rlucene 0.1.0-alpha.3

A Rust port of Apache Lucene, a full-text search engine library
Documentation
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
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
/*
 * 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.
 */
use crate::core::codecs::doc_values_producer::DocValuesProducer;
use crate::core::codecs::fields_producer::FieldsProducer;
use crate::core::codecs::knn_vectors_reader::KnnVectorsReader;
use crate::core::codecs::norms_producer::{NormsProducer, NormsProducerEnum2};
use crate::core::codecs::points_reader::PointsReader;
use crate::core::codecs::stored_fields_reader::StoredFieldsReader;
use crate::core::codecs::term_vectors_reader::TermVectorsReader;
use crate::core::index::codec_reader::{
  CRBits, CRDocValuesProducer, CRFieldsProducer, CRKnnVectorReader, CRNormsProducer,
  CRPointsReader, CRStoredFieldsReader, CRTermVectorsReader, CodecReader,
};

use crate::core::index::field_infos::{EMPTY, FieldInfos};
use crate::core::index::index_writer::is_congruent_sort;
use crate::core::index::multi_sorter::MultiSorter;
use crate::core::index::segment_info::SegmentInfo;
use crate::core::search::task_executor::TaskExecutor;
use crate::core::util::bits::Bits;
use crate::core::util::error::lucene_error::{LuceneError, Result};
use crate::core::util::info_stream::{InfoStream, InfoStreamEnum};
use crate::core::util::long_values::LongValues;
use crate::core::util::packed::PackedInts;
use crate::core::util::packed::packed_long_values::PackedLongValues;
use std::rc::Rc;
use std::sync::Arc;
use std::time::Instant;

type MergeNormsProducer<CR> =
  NormsProducerEnum2<CRNormsProducer<CR>, <CRNormsProducer<CR> as NormsProducer>::MergeInstance>;

/// Holds common state used during segment merging.
///
/// @lucene.experimental
pub struct MergeState<'a, D, CR, DocMaps = Rc<Vec<MergeStateDocMap<CR>>>>
where
  CR: CodecReader,
{
  /// [SegmentInfo] of the newly merged segment.
  pub(crate) segment_info: &'a mut SegmentInfo<D>,
  /// Maps document IDs from old segments to document IDs in the new segment
  pub(crate) doc_maps: DocMaps,
  /// [FieldInfos] of the newly merged segment.
  pub(crate) merge_field_infos: Arc<FieldInfos>,
  /// Stored field producers being merged
  pub(crate) stored_fields_readers: Vec<Option<CRStoredFieldsReader<CR>>>,
  /// Term vector producers being merged
  pub(crate) term_vectors_readers: Vec<Option<CRTermVectorsReader<CR>>>,
  /// Norms producers being merged
  pub(crate) norms_producers: Vec<Option<MergeNormsProducer<CR>>>,
  /// DocValues producers being merged
  pub(crate) doc_values_producers: Vec<Option<CRDocValuesProducer<CR>>>,
  /// Postings to merge
  pub(crate) fields_producers: Vec<Option<CRFieldsProducer<CR>>>,
  /// Point readers to merge
  pub(crate) points_readers: Vec<Option<CRPointsReader<CR>>>,
  /// Vector readers to merge
  pub(crate) knn_vectors_readers: Vec<Option<CRKnnVectorReader<CR>>>,
  /// FieldInfos being merged
  pub(crate) field_infos: Rc<Vec<Arc<FieldInfos>>>,
  /// Live docs for each reader
  pub(crate) live_docs: Vec<Option<CRBits<CR>>>,
  /// Indicates if the index needs to be sorted
  pub(crate) needs_index_sort: bool,
  /// Max docs per reader
  pub(crate) max_docs: smallvec::SmallVec<[i32; 4]>,
  /// [InfoStream] for debugging messages.
  pub(crate) info_stream: Arc<InfoStreamEnum>,
  /// Executor for intra-merge activity.
  pub(crate) intra_merge_task_executor: Arc<TaskExecutor>,
}

