commonware-storage 2026.7.0

Persist and retrieve data from an abstract store.
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
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
//! Segmented journal for fixed-size items.
//!
//! # Format
//!
//! Data is stored in one blob per section. Items are stored sequentially:
//!
//! ```text
//! +--------+--------+--------+----------+
//! | item_0 | item_1 |   ...  | item_n-1 |
//! +--------+--------+--------+----------+
//! ```
//!
//! # Sync
//!
//! Data written to `Journal` may not be immediately persisted to `Storage`. Use the
//! `sync` method to force pending data to be written.
//!
//! # Pruning
//!
//! All data must be assigned to a `section`. This allows pruning entire sections
//! (and their corresponding blobs) independently.

use super::manager::{AppendFactory, Config as ManagerConfig, Manager};
use crate::journal::Error;
use commonware_codec::{CodecFixed, CodecFixedShared, DecodeExt as _, ReadExt as _};
use commonware_runtime::{
    buffer::paged::{CacheRef, Replay},
    Blob, Buf, Handle, Metrics, Storage,
};
use commonware_utils::NZUsize;
use futures::{
    stream::{self, Stream},
    StreamExt,
};
use std::{marker::PhantomData, num::NonZeroUsize};
use tracing::{trace, warn};

/// State for replaying a single section's blob.
struct ReplayState<B: Blob> {
    section: u64,
    replay: Replay<B>,
    position: u64,
    done: bool,
}

/// Configuration for the fixed segmented journal.
#[derive(Clone)]
pub struct Config {
    /// The partition to use for storing blobs.
    pub partition: String,

    /// The page cache to use for caching data.
    pub page_cache: CacheRef,

    /// The size of the write buffer to use for each blob.
    pub write_buffer: NonZeroUsize,
}

/// A segmented journal with fixed-size entries.
///
/// Each section is stored in a separate blob. Within each blob, items are fixed-size.
///
/// # Repair
///
/// Like
/// [sqlite](https://github.com/sqlite/sqlite/blob/8658a8df59f00ec8fcfea336a2a6a4b5ef79d2ee/src/wal.c#L1504-L1505)
/// and
/// [rocksdb](https://github.com/facebook/rocksdb/blob/0c533e61bc6d89fdf1295e8e0bcee4edb3aef401/include/rocksdb/options.h#L441-L445),
/// the first invalid data read will be considered the new end of the journal (and the
/// underlying [Blob] will be truncated to the last valid item). Repair occurs during
/// init by checking each blob's size.
pub struct Journal<E: Storage + Metrics, A: CodecFixed> {
    manager: Manager<E, AppendFactory>,
    _array: PhantomData<A>,
}

impl<E: Storage + Metrics, A: CodecFixedShared> Journal<E, A> {
    /// Size of each entry.
    pub const CHUNK_SIZE: usize = A::SIZE;
    const CHUNK_SIZE_U64: u64 = Self::CHUNK_SIZE as u64;

    /// Initialize a new `Journal` instance.
    ///
    /// All backing blobs are opened but not read during initialization. Use `replay`
    /// to iterate over all items.
    pub async fn init(context: E, cfg: Config) -> Result<Self, Error> {
        let manager_cfg = ManagerConfig {
            partition: cfg.partition,
            factory: AppendFactory {
                write_buffer: cfg.write_buffer,
                page_cache_ref: cfg.page_cache,
            },
        };
        let mut manager = Manager::init(context, manager_cfg).await?;

        // Repair any blobs with trailing bytes (incomplete items from crash)
        let sections: Vec<_> = manager.sections().collect();
        for section in sections {
            let size = manager.size(section)?;
            if !size.is_multiple_of(Self::CHUNK_SIZE_U64) {
                let valid_size = size - (size % Self::CHUNK_SIZE_U64);
                warn!(
                    section,
                    invalid_size = size,
                    new_size = valid_size,
                    "trailing bytes detected: truncating"
                );
                manager.rewind_section(section, valid_size).await?;
                // Startup repair is exceptional; make it durable immediately so callers do not
                // need to track repaired sections separately.
                manager.sync(section).await?;
            }
        }

        Ok(Self {
            manager,
            _array: PhantomData,
        })
    }

    /// Append a new item to the journal in the given section.
    ///
    /// Returns the position of the item within the section (0-indexed).
    pub async fn append(&mut self, section: u64, item: &A) -> Result<u64, Error> {
        let blob = self.manager.get_or_create(section).await?;

        // Encode the item
        let buf = item.encode_mut();
        let offset = blob.append(&buf).await?;
        if !offset.is_multiple_of(Self::CHUNK_SIZE_U64) {
            return Err(Error::InvalidBlobSize(section, offset));
        }
        let position = offset / Self::CHUNK_SIZE_U64;
        trace!(section, position, "appended item");

        Ok(position)
    }

    /// Read the item at the given section and position.
    ///
    /// # Errors
    ///
    /// - [Error::AlreadyPrunedToSection] if the section has been pruned.
    /// - [Error::SectionOutOfRange] if the section doesn't exist.
    /// - [Error::ItemOutOfRange] if the position is beyond the blob size.
    pub async fn get(&self, section: u64, position: u64) -> Result<A, Error> {
        let blob = self
            .manager
            .get(section)?
            .ok_or(Error::SectionOutOfRange(section))?;

        let offset = position
            .checked_mul(Self::CHUNK_SIZE_U64)
            .ok_or(Error::ItemOutOfRange(position))?;

