forked from vigna/fastutil
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathOpenHashBigSet.drv
1558 lines (1275 loc) · 48.5 KB
/
OpenHashBigSet.drv
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
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
* Copyright (C) 2002-2023 Sebastiano Vigna
*
* 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 PACKAGE;
import static it.unimi.dsi.fastutil.BigArrays.copy;
import static it.unimi.dsi.fastutil.BigArrays.fill;
import static it.unimi.dsi.fastutil.BigArrays.set;
import it.unimi.dsi.fastutil.BigArrays;
import it.unimi.dsi.fastutil.Hash;
import it.unimi.dsi.fastutil.Size64;
import it.unimi.dsi.fastutil.HashCommon;
import static it.unimi.dsi.fastutil.HashCommon.bigArraySize;
import static it.unimi.dsi.fastutil.HashCommon.maxFill;
#if KEYS_REFERENCE
import java.util.function.Consumer;
import java.util.stream.Collector;
#endif
import java.util.Collection;
import java.util.Iterator;
import java.util.NoSuchElementException;
/** A type-specific hash big set with with a fast, small-footprint implementation.
*
* <p>Instances of this class use a hash table to represent a big set: the number
* of elements in the set is limited only by the amount of core memory. The table
* (backed by a {@linkplain it.unimi.dsi.fastutil.BigArrays big array}) is
* filled up to a specified <em>load factor</em>, and then doubled in size to
* accommodate new entries. If the table is emptied below <em>one fourth</em>
* of the load factor, it is halved in size; however, the table is never reduced to a
* size smaller than that at creation time: this approach makes it
* possible to create sets with a large capacity in which insertions and
* deletions do not cause immediately rehashing. Moreover, halving is
* not performed when deleting entries from an iterator, as it would interfere
* with the iteration process.
*
* <p>Note that {@link #clear()} does not modify the hash table size.
* Rather, a family of {@linkplain #trim() trimming
* methods} lets you control the size of the table; this is particularly useful
* if you reuse instances of this class.
*
* <p>The methods of this class are about 30% slower than those of the corresponding non-big set.
*
* @see Hash
* @see HashCommon
*/
public class OPEN_HASH_BIG_SET KEY_GENERIC extends ABSTRACT_SET KEY_GENERIC implements java.io.Serializable, Cloneable, Hash, Size64 {
private static final long serialVersionUID = 0L;
private static final boolean ASSERTS = ASSERTS_VALUE;
/** The big array of keys. */
protected transient KEY_GENERIC_TYPE[][] key;
/** The mask for wrapping a position counter. */
protected transient long mask;
/** The mask for wrapping a segment counter. */
protected transient int segmentMask;
/** The mask for wrapping a base counter. */
protected transient int baseMask;
/** Whether this set contains the null key. */
protected transient boolean containsNull;
/** The current table size (always a power of 2). */
protected transient long n;
/** Threshold after which we rehash. It must be the table size times {@link #f}. */
protected transient long maxFill;
/** We never resize below this threshold, which is the construction-time {#n}. */
protected final transient long minN;
/** The acceptable load factor. */
protected final float f;
/** Number of entries in the set. */
protected long size;
/** Initialises the mask values. */
private void initMasks() {
mask = n - 1;
/* Note that either we have more than one segment, and in this case all segments
* are BigArrays.SEGMENT_SIZE long, or we have exactly one segment whose length
* is a power of two. */
segmentMask = key[0].length - 1;
baseMask = key.length - 1;
}
/** Creates a new hash big set.
*
* <p>The actual table size will be the least power of two greater than {@code expected}/{@code f}.
*
* @param expected the expected number of elements in the set.
* @param f the load factor.
*/
SUPPRESS_WARNINGS_KEY_UNCHECKED
public OPEN_HASH_BIG_SET(final long expected, final float f) {
if (f <= 0 || f > 1) throw new IllegalArgumentException("Load factor must be greater than 0 and smaller than or equal to 1");
if (n < 0) throw new IllegalArgumentException("The expected number of elements must be nonnegative");
this.f = f;
minN = n = bigArraySize(expected, f);
maxFill = maxFill(n, f);
key = KEY_GENERIC_BIG_ARRAY_CAST BIG_ARRAYS.newBigArray(n);
initMasks();
}
/** Creates a new hash big set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor.
