-
-
Notifications
You must be signed in to change notification settings - Fork 397
/
bounded_integer_sequences.pyx
1445 lines (1135 loc) · 45.1 KB
/
bounded_integer_sequences.pyx
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
r"""
Sequences of bounded integers
This module provides :class:`BoundedIntegerSequence`, which implements
sequences of bounded integers and is for many (but not all) operations faster
than representing the same sequence as a Python :class:`tuple`.
The underlying data structure is similar to :class:`~sage.misc.bitset.Bitset`,
which means that certain operations are implemented by using fast shift
operations from MPIR. The following boilerplate functions can be
cimported in Cython modules:
- ``cdef bint biseq_init(biseq_t R, mp_size_t l, mp_size_t itemsize) except -1``
Allocate memory for a bounded integer sequence of length ``l`` with
items fitting in ``itemsize`` bits.
- ``cdef inline void biseq_dealloc(biseq_t S)``
Deallocate the memory used by ``S``.
- ``cdef bint biseq_init_copy(biseq_t R, biseq_t S)``
Initialize ``R`` as a copy of ``S``.
- ``cdef tuple biseq_pickle(biseq_t S)``
Return a triple ``(bitset_data, itembitsize, length)`` defining ``S``.
- ``cdef bint biseq_unpickle(biseq_t R, tuple bitset_data, mp_bitcnt_t itembitsize, mp_size_t length) except -1``
Initialise ``R`` from data returned by ``biseq_pickle``.
- ``cdef bint biseq_init_list(biseq_t R, list data, size_t bound) except -1``
Convert a list to a bounded integer sequence, which must not be allocated.
- ``cdef inline Py_hash_t biseq_hash(biseq_t S)``
Hash value for ``S``.
- ``cdef inline bint biseq_richcmp(biseq_t S1, biseq_t S2, int op)``
Comparison of ``S1`` and ``S2``. This takes into account the bound, the
length, and the list of items of the two sequences.
- ``cdef bint biseq_init_concat(biseq_t R, biseq_t S1, biseq_t S2) except -1``
Concatenate ``S1`` and ``S2`` and write the result to ``R``. Does not test
whether the sequences have the same bound!
- ``cdef inline bint biseq_startswith(biseq_t S1, biseq_t S2)``
Is ``S1=S2+something``? Does not check whether the sequences have the same
bound!
- ``cdef mp_size_t biseq_contains(biseq_t S1, biseq_t S2, mp_size_t start) except -2``
Return the position in ``S1`` of ``S2`` as a subsequence of
``S1[start:]``, or ``-1`` if ``S2`` is not a subsequence. Does not check
whether the sequences have the same bound!
- ``cdef mp_size_t biseq_starswith_tail(biseq_t S1, biseq_t S2, mp_size_t start) except -2:``
Return the smallest number ``i`` such that the bounded integer sequence
``S1`` starts with the sequence ``S2[i:]``, where ``start <= i <
S1.length``, or return ``-1`` if no such ``i`` exists.
- ``cdef mp_size_t biseq_index(biseq_t S, size_t item, mp_size_t start) except -2``
Return the position in ``S`` of the item in ``S[start:]``, or ``-1`` if
``S[start:]`` does not contain the item.
- ``cdef size_t biseq_getitem(biseq_t S, mp_size_t index)``
Return ``S[index]``, without checking margins.
- ``cdef size_t biseq_getitem_py(biseq_t S, mp_size_t index)``
Return ``S[index]`` as Python ``int`` or ``long``, without checking margins.
- ``cdef biseq_inititem(biseq_t S, mp_size_t index, size_t item)``
Set ``S[index] = item``, without checking margins and assuming that ``S[index]``
has previously been zero.
- ``cdef inline void biseq_clearitem(biseq_t S, mp_size_t index)``
Set ``S[index] = 0``, without checking margins.
- ``cdef bint biseq_init_slice(biseq_t R, biseq_t S, mp_size_t start, mp_size_t stop, mp_size_t step) except -1``
Initialise ``R`` with ``S[start:stop:step]``.