        // The read validates bounds against the blob's logical size.
        let buf = blob
            .read_at(offset, Self::CHUNK_SIZE)
            .await
            .map_err(|err| match err {
                commonware_runtime::Error::BlobInsufficientLength
                | commonware_runtime::Error::OffsetOverflow => Error::ItemOutOfRange(position),
                err => Error::Runtime(err),
            })?;
        A::decode(buf.coalesce()).map_err(Error::Codec)
    }

    /// Read multiple items from the same section into a caller buffer.
    ///
    /// `buf` must be at least `positions.len() * CHUNK_SIZE` bytes. All positions must be
    /// strictly increasing and within the section's bounds.
    ///
    /// Returns the decoded items and the number served without a blob read (page cache or tip
    /// buffer hits).
    pub async fn get_many(
        &self,
        section: u64,
        positions: &[u64],
        buf: &mut [u8],
    ) -> Result<(Vec<A>, usize), Error> {
        assert!(
            positions.is_sorted_by(|a, b| a < b),
            "positions must be strictly increasing"
        );
        if positions.is_empty() {
            return Ok((Vec::new(), 0));
        }
        assert!(
            buf.len() >= positions.len() * Self::CHUNK_SIZE,
            "get_many requires buf.len() >= positions.len() * CHUNK_SIZE"
        );
        let buf = &mut buf[..positions.len() * Self::CHUNK_SIZE];
        let blob = self
            .manager
            .get(section)?
            .ok_or(Error::SectionOutOfRange(section))?;

        let offsets: Vec<u64> = positions
            .iter()
            .map(|&p| {
                p.checked_mul(Self::CHUNK_SIZE_U64)
                    .ok_or(Error::ItemOutOfRange(p))
            })
            .collect::<Result<_, _>>()?;

        let hits = blob
            .read_many_into(buf, &offsets, NZUsize!(Self::CHUNK_SIZE))
            .await?;

        let mut items = Vec::with_capacity(positions.len());
        for i in 0..positions.len() {
            let slice = &buf[i * Self::CHUNK_SIZE..(i + 1) * Self::CHUNK_SIZE];
            items.push(A::decode(slice).map_err(Error::Codec)?);
        }
        Ok((items, hits))
    }

    /// Get an item if it can be done synchronously (e.g. without I/O), returning `None` otherwise.
    pub fn try_get_sync(&self, section: u64, position: u64) -> Option<A> {
        let blob = self.manager.get(section).ok()??;
        let offset = position.checked_mul(Self::CHUNK_SIZE_U64)?;
        let remaining = blob.size().checked_sub(offset)?;
        if remaining < Self::CHUNK_SIZE_U64 {
            return None;
        }
        let mut buf = vec![0u8; Self::CHUNK_SIZE];
        if !blob.try_read_sync_into(&mut buf, offset) {
            return None;
        }
        A::decode(&buf[..]).ok()
    }

    /// Read the last item in a section, if any.
    ///
    /// Returns `Ok(None)` if the section is empty.
    ///
    /// # Errors
    ///
    /// - [Error::AlreadyPrunedToSection] if the section has been pruned.
    /// - [Error::SectionOutOfRange] if the section doesn't exist.
    pub async fn last(&self, section: u64) -> Result<Option<A>, Error> {
        let blob = self
            .manager
            .get(section)?
            .ok_or(Error::SectionOutOfRange(section))?;

        let size = blob.size();
        if size < Self::CHUNK_SIZE_U64 {
            return Ok(None);
        }

        let last_position = (size / Self::CHUNK_SIZE_U64) - 1;
        let offset = last_position * Self::CHUNK_SIZE_U64;
        let buf = blob.read_at(offset, Self::CHUNK_SIZE).await?;
        A::decode(buf.coalesce()).map_err(Error::Codec).map(Some)
    }

    /// Returns a stream of all items starting from the given section.
    ///
    /// Each item is returned as (section, position, item).
    pub async fn replay(
        &mut self,
        start_section: u64,
        start_position: u64,
        buffer: NonZeroUsize,
    ) -> Result<impl Stream<Item = Result<(u64, u64, A), Error>> + Send + '_, Error> {
        // Pre-create readers from blobs. This validates replay setup but does not allocate
        // `buffer` bytes per blob; page buffers are allocated later by `Replay::ensure`.
        let mut blob_info = Vec::new();
        for (&section, blob) in self.manager.sections_from(start_section) {
            let blob_size = blob.size();
            let mut replay = blob.replay(buffer).await?;
            // For the first section, seek to the start position
            let initial_position = if section == start_section {
                let start = start_position
                    .checked_mul(Self::CHUNK_SIZE_U64)
                    .ok_or(Error::ItemOutOfRange(start_position))?;
                if start > blob_size {
                    return Err(Error::ItemOutOfRange(start_position));
                }
                replay.seek_to(start)?;
                start_position
            } else {
                0
            };
            blob_info.push((section, replay, initial_position));
        }

