forked from apache/lucenenet
/
DocValuesConsumer.cs
540 lines (485 loc) · 21.7 KB
/
DocValuesConsumer.cs
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
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
using System;
using System.Collections.Generic;
using System.Diagnostics;
using System.Linq;
namespace Lucene.Net.Codecs
{
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You 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.
*/
using ArrayUtil = Lucene.Net.Util.ArrayUtil;
using AtomicReader = Lucene.Net.Index.AtomicReader;
using BinaryDocValues = Lucene.Net.Index.BinaryDocValues;
using IBits = Lucene.Net.Util.IBits;
using BytesRef = Lucene.Net.Util.BytesRef;
using FieldInfo = Lucene.Net.Index.FieldInfo;
using FilteredTermsEnum = Lucene.Net.Index.FilteredTermsEnum;
using Int64BitSet = Lucene.Net.Util.Int64BitSet;
using MergeState = Lucene.Net.Index.MergeState;
using NumericDocValues = Lucene.Net.Index.NumericDocValues;
using OrdinalMap = Lucene.Net.Index.MultiDocValues.OrdinalMap;
using SortedDocValues = Lucene.Net.Index.SortedDocValues;
using SortedSetDocValues = Lucene.Net.Index.SortedSetDocValues;
using TermsEnum = Lucene.Net.Index.TermsEnum;
/// <summary>
/// Abstract API that consumes numeric, binary and
/// sorted docvalues. Concrete implementations of this
/// actually do "something" with the docvalues (write it into
/// the index in a specific format).
/// <para/>
/// The lifecycle is:
/// <list type="number">
/// <item><description>DocValuesConsumer is created by
/// <see cref="DocValuesFormat.FieldsConsumer(Index.SegmentWriteState)"/> or
/// <see cref="NormsFormat.NormsConsumer(Index.SegmentWriteState)"/>.</description></item>
/// <item><description><see cref="AddNumericField(FieldInfo, IEnumerable{long?})"/>,
/// <see cref="AddBinaryField(FieldInfo, IEnumerable{BytesRef})"/>,
/// or <see cref="AddSortedField(FieldInfo, IEnumerable{BytesRef}, IEnumerable{long?})"/> are called for each Numeric,
/// Binary, or Sorted docvalues field. The API is a "pull" rather
/// than "push", and the implementation is free to iterate over the
/// values multiple times (<see cref="IEnumerable{T}.GetEnumerator()"/>).</description></item>
/// <item><description>After all fields are added, the consumer is <see cref="Dispose()"/>d.</description></item>
/// </list>
/// <para/>
/// @lucene.experimental
/// </summary>
public abstract class DocValuesConsumer : IDisposable
{
/// <summary>
/// Sole constructor. (For invocation by subclass
/// constructors, typically implicit.)
/// </summary>
protected internal DocValuesConsumer()
{
}
/// <summary>
/// Writes numeric docvalues for a field. </summary>
/// <param name="field"> Field information. </param>
/// <param name="values"> <see cref="IEnumerable{T}"/> of numeric values (one for each document). <c>null</c> indicates
/// a missing value. </param>
/// <exception cref="System.IO.IOException"> If an I/O error occurred. </exception>
public abstract void AddNumericField(FieldInfo field, IEnumerable<long?> values);
/// <summary>
/// Writes binary docvalues for a field. </summary>
/// <param name="field"> Field information. </param>
/// <param name="values"> <see cref="IEnumerable{T}"/> of binary values (one for each document). <c>null</c> indicates
/// a missing value. </param>
/// <exception cref="System.IO.IOException"> If an I/O error occurred. </exception>
public abstract void AddBinaryField(FieldInfo field, IEnumerable<BytesRef> values);
/// <summary>
/// Writes pre-sorted binary docvalues for a field. </summary>
/// <param name="field"> Field information. </param>
/// <param name="values"> <see cref="IEnumerable{T}"/> of binary values in sorted order (deduplicated). </param>
/// <param name="docToOrd"> <see cref="IEnumerable{T}"/> of ordinals (one for each document). <c>-1</c> indicates
/// a missing value. </param>
/// <exception cref="System.IO.IOException"> If an I/O error occurred. </exception>
public abstract void AddSortedField(FieldInfo field, IEnumerable<BytesRef> values, IEnumerable<long?> docToOrd);
/// <summary>
/// Writes pre-sorted set docvalues for a field </summary>
/// <param name="field"> Field information. </param>
/// <param name="values"> <see cref="IEnumerable{T}"/> of binary values in sorted order (deduplicated). </param>
/// <param name="docToOrdCount"> <see cref="IEnumerable{T}"/> of the number of values for each document. A zero ordinal
/// count indicates a missing value. </param>
/// <param name="ords"> <see cref="IEnumerable{T}"/> of ordinal occurrences (<paramref name="docToOrdCount"/>*maxDoc total). </param>
/// <exception cref="System.IO.IOException"> If an I/O error occurred. </exception>
public abstract void AddSortedSetField(FieldInfo field, IEnumerable<BytesRef> values, IEnumerable<long?> docToOrdCount, IEnumerable<long?> ords);
/// <summary>
/// Merges the numeric docvalues from <paramref name="toMerge"/>.