AUTHORS:
- Simon King, Jeroen Demeyer (2014-10): initial version (:trac:`15820`)
"""
# ****************************************************************************
# Copyright (C) 2014 Simon King <simon.king@uni-jena.de>
# Copyright (C) 2014 Jeroen Demeyer <jdemeyer@cage.ugennt.be>
#
# Distributed under the terms of the GNU General Public License (GPL)
# as published by the Free Software Foundation; either version 2 of
# the License, or (at your option) any later version.
# https://www.gnu.org/licenses/
# ****************************************************************************
from cysignals.signals cimport sig_check, sig_on, sig_off
include 'sage/data_structures/bitset.pxi'
from cpython.int cimport PyInt_FromSize_t
from cpython.slice cimport PySlice_GetIndicesEx
from sage.libs.gmp.mpn cimport mpn_rshift, mpn_lshift, mpn_copyi, mpn_ior_n, mpn_zero, mpn_copyd, mpn_cmp
from sage.libs.flint.flint cimport FLINT_BIT_COUNT as BIT_COUNT
from sage.structure.richcmp cimport richcmp_not_equal, rich_to_bool
cimport cython
###################
# Boilerplate
# cdef functions
###################
#
# (De)allocation, copying
#
@cython.overflowcheck
cdef bint biseq_init(biseq_t R, mp_size_t l, mp_bitcnt_t itemsize) except -1:
"""
Allocate memory for a bounded integer sequence of length ``l`` with
items fitting in ``itemsize`` bits.
"""
cdef mp_bitcnt_t totalbitsize
if l:
totalbitsize = l * itemsize
else:
totalbitsize = 1
bitset_init(R.data, totalbitsize)
R.length = l
R.itembitsize = itemsize
R.mask_item = limb_lower_bits_up(itemsize)
cdef inline void biseq_dealloc(biseq_t S):
"""
Deallocate the memory used by ``S``.
"""
bitset_free(S.data)
cdef bint biseq_init_copy(biseq_t R, biseq_t S) except -1:
"""
Initialize ``R`` as a copy of ``S``.
"""
biseq_init(R, S.length, S.itembitsize)
sig_on()
bitset_copy(R.data, S.data)
sig_off()
#
# Pickling
#
cdef tuple biseq_pickle(biseq_t S):
return (bitset_pickle(S.data), S.itembitsize, S.length)
cdef bint biseq_unpickle(biseq_t R, tuple bitset_data, mp_bitcnt_t itembitsize, mp_size_t length) except -1:
biseq_init(R, length, itembitsize)
sig_on()
bitset_unpickle(R.data, bitset_data)
sig_off()
return 1
#
# Conversion
#
cdef bint biseq_init_list(biseq_t R, list data, size_t bound) except -1:
"""
Convert a list into a bounded integer sequence and write the result
into ``R``, which must not be initialised.
INPUT:
- ``data`` -- a list of integers
- ``bound`` -- a number which is the maximal value of an item
"""
cdef mp_size_t index = 0
cdef size_t item_c
biseq_init(R, len(data), BIT_COUNT(bound|<size_t>1))
for item in data:
sig_check()
item_c = item
if item_c > bound:
raise OverflowError("list item {!r} larger than {}".format(item, bound) )
biseq_inititem(R, index, item_c)
index += 1
cdef inline Py_hash_t biseq_hash(biseq_t S):
return S.itembitsize*(<Py_hash_t>1073807360)+bitset_hash(S.data)
cdef inline bint biseq_richcmp(biseq_t S1, biseq_t S2, int op):
if S1.itembitsize != S2.itembitsize:
return richcmp_not_equal(S1.itembitsize, S2.itembitsize, op)
if S1.length != S2.length:
return richcmp_not_equal(S1.length, S2.length, op)
return rich_to_bool(op, bitset_cmp(S1.data, S2.data))
#
# Arithmetics
#
cdef bint biseq_init_concat(biseq_t R, biseq_t S1, biseq_t S2) except -1:
"""
Concatenate two bounded integer sequences ``S1`` and ``S2``.
ASSUMPTION:
- The two sequences must have equivalent bounds, i.e., the items on the
sequences must fit into the same number of bits.
OUTPUT:
The result is written into ``R``, which must not be initialised
"""
biseq_init(R, S1.length + S2.length, S1.itembitsize)
sig_on()
bitset_lshift(R.data, S2.data, S1.length * S1.itembitsize)
bitset_or(R.data, R.data, S1.data)
sig_off()
cdef inline bint biseq_startswith(biseq_t S1, biseq_t S2) except -1:
"""
Tests if bounded integer sequence ``S1`` starts with bounded integer
sequence ``S2``.