        // Stream items as they are read to avoid occupying too much memory.
        // Each blob is processed sequentially, yielding batches of items that are then
        // flattened into individual stream elements.
        Ok(
            stream::iter(blob_info).flat_map(move |(section, replay, initial_position)| {
                stream::unfold(
                    ReplayState {
                        section,
                        replay,
                        position: initial_position,
                        done: false,
                    },
                    move |mut state| async move {
                        if state.done {
                            return None;
                        }

                        let mut batch: Vec<Result<(u64, u64, A), Error>> = Vec::new();
                        loop {
                            // Ensure we have enough data for one item
                            match state.replay.ensure(Self::CHUNK_SIZE).await {
                                Ok(true) => {}
                                Ok(false) => {
                                    // Reader exhausted - we're done with this blob
                                    state.done = true;
                                    return if batch.is_empty() {
                                        None
                                    } else {
                                        Some((batch, state))
                                    };
                                }
                                Err(err) => {
                                    batch.push(Err(Error::Runtime(err)));
                                    state.done = true;
                                    return Some((batch, state));
                                }
                            }

                            // Decode items from buffer
                            while state.replay.remaining() >= Self::CHUNK_SIZE {
                                match A::read(&mut state.replay) {
                                    Ok(item) => {
                                        batch.push(Ok((state.section, state.position, item)));
                                        state.position += 1;
                                    }
                                    Err(err) => {
                                        batch.push(Err(Error::Codec(err)));
                                        state.done = true;
                                        return Some((batch, state));
                                    }
                                }
                            }

                            // Return batch if we have items
                            if !batch.is_empty() {
                                return Some((batch, state));
                            }
                        }
                    },
                )
                .flat_map(stream::iter)
            }),
        )
    }

    /// Sync the given `sections` to storage.
    pub async fn sync(&mut self, sections: impl crate::Sections) -> Result<(), Error> {
        self.manager.sync(sections).await
    }

    /// Start syncing the given `sections` to storage.
    pub async fn start_sync(
        &mut self,
        sections: impl crate::Sections,
    ) -> Result<Handle<()>, Error> {
        self.manager.start_sync(sections).await
    }

    /// Sync all sections to storage.
    pub async fn sync_all(&mut self) -> Result<(), Error> {
        self.manager.sync_all().await
    }

    /// Prune all sections less than `min`. Returns true if any were pruned.
    pub async fn prune(&mut self, min: u64) -> Result<bool, Error> {
        self.manager.prune(min).await
    }

    /// Returns the oldest section number, if any blobs exist.
    pub fn oldest_section(&self) -> Option<u64> {
        self.manager.oldest_section()
    }

    /// Returns the newest section number, if any blobs exist.
    pub fn newest_section(&self) -> Option<u64> {
        self.manager.newest_section()
    }

    /// Returns an iterator over all section numbers.
    pub fn sections(&self) -> impl Iterator<Item = u64> + '_ {
        self.manager.sections()
    }

    /// Returns the number of items in the given section.
    pub fn section_len(&self, section: u64) -> Result<u64, Error> {
        let size = self.manager.size(section)?;
        Ok(size / Self::CHUNK_SIZE_U64)
    }

    /// Returns the byte size of the given section.
    pub fn size(&self, section: u64) -> Result<u64, Error> {
        self.manager.size(section)
    }

    /// Rewind the journal to a specific section and byte offset.
    ///
    /// This truncates the section to the given size. All sections
    /// after `section` are removed.
    pub async fn rewind(&mut self, section: u64, offset: u64) -> Result<(), Error> {
        self.manager.rewind(section, offset).await
    }

    /// Rewind only the given section to a specific byte offset.
    ///
    /// Unlike `rewind`, this does not affect other sections.
    pub async fn rewind_section(&mut self, section: u64, size: u64) -> Result<(), Error> {
        self.manager.rewind_section(section, size).await
    }

    /// Remove all underlying blobs.
    pub async fn destroy(self) -> Result<(), Error> {
        self.manager.destroy().await
    }

    /// Clear all data, resetting the journal to an empty state.
    ///
    /// Unlike `destroy`, this keeps the journal alive so it can be reused.
    pub async fn clear(&mut self) -> Result<(), Error> {
        self.manager.clear().await
    }
}

#[cfg(test)]
mod tests {
    use super::*;
    use commonware_cryptography::{sha256::Digest, Hasher as _, Sha256};
    use commonware_macros::test_traced;
    use commonware_runtime::{
        buffer::paged::CacheRef,
        deterministic,
        mocks::{fail_pending_syncs, release_pending_syncs, DelayedSyncContext, PendingSyncs},
        BufferPooler, Error as RError, Runner, Spawner as _, Supervisor as _,
    };
    use commonware_utils::{NZUsize, NZU16};
    use core::num::NonZeroU16;
    use futures::{pin_mut, StreamExt};
    use std::sync::{
        atomic::{AtomicUsize, Ordering},
        Arc,
    };

    const PAGE_SIZE: NonZeroU16 = NZU16!(44);
    const PAGE_CACHE_SIZE: NonZeroUsize = NZUsize!(3);

    fn test_digest(value: u64) -> Digest {
        Sha256::hash(&value.to_be_bytes())
    }

    fn test_cfg(pooler: &impl BufferPooler) -> Config {
        Config {
            partition: "test-partition".into(),
            page_cache: CacheRef::from_pooler(pooler, PAGE_SIZE, PAGE_CACHE_SIZE),
            write_buffer: NZUsize!(2048),
        }
    }

