vsss-rs 5.4.0

Verifiable Secret Sharing Schemes for splitting, combining and verifying secret shares
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
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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
//! Represents Galois Field of 2^4 elements. This uses constant time operations
//! for all operations as related to shamir secret sharing. Too many implementations
//! use lookup tables which help for speed but leak secret information.
//! No lookup tables are used in this implementation because Cryptographic operations should
//!
//! 1. Ensure runtime is independent of secret data
//! 2. Ensure code access patterns are independent of secret data
//! 3. Ensure data access patterns are independent of secret data

use crate::util::{CtIsNotZero, field_bounded_add, uniform_nonzero_u8};
use crate::*;
use core::borrow::Borrow;
use core::{
    fmt::{self, Binary, Display, Formatter, LowerHex, UpperHex},
    iter::{Product, Sum},
    ops::{
        Add, AddAssign, BitAnd, BitAndAssign, BitOr, BitOrAssign, BitXor, BitXorAssign, Deref,
        DerefMut, Div, DivAssign, Mul, MulAssign, Neg, Sub, SubAssign,
    },
};
use elliptic_curve::ff::{Field, PrimeField};
use rand_core::RngCore;
use subtle::{Choice, ConditionallySelectable, ConstantTimeEq, CtOption};

#[cfg(any(feature = "alloc", feature = "std"))]
use crate::ParticipantIdGeneratorType;
use rand_core::CryptoRng;
#[cfg(feature = "zeroize")]
use zeroize::DefaultIsZeroes;

#[cfg(any(feature = "alloc", feature = "std"))]
type GfShare = DefaultShare<IdentifierGf16, IdentifierGf16>;

/// Represents the finite field GF(2^4) with 16 elements.
/// Elements are stored in the lower nibble of a u8 (values 0x00..=0x0F).
/// Uses the irreducible polynomial x^4 + x + 1 for multiplication.
#[derive(Debug, Default, Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash)]
#[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
#[repr(transparent)]
pub struct Gf16(pub u8);

#[cfg(feature = "zeroize")]
impl DefaultIsZeroes for Gf16 {}

impl Display for Gf16 {
    fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
        write!(f, "{}", self.0)
    }
}

impl LowerHex for Gf16 {
    fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
        write!(f, "{:01x}", self.0)
    }
}

impl UpperHex for Gf16 {
    fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
        write!(f, "{:01X}", self.0)
    }
}

impl Binary for Gf16 {
    fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
        write!(f, "{:04b}", self.0)
    }
}

impl ConditionallySelectable for Gf16 {
    fn conditional_select(a: &Self, b: &Self, choice: Choice) -> Self {
        Gf16(u8::conditional_select(&a.0, &b.0, choice))
    }
}

impl ConstantTimeEq for Gf16 {
    fn ct_eq(&self, other: &Self) -> Choice {
        self.0.ct_eq(&other.0)
    }
}

impl Add for Gf16 {
    type Output = Self;

    fn add(self, rhs: Self) -> Self {
        Gf16(self.0 ^ rhs.0)
    }
}

impl Add<&Gf16> for Gf16 {
    type Output = Gf16;

    fn add(self, rhs: &Gf16) -> Gf16 {
        self + *rhs
    }
}

impl Add<Gf16> for &Gf16 {
    type Output = Gf16;

    fn add(self, rhs: Gf16) -> Gf16 {
        *self + rhs
    }
}

impl Add<&Gf16> for &Gf16 {
    type Output = Gf16;

    fn add(self, rhs: &Gf16) -> Gf16 {
        *self + *rhs
    }
}

impl AddAssign for Gf16 {
    fn add_assign(&mut self, rhs: Self) {
        *self = *self + rhs;
    }
}

impl AddAssign<&Gf16> for Gf16 {
    fn add_assign(&mut self, rhs: &Gf16) {
        *self = *self + *rhs;
    }
}

impl Sub for Gf16 {
    type Output = Self;

    fn sub(self, rhs: Self) -> Self {
        Gf16(self.0 ^ rhs.0)
    }
}

impl Sub<&Gf16> for Gf16 {
    type Output = Gf16;

    fn sub(self, rhs: &Gf16) -> Gf16 {
        Gf16(self.0 ^ rhs.0)
    }
}

impl Sub<Gf16> for &Gf16 {
    type Output = Gf16;

    fn sub(self, rhs: Gf16) -> Gf16 {
        Gf16(self.0 ^ rhs.0)
    }
}

impl Sub<&Gf16> for &Gf16 {
    type Output = Gf16;

    fn sub(self, rhs: &Gf16) -> Gf16 {
        Gf16(self.0 ^ rhs.0)
    }
}

impl SubAssign for Gf16 {
    fn sub_assign(&mut self, rhs: Self) {
        self.0 ^= rhs.0;
    }
}

impl SubAssign<&Gf16> for Gf16 {
    fn sub_assign(&mut self, rhs: &Gf16) {
        self.0 ^= rhs.0;
    }
}

impl Mul for Gf16 {
    type Output = Self;

    fn mul(self, rhs: Self) -> Self {
        Self(gf16_mul(self.0, rhs.0))
    }
}

impl Mul<&Gf16> for Gf16 {
    type Output = Gf16;

    fn mul(self, rhs: &Gf16) -> Gf16 {
        self * *rhs
    }
}

impl Mul<Gf16> for &Gf16 {
    type Output = Gf16;