*
* @param expected the expected number of elements in the hash big set.
*/
public OPEN_HASH_BIG_SET(final long expected) {
this(expected, DEFAULT_LOAD_FACTOR);
}
/** Creates a new hash big set with initial expected {@link Hash#DEFAULT_INITIAL_SIZE} elements
* and {@link Hash#DEFAULT_LOAD_FACTOR} as load factor.
*/
public OPEN_HASH_BIG_SET() {
this(DEFAULT_INITIAL_SIZE, DEFAULT_LOAD_FACTOR);
}
/** Creates a new hash big set copying a given collection.
*
* @param c a {@link Collection} to be copied into the new hash big set.
* @param f the load factor.
*/
public OPEN_HASH_BIG_SET(final Collection<? extends KEY_GENERIC_CLASS> c, final float f) {
this(Size64.sizeOf(c), f);
addAll(c);
}
/** Creates a new hash big set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor
* copying a given collection.
*
* @param c a {@link Collection} to be copied into the new hash big set.
*/
public OPEN_HASH_BIG_SET(final Collection<? extends KEY_GENERIC_CLASS> c) {
this(c, DEFAULT_LOAD_FACTOR);
}
/** Creates a new hash big set copying a given type-specific collection.
*
* @param c a type-specific collection to be copied into the new hash big set.
* @param f the load factor.
*/
public OPEN_HASH_BIG_SET(final COLLECTION KEY_EXTENDS_GENERIC c, final float f) {
this(Size64.sizeOf(c), f);
addAll(c);
}
/** Creates a new hash big set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor
* copying a given type-specific collection.
*
* @param c a type-specific collection to be copied into the new hash big set.
*/
public OPEN_HASH_BIG_SET(final COLLECTION KEY_EXTENDS_GENERIC c) {
this(c, DEFAULT_LOAD_FACTOR);
}
/** Creates a new hash big set using elements provided by a type-specific iterator.
*
* @param i a type-specific iterator whose elements will fill the new hash big set.
* @param f the load factor.
*/
public OPEN_HASH_BIG_SET(final STD_KEY_ITERATOR KEY_EXTENDS_GENERIC i, final float f) {
this(DEFAULT_INITIAL_SIZE, f);
while(i.hasNext()) add(i.NEXT_KEY());
}
/** Creates a new hash big set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor using elements provided by a type-specific iterator.
*
* @param i a type-specific iterator whose elements will fill the new hash big set.
*/
public OPEN_HASH_BIG_SET(final STD_KEY_ITERATOR KEY_EXTENDS_GENERIC i) {
this(i, DEFAULT_LOAD_FACTOR);
}
#if KEYS_PRIMITIVE
/** Creates a new hash big set using elements provided by an iterator.
*
* @param i an iterator whose elements will fill the new hash big set.
* @param f the load factor.
*/
public OPEN_HASH_BIG_SET(final Iterator<?> i, final float f) {
this(ITERATORS.AS_KEY_ITERATOR(i), f);
}
/** Creates a new hash big set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor using elements provided by an iterator.
*
* @param i an iterator whose elements will fill the new hash big set.
*/
public OPEN_HASH_BIG_SET(final Iterator<?> i) {
this(ITERATORS.AS_KEY_ITERATOR(i));
}
#endif
/** Creates a new hash big set and fills it with the elements of a given array.
*
* @param a an array whose elements will be used to fill the new hash big set.
* @param offset the first element to use.
* @param length the number of elements to use.
* @param f the load factor.
*/
public OPEN_HASH_BIG_SET(final KEY_GENERIC_TYPE[] a, final int offset, final int length, final float f) {
this(length < 0 ? 0 : length, f);
ARRAYS.ensureOffsetLength(a, offset, length);
for(int i = 0; i < length; i++) add(a[offset + i]);
}
/** Creates a new hash big set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor and fills it with the elements of a given array.
*
* @param a an array whose elements will be used to fill the new hash big set.
* @param offset the first element to use.
* @param length the number of elements to use.
*/
public OPEN_HASH_BIG_SET(final KEY_GENERIC_TYPE[] a, final int offset, final int length) {
this(a, offset, length, DEFAULT_LOAD_FACTOR);
}
/** Creates a new hash big set copying the elements of an array.