    #[test_traced]
    fn test_segmented_fixed_append_and_get() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg.clone())
                .await
                .expect("failed to init");

            let pos0 = journal
                .append(1, &test_digest(0))
                .await
                .expect("failed to append");
            assert_eq!(pos0, 0);

            let pos1 = journal
                .append(1, &test_digest(1))
                .await
                .expect("failed to append");
            assert_eq!(pos1, 1);

            let pos2 = journal
                .append(2, &test_digest(2))
                .await
                .expect("failed to append");
            assert_eq!(pos2, 0);

            let item0 = journal.get(1, 0).await.expect("failed to get");
            assert_eq!(item0, test_digest(0));

            let item1 = journal.get(1, 1).await.expect("failed to get");
            assert_eq!(item1, test_digest(1));

            let item2 = journal.get(2, 0).await.expect("failed to get");
            assert_eq!(item2, test_digest(2));

            let err = journal.get(1, 2).await;
            assert!(matches!(err, Err(Error::ItemOutOfRange(2))));

            let err = journal.get(3, 0).await;
            assert!(matches!(err, Err(Error::SectionOutOfRange(3))));

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_replay() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("first"), cfg.clone())
                .await
                .expect("failed to init");

            for i in 0u64..10 {
                journal
                    .append(1, &test_digest(i))
                    .await
                    .expect("failed to append");
            }
            for i in 10u64..20 {
                journal
                    .append(2, &test_digest(i))
                    .await
                    .expect("failed to append");
            }

            journal.sync_all().await.expect("failed to sync");
            drop(journal);

            let mut journal = Journal::<_, Digest>::init(context.child("second"), cfg.clone())
                .await
                .expect("failed to re-init");

            let items = {
                let stream = journal
                    .replay(0, 0, NZUsize!(1024))
                    .await
                    .expect("failed to replay");
                pin_mut!(stream);

                let mut items = Vec::new();
                while let Some(result) = stream.next().await {
                    match result {
                        Ok((section, pos, item)) => items.push((section, pos, item)),
                        Err(err) => panic!("replay error: {err}"),
                    }
                }
                items
            };

            assert_eq!(items.len(), 20);
            for (i, item) in items.iter().enumerate().take(10) {
                assert_eq!(item.0, 1);
                assert_eq!(item.1, i as u64);
                assert_eq!(item.2, test_digest(i as u64));
            }
            for (i, item) in items.iter().enumerate().skip(10).take(10) {
                assert_eq!(item.0, 2);
                assert_eq!(item.1, (i - 10) as u64);
                assert_eq!(item.2, test_digest(i as u64));
            }

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_replay_with_start_offset() {
        // Test that replay with a non-zero start_position correctly skips items.
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("first"), cfg.clone())
                .await
                .expect("failed to init");

            // Append 10 items to section 1
            for i in 0u64..10 {
                journal
                    .append(1, &test_digest(i))
                    .await
                    .expect("failed to append");
            }
            // Append 5 items to section 2
            for i in 10u64..15 {
                journal
                    .append(2, &test_digest(i))
                    .await
                    .expect("failed to append");
            }
            journal.sync_all().await.expect("failed to sync");
            drop(journal);

            let mut journal = Journal::<_, Digest>::init(context.child("second"), cfg.clone())
                .await
                .expect("failed to re-init");

            // Replay from section 1, position 5 - should get items 5-9 from section 1 and all of section 2
            {
                let stream = journal
                    .replay(1, 5, NZUsize!(1024))
                    .await
                    .expect("failed to replay");
                pin_mut!(stream);

                let mut items = Vec::new();
                while let Some(result) = stream.next().await {
                    let (section, pos, item) = result.expect("replay error");
                    items.push((section, pos, item));
                }

                assert_eq!(
                    items.len(),
                    10,
                    "Should have 5 items from section 1 + 5 from section 2"
                );

                // Check section 1 items (positions 5-9)
                for (i, (section, pos, item)) in items.iter().enumerate().take(5) {
                    assert_eq!(*section, 1);
                    assert_eq!(*pos, (i + 5) as u64);
                    assert_eq!(*item, test_digest((i + 5) as u64));
                }

                // Check section 2 items (positions 0-4)
                for (i, (section, pos, item)) in items.iter().enumerate().skip(5) {
                    assert_eq!(*section, 2);
                    assert_eq!(*pos, (i - 5) as u64);
                    assert_eq!(*item, test_digest((i + 5) as u64));
                }
            }