    fn mul(self, rhs: Gf16) -> Gf16 {
        *self * rhs
    }
}

impl Mul<&Gf16> for &Gf16 {
    type Output = Gf16;

    fn mul(self, rhs: &Gf16) -> Gf16 {
        *self * *rhs
    }
}

impl MulAssign for Gf16 {
    fn mul_assign(&mut self, rhs: Self) {
        *self = *self * rhs;
    }
}

impl MulAssign<&Gf16> for Gf16 {
    fn mul_assign(&mut self, rhs: &Gf16) {
        *self = *self * *rhs;
    }
}

impl Div for Gf16 {
    type Output = Self;

    fn div(self, rhs: Self) -> Self::Output {
        self * rhs.invert().expect("no division by zero")
    }
}

impl Div<&Gf16> for Gf16 {
    type Output = Gf16;

    fn div(self, rhs: &Gf16) -> Gf16 {
        self / *rhs
    }
}

impl Div<Gf16> for &Gf16 {
    type Output = Gf16;

    fn div(self, rhs: Gf16) -> Gf16 {
        *self / rhs
    }
}

impl Div<&Gf16> for &Gf16 {
    type Output = Gf16;

    fn div(self, rhs: &Gf16) -> Gf16 {
        *self / *rhs
    }
}

impl DivAssign for Gf16 {
    fn div_assign(&mut self, rhs: Self) {
        *self *= rhs.invert().expect("no division by zero");
    }
}

impl DivAssign<&Gf16> for Gf16 {
    fn div_assign(&mut self, rhs: &Gf16) {
        *self *= rhs.invert().expect("no division by zero");
    }
}

impl Neg for Gf16 {
    type Output = Self;

    fn neg(self) -> Self {
        self
    }
}

impl BitAnd for Gf16 {
    type Output = Self;

    fn bitand(self, rhs: Self) -> Self {
        Self(self.0 & rhs.0)
    }
}

impl BitAnd<&Gf16> for Gf16 {
    type Output = Gf16;

    fn bitand(self, rhs: &Gf16) -> Gf16 {
        self & *rhs
    }
}

impl BitAnd<Gf16> for &Gf16 {
    type Output = Gf16;

    fn bitand(self, rhs: Gf16) -> Gf16 {
        *self & rhs
    }
}

impl BitAnd<&Gf16> for &Gf16 {
    type Output = Gf16;

    fn bitand(self, rhs: &Gf16) -> Gf16 {
        *self & *rhs
    }
}

impl BitAndAssign for Gf16 {
    fn bitand_assign(&mut self, rhs: Self) {
        self.0 &= rhs.0;
    }
}

impl BitAndAssign<&Gf16> for Gf16 {
    fn bitand_assign(&mut self, rhs: &Gf16) {
        self.0 &= rhs.0;
    }
}

impl BitOr for Gf16 {
    type Output = Self;

    fn bitor(self, rhs: Self) -> Self {
        Self(self.0 | rhs.0)
    }
}

impl BitOr<&Gf16> for Gf16 {
    type Output = Gf16;

    fn bitor(self, rhs: &Gf16) -> Gf16 {
        self | *rhs
    }
}

impl BitOr<Gf16> for &Gf16 {
    type Output = Gf16;

    fn bitor(self, rhs: Gf16) -> Gf16 {
        *self | rhs
    }
}

impl BitOr<&Gf16> for &Gf16 {
    type Output = Gf16;

    fn bitor(self, rhs: &Gf16) -> Gf16 {
        *self | *rhs
    }
}

impl BitOrAssign for Gf16 {
    fn bitor_assign(&mut self, rhs: Self) {
        self.0 |= rhs.0;
    }
}

impl BitOrAssign<&Gf16> for Gf16 {
    fn bitor_assign(&mut self, rhs: &Gf16) {
        self.0 |= rhs.0;
    }
}

impl BitXor for Gf16 {
    type Output = Self;

    fn bitxor(self, rhs: Self) -> Self {
        Self(self.0 ^ rhs.0)
    }
}

impl BitXor<&Gf16> for Gf16 {
    type Output = Gf16;

    fn bitxor(self, rhs: &Gf16) -> Gf16 {
        self ^ *rhs
    }
}

impl BitXor<Gf16> for &Gf16 {
    type Output = Gf16;

    fn bitxor(self, rhs: Gf16) -> Gf16 {
        *self ^ rhs
    }
}

impl BitXor<&Gf16> for &Gf16 {
    type Output = Gf16;

    fn bitxor(self, rhs: &Gf16) -> Gf16 {
        *self ^ *rhs
    }
}

impl BitXorAssign for Gf16 {
    fn bitxor_assign(&mut self, rhs: Self) {
        self.0 ^= rhs.0;
    }
}

impl BitXorAssign<&Gf16> for Gf16 {
    fn bitxor_assign(&mut self, rhs: &Gf16) {
        self.0 ^= rhs.0;
    }
}

impl<T: Borrow<Gf16>> Sum<T> for Gf16 {
    fn sum<I: Iterator<Item = T>>(iter: I) -> Self {
        iter.fold(Self(0), |acc, x| acc + x.borrow())
    }
}