*
* @param a an array to be copied into the new hash big set.
* @param f the load factor.
*/
public OPEN_HASH_BIG_SET(final KEY_GENERIC_TYPE[] a, final float f) {
this(a, 0, a.length, f);
}
/** Creates a new hash big set with {@link Hash#DEFAULT_LOAD_FACTOR} as load factor
* copying the elements of an array.
*
* @param a an array to be copied into the new hash big set.
*/
public OPEN_HASH_BIG_SET(final KEY_GENERIC_TYPE[] a) {
this(a, DEFAULT_LOAD_FACTOR);
}
#ifndef Custom
#if KEYS_INT_LONG_DOUBLE
/** Collects the result of a primitive {@code Stream} into a new big hash set.
*
* <p>This method performs a terminal operation on the given {@code Stream}
*
* @apiNote Taking a primitive stream instead of returning something like a
* {@link java.util.stream.Collector Collector} is necessary because there is no
* primitive {@code Collector} equivalent in the Java API.
*/
public static KEY_GENERIC OPEN_HASH_BIG_SET KEY_GENERIC toBigSet(JDK_PRIMITIVE_STREAM stream) {
return stream.collect(
OPEN_HASH_BIG_SET::new,
OPEN_HASH_BIG_SET::add,
OPEN_HASH_BIG_SET::addAll);
}
/** Collects the result of a primitive {@code Stream} into a new big hash set.
*
* <p>This method performs a terminal operation on the given {@code Stream}
*
* @apiNote Taking a primitive stream instead returning something like a
* {@link java.util.stream.Collector Collector} is necessary because there is no
* primitive {@code Collector} equivalent in the Java API.
*/
public static KEY_GENERIC OPEN_HASH_BIG_SET KEY_GENERIC toBigSetWithExpectedSize(JDK_PRIMITIVE_STREAM stream, long expectedSize) {
return stream.collect(
() -> new OPEN_HASH_BIG_SET KEY_GENERIC(expectedSize),
OPEN_HASH_BIG_SET::add,
OPEN_HASH_BIG_SET::addAll);
}
#elif KEYS_REFERENCE
// Collector wants a function that returns the collection being added to.
private OPEN_HASH_BIG_SET KEY_GENERIC combine(OPEN_HASH_BIG_SET KEY_EXTENDS_GENERIC toAddFrom) {
addAll(toAddFrom);
return this;
}
private static final Collector<KEY_TYPE, ?, OPEN_HASH_BIG_SET<KEY_TYPE>> TO_SET_COLLECTOR =
Collector.of(
OPEN_HASH_BIG_SET::new,
OPEN_HASH_BIG_SET::add,
OPEN_HASH_BIG_SET::combine);
/** Returns a {@link Collector} that collects a {@code Stream}'s elements into a new big hash set. */
SUPPRESS_WARNINGS_KEY_UNCHECKED_RAWTYPES
public static KEY_GENERIC Collector<KEY_GENERIC_TYPE, ?, OPEN_HASH_BIG_SET KEY_GENERIC> toBigSet() {
return (Collector) TO_SET_COLLECTOR;
}
/** Returns a {@link Collector} that collects a {@code Stream}'s elements into a new big hash set. */
public static KEY_GENERIC Collector<KEY_GENERIC_TYPE, ?, OPEN_HASH_BIG_SET KEY_GENERIC> toBigSetWithExpectedSize(long expectedSize) {
return Collector.of(
() -> new OPEN_HASH_BIG_SET KEY_GENERIC(expectedSize),
OPEN_HASH_BIG_SET::add,
OPEN_HASH_BIG_SET::combine);
}
#endif
#endif
private long realSize() {
return containsNull ? size - 1 : size;
}
/** Ensures that this big set can hold a certain number of elements without rehashing.
*
* @param capacity a number of elements; there will be no rehashing unless
* the set {@linkplain #size64() size} exceeds this number.