ASSUMPTION:
- The two sequences must have equivalent bounds, i.e., the items on the
sequences must fit into the same number of bits. This condition is not
tested.
"""
if S2.length > S1.length:
return False
if S2.length == 0:
return True
sig_on()
ret = mpn_equal_bits(S1.data.bits, S2.data.bits, S2.data.size)
sig_off()
return ret
cdef mp_size_t biseq_index(biseq_t S, size_t item, mp_size_t start) except -2:
"""
Returns the position in ``S`` of an item in ``S[start:]``, or -1 if
``S[start:]`` does not contain the item.
"""
cdef mp_size_t index
sig_on()
for index from start <= index < S.length:
if biseq_getitem(S, index) == item:
sig_off()
return index
sig_off()
return -1
cdef inline size_t biseq_getitem(biseq_t S, mp_size_t index):
"""
Get item ``S[index]``, without checking margins.
"""
cdef mp_bitcnt_t limb_index, bit_index
bit_index = (<mp_bitcnt_t>index) * S.itembitsize
limb_index = bit_index // GMP_LIMB_BITS
bit_index %= GMP_LIMB_BITS
cdef mp_limb_t out
out = (S.data.bits[limb_index]) >> bit_index
if bit_index + S.itembitsize > GMP_LIMB_BITS:
# Our item is stored using 2 limbs, add the part from the upper limb
out |= (S.data.bits[limb_index+1]) << (GMP_LIMB_BITS - bit_index)
return out & S.mask_item
cdef biseq_getitem_py(biseq_t S, mp_size_t index):
"""
Get item ``S[index]`` as a Python ``int`` or ``long``, without
checking margins.
"""
cdef size_t out = biseq_getitem(S, index)
return PyInt_FromSize_t(out)
cdef inline void biseq_inititem(biseq_t S, mp_size_t index, size_t item):
"""
Set ``S[index] = item``, without checking margins.
Note that it is assumed that ``S[index] == 0`` before the assignment.
"""
cdef mp_bitcnt_t limb_index, bit_index
bit_index = (<mp_bitcnt_t>index) * S.itembitsize
limb_index = bit_index // GMP_LIMB_BITS
bit_index %= GMP_LIMB_BITS
S.data.bits[limb_index] |= (item << bit_index)
# Have some bits been shifted out of bound?
if bit_index + S.itembitsize > GMP_LIMB_BITS:
# Our item is stored using 2 limbs, add the part from the upper limb
S.data.bits[limb_index+1] |= (item >> (GMP_LIMB_BITS - bit_index))
cdef inline void biseq_clearitem(biseq_t S, mp_size_t index):
"""
Set ``S[index] = 0``, without checking margins.
In contrast to ``biseq_inititem``, the previous content of ``S[index]``
will be erased.
"""
cdef mp_bitcnt_t limb_index, bit_index
bit_index = (<mp_bitcnt_t>index) * S.itembitsize
limb_index = bit_index // GMP_LIMB_BITS
bit_index %= GMP_LIMB_BITS
S.data.bits[limb_index] &= ~(S.mask_item << bit_index)
# Have some bits been shifted out of bound?
if bit_index + S.itembitsize > GMP_LIMB_BITS:
# Our item is stored using 2 limbs, add the part from the upper limb
S.data.bits[limb_index+1] &= ~(S.mask_item >> (GMP_LIMB_BITS - bit_index))
cdef bint biseq_init_slice(biseq_t R, biseq_t S, mp_size_t start, mp_size_t stop, mp_size_t step) except -1:
"""
Create the slice ``S[start:stop:step]`` as bounded integer sequence
and write the result to ``R``, which must not be initialised.
"""
cdef mp_size_t length = 0
if step > 0:
if stop > start:
length = ((stop-start-1)//step)+1
else:
if stop < start:
length = ((stop-start+1)//step)+1
biseq_init(R, length, S.itembitsize)
if not length:
return 0
if step == 1:
# Slicing essentially boils down to a shift operation.
sig_on()
bitset_rshift(R.data, S.data, start*S.itembitsize)
sig_off()
return 0
# In the general case, we move item by item.
cdef mp_size_t src_index = start
cdef mp_size_t tgt_index
sig_on()
for tgt_index in range(length):
biseq_inititem(R, tgt_index, biseq_getitem(S, src_index))
src_index += step
sig_off()
cdef mp_size_t biseq_contains(biseq_t S1, biseq_t S2, mp_size_t start) except -2:
"""
Tests if the bounded integer sequence ``S1[start:]`` contains a
sub-sequence ``S2``.