/// <para>
/// The default implementation calls <see cref="AddNumericField(FieldInfo, IEnumerable{long?})"/>, passing
/// an <see cref="IEnumerable{T}"/> that merges and filters deleted documents on the fly.</para>
/// </summary>
public virtual void MergeNumericField(FieldInfo fieldInfo, MergeState mergeState, IList<NumericDocValues> toMerge, IList<IBits> docsWithField)
{
AddNumericField(fieldInfo, GetMergeNumericFieldEnumerable(fieldInfo, mergeState, toMerge, docsWithField));
}
private IEnumerable<long?> GetMergeNumericFieldEnumerable(FieldInfo fieldinfo, MergeState mergeState, IList<NumericDocValues> toMerge, IList<IBits> docsWithField)
{
int readerUpto = -1;
int docIDUpto = 0;
AtomicReader currentReader = null;
NumericDocValues currentValues = null;
IBits currentLiveDocs = null;
IBits currentDocsWithField = null;
while (true)
{
if (readerUpto == toMerge.Count)
{
yield break;
}
if (currentReader == null || docIDUpto == currentReader.MaxDoc)
{
readerUpto++;
if (readerUpto < toMerge.Count)
{
currentReader = mergeState.Readers[readerUpto];
currentValues = toMerge[readerUpto];
currentDocsWithField = docsWithField[readerUpto];
currentLiveDocs = currentReader.LiveDocs;
}
docIDUpto = 0;
continue;
}
if (currentLiveDocs == null || currentLiveDocs.Get(docIDUpto))
{
long? nextValue;
if (currentDocsWithField.Get(docIDUpto))
{
nextValue = currentValues.Get(docIDUpto);
}
else
{
nextValue = null;
}
docIDUpto++;
yield return nextValue;
continue;
}
docIDUpto++;
}
}
/// <summary>
/// Merges the binary docvalues from <paramref name="toMerge"/>.
/// <para>
/// The default implementation calls <see cref="AddBinaryField(FieldInfo, IEnumerable{BytesRef})"/>, passing
/// an <see cref="IEnumerable{T}"/> that merges and filters deleted documents on the fly.</para>
/// </summary>
public virtual void MergeBinaryField(FieldInfo fieldInfo, MergeState mergeState, IList<BinaryDocValues> toMerge, IList<IBits> docsWithField)
{
AddBinaryField(fieldInfo, GetMergeBinaryFieldEnumerable(fieldInfo, mergeState, toMerge, docsWithField));
}
private IEnumerable<BytesRef> GetMergeBinaryFieldEnumerable(FieldInfo fieldInfo, MergeState mergeState, IList<BinaryDocValues> toMerge, IList<IBits> docsWithField)
{
int readerUpto = -1;
int docIDUpto = 0;
AtomicReader currentReader = null;
BinaryDocValues currentValues = null;
IBits currentLiveDocs = null;
IBits currentDocsWithField = null;
while (true)
{
if (readerUpto == toMerge.Count)
{
yield break;
}
if (currentReader == null || docIDUpto == currentReader.MaxDoc)
{
readerUpto++;
if (readerUpto < toMerge.Count)
{
currentReader = mergeState.Readers[readerUpto];
currentValues = toMerge[readerUpto];
currentDocsWithField = docsWithField[readerUpto];
currentLiveDocs = currentReader.LiveDocs;
}
docIDUpto = 0;
continue;
}
if (currentLiveDocs == null || currentLiveDocs.Get(docIDUpto))
{
var nextValue = new BytesRef();
if (currentDocsWithField.Get(docIDUpto))
{
currentValues.Get(docIDUpto, nextValue);
}
else
{
nextValue = null;
}
docIDUpto++;
yield return nextValue;
continue;
}
docIDUpto++;
}
}
/// <summary>
/// Merges the sorted docvalues from <paramref name="toMerge"/>.