*/
public void ensureCapacity(final long capacity) {
final long needed = bigArraySize(capacity, f);
if (needed > n) rehash(needed);
}
@Override
public boolean addAll(Collection<? extends KEY_GENERIC_CLASS> c) {
final long size = Size64.sizeOf(c);
// The resulting collection will be at least c.size() big
if (f <= .5) ensureCapacity(size); // The resulting collection will be sized for c.size() elements
else ensureCapacity(size64() + size); // The resulting collection will be sized for size() + c.size() elements
return super.addAll(c);
}
#if KEYS_PRIMITIVE
@Override
public boolean addAll(COLLECTION c) {
final long size = Size64.sizeOf(c);
if (f <= .5) ensureCapacity(size); // The resulting collection will be size for c.size() elements
else ensureCapacity(size64() + size); // The resulting collection will be sized for size() + c.size() elements
return super.addAll(c);
}
#endif
@Override
public boolean add(final KEY_GENERIC_TYPE k) {
int displ, base;
if (KEY_IS_NULL(k)) {
if (containsNull) return false;
containsNull = true;
}
else {
KEY_GENERIC_TYPE curr;
final KEY_GENERIC_TYPE[][] key = this.key;
final long h = KEY2LONGHASH(k);
// The starting point.
if (! KEY_IS_NULL(curr = key[base = (int)((h & mask) >>> BigArrays.SEGMENT_SHIFT)][displ = (int)(h & segmentMask)])) {
if (KEY_EQUALS_NOT_NULL(curr, k)) return false;
while(! KEY_IS_NULL(curr = key[base = (base + ((displ = (displ + 1) & segmentMask) == 0 ? 1 : 0)) & baseMask][displ]))
if (KEY_EQUALS_NOT_NULL(curr, k)) return false;
}
key[base][displ] = k;
}
if (size++ >= maxFill) rehash(2 * n);
if (ASSERTS) checkTable();
return true;
}
#if KEY_CLASS_Object
/** Add a random element if not present, get the existing value if already present.
*
* This is equivalent to (but faster than) doing a:
* <pre>
* K exist = set.get(k);
* if (exist == null) {
* set.add(k);
* exist = k;
* }
* </pre>
*/
public KEY_GENERIC_TYPE addOrGet(final KEY_GENERIC_TYPE k) {
int displ, base;
if (KEY_IS_NULL(k)) {
if (containsNull) return null;
containsNull = true;
}
else {
KEY_GENERIC_TYPE curr;
final KEY_GENERIC_TYPE[][] key = this.key;
final long h = KEY2LONGHASH(k);
// The starting point.
if (! KEY_IS_NULL(curr = key[base = (int)((h & mask) >>> BigArrays.SEGMENT_SHIFT)][displ = (int)(h & segmentMask)])) {
if (KEY_EQUALS_NOT_NULL(curr, k)) return curr;
while(! KEY_IS_NULL(curr = key[base = (base + ((displ = (displ + 1) & segmentMask) == 0 ? 1 : 0)) & baseMask][displ]))
if (KEY_EQUALS_NOT_NULL(curr, k)) return curr;
}
key[base][displ] = k;
}
if (size++ >= maxFill) rehash(2 * n);
if (ASSERTS) checkTable();
return k;
}
#endif
/** Shifts left entries with the specified hash code, starting at the specified position,
* and empties the resulting free entry.
*
* @param pos a starting position.
*/
protected final void shiftKeys(long pos) {
// Shift entries with the same hash.
long last, slot;
final KEY_GENERIC_TYPE[][] key = this.key;
for(;;) {
pos = ((last = pos) + 1) & mask;
for(;;) {
if (KEY_IS_NULL(BigArrays.get(key, pos))) {
set(key, last, KEY_NULL);
return;
}
slot = KEY2LONGHASH(BigArrays.get(key, pos)) & mask;
if (last <= pos ? last >= slot || slot > pos : last >= slot && slot > pos) break;
pos = (pos + 1) & mask;
}
set(key, last, BigArrays.get(key, pos));
}
}
private boolean removeEntry(final int base, final int displ) {
size--;
shiftKeys(base * (long)BigArrays.SEGMENT_SIZE + displ);
if (n > minN && size < maxFill / 4 && n > DEFAULT_INITIAL_SIZE) rehash(n / 2);
return true;
}
private boolean removeNullEntry() {
containsNull = false;
size--;
if (n > minN && size < maxFill / 4 && n > DEFAULT_INITIAL_SIZE) rehash(n / 2);
return true;
}
@Override
public boolean remove(final KEY_TYPE k) {
if (KEY_IS_NULL(k)) {
if (containsNull) return removeNullEntry();
return false;
}
KEY_GENERIC_TYPE curr;
final KEY_GENERIC_TYPE[][] key = this.key;
final long h = KEY2LONGHASH(k);
int displ, base;
// The starting point.