INPUT:
- ``S1``, ``S2`` -- two bounded integer sequences
- ``start`` -- integer, start index
OUTPUT:
The smallest index ``i >= start`` such that ``S1[i:]`` starts with
``S2``, or ``-1`` if ``S1[start:]`` does not contain ``S2``.
ASSUMPTION:
- The two sequences must have equivalent bounds, i.e., the items on the
sequences must fit into the same number of bits. This condition is not
tested.
"""
if S2.length == 0:
return start
cdef mp_size_t index
sig_on()
for index from start <= index <= S1.length-S2.length:
if mpn_equal_bits_shifted(S2.data.bits, S1.data.bits,
S2.length*S2.itembitsize, index*S2.itembitsize):
sig_off()
return index
sig_off()
return -1
cdef mp_size_t biseq_startswith_tail(biseq_t S1, biseq_t S2, mp_size_t start) except -2:
"""
Return the smallest index ``i`` such that the bounded integer sequence
``S1`` starts with the sequence ``S2[i:]``, where ``start <= i <
S2.length``.
INPUT:
- ``S1``, ``S2`` -- two bounded integer sequences
- ``start`` -- integer, start index
OUTPUT:
The smallest index ``i >= start`` such that ``S1`` starts with ``S2[i:],
or ``-1`` if no such ``i < S2.length`` exists.
ASSUMPTION:
- The two sequences must have equivalent bounds, i.e., the items on the
sequences must fit into the same number of bits. This condition is not
tested.
"""
# Increase start if S1 is too short to contain S2[start:]
if S1.length < S2.length - start:
start = S2.length - S1.length
cdef mp_size_t index
sig_on()
for index from start <= index < S2.length:
if mpn_equal_bits_shifted(S1.data.bits, S2.data.bits,
(S2.length - index)*S2.itembitsize, index*S2.itembitsize):
sig_off()
return index
sig_off()
return -1
###########################################
# A cdef class that wraps the above, and
# behaves like a tuple
from sage.rings.integer cimport smallInteger
cdef class BoundedIntegerSequence:
"""
A sequence of non-negative uniformly bounded integers.
INPUT:
- ``bound`` -- non-negative integer. When zero, a :class:`ValueError`
will be raised. Otherwise, the given bound is replaced by the
power of two that is at least the given bound.
- ``data`` -- a list of integers.
EXAMPLES:
We showcase the similarities and differences between bounded integer
sequences and lists respectively tuples.
To distinguish from tuples or lists, we use pointed brackets for the
string representation of bounded integer sequences::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: S = BoundedIntegerSequence(21, [2, 7, 20]); S
<2, 7, 20>
Each bounded integer sequence has a bound that is a power of two, such
that all its item are less than this bound::
sage: S.bound()
32
sage: BoundedIntegerSequence(16, [2, 7, 20])
Traceback (most recent call last):
...