/// <para>
/// The default implementation calls <see cref="AddSortedField(FieldInfo, IEnumerable{BytesRef}, IEnumerable{long?})"/>, passing
/// an <see cref="IEnumerable{T}"/> that merges ordinals and values and filters deleted documents.</para>
/// </summary>
public virtual void MergeSortedField(FieldInfo fieldInfo, MergeState mergeState, IList<SortedDocValues> toMerge)
{
AtomicReader[] readers = mergeState.Readers.ToArray();
SortedDocValues[] dvs = toMerge.ToArray();
// step 1: iterate thru each sub and mark terms still in use
var liveTerms = new TermsEnum[dvs.Length];
for (int sub = 0; sub < liveTerms.Length; sub++)
{
AtomicReader reader = readers[sub];
SortedDocValues dv = dvs[sub];
IBits liveDocs = reader.LiveDocs;
if (liveDocs == null)
{
liveTerms[sub] = dv.GetTermsEnum();
}
else
{
var bitset = new Int64BitSet(dv.ValueCount);
for (int i = 0; i < reader.MaxDoc; i++)
{
if (liveDocs.Get(i))
{
int ord = dv.GetOrd(i);
if (ord >= 0)
{
bitset.Set(ord);
}
}
}
liveTerms[sub] = new BitsFilteredTermsEnum(dv.GetTermsEnum(), bitset);
}
}
// step 2: create ordinal map (this conceptually does the "merging")
var map = new OrdinalMap(this, liveTerms);
// step 3: add field
AddSortedField(fieldInfo, GetMergeSortValuesEnumerable(map, dvs),
// doc -> ord
GetMergeSortedFieldDocToOrdEnumerable(readers, dvs, map)
);
}
private IEnumerable<BytesRef> GetMergeSortValuesEnumerable(OrdinalMap map, SortedDocValues[] dvs)
{
int currentOrd = 0;
while (currentOrd < map.ValueCount)
{
var scratch = new BytesRef();
int segmentNumber = map.GetFirstSegmentNumber(currentOrd);
var segmentOrd = (int)map.GetFirstSegmentOrd(currentOrd);
dvs[segmentNumber].LookupOrd(segmentOrd, scratch);
currentOrd++;
yield return scratch;
}
}
private IEnumerable<long?> GetMergeSortedFieldDocToOrdEnumerable(AtomicReader[] readers, SortedDocValues[] dvs, OrdinalMap map)
{
int readerUpTo = -1;
int docIDUpTo = 0;
AtomicReader currentReader = null;
IBits currentLiveDocs = null;
while (true)
{
if (readerUpTo == readers.Length)
{
yield break;
}
if (currentReader == null || docIDUpTo == currentReader.MaxDoc)
{
readerUpTo++;
if (readerUpTo < readers.Length)
{
currentReader = readers[readerUpTo];
currentLiveDocs = currentReader.LiveDocs;
}
docIDUpTo = 0;
continue;
}
if (currentLiveDocs == null || currentLiveDocs.Get(docIDUpTo))
{
int segOrd = dvs[readerUpTo].GetOrd(docIDUpTo);
docIDUpTo++;
yield return segOrd == -1 ? -1 : map.GetGlobalOrd(readerUpTo, segOrd);
continue;
}
docIDUpTo++;
}
}
/// <summary>
/// Merges the sortedset docvalues from <paramref name="toMerge"/>.