/// Access to the portion of a [`MergeState`] used by per-field codecs.
///
/// Per-field codecs restrict field infos and fields producers to one format's fields.
/// This trait exposes that restricted view while keeping the producer type
/// statically dispatched.
pub trait MergeStateAccess {
  type FieldsProducer: FieldsProducer;
  type DocValuesProducer: DocValuesProducer;
  type LiveDocs: Bits;
  type DocMap: DocMap;

  fn fields_producers(&self) -> &[Option<Self::FieldsProducer>];

  fn doc_values_producers(&self) -> &[Option<Self::DocValuesProducer>];

  fn doc_maps(&self) -> &Rc<Vec<Self::DocMap>>;

  fn doc_map(&self, index: usize) -> SharedDocMap<Self::DocMap> {
    SharedDocMap::new(self.doc_maps().clone(), index)
  }

  fn merge_field_infos(&self) -> &Arc<FieldInfos>;

  fn field_infos(&self) -> &[Arc<FieldInfos>];

  fn live_docs(&self) -> &[Option<Self::LiveDocs>];

  fn needs_index_sort(&self) -> bool;

  fn max_docs(&self) -> &[i32];

  fn intra_merge_task_executor(&self) -> &Arc<TaskExecutor>;

  fn get_meta(&self) -> MergeStateMeta<Self::DocMap>;
}

impl<D, CR> MergeStateAccess for MergeState<'_, D, CR>
where
  CR: CodecReader,
{
  type FieldsProducer = CRFieldsProducer<CR>;
  type DocValuesProducer = CRDocValuesProducer<CR>;
  type LiveDocs = CRBits<CR>;
  type DocMap = MergeStateDocMap<CR>;

  fn fields_producers(&self) -> &[Option<Self::FieldsProducer>] {
    &self.fields_producers
  }

  fn doc_values_producers(&self) -> &[Option<Self::DocValuesProducer>] {
    &self.doc_values_producers
  }

  fn doc_maps(&self) -> &Rc<Vec<Self::DocMap>> {
    &self.doc_maps
  }

  fn merge_field_infos(&self) -> &Arc<FieldInfos> {
    &self.merge_field_infos
  }

  fn field_infos(&self) -> &[Arc<FieldInfos>] {
    &self.field_infos
  }

  fn live_docs(&self) -> &[Option<Self::LiveDocs>] {
    &self.live_docs
  }

  fn needs_index_sort(&self) -> bool {
    self.needs_index_sort
  }

  fn max_docs(&self) -> &[i32] {
    &self.max_docs
  }

  fn intra_merge_task_executor(&self) -> &Arc<TaskExecutor> {
    &self.intra_merge_task_executor
  }

  fn get_meta(&self) -> MergeStateMeta<Self::DocMap> {
    MergeState::get_meta(self)
  }
}

impl<'a, D, CR> MergeState<'a, D, CR>
where
  CR: CodecReader,
{
  /// Sole constructor.
  pub(crate) fn new(
    readers: &'a [CR],
    segment_info: &'a mut SegmentInfo<D>,
    info_stream: Arc<InfoStreamEnum>,
    intra_merge_task_executor: Arc<TaskExecutor>,
  ) -> Result<Self>
  where
    CR: CodecReader,
  {
    verify_index_sort(readers, segment_info)?;

    let num_readers = readers.len();

    let mut max_docs = smallvec::SmallVec::with_capacity(num_readers);
    let mut fields_producers = Vec::with_capacity(num_readers);
    let mut norms_producers = Vec::with_capacity(num_readers);
    let mut stored_fields_readers = Vec::with_capacity(num_readers);
    let mut term_vectors_readers = Vec::with_capacity(num_readers);
    let mut points_readers = Vec::with_capacity(num_readers);
    let mut knn_vectors_readers = Vec::with_capacity(num_readers);
    let mut doc_values_producers = Vec::with_capacity(num_readers);
    let mut field_infos = Vec::with_capacity(num_readers);
    let mut live_docs = Vec::with_capacity(num_readers);

    let mut num_docs = 0;

    for reader in readers {
      max_docs.push(reader.max_doc()?);
      live_docs.push(reader.get_live_docs()?);
      field_infos.push(reader.get_field_infos()?);

      let norms = if let Some(norms_reader) = reader.get_norms_reader()? {
        if let Some(n) = norms_reader.get_merge_instance()? {
          Some(NormsProducerEnum2::B(n))
        } else {
          Some(NormsProducerEnum2::A(norms_reader))
        }
      } else {
        None
      };
      norms_producers.push(norms);