OverflowError: list item 20 larger than 15
Bounded integer sequences are iterable, and we see that we can recover the
originally given list::
sage: L = [randint(0,31) for i in range(5000)]
sage: S = BoundedIntegerSequence(32, L)
sage: list(L) == L
True
Getting items and slicing works in the same way as for lists::
sage: n = randint(0,4999)
sage: S[n] == L[n]
True
sage: m = randint(0,1000)
sage: n = randint(3000,4500)
sage: s = randint(1, 7)
sage: list(S[m:n:s]) == L[m:n:s]
True
sage: list(S[n:m:-s]) == L[n:m:-s]
True
The :meth:`index` method works different for bounded integer sequences and
tuples or lists. If one asks for the index of an item, the behaviour is
the same. But we can also ask for the index of a sub-sequence::
sage: L.index(L[200]) == S.index(L[200])
True
sage: S.index(S[100:2000]) # random
100
Similarly, containment tests work for both items and sub-sequences::
sage: S[200] in S
True
sage: S[200:400] in S
True
sage: S[200]+S.bound() in S
False
Bounded integer sequences are immutable, and thus copies are
identical. This is the same for tuples, but of course not for lists::
sage: T = tuple(S)
sage: copy(T) is T
True
sage: copy(S) is S
True
sage: copy(L) is L
False
Concatenation works in the same way for lists, tuples and bounded
integer sequences::
sage: M = [randint(0,31) for i in range(5000)]
sage: T = BoundedIntegerSequence(32, M)
sage: list(S+T)==L+M
True
sage: list(T+S)==M+L
True
sage: (T+S == S+T) == (M+L == L+M)
True
However, comparison works different for lists and bounded integer
sequences. Bounded integer sequences are first compared by bound, then by
length, and eventually by *reverse* lexicographical ordering::
sage: S = BoundedIntegerSequence(21, [4,1,6,2,7,20,9])
sage: T = BoundedIntegerSequence(51, [4,1,6,2,7,20])
sage: S < T # compare by bound, not length
True
sage: T < S
False
sage: S.bound() < T.bound()
True
sage: len(S) > len(T)
True
::
sage: T = BoundedIntegerSequence(21, [0,0,0,0,0,0,0,0])
sage: S < T # compare by length, not lexicographically
True
sage: T < S
False
sage: list(T) < list(S)
True
sage: len(T) > len(S)
True
::
sage: T = BoundedIntegerSequence(21, [4,1,5,2,8,20,9])
sage: T > S # compare by reverse lexicographic ordering...
True
sage: S > T
False
sage: len(S) == len(T)
True
sage: list(S) > list(T) # direct lexicographic ordering is different
True
TESTS:
We test against various corner cases::
sage: BoundedIntegerSequence(16, [2, 7, -20])
Traceback (most recent call last):
...
OverflowError: can't convert negative value to size_t
sage: BoundedIntegerSequence(1, [0, 0, 0])
<0, 0, 0>
sage: BoundedIntegerSequence(1, [0, 1, 0])
Traceback (most recent call last):
...
OverflowError: list item 1 larger than 0
sage: BoundedIntegerSequence(0, [0, 1, 0])
Traceback (most recent call last):
...
ValueError: positive bound expected
sage: BoundedIntegerSequence(2, [])
<>
sage: BoundedIntegerSequence(2, []) == BoundedIntegerSequence(4, []) # The bounds differ
False
sage: BoundedIntegerSequence(16, [2, 7, 4])[1:1]
<>
"""
def __cinit__(self, *args, **kwds):
"""
Allocate memory for underlying data
INPUT:
- ``bound``, non-negative integer
- ``data``, ignored
.. WARNING::
If ``bound=0`` then no allocation is done. Hence, this should
only be done internally, when calling :meth:`__new__` without :meth:`__init__`.
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: BoundedIntegerSequence(21, [4,1,6,2,7,20,9]) # indirect doctest
<4, 1, 6, 2, 7, 20, 9>
"""
# In __init__, we'll raise an error if the bound is 0.
self.data.data.bits = NULL
def __dealloc__(self):
"""
Free the memory from underlying data
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: S = BoundedIntegerSequence(21, [4,1,6,2,7,20,9])
sage: del S # indirect doctest
"""
biseq_dealloc(self.data)
def __init__(self, bound, data):
"""
INPUT:
- ``bound`` -- positive integer. The given bound is replaced by
the next power of two that is greater than the given bound.
- ``data`` -- a list of non-negative integers, all less than
``bound``.
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: L = [randint(0,26) for i in range(5000)]
sage: S = BoundedIntegerSequence(57, L) # indirect doctest
sage: list(S) == L
True
sage: S = BoundedIntegerSequence(11, [4,1,6,2,7,4,9]); S
<4, 1, 6, 2, 7, 4, 9>
sage: S.bound()
16
Non-positive bounds or bounds which are too large result in errors::
sage: BoundedIntegerSequence(-1, L)
Traceback (most recent call last):
...
ValueError: positive bound expected
sage: BoundedIntegerSequence(0, L)
Traceback (most recent call last):
...
ValueError: positive bound expected
sage: BoundedIntegerSequence(2^64+1, L)
Traceback (most recent call last):
...
OverflowError: ... int too large to convert...
