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
/*
 * 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::DefaultTermVectorsFormat;
use crate::core::codecs::term_vectors_format::TermVectorsFormat;
use crate::core::codecs::term_vectors_reader::TermVectorsReader;
use crate::core::index::bytes_ref::BytesRefValue;
use crate::core::index::codec_reader::CodecReader;
use crate::core::index::field_info::FieldInfo;
use crate::core::index::fields::Fields;
use crate::core::index::merge_state::SharedDocMap;
use crate::core::index::merge_state::{DocMap, MergeState, MergeStateMeta};
use crate::core::index::postings_enum::{OFFSETS, PAYLOADS, PostingsEnum};
use crate::core::index::term_vectors::TermVectors;
use crate::core::index::terms::Terms;
use crate::core::index::terms_enum::TermsEnum;
use crate::core::index::{BytesRef, DocIDMerger, Sub, SubBase, of};
use crate::core::search::doc_id_set_iterator::DocIdSetIterator;
use crate::core::search::doc_id_set_iterator::NO_MORE_DOCS;
use crate::core::store::DataInput;
use crate::core::util::accountable::Accountable;
use crate::core::util::array_util::ArrayUtil;
use crate::core::util::bytes_ref_iterator::BytesRefIterator;
use crate::core::util::close::Closeable;
use crate::core::util::error::lucene_error::{LuceneError, Result};
use crate::core::util::iterator::IteratorExt;

struct TermVectorsMergeSub<DM> {
  reader_index: usize,
  max_doc: i32,
  doc_id: i32,
  doc_map: SharedDocMap<DM>,
}

impl<DM> TermVectorsMergeSub<DM> {
  fn new(doc_map: SharedDocMap<DM>, reader_index: usize, max_doc: i32) -> Self {
    Self {
      reader_index,
      max_doc,
      doc_id: -1,
      doc_map,
    }
  }
}

impl<DM> SubBase for TermVectorsMergeSub<DM>
where
  DM: DocMap,
{
  fn next_doc(&mut self) -> Result<i32> {
    self.doc_id += 1;
    if self.doc_id == self.max_doc {
      Ok(NO_MORE_DOCS)
    } else {
      Ok(self.doc_id)
    }
  }

  type DocMap = DM;

  fn get_doc_map(&self) -> Result<&Self::DocMap> {
    Ok(&self.doc_map)
  }
}

pub trait TermVectorsWriter: Accountable + Closeable {
  fn start_document(&mut self, num_vector_fields: i32) -> Result<()>;

  fn finish_document(&mut self) -> Result<()> {
    Ok(())
  }
  fn start_field(
    &mut self,
    field_info: &FieldInfo,
    num_terms: usize,
    positions: bool,
    offsets: bool,
    payloads: bool,
  ) -> Result<()>;

  fn finish_field(&mut self) -> Result<()> {
    Ok(())
  }

  fn start_term(&mut self, term: &BytesRef<&[u8]>, freq: i32) -> Result<()>;

  fn finish_term(&mut self) -> Result<()> {
    Ok(())
  }

  fn add_position(
    &mut self,
    position: i32,
    start_offset: i32,
    end_offset: i32,
    payload: Option<&BytesRef<&[u8]>>,
  ) -> Result<()>;

  /// Called before [`Closeable::close`], passing in the number of documents
  /// that were written. Note that this is intentionally redundant (equivalent
  /// to the number of calls to [`Self::start_document`]), but a codec should
  /// check that this is the case to detect the JRE bug described in
  /// LUCENE-1282.
  fn finish(&mut self, num_docs: i32) -> Result<()>;