impl<T: Borrow<Gf16>> Product<T> for Gf16 {
    fn product<I: Iterator<Item = T>>(iter: I) -> Self {
        iter.fold(Self(1), |acc, x| acc * x.borrow())
    }
}

impl Field for Gf16 {
    const ZERO: Self = Self(0);
    const ONE: Self = Self(1);

    fn random(mut rng: impl RngCore) -> Self {
        // Uniform over the full field {0, 1, ..., 15}. The prior
        // `(b & 0x0E) + 1` forced the low bit and yielded only odd
        // nibbles {1,3,...,15}; that bias leaks into polynomial
        // coefficients used by Shamir secret sharing (audit finding #1).
        Self(rng.next_u32() as u8 & 0x0F)
    }

    fn square(&self) -> Self {
        self * self
    }

    fn double(&self) -> Self {
        self + self
    }

    fn invert(&self) -> CtOption<Self> {
        // Compute a^(2^4 - 2) = a^14 = a^(-1) in GF(2^4).
        // Loop pattern: after k iterations, z = a^(2^(k+1) - 1).
        // After 2 iterations: z = a^7. Final square: a^14.
        let mut z = self.0;
        for _ in 0..2 {
            z = gf16_mul(z, z);
            z = gf16_mul(z, self.0);
        }
        CtOption::new(Self(gf16_mul(z, z)), self.0.ct_is_not_zero())
    }

    fn sqrt_ratio(num: &Self, div: &Self) -> (Choice, Self) {
        let p = 0xfu8; // |GF(16)*| = 15
        let pm1d2 = (p - 1) >> 1; // 7
        let pp2d4 = (p + 2) >> 2; // 4

        // z = 2 (x) is a primitive root: gf16_pow(2, 7) = 11 != 1
        let z = (2..=p).find(|z| gf16_pow(*z, pm1d2) != 1).unwrap();

        let a = gf16_mul(num.0, div.0);
        let mut c = gf16_pow(a, pp2d4);
        let mut t = gf16_pow(a, pm1d2);
        let mut r = gf16_pow(z, pm1d2);

        let mut m = t;
        let mut i = 1usize;
        while m != 1 && m != 0 {
            let mut temp = m;
            for _ in 1..i {
                temp = gf16_mul(temp, temp);
            }
            let mut j = 0usize;
            while temp != 1 && temp != 0 && j < 4 {
                temp = gf16_mul(temp, temp);
                j += 1;
            }
            if i <= j {
                break;
            }
            let b = gf16_pow(r, 1u8 << (i - j - 1));
            c = gf16_mul(c, b);
            r = gf16_mul(b, b);
            t = gf16_mul(t, r);
            m = t;
            i = j;
        }
        let is_square = gf16_pow(c, 2).ct_eq(&c);
        (is_square, Self(c))
    }
}

impl From<u8> for Gf16 {
    fn from(val: u8) -> Self {
        Gf16(val)
    }
}

impl From<Gf16> for u8 {
    fn from(val: Gf16) -> u8 {
        val.0
    }
}

impl From<u16> for Gf16 {
    fn from(val: u16) -> Self {
        Gf16(val as u8)
    }
}

impl From<Gf16> for u16 {
    fn from(val: Gf16) -> u16 {
        val.0 as u16
    }
}

impl From<u32> for Gf16 {
    fn from(val: u32) -> Self {
        Gf16(val as u8)
    }
}

impl From<Gf16> for u32 {
    fn from(val: Gf16) -> u32 {
        val.0 as u32
    }
}

impl From<u64> for Gf16 {
    fn from(val: u64) -> Self {
        Gf16(val as u8)
    }
}

impl From<Gf16> for u64 {
    fn from(val: Gf16) -> u64 {
        val.0 as u64
    }
}

impl From<u128> for Gf16 {
    fn from(val: u128) -> Self {
        Gf16(val as u8)
    }
}

impl From<Gf16> for u128 {
    fn from(val: Gf16) -> u128 {
        val.0 as u128
    }
}

impl PrimeField for Gf16 {
    type Repr = [u8; 1];

    fn from_repr(repr: Self::Repr) -> CtOption<Self> {
        CtOption::new(Self(repr[0]), Choice::from(1u8))
    }

    fn to_repr(&self) -> Self::Repr {
        [self.0]
    }

    fn is_odd(&self) -> Choice {
        (self.0 & 1).ct_eq(&1)
    }

    const MODULUS: &'static str = "";
    const NUM_BITS: u32 = 4;
    const CAPACITY: u32 = 3;
    /// TWO_INV = 2^(-1) in GF(2^4): 2 * 9 = 1, so TWO_INV = 9.
    const TWO_INV: Self = Self(9);
    /// x = 2 is a primitive root of GF(2^4): powers of 2 generate all 15 non-zero elements.
    const MULTIPLICATIVE_GENERATOR: Self = Self(2);
    /// S = 0 since |GF(16)*| = 15 is odd (2-adic valuation of 15 is 0).
    const S: u32 = 0;
    /// The 2^S = 1-st root of unity is 1 (trivially).
    const ROOT_OF_UNITY: Self = Self(1);
    const ROOT_OF_UNITY_INV: Self = Self(1);
    /// DELTA = MULTIPLICATIVE_GENERATOR^(2^S) = 2^1 = 2.
    const DELTA: Self = Self(2);
}

impl Gf16 {
    /// Raise the element to the power of `exp` in GF(2^4).
    pub fn pow(&self, exp: u8) -> Self {
        Self(gf16_pow(self.0, exp))
    }