/// <para>
/// The default implementation calls <see cref="AddSortedSetField(FieldInfo, IEnumerable{BytesRef}, IEnumerable{long?}, IEnumerable{long?})"/>, passing
/// an <see cref="IEnumerable{T}"/> that merges ordinals and values and filters deleted documents.</para>
/// </summary>
public virtual void MergeSortedSetField(FieldInfo fieldInfo, MergeState mergeState, IList<SortedSetDocValues> toMerge)
{
var readers = mergeState.Readers.ToArray();
var dvs = toMerge.ToArray();
// step 1: iterate thru each sub and mark terms still in use
var liveTerms = new TermsEnum[dvs.Length];
for (int sub = 0; sub < liveTerms.Length; sub++)
{
var reader = readers[sub];
var dv = dvs[sub];
var liveDocs = reader.LiveDocs;
if (liveDocs == null)
{
liveTerms[sub] = dv.GetTermsEnum();
}
else
{
var bitset = new Int64BitSet(dv.ValueCount);
for (int i = 0; i < reader.MaxDoc; i++)
{
if (liveDocs.Get(i))
{
dv.SetDocument(i);
long ord;
while ((ord = dv.NextOrd()) != SortedSetDocValues.NO_MORE_ORDS)
{
bitset.Set(ord);
}
}
}
liveTerms[sub] = new BitsFilteredTermsEnum(dv.GetTermsEnum(), bitset);
}
}
// step 2: create ordinal map (this conceptually does the "merging")
var map = new OrdinalMap(this, liveTerms);
// step 3: add field
AddSortedSetField(fieldInfo, GetMergeSortedSetValuesEnumerable(map, dvs),
// doc -> ord count
GetMergeSortedSetDocToOrdCountEnumerable(readers, dvs),
// ords
GetMergeSortedSetOrdsEnumerable(readers, dvs, map)
);
}
private IEnumerable<BytesRef> GetMergeSortedSetValuesEnumerable(OrdinalMap map, SortedSetDocValues[] dvs)
{
long currentOrd = 0;
while (currentOrd < map.ValueCount)
{
int segmentNumber = map.GetFirstSegmentNumber(currentOrd);
long segmentOrd = map.GetFirstSegmentOrd(currentOrd);
var scratch = new BytesRef();
dvs[segmentNumber].LookupOrd(segmentOrd, scratch);
currentOrd++;
yield return scratch;
}
}
private IEnumerable<long?> GetMergeSortedSetDocToOrdCountEnumerable(AtomicReader[] readers, SortedSetDocValues[] dvs)
{
int readerUpto = -1;
int docIDUpto = 0;
AtomicReader currentReader = null;
IBits currentLiveDocs = null;
while (true)
{
if (readerUpto == readers.Length)
{
yield break;
}
if (currentReader == null || docIDUpto == currentReader.MaxDoc)
{
readerUpto++;
if (readerUpto < readers.Length)
{
currentReader = readers[readerUpto];
currentLiveDocs = currentReader.LiveDocs;
}
docIDUpto = 0;
continue;
}
if (currentLiveDocs == null || currentLiveDocs.Get(docIDUpto))
{
SortedSetDocValues dv = dvs[readerUpto];
dv.SetDocument(docIDUpto);
long value = 0;
while (dv.NextOrd() != SortedSetDocValues.NO_MORE_ORDS)
{
value++;
}
docIDUpto++;
yield return value;
continue;
}
docIDUpto++;
}
}
private IEnumerable<long?> GetMergeSortedSetOrdsEnumerable(AtomicReader[] readers, SortedSetDocValues[] dvs, OrdinalMap map)
{
int readerUpto = -1;
int docIDUpto = 0;
AtomicReader currentReader = null;
IBits currentLiveDocs = null;
var ords = new long[8];
int ordUpto = 0;
int ordLength = 0;
while (true)
{
if (readerUpto == readers.Length)
{
yield break;
}
if (ordUpto < ordLength)
{
var value = ords[ordUpto];
ordUpto++;
yield return value;
continue;
}
if (currentReader == null || docIDUpto == currentReader.MaxDoc)
{
readerUpto++;
if (readerUpto < readers.Length)
{
currentReader = readers[readerUpto];
currentLiveDocs = currentReader.LiveDocs;
}
docIDUpto = 0;
continue;
}
if (currentLiveDocs == null || currentLiveDocs.Get(docIDUpto))
{
Debug.Assert(docIDUpto < currentReader.MaxDoc);
SortedSetDocValues dv = dvs[readerUpto];
dv.SetDocument(docIDUpto);
ordUpto = ordLength = 0;
long ord;
while ((ord = dv.NextOrd()) != SortedSetDocValues.NO_MORE_ORDS)
{
if (ordLength == ords.Length)
{
ords = ArrayUtil.Grow(ords, ordLength + 1);
}
ords[ordLength] = map.GetGlobalOrd(readerUpto, ord);
ordLength++;
}
docIDUpto++;
continue;
}
docIDUpto++;
}
}
// TODO: seek-by-ord to nextSetBit
internal class BitsFilteredTermsEnum : FilteredTermsEnum
{
internal readonly Int64BitSet liveTerms;
internal BitsFilteredTermsEnum(TermsEnum @in, Int64BitSet liveTerms)
: base(@in, false)
{
Debug.Assert(liveTerms != null);
this.liveTerms = liveTerms;
}
protected override AcceptStatus Accept(BytesRef term)
{
return liveTerms.Get(Ord) ? AcceptStatus.YES : AcceptStatus.NO;
}
}
/// <summary>
/// Disposes all resources used by this object.
/// </summary>
public void Dispose()
{
Dispose(true);
GC.SuppressFinalize(this);
}
/// <summary>
/// Implementations must override and should dispose all resources used by this instance.
/// </summary>
protected abstract void Dispose(bool disposing);
}
}