, err
}
r.iterator.Seek(r.keyBuf)
key, val, valid := r.iterator.Current()
if valid {
err := tfr.parseKDoc(key, r.term)
if err != nil {
return nil, err
}
err = tfr.parseV(val, r.includeTermVectors)
if err != nil {
return nil, err
}
rv = preAlloced
if rv == nil {
rv = &index.TermFieldDoc{}
}
rv.ID = append(rv.ID, tfr.doc...)
rv.Freq = tfr.freq
rv.Norm = float64(tfr.norm)
if tfr.vectors != nil {
rv.Vectors = r.indexReader.index.termFieldVectorsFromTermVectors(tfr.vectors)
}
return rv, nil
}
}
return nil, nil
}
func (r *UpsideDownCouchTermFieldReader) Close() error {
if r.indexReader != nil {
atomic.AddUint64(&r.indexReader.index.stats.termSearchersFinished, uint64(1))
}
if r.iterator != nil {
return r.iterator.Close()
}
return nil
}
type UpsideDownCouchDocIDReader struct {
indexReader *IndexReader
iterator store.KVIterator
only []string
onlyPos int
onlyMode bool
}
func newUpsideDownCouchDocIDReader(indexReader *IndexReader) (*UpsideDownCouchDocIDReader, error) {
startBytes := []byte{0x0}
endBytes := []byte{0xff}
bisr := NewBackIndexRow(startBytes, nil, nil)
bier := NewBackIndexRow(endBytes, nil, nil)
it := indexReader.kvreader.RangeIterator(bisr.Key(), bier.Key())
return &UpsideDownCouchDocIDReader{
indexReader: indexReader,
iterator: it,
}, nil
}
func newUpsideDownCouchDocIDReaderOnly(indexReader *IndexReader, ids []string) (*UpsideDownCouchDocIDReader, error) {
// we don't actually own the list of ids, so if before we sort we must copy
idsCopy := make([]string, len(ids))
copy(idsCopy, ids)
// ensure ids are sorted
sort.Strings(idsCopy)
startBytes := []byte{0x0}
if len(idsCopy) > 0 {
startBytes = []byte(idsCopy[0])
}
endBytes := []byte{0xff}
if len(idsCopy) > 0 {
endBytes = incrementBytes([]byte(idsCopy[len(idsCopy)-1]))
}
bisr := NewBackIndexRow(startBytes, nil, nil)
bier := NewBackIndexRow(endBytes, nil, nil)
it := indexReader.kvreader.RangeIterator(bisr.Key(), bier.Key())
return &UpsideDownCouchDocIDReader{
indexReader: indexReader,
iterator: it,
only: idsCopy,
onlyMode: true,
}, nil
}
func (r *UpsideDownCouchDocIDReader) Next() (index.IndexInternalID, error) {
key, val, valid := r.iterator.Current()
if r.onlyMode {
var rv index.IndexInternalID
for valid && r.onlyPos < len(r.only) {
br, err := NewBackIndexRowKV(key, val)
if err != nil {
return nil, err
}
if !bytes.Equal(br.doc, []byte(r.only[r.onlyPos])) {
ok := r.nextOnly()
if !ok {
return nil, nil
}
r.iterator.Seek(NewBackIndexRow([]byte(r.only[r.onlyPos]), nil, nil).Key())
key, val, valid = r.iterator.Current()
continue
} else {
rv = append([]byte(nil), br.doc...)
break
}
}
if valid && r.onlyPos < len(r.only) {
ok := r.nextOnly()
if ok {
r.iterator.Seek(NewBackIndexRow([]byte(r.only[r.onlyPos]), nil, nil).Key())
}
return rv, nil
}
} else {
if valid {
br, err := NewBackIndexRowKV(key, val)
if err != nil {
return nil, err
}
rv := append([]byte(nil), br.doc...)
r.iterator.Next()
return rv, nil
}
}
return nil, nil
}
func (r *UpsideDownCouchDocIDReader) Advance(docID index.IndexInternalID) (index.IndexInternalID, error) {
if r.onlyMode {
r.onlyPos = sort.SearchStrings(r.only, string(docID))
if r.onlyPos >= len(r.only) {
// advanced to key after our last only key
return nil, nil
}
r.iterator.Seek(NewBackIndexRow([]byte(r.only[r.onlyPos]), nil, nil).Key())
key, val, valid := r.iterator.Current()
var rv index.IndexInternalID
for valid && r.onlyPos < len(r.only) {
br, err := NewBackIndexRowKV(key, val)
if err != nil {
return nil, err
}
if !bytes.Equal(br.doc, []byte(r.only[r.onlyPos])) {
// the only key we seek'd to didn't exist
// now look for the closest key that did exist in only
r.onlyPos = sort.SearchStrings(r.only, string(br.doc))
if r.onlyPos >= len(r.only) {
// advanced to key after our last only key
return nil, nil
}
// now seek to this new only key
r.iterator.Seek(NewBackIndexRow([]byte(r.only[r.onlyPos]), nil, nil).Key())
key, val, valid = r.iterator.Current()
continue
} else {
rv = append([]byte(nil), br.doc...)
break
}
}
if valid && r.onlyPos < len(r.only) {
ok := r.nextOnly()
if ok {
r.iterator.Seek(NewBackIndexRow([]byte(r.only[r.onlyPos]), nil, nil).Key())
}
return rv, nil
}
} else {
bir := NewBackIndexRow(docID, nil, nil)
r.iterator.Seek(bir.Key())
key, val, valid := r.iterator.Current()
if valid {
br, err := NewBackIndexRowKV(key, val)
if err != nil {
return nil, err
}
rv := append([]byte(nil), br.doc...)
r.iterator.Next()
return rv, nil
}
}
return nil, nil
}
func (r *UpsideDownCouchDocIDReader) Close() error {
return r.iterator.Close()
}
// move the r.only pos forward one, skipping duplicates
// return true if there is more data, or false if we got to the end of the list
func (r *UpsideDownCouchDocIDReader) nextOnly() bool {
// advance 1 position, until we see a different key
// it's already sorted, so this skips duplicates
start := r.onlyPos
r.onlyPos++
for r.onlyPos < len(r.only) && r.only[r.onlyPos] == r.only[start] {
start = r.onlyPos
r.onlyPos++
}
// inidicate if we got to the end of the list
return r.onlyPos < len(r.only)
}
|