    #[cfg(any(feature = "alloc", feature = "std"))]
    /// Split a byte array into shares using GF(2^4) arithmetic.
    ///
    /// Each input byte is treated as two 4-bit nibbles (low = bits 0-3, high = bits 4-7).
    /// Both nibbles are independently shared as GF(16) elements and packed back into
    /// a single share byte, preserving the 1:1 byte ratio between secret and share data.
    /// Maximum of 15 shares (the field has 15 non-zero elements).
    pub fn split_array<B: AsRef<[u8]>>(
        threshold: usize,
        limit: usize,
        secret: B,
        rng: impl RngCore + CryptoRng,
    ) -> VsssResult<Vec<Vec<u8>>> {
        Self::split_array_with_participant_generators(
            threshold,
            limit,
            secret,
            rng,
            &[ParticipantIdGeneratorType::default()],
        )
    }

    #[cfg(any(feature = "alloc", feature = "std"))]
    /// Split a byte array into shares using the participant number generator.
    pub fn split_array_with_participant_generators<B: AsRef<[u8]>>(
        threshold: usize,
        limit: usize,
        secret: B,
        mut rng: impl RngCore + CryptoRng,
        participant_generators: &[ParticipantIdGeneratorType<IdentifierGf16>],
    ) -> VsssResult<Vec<Vec<u8>>> {
        if limit > 15 {
            return Err(Error::InvalidSizeRequest);
        }
        let secret = secret.as_ref();
        if secret.is_empty() {
            return Err(Error::InvalidSecret);
        }
        let mut shares = Vec::with_capacity(limit);

        let collection = ParticipantIdGeneratorCollection::from(participant_generators);
        let mut participant_id_iter = collection.iter();

        for _ in 0..limit {
            let id = participant_id_iter
                .next()
                .ok_or(Error::NotEnoughShareIdentifiers)?;
            let mut inner = Vec::with_capacity(limit + 1);
            inner.push(id.0.0);
            shares.push(inner);
        }
        for b in secret {
            // Each byte is split into two nibbles and shared independently.
            // The low nibble and high nibble are each a GF(16) element (0..=15).
            let lo = IdentifierGf16(Gf16(*b & 0x0f));
            let hi = IdentifierGf16(Gf16((*b >> 4) & 0x0f));

            let lo_shares = shamir::split_secret_with_participant_generator::<GfShare>(
                threshold,
                limit,
                &lo,
                &mut rng,
                participant_generators,
            )?;
            let hi_shares = shamir::split_secret_with_participant_generator::<GfShare>(
                threshold,
                limit,
                &hi,
                &mut rng,
                participant_generators,
            )?;
            // Pack both nibble-shares into a single byte per participant.
            for (share, (lo_s, hi_s)) in shares
                .iter_mut()
                .zip(lo_shares.iter().zip(hi_shares.iter()))
            {
                share.push((hi_s.value.0.0 << 4) | lo_s.value.0.0);
            }
        }
        Ok(shares)
    }

    #[cfg(any(feature = "alloc", feature = "std"))]
    /// Combine shares into a byte array.
    pub fn combine_array<B: AsRef<[Vec<u8>]>>(shares: B) -> VsssResult<Vec<u8>> {
        let shares = shares.as_ref();

        Self::are_shares_valid(shares)?;

        let mut secret = Vec::with_capacity(shares[0].len() - 1);
        let mut lo_inner = Vec::<GfShare>::with_capacity(shares.len());
        let mut hi_inner = Vec::<GfShare>::with_capacity(shares.len());

        for share in shares {
            lo_inner.push(DefaultShare {
                identifier: IdentifierGf16(Gf16(share[0])),
                value: IdentifierGf16(Gf16(0u8)),
            });
            hi_inner.push(DefaultShare {
                identifier: IdentifierGf16(Gf16(share[0])),
                value: IdentifierGf16(Gf16(0u8)),
            });
        }
        for i in 1..shares[0].len() {
            for ((lo_s, hi_s), share) in lo_inner
                .iter_mut()
                .zip(hi_inner.iter_mut())
                .zip(shares.iter())
            {
                lo_s.value = IdentifierGf16(Gf16(share[i] & 0x0f));
                hi_s.value = IdentifierGf16(Gf16((share[i] >> 4) & 0x0f));
            }
            let lo = lo_inner.combine()?.0.0 & 0x0f;
            let hi = hi_inner.combine()?.0.0 & 0x0f;
            secret.push((hi << 4) | lo);
        }
        Ok(secret)
    }

    #[cfg(any(feature = "alloc", feature = "std"))]
    fn are_shares_valid(shares: &[Vec<u8>]) -> VsssResult<()> {
        if shares.len() < 2 {
            return Err(Error::SharingMinThreshold);
        }
        if shares[0].len() < 2 {
            return Err(Error::InvalidShare);
        }
        if shares[1..].iter().any(|s| s.len() != shares[0].len()) {
            return Err(Error::InvalidShare);
        }
        Ok(())
    }
}