      let doc_values = if let Some(dv_reader) = reader.get_doc_values_reader()? {
        if let Some(dv) = dv_reader.get_merge_instance()? {
          Some(dv)
        } else {
          Some(dv_reader)
        }
      } else {
        None
      };
      doc_values_producers.push(doc_values);

      let stored_fields = if let Some(stored_reader) = reader.get_fields_reader()? {
        if let Some(stored_fields) = stored_reader.get_merge_instance()? {
          Some(stored_fields)
        } else {
          Some(stored_reader)
        }
      } else {
        None
      };

      stored_fields_readers.push(stored_fields);

      let term_vectors = if let Some(tv_reader) = reader.get_term_vectors_reader()? {
        if let Some(term_vectors) = tv_reader.get_merge_instance()? {
          Some(term_vectors)
        } else {
          Some(tv_reader)
        }
      } else {
        None
      };
      term_vectors_readers.push(term_vectors);

      let postings = if let Some(postings_reader) = reader.get_postings_reader()? {
        if let Some(p) = postings_reader.get_merge_instance()? {
          Some(p)
        } else {
          Some(postings_reader)
        }
      } else {
        None
      };
      fields_producers.push(postings);

      let points = if let Some(points_reader) = reader.get_points_reader()? {
        if let Some(p) = points_reader.get_merge_instance()? {
          Some(p)
        } else {
          Some(points_reader)
        }
      } else {
        None
      };
      points_readers.push(points);

      let knn_vectors = if let Some(knn_vectors_reader) = reader.get_vector_reader()? {
        if let Some(v) = knn_vectors_reader.get_merge_instance()? {
          Some(v)
        } else {
          Some(knn_vectors_reader)
        }
      } else {
        None
      };
      knn_vectors_readers.push(knn_vectors);

      num_docs += reader.num_docs()?;
    }

    segment_info.set_max_doc(num_docs)?;

    // The construction stage owns all resources in their usual declaration order.
    let doc_maps = ();
    // let doc_maps = build_doc_maps(readers, segment_info.index_sort());

    let mut merge_state = MergeState {
      segment_info,
      doc_maps,
      merge_field_infos: Arc::clone(&EMPTY),
      stored_fields_readers,
      term_vectors_readers,
      norms_producers,
      doc_values_producers,
      fields_producers,
      points_readers,
      knn_vectors_readers,
      field_infos: Rc::new(field_infos),
      live_docs,
      needs_index_sort: false,
      max_docs,
      info_stream,
      intra_merge_task_executor,
    };
    let doc_maps = merge_state.build_doc_maps(readers)?;
    Ok(merge_state.with_doc_maps(doc_maps))
  }
  pub(crate) fn doc_map(&self, index: usize) -> SharedDocMap<MergeStateDocMap<CR>> {
    SharedDocMap::new(self.doc_maps.clone(), index)
  }

  pub(crate) fn get_meta(&self) -> MergeStateMeta<MergeStateDocMap<CR>> {
    MergeStateMeta {
      fields_producers_len: self.fields_producers.len(),
      doc_maps: self.doc_maps.clone(),
      needs_index_sort: self.needs_index_sort,
      merge_field_infos: self.merge_field_infos.clone(),
      field_infos: self.field_infos.clone(),
    }
  }
}

impl<D, CR, DocMaps> MergeState<'_, D, CR, DocMaps>
where
  CR: CodecReader,
{
  fn build_doc_maps(&mut self, readers: &[CR]) -> Result<Rc<Vec<MergeStateDocMap<CR>>>>
  where
    CR: CodecReader,
  {
    let v = if let Some(ref sort) = self.segment_info.index_sort {
      // do a merge sort of the incoming leaves:
      let t0 = Instant::now();
      match MultiSorter::sort(sort, readers)? {
        None => {
          // already sorted, fall back to deletion-only mapping
          build_deletion_doc_maps(readers)?
        },
        Some(result) => {
          self.needs_index_sort = true;

          let t1 = Instant::now();
          if self.info_stream.is_enabled("SM") {
            let elapsed = t1.duration_since(t0).as_secs_f64() * 1000.0;
            self.info_stream.message(
              "SM",
              &format!("{:.2} msec to build merge sorted DocMaps", elapsed),
            )?;
          }
          result
        },
      }
    } else {
      // no index sort ... we only must map around deletions, and rebase to the merged segment's
      // docID space
      build_deletion_doc_maps(readers)?
    };
    Ok(Rc::new(v))
  }
}