if (KEY_IS_NULL(curr = key[base = (int)((h & mask) >>> BigArrays.SEGMENT_SHIFT)][displ = (int)(h & segmentMask)])) return false;
if (KEY_EQUALS_NOT_NULL(curr, k)) return removeEntry(base, displ);
while(true) {
if (KEY_IS_NULL(curr = key[base = (base + ((displ = (displ + 1) & segmentMask) == 0 ? 1 : 0)) & baseMask][displ])) return false;
if (KEY_EQUALS_NOT_NULL(curr, k)) return removeEntry(base, displ);
}
}
@Override
public boolean contains(final KEY_TYPE k) {
if (KEY_IS_NULL(k)) return containsNull;
KEY_GENERIC_TYPE curr;
final KEY_GENERIC_TYPE[][] key = this.key;
final long h = KEY2LONGHASH(k);
int displ, base;
// The starting point.
if (KEY_IS_NULL(curr = key[base = (int)((h & mask) >>> BigArrays.SEGMENT_SHIFT)][displ = (int)(h & segmentMask)])) return false;
if (KEY_EQUALS_NOT_NULL(curr, k)) return true;
while(true) {
if (KEY_IS_NULL(curr = key[base = (base + ((displ = (displ + 1) & segmentMask) == 0 ? 1 : 0)) & baseMask][displ])) return false;
if (KEY_EQUALS_NOT_NULL(curr, k)) return true;
}
}
#if KEY_CLASS_Object
/** Returns the element of this set that is equal to the given key, or {@code null}.
* @return the element of this set that is equal to the given key, or {@code null}.
*/
public K get(final KEY_TYPE k) {
if (k == null) return null; // This is correct independently of the value of containsNull
KEY_GENERIC_TYPE curr;
final KEY_GENERIC_TYPE[][] key = this.key;
final long h = KEY2LONGHASH(k);
int displ, base;
// The starting point.
if (KEY_IS_NULL(curr = key[base = (int)((h & mask) >>> BigArrays.SEGMENT_SHIFT)][displ = (int)(h & segmentMask)])) return null;
if (KEY_EQUALS_NOT_NULL(curr, k)) return curr;
while(true) {
if (KEY_IS_NULL(curr = key[base = (base + ((displ = (displ + 1) & segmentMask) == 0 ? 1 : 0)) & baseMask][displ])) return null;
if (KEY_EQUALS_NOT_NULL(curr, k)) return curr;
}
}
#endif
/* Removes all elements from this set.
*
*/
/** {@inheritDoc}
*
* <p>To increase object reuse, this method does not change the table size.
* If you want to reduce the table size, you must use {@link #trim(long)}.
*/
@Override
public void clear() {
if (size == 0) return;
size = 0;
containsNull = false;
fill(key, KEY_NULL);
}
/** An iterator over a hash big set. */
private class SetIterator implements KEY_ITERATOR KEY_GENERIC {
/** The base of the last entry returned, if positive or zero; initially, the number of components
of the key array. If negative, the last element returned was
that of index {@code - base - 1} from the {@link #wrapped} list. */
int base = key.length;
/** The displacement of the last entry returned; initially, zero. */
int displ;
/** The index of the last entry that has been returned (or {@link Long#MIN_VALUE} if {@link #base} is negative).
It is -1 if either we did not return an entry yet, or the last returned entry has been removed. */
long last = -1;
/** A downward counter measuring how many entries must still be returned. */
long c = size;
/** A boolean telling us whether we should return the null key. */
boolean mustReturnNull = OPEN_HASH_BIG_SET.this.containsNull;
/** A lazily allocated list containing elements that have wrapped around the table because of removals. */
ARRAY_LIST KEY_GENERIC wrapped;
@Override
public boolean hasNext() { return c != 0; }
@Override
public KEY_GENERIC_TYPE NEXT_KEY() {
if (! hasNext()) throw new NoSuchElementException();
c--;
if (mustReturnNull) {
mustReturnNull = false;
last = n;
return KEY_NULL;
}
final KEY_GENERIC_TYPE[][] key = OPEN_HASH_BIG_SET.this.key;
for(;;) {
if (displ == 0 && base <= 0) {
// We are just enumerating elements from the wrapped list.
last = Long.MIN_VALUE;
return wrapped.GET_KEY(- (--base) - 1);
}
if (displ-- == 0) displ = key[--base].length - 1;
final KEY_GENERIC_TYPE k = key[base][displ];
if (! KEY_IS_NULL(k)) {
last = base * (long)BigArrays.SEGMENT_SIZE + displ;
return k;
}
}
}
/** Shifts left entries with the specified hash code, starting at the specified position,
* and empties the resulting free entry.