  /// Called by [`IndexWriter`](crate::core::index::index_writer::IndexWriter) when writing new segments.
  ///
  /// This is an expert API that allows the codec to consume positions and
  /// offsets directly from the indexer.
  ///
  /// The default implementation calls [`Self::add_position`], but
  /// implementations can override this if they want to efficiently write all
  /// the positions, then all the offsets, for example.
  ///
  /// NOTE: This API is extremely expert and subject to change or removal!!!
  fn add_prox<DI, DI2>(
    &mut self,
    num_prox: usize,
    positions: Option<&mut DI>,
    offsets: Option<&mut DI2>,
  ) -> Result<()>
  where
    DI: DataInput,
    DI2: DataInput,
  {
    TermVectorsWriterDefaults::add_prox(self, num_prox, positions, offsets)
  }
  /// Merges in the term vectors from the readers in `merge_state`. The default
  /// implementation skips over deleted documents, and uses
  /// [`Self::start_document`], [`Self::start_field`], [`Self::start_term`],
  /// [`Self::add_position`], and [`Self::finish`], returning the number of
  /// documents that were written. Implementations can override this method for
  /// more sophisticated merging (bulk-byte copying, etc).
  fn merge<D, CR>(&mut self, merge_state: &mut MergeState<D, CR>) -> Result<i32>
  where
    CR: CodecReader,
    Self: Sized,
  {
    TermVectorsWriterDefaults::merge(self, merge_state)
  }

  /// Safe (but, slowish) default method to write every vector field in the document.
  fn add_all_doc_vectors<F, DM>(
    &mut self,
    vectors: Option<&F>,
    merge_state: &MergeStateMeta<DM>,
  ) -> Result<()>
  where
    F: Fields,
    DM: DocMap,
  {
    let Some(vectors) = vectors else {
      self.start_document(0)?;
      self.finish_document()?;
      return Ok(());
    };

    let mut num_fields = vectors.size()?;
    if num_fields == -1 {
      // count manually
      num_fields = 0;
      let mut it = vectors.iterator()?;
      while it.has_next()? {
        it.next()?;
        num_fields += 1;
      }
    }

    self.start_document(num_fields)?;

    #[cfg(debug_assertions)]
    let mut last_field_name: Option<&String> = None;
    let mut docs_and_positions_reuse = None;

    let mut field_count = 0;

    let mut fields_iter = vectors.iterator()?;
    while let Some(field_name) = fields_iter.next()? {
      field_count += 1;

      let field_info = merge_state
        .merge_field_infos
        .field_info_by_name(field_name)?
        .ok_or_else(|| LuceneError::illegal_state("missing FieldInfo"))?;

      #[cfg(debug_assertions)]
      {
        if let Some(last) = last_field_name {
          debug_assert!(
            field_name > last,
            "lastFieldName={} fieldName={}",
            last,
            field_name
          );
        }
        last_field_name = Some(field_name);
      }

      let Some(terms) = vectors.terms(field_name)? else {
        // Fields iterator should not lie
        continue;
      };

      let has_positions = terms.has_positions();
      let has_offsets = terms.has_offsets();
      let has_payloads = terms.has_payloads();
      debug_assert!(!has_payloads || has_positions);

      let mut num_terms = terms.size()? as i32;
      if num_terms == -1 {
        num_terms = 0;
        let mut terms_enum = terms.iterator()?;
        // count manually. It is stupid, but needed, as Terms.size() is not a mandatory statistics
        // function
        while terms_enum.next()?.is_some() {
          num_terms += 1;
        }
      }

      self.start_field(
        field_info.as_ref(),
        num_terms as usize,
        has_positions,
        has_offsets,
        has_payloads,
      )?;

      let mut terms_enum = terms.iterator()?;
      let mut term_count = 0;

      loop {
        let next = terms_enum.next()?;
        if next.is_none() {
          break;
        }
        drop(next);
        term_count += 1;

        let freq = terms_enum.total_term_freq()? as i32;
        self.start_term(&terms_enum.term()?.as_bytes_ref(), freq)?;

        if has_positions || has_offsets {
          let mut docs_and_positions_enum = terms_enum
            .postings_with_flags(docs_and_positions_reuse.take(), (OFFSETS | PAYLOADS) as i32)?;