impl<'a, D, CR> MergeState<'a, D, CR, ()>
where
  CR: CodecReader,
{
  fn with_doc_maps(self, doc_maps: Rc<Vec<MergeStateDocMap<CR>>>) -> MergeState<'a, D, CR> {
    MergeState {
      segment_info: self.segment_info,
      doc_maps,
      merge_field_infos: self.merge_field_infos,
      stored_fields_readers: self.stored_fields_readers,
      term_vectors_readers: self.term_vectors_readers,
      norms_producers: self.norms_producers,
      doc_values_producers: self.doc_values_producers,
      fields_producers: self.fields_producers,
      points_readers: self.points_readers,
      knn_vectors_readers: self.knn_vectors_readers,
      field_infos: self.field_infos,
      live_docs: self.live_docs,
      needs_index_sort: self.needs_index_sort,
      max_docs: self.max_docs,
      info_stream: self.info_stream,
      intra_merge_task_executor: self.intra_merge_task_executor,
    }
  }
}

pub type MergeStateDocMap<CR> = MergeStateDocMapImpl<CRBits<CR>>;

pub struct MergeStateDocMapImpl<B> {
  live_docs: Option<B>,
  hook: MergeStateDocMapHook,
}

enum MergeStateDocMapHook {
  Sorted {
    remapped: PackedLongValues,
  },
  Deletions {
    del_doc_map: Option<PackedLongValues>,
    doc_base: i32,
  },
}

impl<B> MergeStateDocMapImpl<B> {
  pub(crate) fn new_sorted(live_docs: Option<B>, remapped: PackedLongValues) -> Self {
    Self {
      live_docs,
      hook: MergeStateDocMapHook::Sorted { remapped },
    }
  }

  fn new_deletions(
    live_docs: Option<B>,
    del_doc_map: Option<PackedLongValues>,
    doc_base: i32,
  ) -> Self {
    Self {
      live_docs,
      hook: MergeStateDocMapHook::Deletions {
        del_doc_map,
        doc_base,
      },
    }
  }
}

impl<B> DocMap for MergeStateDocMapImpl<B>
where
  B: Bits,
{
  fn get(&self, doc_id: i32) -> Result<i32> {
    let doc_index = doc_id as usize;
    match &self.hook {
      MergeStateDocMapHook::Sorted { remapped } => {
        if match self.live_docs {
          None => true,
          Some(ref bits) => bits.get(doc_index)?,
        } {
          Ok(remapped.get(doc_index)? as i32)
        } else {
          Ok(-1)
        }
      },
      MergeStateDocMapHook::Deletions {
        del_doc_map,
        doc_base,
      } => match (&self.live_docs, del_doc_map) {
        (None, None) => Ok(doc_base + doc_id),
        (Some(bits), Some(map)) => {
          if bits.get(doc_index)? {
            Ok(doc_base + map.get(doc_index)? as i32)
          } else {
            Ok(-1)
          }
        },
        _ => Err(LuceneError::illegal_state("should not be here")),
      },
    }
  }
}

impl<B: Bits> DocMap for &MergeStateDocMapImpl<B> {
  fn get(&self, doc_id: i32) -> Result<i32> {
    DocMap::get(*self, doc_id)
  }
}

// Remap docIDs around deletions
fn build_deletion_doc_maps<CR>(readers: &[CR]) -> Result<Vec<MergeStateDocMap<CR>>>
where
  CR: CodecReader,
{
  let mut total_docs: i32 = 0;
  let num_readers = readers.len();
  let mut doc_maps = Vec::with_capacity(num_readers);

  for reader in readers.iter() {
    let live_docs = reader.get_live_docs()?;

    let del_doc_map = if let Some(ref bits) = live_docs {
      Some(remove_deletes(reader.max_doc()?, bits)?)
    } else {
      None
    };

    let doc_base = total_docs;

    doc_maps.push(MergeStateDocMapImpl::new_deletions(
      live_docs,
      del_doc_map,
      doc_base,
    ));

    total_docs += reader.num_docs()?;
  }

  Ok(doc_maps)
}
fn verify_index_sort<CR, D>(readers: &[CR], segment_info: &SegmentInfo<D>) -> Result<()>
where
  CR: CodecReader,
{
  let index_sort = match segment_info.index_sort_ref() {
    Some(sort) => sort,
    None => return Ok(()),
  };