*
* @param pos a starting position.
*/
private final void shiftKeys(long pos) {
// Shift entries with the same hash.
long last, slot;
KEY_GENERIC_TYPE curr;
final KEY_GENERIC_TYPE[][] key = OPEN_HASH_BIG_SET.this.key;
for(;;) {
pos = ((last = pos) + 1) & mask;
for(;;) {
if(KEY_IS_NULL(curr = BigArrays.get(key, pos))) {
set(key, last, KEY_NULL);
return;
}
slot = KEY2LONGHASH(curr) & mask;
if (last <= pos ? last >= slot || slot > pos : last >= slot && slot > pos) break;
pos = (pos + 1) & mask;
}
if (pos < last) { // Wrapped entry.
if (wrapped == null) wrapped = new ARRAY_LIST KEY_GENERIC_DIAMOND();
wrapped.add(BigArrays.get(key, pos));
}
set(key, last, curr);
}
}
@Override
public void remove() {
if (last == -1) throw new IllegalStateException();
if (last == n) OPEN_HASH_BIG_SET.this.containsNull = false;
else if (base >= 0) shiftKeys(last);
else {
// We're removing wrapped entries.
#if KEYS_REFERENCE
OPEN_HASH_BIG_SET.this.remove(wrapped.set(- base - 1, null));
#else
OPEN_HASH_BIG_SET.this.remove(wrapped.GET_KEY(- base - 1));
#endif
last = -1; // Note that we must not decrement size
return;
}
size--;
last = -1; // You can no longer remove this entry.
if (ASSERTS) checkTable();
}
}
@Override
public KEY_ITERATOR KEY_GENERIC iterator() {
return new SetIterator();
}
private class SetSpliterator implements KEY_SPLITERATOR KEY_GENERIC {
/* For the sake of keeping things at least somewhat simple
* (aka. my sanity), the spliterator will NOT handle the indexing
* of the subarrays directly, like iterator does. Instead, it will
* delegate to BigArrays and have only a single, unified index it
* will fence on. This is probably less effecient, but it avoids having
* to track what it means to split on two sets of indexes.
* This may change in the future if the performance hit high.
*/
private static final int POST_SPLIT_CHARACTERISTICS = SPLITERATORS.SET_SPLITERATOR_CHARACTERISTICS & ~java.util.Spliterator.SIZED;
/** The index (which bucket) of the next item to give to the action. */
long pos = 0;
/** The maximum bucket (exclusive) to iterate to */
long max = n;
/** An upwards counter counting how many we have given */
long c = 0;
/** A boolean telling us whether we should return the null key. */
boolean mustReturnNull = OPEN_HASH_BIG_SET.this.containsNull;
boolean hasSplit = false;
SetSpliterator() {}
SetSpliterator(long pos, long max, boolean mustReturnNull, boolean hasSplit) {
this.pos = pos;
this.max = max;
this.mustReturnNull = mustReturnNull;
this.hasSplit = hasSplit;
}
@Override
public boolean tryAdvance(final METHOD_ARG_KEY_CONSUMER action) {
if (mustReturnNull) {
mustReturnNull = false;
++c;
action.accept(KEY_NULL);
return true;
}
final KEY_GENERIC_TYPE key[][] = OPEN_HASH_BIG_SET.this.key;
while (pos < max) {
KEY_GENERIC_TYPE gotten = BigArrays.get(key, pos);
if (! KEY_IS_NULL(gotten)) {
++c;
++pos;
action.accept(gotten);
return true;
} else {
++pos;
}
}
return false;
}
@Override
public void forEachRemaining(final METHOD_ARG_KEY_CONSUMER action) {
if (mustReturnNull) {
mustReturnNull = false;
action.accept(KEY_NULL);
++c;
}
final KEY_GENERIC_TYPE key[][] = OPEN_HASH_BIG_SET.this.key;
while (pos < max) {
KEY_GENERIC_TYPE gotten = BigArrays.get(key, pos);
if (! KEY_IS_NULL(gotten)) {
action.accept(gotten);
++c;
}
++pos;
}
}
@Override
public int characteristics() {
return hasSplit ? POST_SPLIT_CHARACTERISTICS : SPLITERATORS.SET_SPLITERATOR_CHARACTERISTICS;
}
@Override
public long estimateSize() {
if (!hasSplit) {
// Root spliterator; we know how many are remaining.