          let doc_id = docs_and_positions_enum.next_doc()?;
          debug_assert!(doc_id != NO_MORE_DOCS);
          debug_assert!(docs_and_positions_enum.freq()? == freq);

          for _ in 0..freq {
            let pos = docs_and_positions_enum.next_position()?;
            let start_offset = docs_and_positions_enum.start_offset()?;
            let end_offset = docs_and_positions_enum.end_offset()?;
            let payload = docs_and_positions_enum.get_payload()?;

            debug_assert!(!has_positions || pos >= 0);
            let payload = payload.as_ref().map(BytesRefValue::as_bytes_ref);
            self.add_position(pos, start_offset, end_offset, payload.as_ref())?;
          }
          docs_and_positions_reuse = Some(docs_and_positions_enum);
        }

        self.finish_term()?;
      }

      debug_assert!(term_count == num_terms);
      self.finish_field()?;
    }

    debug_assert!(field_count == num_fields);
    self.finish_document()?;

    Ok(())
  }
}

pub struct TermVectorsWriterDefaults;

impl TermVectorsWriterDefaults {
  pub fn add_prox<W, DI, DI2>(
    writer: &mut W,
    num_prox: usize,
    mut positions: Option<&mut DI>,
    mut offsets: Option<&mut DI2>,
  ) -> Result<()>
  where
    W: TermVectorsWriter + ?Sized,
    DI: DataInput,
    DI2: DataInput,
  {
    let mut position = 0;
    let mut last_offset = 0;

    for _ in 0..num_prox {
      let bytes;
      let this_payload = if let Some(pos_input) = positions.as_mut() {
        let code = pos_input.read_vint()?;
        position += (code as u32 >> 1) as i32;

        if code & 1 != 0 {
          let payload_len = pos_input.read_vint()? as usize;
          if payload_len > ArrayUtil::MAX_ARRAY_LENGTH {
            ArrayUtil::oversize(payload_len, 1)?;
          }
          bytes = pos_input.get_bytes(payload_len)?;
          Some(BytesRef {
            bytes: bytes.as_ref(),
            offset: 0,
            length: payload_len,
          })
        } else {
          None
        }
      } else {
        position = -1;
        None
      };

      let (start_offset, end_offset) = if let Some(off_input) = offsets.as_mut() {
        let start = last_offset + off_input.read_vint()?;
        let end = start + off_input.read_vint()?;
        last_offset = end;
        (start, end)
      } else {
        (-1, -1)
      };

      writer.add_position(position, start_offset, end_offset, this_payload.as_ref())?;
    }

    Ok(())
  }

  pub fn merge<W, D, CR>(writer: &mut W, merge_state: &mut MergeState<D, CR>) -> Result<i32>
  where
    W: TermVectorsWriter,
    CR: CodecReader,
  {
    let mut subs = Vec::with_capacity(merge_state.term_vectors_readers.len());
    for i in 0..merge_state.term_vectors_readers.len() {
      if let Some(reader) = &merge_state.term_vectors_readers[i] {
        reader.check_integrity()?;
      }
      subs.push(Sub::new(TermVectorsMergeSub::new(
        merge_state.doc_map(i),
        i,
        merge_state.max_docs[i],
      )));
    }

    let mut doc_id_merger = of(subs, merge_state.needs_index_sort)?;
    let merge_state_meta = merge_state.get_meta();
    let mut doc_count = 0;
    while let Some(sub_index) = doc_id_merger.next()? {
      let sub = &doc_id_merger.get_subs()[sub_index].sub;