/// Constant-time multiplication in GF(2^4) modulo x^4 + x + 1.
///
/// Uses the i8 arithmetic trick for constant-time operation:
/// `-(b & 1)` in signed arithmetic gives 0 (b even) or -1/0xFF (b odd),
/// serving as a branchless conditional mask.
/// Reduction: when bit 3 overflows on left-shift, XOR with 0x03 (= x + 1,
/// the non-leading terms of the irreducible polynomial x^4 + x + 1).
fn gf16_mul(a: u8, b: u8) -> u8 {
    let mut a = (a & 0x0f) as i8;
    let mut b = (b & 0x0f) as i8;
    let mut r = 0i8;
    for _ in 0..4 {
        // If the LSB of b is 1, XOR a into the accumulator (constant-time via -(b&1)).
        r ^= a & -(b & 1);
        b >>= 1;
        // Extract bit 3 of a (the overflow bit after a left-shift).
        let hi = (a as u8 >> 3) as i8 & 1;
        // Create an all-ones mask if bit 3 was set, zero otherwise.
        let t = -hi;
        // Shift a left by 1, masking to 4 bits.
        a = (a << 1) & 0x0f;
        // If bit 3 was set, reduce by XOR with x+1 = 0x03 (the non-leading
        // terms of the irreducible polynomial x^4 + x + 1).
        a ^= 0x03i8 & t;
    }
    (r & 0x0f) as u8
}

/// Constant-time exponentiation in GF(2^4).
///
/// Computes `base^exp` using the left-to-right binary (MSB-first) square-and-multiply
/// method, processing the 4 significant bits of the exponent.
fn gf16_pow(base: u8, exp: u8) -> u8 {
    let mut result = 1u8;
    let base = base & 0x0f;
    // Process bits from MSB (bit 3) down to LSB (bit 0).
    for i in (0..4).rev() {
        result = gf16_mul(result, result);
        let tmp = gf16_mul(result, base);
        let allow = ((exp >> i) & 1).ct_eq(&1);
        result.conditional_assign(&tmp, allow);
    }
    result.conditional_assign(&1u8, exp.ct_eq(&0));
    result
}

#[derive(Debug, Copy, Clone, Default, Eq, PartialEq, Ord, PartialOrd, Hash)]
#[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
#[repr(transparent)]
/// Represents an identifier in the Galois Field GF(2^4).
///
/// Used solely for Sequential Participant ID generation,
/// since GF16 addition = xor i.e. identifiers just oscillate between
/// the start number and the incremented number instead of adding.
pub struct IdentifierGf16(pub Gf16);

impl Display for IdentifierGf16 {
    fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
        write!(f, "{}", self.0)
    }
}

#[cfg(feature = "zeroize")]
impl DefaultIsZeroes for IdentifierGf16 {}

impl Deref for IdentifierGf16 {
    type Target = Gf16;

    fn deref(&self) -> &Self::Target {
        &self.0
    }
}

impl DerefMut for IdentifierGf16 {
    fn deref_mut(&mut self) -> &mut Self::Target {
        &mut self.0
    }
}

impl AsRef<Gf16> for IdentifierGf16 {
    fn as_ref(&self) -> &Gf16 {
        &self.0
    }
}

impl AsMut<Gf16> for IdentifierGf16 {
    fn as_mut(&mut self) -> &mut Gf16 {
        &mut self.0
    }
}

impl From<Gf16> for IdentifierGf16 {
    fn from(val: Gf16) -> Self {
        IdentifierGf16(val)
    }
}

impl From<&IdentifierGf16> for IdentifierGf16 {
    fn from(val: &IdentifierGf16) -> Self {
        IdentifierGf16(val.0)
    }
}

impl Mul<&IdentifierGf16> for IdentifierGf16 {
    type Output = IdentifierGf16;

    fn mul(self, rhs: &IdentifierGf16) -> IdentifierGf16 {
        IdentifierGf16(self.0 * rhs.0)
    }
}

impl ShareElement for IdentifierGf16 {
    type Serialization = [u8; 1];

    type Inner = Gf16;

    fn random(mut rng: impl RngCore + CryptoRng) -> Self {
        // x-coordinate of a Shamir evaluation point; zero is reserved
        // for the secret (f(0)), so MUST be non-zero. Uniform over 1..=15.
        Self(Gf16(uniform_nonzero_u8(rng.next_u32(), 15)))
    }

    fn zero() -> Self {
        Self(Gf16::ZERO)
    }

    fn one() -> Self {
        Self(Gf16::ONE)
    }

    fn is_zero(&self) -> Choice {
        self.0.is_zero()
    }

    fn serialize(&self) -> Self::Serialization {
        [self.0.0]
    }

    fn deserialize(serialized: &Self::Serialization) -> VsssResult<Self> {
        Ok(Self(Gf16(serialized[0])))
    }

    fn from_slice(slice: &[u8]) -> VsssResult<Self> {
        if slice.len() != 1 {
            return Err(Error::InvalidShareElement);
        }
        Ok(Self(Gf16(slice[0])))
    }