            // Replay from section 1, position 9 - should get only item 9 from section 1 and all of section 2
            {
                let stream = journal
                    .replay(1, 9, NZUsize!(1024))
                    .await
                    .expect("failed to replay");
                pin_mut!(stream);

                let mut items = Vec::new();
                while let Some(result) = stream.next().await {
                    let (section, pos, item) = result.expect("replay error");
                    items.push((section, pos, item));
                }

                assert_eq!(
                    items.len(),
                    6,
                    "Should have 1 item from section 1 + 5 from section 2"
                );
                assert_eq!(items[0], (1, 9, test_digest(9)));
                for (i, (section, pos, item)) in items.iter().enumerate().skip(1) {
                    assert_eq!(*section, 2);
                    assert_eq!(*pos, (i - 1) as u64);
                    assert_eq!(*item, test_digest((i + 9) as u64));
                }
            }

            // Replay from section 2, position 3 - should get only items 3-4 from section 2
            {
                let stream = journal
                    .replay(2, 3, NZUsize!(1024))
                    .await
                    .expect("failed to replay");
                pin_mut!(stream);

                let mut items = Vec::new();
                while let Some(result) = stream.next().await {
                    let (section, pos, item) = result.expect("replay error");
                    items.push((section, pos, item));
                }

                assert_eq!(items.len(), 2, "Should have 2 items from section 2");
                assert_eq!(items[0], (2, 3, test_digest(13)));
                assert_eq!(items[1], (2, 4, test_digest(14)));
            }

            // Replay from position past the end should return ItemOutOfRange error
            let result = journal.replay(1, 100, NZUsize!(1024)).await;
            assert!(matches!(result, Err(Error::ItemOutOfRange(100))));
            drop(result);

            let result = journal.replay(1, u64::MAX, NZUsize!(1024)).await;
            assert!(matches!(result, Err(Error::ItemOutOfRange(u64::MAX))));
            drop(result);

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_prune() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg.clone())
                .await
                .expect("failed to init");

            for section in 1u64..=5 {
                journal
                    .append(section, &test_digest(section))
                    .await
                    .expect("failed to append");
            }
            journal.sync_all().await.expect("failed to sync");

            journal.prune(3).await.expect("failed to prune");

            let err = journal.get(1, 0).await;
            assert!(matches!(err, Err(Error::AlreadyPrunedToSection(3))));

            let err = journal.get(2, 0).await;
            assert!(matches!(err, Err(Error::AlreadyPrunedToSection(3))));

            let item = journal.get(3, 0).await.expect("should exist");
            assert_eq!(item, test_digest(3));

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_rewind() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg.clone())
                .await
                .expect("failed to init");

            // Create sections 1, 2, 3
            for section in 1u64..=3 {
                journal
                    .append(section, &test_digest(section))
                    .await
                    .expect("failed to append");
            }
            journal.sync_all().await.expect("failed to sync");

            // Verify all sections exist
            for section in 1u64..=3 {
                let size = journal.size(section).expect("failed to get size");
                assert!(size > 0, "section {section} should have data");
            }

            // Rewind to section 1 (should remove sections 2, 3)
            let size = journal.size(1).expect("failed to get size");
            journal.rewind(1, size).await.expect("failed to rewind");

            // Verify section 1 still has data
            let size = journal.size(1).expect("failed to get size");
            assert!(size > 0, "section 1 should still have data");

            // Verify sections 2, 3 are removed
            for section in 2u64..=3 {
                let size = journal.size(section).expect("failed to get size");
                assert_eq!(size, 0, "section {section} should be removed");
            }

            // Verify data in section 1 is still readable
            let item = journal.get(1, 0).await.expect("failed to get");
            assert_eq!(item, test_digest(1));

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_rewind_max_section() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg.clone())
                .await
                .expect("failed to init");

            // Append to the maximal section. `section + 1` has no representable successor.
            journal
                .append(u64::MAX, &test_digest(0))
                .await
                .expect("failed to append");
            journal.sync_all().await.expect("failed to sync");

            // Rewinding the maximal section removes no sections above it and must not panic.
            let size = journal.size(u64::MAX).expect("failed to get size");
            journal
                .rewind(u64::MAX, size)
                .await
                .expect("failed to rewind");

            // The section is intact and readable.
            assert_eq!(journal.size(u64::MAX).expect("failed to get size"), size);
            assert_eq!(journal.get(u64::MAX, 0).await.unwrap(), test_digest(0));

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_rewind_many_sections() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg.clone())
                .await
                .expect("failed to init");

            // Create sections 1-10
            for section in 1u64..=10 {
                journal
                    .append(section, &test_digest(section))
                    .await
                    .expect("failed to append");
            }
            journal.sync_all().await.expect("failed to sync");

            // Rewind to section 5 (should remove sections 6-10)
            let size = journal.size(5).expect("failed to get size");
            journal.rewind(5, size).await.expect("failed to rewind");

            // Verify sections 1-5 still have data
            for section in 1u64..=5 {
                let size = journal.size(section).expect("failed to get size");
                assert!(size > 0, "section {section} should still have data");
            }

            // Verify sections 6-10 are removed
            for section in 6u64..=10 {
                let size = journal.size(section).expect("failed to get size");
                assert_eq!(size, 0, "section {section} should be removed");
            }

            // Verify data integrity via replay
            {
                let stream = journal
                    .replay(0, 0, NZUsize!(1024))
                    .await
                    .expect("failed to replay");
                pin_mut!(stream);
                let mut items = Vec::new();
                while let Some(result) = stream.next().await {
                    let (section, _, item) = result.expect("failed to read");
                    items.push((section, item));
                }
                assert_eq!(items.len(), 5);
                for (i, (section, item)) in items.iter().enumerate() {
                    assert_eq!(*section, (i + 1) as u64);
                    assert_eq!(*item, test_digest((i + 1) as u64));
                }
            }