      // NOTE: it's very important to first assign to vectors then pass it to
      // termVectorsWriter.addAllDocVectors; see LUCENE-1282
      let vectors = match merge_state.term_vectors_readers[sub.reader_index].as_mut() {
        Some(reader) => reader.get(sub.doc_id)?,
        None => None,
      };
      writer.add_all_doc_vectors(vectors.as_ref(), &merge_state_meta)?;
      doc_count += 1;
    }
    writer.finish(doc_count)?;
    Ok(doc_count)
  }
}

pub type DefaultTermVectorsWriter<D> =
  <DefaultTermVectorsFormat as TermVectorsFormat>::TermVectorsWriter<D>;

pub enum TermVectorsWriterEnum2<A, B> {
  A(A),
  B(B),
}

impl<A, B> Closeable for TermVectorsWriterEnum2<A, B>
where
  A: Closeable,
  B: Closeable,
{
  fn close(&mut self) -> Result<()> {
    match self {
      Self::A(inner) => inner.close(),
      Self::B(inner) => inner.close(),
    }
  }
}

impl<A, B> Accountable for TermVectorsWriterEnum2<A, B>
where
  A: Accountable,
  B: Accountable,
{
  fn ram_bytes_used(&self) -> Result<i64> {
    match self {
      Self::A(inner) => inner.ram_bytes_used(),
      Self::B(inner) => inner.ram_bytes_used(),
    }
  }
}

impl<A, B> TermVectorsWriter for TermVectorsWriterEnum2<A, B>
where
  A: TermVectorsWriter,
  B: TermVectorsWriter,
{
  fn start_document(&mut self, num_vector_fields: i32) -> Result<()> {
    match self {
      Self::A(inner) => inner.start_document(num_vector_fields),
      Self::B(inner) => inner.start_document(num_vector_fields),
    }
  }

  fn finish_document(&mut self) -> Result<()> {
    match self {
      Self::A(inner) => inner.finish_document(),
      Self::B(inner) => inner.finish_document(),
    }
  }

  fn start_field(
    &mut self,
    field_info: &FieldInfo,
    num_terms: usize,
    positions: bool,
    offsets: bool,
    payloads: bool,
  ) -> Result<()> {
    match self {
      Self::A(inner) => inner.start_field(field_info, num_terms, positions, offsets, payloads),
      Self::B(inner) => inner.start_field(field_info, num_terms, positions, offsets, payloads),
    }
  }

  fn finish_field(&mut self) -> Result<()> {
    match self {
      Self::A(inner) => inner.finish_field(),
      Self::B(inner) => inner.finish_field(),
    }
  }

  fn start_term(&mut self, term: &BytesRef<&[u8]>, freq: i32) -> Result<()> {
    match self {
      Self::A(inner) => inner.start_term(term, freq),
      Self::B(inner) => inner.start_term(term, freq),
    }
  }

  fn finish_term(&mut self) -> Result<()> {
    match self {
      Self::A(inner) => inner.finish_term(),
      Self::B(inner) => inner.finish_term(),
    }
  }

  fn add_position(
    &mut self,
    position: i32,
    start_offset: i32,
    end_offset: i32,
    payload: Option<&BytesRef<&[u8]>>,
  ) -> Result<()> {
    match self {
      Self::A(inner) => inner.add_position(position, start_offset, end_offset, payload),
      Self::B(inner) => inner.add_position(position, start_offset, end_offset, payload),
    }
  }

  fn finish(&mut self, num_docs: i32) -> Result<()> {
    match self {
      Self::A(inner) => inner.finish(num_docs),
      Self::B(inner) => inner.finish(num_docs),
    }
  }

  fn add_prox<DI, DI2>(
    &mut self,
    num_prox: usize,
    positions: Option<&mut DI>,
    offsets: Option<&mut DI2>,
  ) -> Result<()>
  where
    DI: DataInput,
    DI2: DataInput,
  {
    match self {
      Self::A(inner) => inner.add_prox(num_prox, positions, offsets),
      Self::B(inner) => inner.add_prox(num_prox, positions, offsets),
    }
  }

  fn merge<D, CR>(&mut self, merge_state: &mut MergeState<D, CR>) -> Result<i32>
  where
    CR: CodecReader,
  {
    match self {
      Self::A(inner) => inner.merge(merge_state),
      Self::B(inner) => inner.merge(merge_state),
    }
  }
}