    #[cfg(any(feature = "alloc", feature = "std"))]
    fn to_vec(&self) -> Vec<u8> {
        vec![self.0.0]
    }
}

impl ShareIdentifier for IdentifierGf16 {
    fn inc(&mut self, increment: &Self) {
        // Zero-on-overflow keyed to the GF(16) range (audit finding #3).
        // Prior `saturating_add` pinned at 15 (duplicates) and on larger
        // increments saturated to 255 — outside GF(16). See
        // `field_bounded_add` for the branch-free mask + halt-marker
        // contract shared with `IdentifierGf256::inc`.
        self.0.0 = field_bounded_add(self.0.0, increment.0.0, 16);
    }

    fn invert(&self) -> VsssResult<Self> {
        Option::from(self.0.invert())
            .map(Self)
            .ok_or(Error::InvalidShareElement)
    }

    fn random_coefficient(rng: impl RngCore + CryptoRng) -> Self {
        // Bypass the "+1" non-zero x-sampler in `ShareElement::random`
        // and draw directly from `Gf16` — uniform over 0..=15.
        Self(Gf16::random(rng))
    }
}

impl IdentifierGf16 {
    /// Returns additive identity.
    pub const ZERO: Self = Self(Gf16(0));
    /// Returns multiplicative identity.
    pub const ONE: Self = Self(Gf16(1));
}

#[cfg(test)]
#[cfg(any(feature = "alloc", feature = "std"))]
mod tests {
    use super::gf16_cmp;
    use super::*;
    use crate::shamir;
    use crate::{ParticipantIdGeneratorCollection, ParticipantIdGeneratorType};
    use rand::{Rng, SeedableRng};
    use rand_chacha::ChaCha8Rng;
    use std::collections::HashSet;
    use std::prelude::v1::Vec;

    #[test]
    fn compatibility() {
        let mut rng = ChaCha8Rng::from_seed([57u8; 32]);
        for _ in 0..1000 {
            let a = rng.r#gen::<u8>() & 0x0f;
            let b = rng.r#gen::<u8>() & 0x0f;
            let y = Gf16(a);
            let z = Gf16(b);

            assert_eq!((y * z).0, gf16_cmp::gf16_mul(a, b));
        }
        rng = ChaCha8Rng::from_entropy();
        for _ in 0..1000 {
            let a = rng.r#gen::<u8>() & 0x0f;
            let b = rng.r#gen::<u8>() & 0x0f;
            let y = Gf16(a);
            let z = Gf16(b);

            assert_eq!((y * z).0, gf16_cmp::gf16_mul(a, b));
        }

        // Verify all 15*15 = 225 non-zero products exhaustively
        for a in 1u8..=15 {
            for b in 1u8..=15 {
                assert_eq!(
                    gf16_mul(a, b),
                    gf16_cmp::gf16_mul(a, b),
                    "gf16_mul({a}, {b}) mismatch"
                );
            }
        }

        let mut rng = ChaCha8Rng::from_seed([57u8; 32]);
        for _ in 0..15 {
            let mut a = rng.r#gen::<u8>() & 0x0f;
            while a == 0 {
                a = rng.r#gen::<u8>() & 0x0f;
            }
            let y = Gf16(a);

            assert_eq!(y.invert().unwrap().0, gf16_cmp::gf16_inv(a));
        }
    }

    #[test]
    fn known_values() {
        // Powers of 2 (= x) in GF(2^4) with poly x^4+x+1:
        // x^1=2, x^2=4, x^3=8, x^4=3, x^5=6, x^6=12, x^7=11,
        // x^8=5, x^9=10, x^10=7, x^11=14, x^12=15, x^13=13, x^14=9, x^15=1
        let powers: [u8; 15] = [2, 4, 8, 3, 6, 12, 11, 5, 10, 7, 14, 15, 13, 9, 1];
        let mut val = Gf16(1);
        let generator = Gf16(2);
        for &expected in &powers {
            val = val * generator;
            assert_eq!(val.0, expected);
        }

        // 2^(-1) = 9
        assert_eq!(Gf16(2).invert().unwrap().0, 9);
        // 3^(-1): 3 = x^4, so 3^(-1) = 2^(-4) = 2^11 = 14
        assert_eq!(Gf16(3).invert().unwrap().0, 14);
        // 1^(-1) = 1
        assert_eq!(Gf16(1).invert().unwrap().0, 1);
        // 0 has no inverse
        assert!(bool::from(Gf16(0).invert().is_none()));
    }

    #[test]
    fn shamir() {
        let mut rng = ChaCha8Rng::from_seed([57u8; 32]);
        for i in 1u8..=15 {
            let secret = IdentifierGf16(Gf16(i));
            let shares = shamir::split_secret::<GfShare>(3, 5, &secret, &mut rng).unwrap();
            assert_eq!(shares[0].identifier.0.0, 1);
            assert_eq!(shares[1].identifier.0.0, 2);
            assert_eq!(shares[2].identifier.0.0, 3);
            assert_eq!(shares[3].identifier.0.0, 4);
            assert_eq!(shares[4].identifier.0.0, 5);
            let res = &shares[0..3].to_vec().combine();
            assert!(
                res.is_ok(),
                "Failed at iteration {}, secret: {}",
                i,
                secret.0.0
            );
            assert_eq!(
                res.unwrap(),
                secret,
                "Failed at iteration {}, secret: {}",
                i,
                secret.0.0
            );
        }
        rng = ChaCha8Rng::from_entropy();
        for i in 1u8..=15 {
            let secret = IdentifierGf16(Gf16(i));
            let shares = shamir::split_secret::<GfShare>(3, 5, &secret, &mut rng).unwrap();
            assert_eq!(shares[0].identifier.0.0, 1);
            assert_eq!(shares[1].identifier.0.0, 2);
            assert_eq!(shares[2].identifier.0.0, 3);
            assert_eq!(shares[3].identifier.0.0, 4);
            assert_eq!(shares[4].identifier.0.0, 5);
            let res = &shares[2..].to_vec().combine();
            assert_eq!(res.unwrap(), secret);
        }
    }