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_rewind_persistence() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);

            // Create sections 1-5
            let mut journal = Journal::init(context.child("first"), cfg.clone())
                .await
                .expect("failed to init");
            for section in 1u64..=5 {
                journal
                    .append(section, &test_digest(section))
                    .await
                    .expect("failed to append");
            }
            journal.sync_all().await.expect("failed to sync");

            // Rewind to section 2
            let size = journal.size(2).expect("failed to get size");
            journal.rewind(2, size).await.expect("failed to rewind");
            journal.sync_all().await.expect("failed to sync");
            drop(journal);

            // Re-init and verify only sections 1-2 exist
            let journal = Journal::<_, Digest>::init(context.child("second"), cfg.clone())
                .await
                .expect("failed to re-init");

            // Verify sections 1-2 have data
            for section in 1u64..=2 {
                let size = journal.size(section).expect("failed to get size");
                assert!(size > 0, "section {section} should have data after restart");
            }

            // Verify sections 3-5 are gone
            for section in 3u64..=5 {
                let size = journal.size(section).expect("failed to get size");
                assert_eq!(size, 0, "section {section} should be gone after restart");
            }

            // Verify data integrity
            let item1 = journal.get(1, 0).await.expect("failed to get");
            assert_eq!(item1, test_digest(1));
            let item2 = journal.get(2, 0).await.expect("failed to get");
            assert_eq!(item2, test_digest(2));

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_corruption_recovery() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("first"), cfg.clone())
                .await
                .expect("failed to init");

            for i in 0u64..5 {
                journal
                    .append(1, &test_digest(i))
                    .await
                    .expect("failed to append");
            }
            journal.sync_all().await.expect("failed to sync");
            drop(journal);

            let (blob, size) = context
                .open(&cfg.partition, &1u64.to_be_bytes())
                .await
                .expect("failed to open blob");
            blob.resize(size - 1).await.expect("failed to truncate");
            blob.sync().await.expect("failed to sync");

            let mut journal = Journal::<_, Digest>::init(context.child("second"), cfg.clone())
                .await
                .expect("failed to re-init");

            let count = {
                let stream = journal
                    .replay(0, 0, NZUsize!(1024))
                    .await
                    .expect("failed to replay");
                pin_mut!(stream);

                let mut count = 0;
                while let Some(result) = stream.next().await {
                    result.expect("should be ok");
                    count += 1;
                }
                count
            };
            assert_eq!(count, 4);

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_persistence() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);

            // Create and populate journal
            let mut journal = Journal::init(context.child("first"), cfg.clone())
                .await
                .expect("failed to init");

            for i in 0u64..5 {
                journal
                    .append(1, &test_digest(i))
                    .await
                    .expect("failed to append");
            }
            journal.sync_all().await.expect("failed to sync");
            drop(journal);

            // Reopen and verify data persisted
            let journal = Journal::<_, Digest>::init(context.child("second"), cfg)
                .await
                .expect("failed to re-init");

            for i in 0u64..5 {
                let item = journal.get(1, i).await.expect("failed to get");
                assert_eq!(item, test_digest(i));
            }

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_sync() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("first"), cfg.clone())
                .await
                .expect("failed to init");

            // One sub-page item per section stays buffered until synced.
            for section in 1u64..=3 {
                journal
                    .append(section, &test_digest(section))
                    .await
                    .expect("failed to append");
            }

            // Sync sections 1 and 3; a nonexistent section (99) is skipped, not an error.
            journal
                .sync(&[1, 3, 99])
                .await
                .expect("failed to sync sections");
            drop(journal);

            // Only the synced sections survive the unclean drop.
            let journal = Journal::<_, Digest>::init(context.child("second"), cfg)
                .await
                .expect("failed to re-init");
            assert_eq!(
                journal.get(1, 0).await.expect("section 1 durable"),
                test_digest(1)
            );
            assert_eq!(
                journal.get(3, 0).await.expect("section 3 durable"),
                test_digest(3)
            );
            assert!(matches!(
                journal.get(2, 0).await,
                Err(Error::ItemOutOfRange(0)) | Err(Error::SectionOutOfRange(2))
            ));

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_section_len() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg.clone())
                .await
                .expect("failed to init");

            assert_eq!(journal.section_len(1).unwrap(), 0);

            for i in 0u64..5 {
                journal
                    .append(1, &test_digest(i))
                    .await
                    .expect("failed to append");
            }

            assert_eq!(journal.section_len(1).unwrap(), 5);
            assert_eq!(journal.section_len(2).unwrap(), 0);

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_non_contiguous_sections() {
        // Test that sections with gaps in numbering work correctly.
        // Sections 1, 5, 10 should all be independent and accessible.
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("first"), cfg.clone())
                .await
                .expect("failed to init");