We are testing the corner case of the maximal possible bound::
sage: S = BoundedIntegerSequence(2*(sys.maxsize+1), [8, 8, 26, 18, 18, 8, 22, 4, 17, 22, 22, 7, 12, 4, 1, 7, 21, 7, 10, 10])
sage: S
<8, 8, 26, 18, 18, 8, 22, 4, 17, 22, 22, 7, 12, 4, 1, 7, 21, 7, 10, 10>
Items that are too large::
sage: BoundedIntegerSequence(100, [2^256])
Traceback (most recent call last):
...
OverflowError: ... int too large to convert...
sage: BoundedIntegerSequence(100, [100])
Traceback (most recent call last):
...
OverflowError: list item 100 larger than 99
Bounds that are too large::
sage: BoundedIntegerSequence(2^256, [200])
Traceback (most recent call last):
...
OverflowError: ... int too large to convert...
"""
if bound <= 0:
raise ValueError("positive bound expected")
biseq_init_list(self.data, data, bound-1)
def __copy__(self):
"""
:class:`BoundedIntegerSequence` is immutable, copying returns ``self``.
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: S = BoundedIntegerSequence(21, [4,1,6,2,7,20,9])
sage: copy(S) is S
True
"""
return self
def __reduce__(self):
"""
Pickling of :class:`BoundedIntegerSequence`
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: L = [randint(0,26) for i in range(5000)]
sage: S = BoundedIntegerSequence(32, L)
sage: loads(dumps(S)) == S # indirect doctest
True
TESTS:
The discussion at :trac:`15820` explains why the following is a good test::
sage: X = BoundedIntegerSequence(21, [4,1,6,2,7,2,3])
sage: S = BoundedIntegerSequence(21, [0,0,0,0,0,0,0])
sage: loads(dumps(X+S))
<4, 1, 6, 2, 7, 2, 3, 0, 0, 0, 0, 0, 0, 0>
sage: loads(dumps(X+S)) == X+S
True
sage: T = BoundedIntegerSequence(21, [0,4,0,1,0,6,0,2,0,7,0,2,0,3,0,0,0,0,0,0,0,0,0,0,0,0,0,0])
sage: T[1::2]
<4, 1, 6, 2, 7, 2, 3, 0, 0, 0, 0, 0, 0, 0>
sage: T[1::2] == X+S
True
sage: loads(dumps(X[1::2])) == X[1::2]
True
"""
return NewBISEQ, biseq_pickle(self.data)
def __len__(self):
"""
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: L = [randint(0,26) for i in range(5000)]
sage: S = BoundedIntegerSequence(57, L) # indirect doctest
sage: len(S) == len(L)
True
"""
return self.data.length
def __nonzero__(self):
"""
A bounded integer sequence is nonzero if and only if its length is nonzero.
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: S = BoundedIntegerSequence(13, [0,0,0])
sage: bool(S)
True
sage: bool(S[1:1])
False
"""
return self.data.length!=0
def __repr__(self):
"""
String representation.
To distinguish it from Python tuples or lists, we use pointed brackets
as delimiters.
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: BoundedIntegerSequence(21, [4,1,6,2,7,20,9]) # indirect doctest
<4, 1, 6, 2, 7, 20, 9>
sage: BoundedIntegerSequence(21, [0,0]) + BoundedIntegerSequence(21, [0,0])
<0, 0, 0, 0>
"""
return "<" + ", ".join(str(x) for x in self) + ">"
def bound(self):
"""
Return the bound of this bounded integer sequence.
All items of this sequence are non-negative integers less than the
returned bound. The bound is a power of two.
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: S = BoundedIntegerSequence(21, [4,1,6,2,7,20,9])
sage: T = BoundedIntegerSequence(51, [4,1,6,2,7,20,9])
sage: S.bound()
32
sage: T.bound()
64
"""
return smallInteger(1) << self.data.itembitsize
def __iter__(self):
"""
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: L = [randint(0,26) for i in range(5000)]
sage: S = BoundedIntegerSequence(27, L)
sage: list(S) == L # indirect doctest
True
TESTS::
sage: list(BoundedIntegerSequence(1, []))
[]
The discussion at :trac:`15820` explains why this is a good test::
sage: S = BoundedIntegerSequence(21, [0,0,0,0,0,0,0])
sage: X = BoundedIntegerSequence(21, [4,1,6,2,7,2,3])
sage: list(X)
[4, 1, 6, 2, 7, 2, 3]
sage: list(X+S)
[4, 1, 6, 2, 7, 2, 3, 0, 0, 0, 0, 0, 0, 0]
sage: list(BoundedIntegerSequence(21, [0,0]) + BoundedIntegerSequence(21, [0,0]))
[0, 0, 0, 0]
"""
cdef mp_size_t index
for index in range(self.data.length):
yield biseq_getitem_py(self.data, index)
def __getitem__(self, index):
"""
Get single items or slices.