    #[test]
    fn split_array() {
        let mut rng = ChaCha8Rng::from_seed([57u8; 32]);
        let secret = b"Hello World!";
        let shares = Gf16::split_array(3, 5, secret, &mut rng).unwrap();
        assert_eq!(shares.len(), 5);

        let res = Gf16::combine_array(&shares[..3]);
        assert_eq!(res.unwrap(), secret);

        let p = ParticipantIdGeneratorType::Sequential {
            start: IdentifierGf16(Gf16(1)),
            increment: IdentifierGf16(Gf16(1)),
            count: 5,
        };
        let shares =
            Gf16::split_array_with_participant_generators(3, 5, secret, &mut rng, &[p]).unwrap();
        assert_eq!(shares.len(), 5);

        let res = Gf16::combine_array(&shares[..3]);
        let secret2 = res.unwrap();
        assert_eq!(secret2, secret);

        let res = Gf16::combine_array(&[shares[4].clone(), shares[1].clone(), shares[3].clone()]);
        let secret2 = res.unwrap();
        assert_eq!(secret2, secret);
    }

    #[test]
    fn combine_fuzz() {
        let res = Gf16::combine_array(&[vec![], vec![]]);
        assert!(res.is_err());
        let res = Gf16::combine_array(&[vec![1u8, 8u8], vec![2u8]]);
        assert!(res.is_err());

        let mut rng = ChaCha8Rng::from_entropy();
        for _ in 0..25 {
            let threshold = (rng.r#gen::<u8>() & 0x0f).saturating_add(1);

            let mut shares = Vec::with_capacity(threshold as usize);
            for i in 0..threshold {
                let share = vec![i; (rng.r#gen::<usize>() % 16) + 1];
                shares.push(share);
            }
            assert!(Gf16::combine_array(shares).is_err());
        }
    }

    /// Audit finding #1 (GF(16) analogue): `Gf16::random` must be
    /// uniform over the entire field, including even nibbles. The
    /// prior `(b & 0x0E) + 1` only produced odd values 1, 3, ..., 15.
    #[test]
    fn poc1_biased_gf16() {
        // 128 draws comfortably exceeds the ~10 needed to see even and
        // zero under a uniform distribution over 16 values.
        let mut rng = ChaCha8Rng::from_seed([57u8; 32]);
        let mut seen_even = false;
        let mut seen_zero = false;
        for _ in 0..128 {
            let x = <Gf16 as Field>::random(&mut rng).0;
            assert!(x < 16, "Gf16::random out of range: {x}");
            if x % 2 == 0 {
                seen_even = true;
            }
            if x == 0 {
                seen_zero = true;
            }
        }
        assert!(
            seen_even,
            "Gf16::random produced no even values — bias regression"
        );
        assert!(
            seen_zero,
            "Gf16::random produced no zero values over 10k draws"
        );
    }

    /// Audit finding #1 (GF(16) analogue): x-identifier must be
    /// non-zero and within the field range.
    #[test]
    fn poc1_identifier_gf16_nonzero() {
        // 1024 draws is sufficient to catch any biased sampler that can
        // produce zero or out-of-range values.
        let mut rng = ChaCha8Rng::from_seed([0xA5u8; 32]);
        for _ in 0..1024 {
            let id = IdentifierGf16::random(&mut rng);
            assert_ne!(id.0.0, 0, "IdentifierGf16::random yielded zero");
            assert!(
                id.0.0 < 16,
                "IdentifierGf16::random out of range: {}",
                id.0.0
            );
        }
    }

    /// Audit finding #3 (GF(16) analogue): sequential identifier
    /// generation must halt cleanly at 15 instead of saturating to
    /// an out-of-field value or pinning at 15.
    #[test]
    fn poc3_gf16_inc_bounds_check() {
        // Start near the upper boundary. Old saturating code would
        // pin at 15 (or leak values outside GF(16) with larger
        // increments); new code halts the stream after 15.
        let start = IdentifierGf16(Gf16(13));
        let inc = IdentifierGf16(Gf16(1));
        let seq = ParticipantIdGeneratorType::Sequential {
            start,
            increment: inc,
            count: 10,
        };
        let generators = [seq];
        let collection = ParticipantIdGeneratorCollection::from(&generators[..]);
        let ids: Vec<_> = collection.iter().collect();

        let mut seen = HashSet::new();
        for id in &ids {
            assert!(
                seen.insert(id.0.0),
                "duplicate identifier emitted: {}",
                id.0.0
            );
            assert!(
                id.0.0 > 0 && id.0.0 < 16,
                "id out of GF(16) range: {}",
                id.0.0
            );
        }
        assert!(
            ids.len() <= 3,
            "generator emitted {} ids past GF(16) boundary — saturating_add regression",
            ids.len()
        );
    }