            // Create sections with gaps: 1, 5, 10
            journal
                .append(1, &test_digest(100))
                .await
                .expect("failed to append");
            journal
                .append(5, &test_digest(500))
                .await
                .expect("failed to append");
            journal
                .append(10, &test_digest(1000))
                .await
                .expect("failed to append");
            journal.sync_all().await.expect("failed to sync");

            // Verify random access to each section
            assert_eq!(journal.get(1, 0).await.unwrap(), test_digest(100));
            assert_eq!(journal.get(5, 0).await.unwrap(), test_digest(500));
            assert_eq!(journal.get(10, 0).await.unwrap(), test_digest(1000));

            // Verify non-existent sections return appropriate errors
            for missing_section in [0u64, 2, 3, 4, 6, 7, 8, 9, 11] {
                let result = journal.get(missing_section, 0).await;
                assert!(
                    matches!(result, Err(Error::SectionOutOfRange(_))),
                    "Expected SectionOutOfRange for section {}, got {:?}",
                    missing_section,
                    result
                );
            }

            // Drop and reopen to test replay
            drop(journal);
            let mut journal = Journal::<_, Digest>::init(context.child("second"), cfg.clone())
                .await
                .expect("failed to re-init");

            // Replay and verify all items in order
            {
                let stream = journal
                    .replay(0, 0, NZUsize!(1024))
                    .await
                    .expect("failed to replay");
                pin_mut!(stream);

                let mut items = Vec::new();
                while let Some(result) = stream.next().await {
                    let (section, _, item) = result.expect("replay error");
                    items.push((section, item));
                }

                assert_eq!(items.len(), 3, "Should have 3 items");
                assert_eq!(items[0], (1, test_digest(100)));
                assert_eq!(items[1], (5, test_digest(500)));
                assert_eq!(items[2], (10, test_digest(1000)));
            }

            // Test replay starting from middle section (5)
            {
                let stream = journal
                    .replay(5, 0, NZUsize!(1024))
                    .await
                    .expect("failed to replay from section 5");
                pin_mut!(stream);

                let mut items = Vec::new();
                while let Some(result) = stream.next().await {
                    let (section, _, item) = result.expect("replay error");
                    items.push((section, item));
                }

                assert_eq!(items.len(), 2, "Should have 2 items from section 5 onwards");
                assert_eq!(items[0], (5, test_digest(500)));
                assert_eq!(items[1], (10, test_digest(1000)));
            }

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_empty_section_in_middle() {
        // Test that replay correctly handles an empty section between sections with data.
        // Section 1 has data, section 2 is empty, section 3 has data.
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("first"), cfg.clone())
                .await
                .expect("failed to init");

            // Append to section 1
            journal
                .append(1, &test_digest(100))
                .await
                .expect("failed to append");

            // Create section 2 but make it empty via rewind
            journal
                .append(2, &test_digest(200))
                .await
                .expect("failed to append");
            journal.sync(2).await.expect("failed to sync");
            journal
                .rewind_section(2, 0)
                .await
                .expect("failed to rewind");

            // Append to section 3
            journal
                .append(3, &test_digest(300))
                .await
                .expect("failed to append");

            journal.sync_all().await.expect("failed to sync");

            // Verify section lengths
            assert_eq!(journal.section_len(1).unwrap(), 1);
            assert_eq!(journal.section_len(2).unwrap(), 0);
            assert_eq!(journal.section_len(3).unwrap(), 1);

            // Drop and reopen to test replay
            drop(journal);
            let mut journal = Journal::<_, Digest>::init(context.child("second"), cfg.clone())
                .await
                .expect("failed to re-init");

            // Replay all - should get items from sections 1 and 3, skipping empty section 2
            {
                let stream = journal
                    .replay(0, 0, NZUsize!(1024))
                    .await
                    .expect("failed to replay");
                pin_mut!(stream);

                let mut items = Vec::new();
                while let Some(result) = stream.next().await {
                    let (section, _, item) = result.expect("replay error");
                    items.push((section, item));
                }

                assert_eq!(
                    items.len(),
                    2,
                    "Should have 2 items (skipping empty section)"
                );
                assert_eq!(items[0], (1, test_digest(100)));
                assert_eq!(items[1], (3, test_digest(300)));
            }

            // Replay starting from empty section 2 - should get only section 3
            {
                let stream = journal
                    .replay(2, 0, NZUsize!(1024))
                    .await
                    .expect("failed to replay from section 2");
                pin_mut!(stream);

                let mut items = Vec::new();
                while let Some(result) = stream.next().await {
                    let (section, _, item) = result.expect("replay error");
                    items.push((section, item));
                }

                assert_eq!(items.len(), 1, "Should have 1 item from section 3");
                assert_eq!(items[0], (3, test_digest(300)));
            }

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_truncation_recovery_across_page_boundary() {
        // Test that truncating a single byte from a blob that has items straddling a page boundary
        // correctly recovers by removing the incomplete item.
        //
        // With PAGE_SIZE=44 and ITEM_SIZE=32:
        // - Item 0: bytes 0-31
        // - Item 1: bytes 32-63 (straddles page boundary at 44)
        // - Item 2: bytes 64-95 (straddles page boundary at 88)
        //
        // After 3 items we have 96 bytes = 2 full pages + 8 bytes. Truncating 1 byte leaves 95
        // bytes, which is not a multiple of 32. Recovery should truncate to 64 bytes (2 complete
        // items).
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("first"), cfg.clone())
                .await
                .expect("failed to init");