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: S = BoundedIntegerSequence(21, [4,1,6,2,7,20,9])
sage: S[2]
6
sage: S[1::2]
<1, 2, 20>
sage: S[-1::-2]
<9, 7, 6, 4>
TESTS::
sage: S = BoundedIntegerSequence(10^8, list(range(9)))
sage: S[-1]
8
sage: S[8]
8
sage: S[9]
Traceback (most recent call last):
...
IndexError: index out of range
sage: S[-10]
Traceback (most recent call last):
...
IndexError: index out of range
sage: S[2^63]
Traceback (most recent call last):
...
OverflowError: ... int too large to convert to ...
::
sage: S[-1::-2]
<8, 6, 4, 2, 0>
sage: S[1::2]
<1, 3, 5, 7>
::
sage: L = [randint(0,26) for i in range(5000)]
sage: S = BoundedIntegerSequence(27, L)
sage: S[1234] == L[1234]
True
sage: list(S[100:2000:3]) == L[100:2000:3]
True
sage: list(S[3000:10:-7]) == L[3000:10:-7]
True
sage: S[:] == S
True
sage: S[:] is S
True
::
sage: S = BoundedIntegerSequence(21, [0,0,0,0,0,0,0])
sage: X = BoundedIntegerSequence(21, [4,1,6,2,7,2,3])
sage: (X+S)[6]
3
sage: (X+S)[10]
0
sage: (X+S)[12:]
<0, 0>
::
sage: S[2:2] == X[4:2]
True
::
sage: S = BoundedIntegerSequence(6, [3, 5, 3, 1, 5, 2, 2, 5, 3, 3, 4])
sage: S[10]
4
::
sage: B = BoundedIntegerSequence(27, [8, 8, 26, 18, 18, 8, 22, 4, 17, 22, 22, 7, 12, 4, 1, 7, 21, 7, 10, 10])
sage: B[8:]
<17, 22, 22, 7, 12, 4, 1, 7, 21, 7, 10, 10>
::
sage: B1 = BoundedIntegerSequence(8, [0,7])
sage: B2 = BoundedIntegerSequence(8, [2,1,4])
sage: B1[0:1]+B2
<0, 2, 1, 4>
"""
cdef BoundedIntegerSequence out
cdef Py_ssize_t start, stop, step, slicelength
if isinstance(index, slice):
PySlice_GetIndicesEx(index, self.data.length, &start, &stop, &step, &slicelength)
if start==0 and stop==self.data.length and step==1:
return self
out = BoundedIntegerSequence.__new__(BoundedIntegerSequence, 0, None)
biseq_init_slice(out.data, self.data, start, stop, step)
return out
cdef Py_ssize_t ind
try:
ind = index
except TypeError:
raise TypeError("Sequence index must be integer or slice")
if ind < 0:
ind += self.data.length
if ind < 0 or ind >= self.data.length:
raise IndexError("index out of range")
return biseq_getitem_py(self.data, ind)
def __contains__(self, other):
"""
Tells whether this bounded integer sequence contains an item or a sub-sequence
EXAMPLES::
sage: from sage.data_structures.bounded_integer_sequences import BoundedIntegerSequence
sage: S = BoundedIntegerSequence(21, [4,1,6,2,7,20,9])
sage: 6 in S
True
sage: BoundedIntegerSequence(21, [2, 7, 20]) in S
True
The bound of the sequences matters::
sage: BoundedIntegerSequence(51, [2, 7, 20]) in S
False
::
sage: 6+S.bound() in S
False
sage: S.index(6+S.bound())
Traceback (most recent call last):
...
ValueError: 38 is not in sequence
TESTS:
The discussion at :trac:`15820` explains why the following are good tests::