  for leaf in readers {
    let segment_sort = leaf.get_metadata()?.get_sort();
    if !segment_sort
      .as_ref()
      .map(|s| is_congruent_sort(index_sort, s))
      .unwrap_or(false)
    {
      return Err(LuceneError::illegal_argument(format!(
        "index sort mismatch: merged segment has sort={} but to-be-merged segment has sort={}",
        index_sort,
        segment_sort
          .as_ref()
          .map(|s| s.to_string())
          .unwrap_or_else(|| "null".to_string())
      )));
    }
  }

  Ok(())
}

pub(crate) fn remove_deletes<B>(max_doc: i32, live_docs: &B) -> Result<PackedLongValues>
where
  B: Bits,
{
  let mut builder = PackedLongValues::monotonic_long_values_builder_default(PackedInts::COMPACT)?;

  let mut del = 0;
  for i in 0..max_doc {
    builder.add(i as i64 - del)?;
    if !live_docs.get(i as usize)? {
      del += 1;
    }
  }
  builder.build()
}

/// A map of doc IDs.
pub trait DocMap {
  /// Return the mapped docID or -1 if the given doc is not mapped.
  fn get(&self, doc_id: i32) -> Result<i32>;
}
/// An independently owned view of one immutable map in a shared directory.
///
/// Retaining a view keeps the whole directory alive without allocating a box per map.
pub struct SharedDocMap<DM> {
  directory: Rc<Vec<DM>>,
  index: usize,
}

impl<DM> SharedDocMap<DM> {
  /// Retain the directory and select one immutable map without copying its data.
  ///
  /// # Panics
  ///
  /// Panics if `index` is outside `directory`.
  pub fn new(directory: Rc<Vec<DM>>, index: usize) -> Self {
    let _ = &directory[index];
    Self { directory, index }
  }

  pub(crate) fn segment_index(&self) -> usize {
    self.index
  }
}

impl<DM> Clone for SharedDocMap<DM> {
  fn clone(&self) -> Self {
    Self {
      directory: self.directory.clone(),
      index: self.index,
    }
  }
}

impl<DM> std::ops::Deref for SharedDocMap<DM> {
  type Target = DM;
  fn deref(&self) -> &DM {
    &self.directory[self.index]
  }
}

impl<DM> AsRef<DM> for SharedDocMap<DM> {
  fn as_ref(&self) -> &DM {
    self
  }
}

impl<DM: DocMap> DocMap for SharedDocMap<DM> {
  fn get(&self, doc_id: i32) -> Result<i32> {
    self.as_ref().get(doc_id)
  }
}

impl<T> DocMap for Rc<T>
where
  T: DocMap,
{
  fn get(&self, doc_id: i32) -> Result<i32> {
    (**self).get(doc_id)
  }
}
macro_rules! either_doc_map {
    ($vis:vis $name:ident { $( $Variant:ident : $T:ident ),+ $(,)? }) => {
        $vis enum $name<$( $T ),+> {
            $( $Variant($T), )+
        }

        impl<$( $T ),+> DocMap for $name<$( $T ),+>
        where
            $( $T: DocMap ),+
        {
            #[inline]
            fn get(&self, doc_id: i32) -> Result<i32> {
                match self {
                    $( Self::$Variant(inner) => inner.get(doc_id), )+
                }
            }
        }
    };
}
either_doc_map!(pub DocMapEnum2 { A: A, B: B});

// for shared
pub struct MergeStateMeta<DM> {
  pub(crate) fields_producers_len: usize,
  pub(crate) doc_maps: Rc<Vec<DM>>,
  pub needs_index_sort: bool,
  pub merge_field_infos: Arc<FieldInfos>,
  pub field_infos: Rc<Vec<Arc<FieldInfos>>>,
}
impl<DM> Clone for MergeStateMeta<DM> {
  fn clone(&self) -> Self {
    Self {
      fields_producers_len: self.fields_producers_len,
      doc_maps: self.doc_maps.clone(),
      needs_index_sort: self.needs_index_sort,
      merge_field_infos: self.merge_field_infos.clone(),
      field_infos: self.field_infos.clone(),
    }
  }
}