return size - c;
} else {
// After we split, we can no longer know exactly how many we have (or at least not efficiently).
// (size / n) * (max - pos) aka currentTableDensity * numberOfBucketsLeft seems like a good estimate.
return Math.min(size - c, (long)(((double)realSize() / n) * (max - pos)) + (mustReturnNull ? 1 : 0));
}
}
@Override
public SetSpliterator trySplit() {
if (pos >= max - 1) return null;
long retLen = (max - pos) >> 1;
if (retLen <= 1) return null;
long myNewPos = pos + retLen;
// Align to an outer array boundary if possible
// We add/subtract one to the bounds to ensure the new pos will always shrink the range
myNewPos = BigArrays.nearestSegmentStart(myNewPos, pos + 1, max - 1);
long retPos = pos;
long retMax = myNewPos;
// Since null is returned first, and the convention is that the returned split is the prefix of elements,
// the split will take care of returning null (if needed), and we won't return it anymore.
SetSpliterator split = new SetSpliterator(retPos, retMax, mustReturnNull, true);
this.pos = myNewPos;
this.mustReturnNull = false;
this.hasSplit = true;
return split;
}
@Override
public long skip(long n) {
if (n < 0) throw new IllegalArgumentException("Argument must be nonnegative: " + n);
if (n == 0) return 0;
long skipped = 0;
if (mustReturnNull) {
mustReturnNull = false;
++skipped;
--n;
}
final KEY_GENERIC_TYPE key[][] = OPEN_HASH_BIG_SET.this.key;
while (pos < max && n > 0) {
if (! KEY_IS_NULL(BigArrays.get(key, pos++))) {
++skipped;
--n;
}
}
return skipped;
}
}
@Override
public KEY_SPLITERATOR KEY_GENERIC spliterator() {
return new SetSpliterator();
}
@Override
public void forEach(final METHOD_ARG_KEY_CONSUMER action) {
if (containsNull) {
action.accept(KEY_NULL);
}
long pos = 0;
final long max = n;
final KEY_GENERIC_TYPE key[][] = this.key;
while (pos < max) {
KEY_GENERIC_TYPE gotten = BigArrays.get(key, pos++);
if (! KEY_IS_NULL(gotten)) {
action.accept(gotten);
}
}
}
/** Rehashes this set, making the table as small as possible.
*
* <p>This method rehashes the table to the smallest size satisfying the
* load factor. It can be used when the set will not be changed anymore, so
* to optimize access speed and size.
*
* <p>If the table size is already the minimum possible, this method
* does nothing.
*
* @return true if there was enough memory to trim the set.
* @see #trim(long)
*/
public boolean trim() {
return trim(size);
}
/** Rehashes this set if the table is too large.
*
* <p>Let <var>N</var> be the smallest table size that can hold
* <code>max(n,{@link #size64()})</code> entries, still satisfying the load factor. If the current
* table size is smaller than or equal to <var>N</var>, this method does
* nothing. Otherwise, it rehashes this set in a table of size
* <var>N</var>.
*
* <p>This method is useful when reusing sets. {@linkplain #clear() Clearing a
* set} leaves the table size untouched. If you are reusing a set
* many times, you can call this method with a typical
* size to avoid keeping around a very large table just
* because of a few large transient sets.
*
* @param n the threshold for the trimming.
* @return true if there was enough memory to trim the set.
* @see #trim()
*/
public boolean trim(final long n) {
final long l = bigArraySize(n, f);
if (l >= this.n || size > maxFill(l, f)) return true;
try {
rehash(l);
}
catch(OutOfMemoryError cantDoIt) { return false; }
return true;
}
/** Resizes the set.