    /// Direct observation of audit finding #2 over GF(16): polynomial
    /// `fill` must produce zero coefficients. With only 16 field
    /// elements, P(zero)=1/16 per coefficient — zeros should appear
    /// frequently across runs.
    #[test]
    fn zero_coefficients_actually_occur() {
        use crate::Polynomial;
        let mut rng = ChaCha8Rng::from_seed([0x7Fu8; 32]);
        let intercept = IdentifierGf16(Gf16(0x05));
        let threshold = 8usize;
        let runs = 50;
        let mut zero_coef_count = 0usize;
        for _ in 0..runs {
            let mut poly: Vec<GfShare> = <Vec<GfShare> as Polynomial<GfShare>>::create(threshold);
            poly.fill(&intercept, &mut rng, threshold).unwrap();
            for coef in &poly[1..threshold] {
                if coef.identifier.0.0 == 0 {
                    zero_coef_count += 1;
                }
            }
        }
        // 50 runs × 7 coefficients = 350 draws; under uniform GF(16)
        // expected zeros ≈ 350/16 ≈ 21. Prior biased fill yields 0.
        assert!(
            zero_coef_count > 0,
            "No zero coefficient across {runs} fills × {} slots — coefficient sampling still biased against zero",
            threshold - 1,
        );
    }

    /// A zero secret over GF(16) must split and reconstruct correctly.
    /// The existing `shamir` test iterates secrets 1..=15 only.
    #[test]
    fn zero_secret_round_trip() {
        let mut rng = ChaCha8Rng::from_seed([0xC3u8; 32]);
        let zero_secret = IdentifierGf16(Gf16(0));
        let shares = shamir::split_secret::<GfShare>(3, 5, &zero_secret, &mut rng).unwrap();
        let recovered = shares[..3].to_vec().combine().unwrap();
        assert_eq!(recovered, zero_secret, "zero-secret round-trip failed");
        let recovered2 = shares[2..].to_vec().combine().unwrap();
        assert_eq!(recovered2, zero_secret);
    }

    /// Shares whose nibble values are zero must still round-trip via
    /// `split_array`/`combine_array`. Over GF(16) with many reps, zero
    /// nibbles are extremely common.
    #[test]
    fn zero_valued_shares_round_trip() {
        let mut rng = ChaCha8Rng::from_seed([0x5Au8; 32]);
        let secret = b"The quick brown fox jumps over the lazy dog";
        let runs = 20;
        let mut saw_zero_nibble = false;
        for _ in 0..runs {
            let shares = Gf16::split_array(5, 8, secret, &mut rng).unwrap();
            for s in &shares {
                for &b in &s[1..] {
                    if (b & 0x0F) == 0 || (b >> 4) == 0 {
                        saw_zero_nibble = true;
                    }
                }
            }
            let recovered = Gf16::combine_array(&shares[..5]).unwrap();
            assert_eq!(
                &recovered[..],
                secret,
                "combine failed over GF(16) with zero-valued nibbles"
            );
        }
        assert!(
            saw_zero_nibble,
            "No zero nibble observed — statistical regression"
        );
    }

    /// Every share identifier emitted by GF(16) `split_secret` is
    /// non-zero (zero reserved for f(0) = secret).
    #[test]
    fn no_share_identifier_is_zero() {
        let mut rng = ChaCha8Rng::from_seed([0xDEu8; 32]);
        for _ in 0..50 {
            let secret = IdentifierGf16(Gf16(rng.r#gen::<u8>() & 0x0F));
            let shares = shamir::split_secret::<GfShare>(3, 5, &secret, &mut rng).unwrap();
            for s in &shares {
                assert_ne!(
                    s.identifier.0.0, 0,
                    "zero identifier produced by GF(16) split_secret",
                );
            }
        }
    }
}

#[cfg(test)]
#[cfg(any(feature = "alloc", feature = "std"))]
mod gf16_cmp {
    // Reference implementation using carryless multiplication with explicit reduction.
    // Based on: https://github.com/mikelodder7/mayo/blob/main/src/gf16.rs
    // Irreducible polynomial: x^4 + x + 1.
    // Reduction rule: x^4 = x + 1, x^5 = x^2 + x, x^6 = x^3 + x^2.
    // For the degree-6 product of two degree-3 polynomials, high nibble bits 4-6
    // each reduce to two lower-bit contributions:
    //   bit 4 -> bits 0,1 (x^4 = x+1)
    //   bit 5 -> bits 1,2 (x^5 = x^2+x)
    //   bit 6 -> bits 2,3 (x^6 = x^3+x^2)
    // So: result = (low nibble) ^ (high_nibble >> 4) ^ (high_nibble >> 3).

    /// Multiply two GF(16) elements mod x^4 + x + 1.
    pub fn gf16_mul(a: u8, b: u8) -> u8 {
        let mut p: u8 = 0;
        p ^= (a & 1).wrapping_mul(b);
        p ^= (a & 2).wrapping_mul(b);
        p ^= (a & 4).wrapping_mul(b);
        p ^= (a & 8).wrapping_mul(b);
        let top_p = p & 0xf0;
        (p ^ (top_p >> 4) ^ (top_p >> 3)) & 0x0f
    }

    /// Compute a^(-1) in GF(16) via a^14 = a^8 * a^4 * a^2.
    pub fn gf16_inv(a: u8) -> u8 {
        let a2 = gf16_mul(a, a);
        let a4 = gf16_mul(a2, a2);
        let a8 = gf16_mul(a4, a4);
        let a6 = gf16_mul(a2, a4);
        gf16_mul(a8, a6)
    }
}