*
* <p>This method implements the basic rehashing strategy, and may be
* overriden by subclasses implementing different rehashing strategies (e.g.,
* disk-based rehashing). However, you should not override this method
* unless you understand the internal workings of this class.
*
* @param newN the new size
*/
SUPPRESS_WARNINGS_KEY_UNCHECKED
protected void rehash(final long newN) {
final KEY_GENERIC_TYPE key[][] = this.key;
final KEY_GENERIC_TYPE newKey[][] = KEY_GENERIC_BIG_ARRAY_CAST BIG_ARRAYS.newBigArray(newN);
final long mask = newN - 1; // Note that this is used by the hashing macro
final int newSegmentMask = newKey[0].length - 1;
final int newBaseMask = newKey.length - 1;
int base = 0, displ = 0, b, d;
long h;
KEY_GENERIC_TYPE k;
for(long i = realSize(); i-- != 0;) {
while(KEY_IS_NULL(key[base][displ])) base = (base + ((displ = (displ + 1) & segmentMask) == 0 ? 1 : 0));
k = key[base][displ];
h = KEY2LONGHASH(k);
// The starting point.
if (! KEY_IS_NULL(newKey[b = (int)((h & mask) >>> BigArrays.SEGMENT_SHIFT)][d = (int)(h & newSegmentMask)]))
while(! KEY_IS_NULL(newKey[b = (b + ((d = (d + 1) & newSegmentMask) == 0 ? 1 : 0)) & newBaseMask][d]));
newKey[b][d] = k;
base = (base + ((displ = (displ + 1) & segmentMask) == 0 ? 1 : 0));
}
this.n = newN;
this.key = newKey;
initMasks();
maxFill = maxFill(n, f);
}
@Deprecated
@Override
public int size() {
return (int)Math.min(Integer.MAX_VALUE, size);
}
@Override
public long size64() {
return size;
}
@Override
public boolean isEmpty() {
return size == 0;
}
/** Returns a deep copy of this big set.
*
* <p>This method performs a deep copy of this big hash set; the data stored in the
* set, however, is not cloned. Note that this makes a difference only for object keys.
*
* @return a deep copy of this big set.
*/
@Override
SUPPRESS_WARNINGS_KEY_UNCHECKED
public OPEN_HASH_BIG_SET KEY_GENERIC clone() {
OPEN_HASH_BIG_SET KEY_GENERIC c;
try {
c = (OPEN_HASH_BIG_SET KEY_GENERIC)super.clone();
}
catch(CloneNotSupportedException cantHappen) {
throw new InternalError();
}
c.key = copy(key);
c.containsNull = containsNull;
return c;
}
/** Returns a hash code for this set.
*
* This method overrides the generic method provided by the superclass.
* Since {@code equals()} is not overriden, it is important
* that the value returned by this method is the same value as
* the one returned by the overriden method.
*
* @return a hash code for this set.
*/
@Override
public int hashCode() {
final KEY_GENERIC_TYPE key[][] = this.key;
int h = 0, base = 0, displ = 0;
for(long j = realSize(); j-- != 0;) {
while(KEY_IS_NULL(key[base][displ])) base = (base + ((displ = (displ + 1) & segmentMask) == 0 ? 1 : 0));
#if KEYS_REFERENCE
if (this != key[base][displ])
#endif
h += KEY2JAVAHASH_NOT_NULL(key[base][displ]);
base = (base + ((displ = (displ + 1) & segmentMask) == 0 ? 1 : 0));
}
return h;
}
private void writeObject(java.io.ObjectOutputStream s) throws java.io.IOException {
final KEY_ITERATOR KEY_GENERIC i = iterator();
s.defaultWriteObject();
for(long j = size; j-- != 0;) s.WRITE_KEY(i.NEXT_KEY());
}
SUPPRESS_WARNINGS_KEY_UNCHECKED
private void readObject(java.io.ObjectInputStream s) throws java.io.IOException, ClassNotFoundException {
s.defaultReadObject();
n = bigArraySize(size, f);
maxFill = maxFill(n, f);
final KEY_GENERIC_TYPE[][] key = this.key = KEY_GENERIC_BIG_ARRAY_CAST BIG_ARRAYS.newBigArray(n);
initMasks();