            // Append 3 items (just over 2 pages worth)
            for i in 0u64..3 {
                journal
                    .append(1, &test_digest(i))
                    .await
                    .expect("failed to append");
            }
            journal.sync_all().await.expect("failed to sync");

            // Verify all 3 items are readable
            for i in 0u64..3 {
                let item = journal.get(1, i).await.expect("failed to get");
                assert_eq!(item, test_digest(i));
            }
            drop(journal);

            // Truncate the blob by exactly 1 byte to simulate partial write
            let (blob, size) = context
                .open(&cfg.partition, &1u64.to_be_bytes())
                .await
                .expect("failed to open blob");
            blob.resize(size - 1).await.expect("failed to truncate");
            blob.sync().await.expect("failed to sync");
            drop(blob);

            // Reopen journal - should recover by truncating last page due to failed checksum, and
            // end up with a correct blob size due to partial-item trimming.
            let journal = Journal::<_, Digest>::init(context.child("second"), cfg.clone())
                .await
                .expect("failed to re-init");

            // Verify section now has only 2 items
            assert_eq!(journal.section_len(1).unwrap(), 2);

            // Verify size is the expected multiple of ITEM_SIZE (this would fail if we didn't trim
            // items and just relied on page-level checksum recovery).
            assert_eq!(journal.size(1).unwrap(), 64);

            // Items 0 and 1 should still be readable
            let item0 = journal.get(1, 0).await.expect("failed to get item 0");
            assert_eq!(item0, test_digest(0));
            let item1 = journal.get(1, 1).await.expect("failed to get item 1");
            assert_eq!(item1, test_digest(1));

            // Item 2 should return ItemOutOfRange
            let err = journal.get(1, 2).await;
            assert!(
                matches!(err, Err(Error::ItemOutOfRange(2))),
                "expected ItemOutOfRange(2), got {:?}",
                err
            );

            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_journal_clear() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = Config {
                partition: "clear-test".into(),
                page_cache: CacheRef::from_pooler(&context, PAGE_SIZE, PAGE_CACHE_SIZE),
                write_buffer: NZUsize!(1024),
            };

            let mut journal: Journal<_, Digest> =
                Journal::init(context.child("journal"), cfg.clone())
                    .await
                    .expect("Failed to initialize journal");

            // Append items across multiple sections
            for section in 0..5u64 {
                for i in 0..10u64 {
                    journal
                        .append(section, &test_digest(section * 1000 + i))
                        .await
                        .expect("Failed to append");
                }
                journal.sync(section).await.expect("Failed to sync");
            }

            // Verify we have data
            assert_eq!(journal.get(0, 0).await.unwrap(), test_digest(0));
            assert_eq!(journal.get(4, 0).await.unwrap(), test_digest(4000));

            // Clear the journal
            journal.clear().await.expect("Failed to clear");

            // After clear, all reads should fail
            for section in 0..5u64 {
                assert!(matches!(
                    journal.get(section, 0).await,
                    Err(Error::SectionOutOfRange(s)) if s == section
                ));
            }

            // Append new data after clear
            for i in 0..5u64 {
                journal
                    .append(10, &test_digest(i * 100))
                    .await
                    .expect("Failed to append after clear");
            }
            journal.sync(10).await.expect("Failed to sync after clear");

            // New data should be readable
            assert_eq!(journal.get(10, 0).await.unwrap(), test_digest(0));

            // Old sections should still be missing
            assert!(matches!(
                journal.get(0, 0).await,
                Err(Error::SectionOutOfRange(0))
            ));

            journal.destroy().await.unwrap();
        });
    }

    #[test_traced]
    fn test_last_missing_section_returns_error() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let journal = Journal::<_, Digest>::init(context.child("storage"), cfg.clone())
                .await
                .expect("failed to init");

            assert!(matches!(
                journal.last(0).await,
                Err(Error::SectionOutOfRange(0))
            ));
            assert!(matches!(
                journal.last(99).await,
                Err(Error::SectionOutOfRange(99))
            ));

            journal.destroy().await.unwrap();
        });
    }

    #[test_traced]
    fn test_last_after_rewind_to_zero() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg.clone())
                .await
                .expect("failed to init");

            journal.append(0, &test_digest(0)).await.unwrap();
            journal.append(0, &test_digest(1)).await.unwrap();
            journal.sync(0).await.unwrap();

            assert!(journal.last(0).await.unwrap().is_some());

            journal.rewind(0, 0).await.unwrap();
            assert_eq!(journal.last(0).await.unwrap(), None);

            journal.destroy().await.unwrap();
        });
    }

    #[test_traced]
    fn test_last_pruned_section_returns_error() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::<_, Digest>::init(context.child("storage"), cfg.clone())
                .await
                .expect("failed to init");

            journal.append(0, &test_digest(0)).await.unwrap();
            journal.append(1, &test_digest(1)).await.unwrap();
            journal.sync_all().await.unwrap();

            journal.prune(1).await.unwrap();

            assert!(matches!(
                journal.last(0).await,
                Err(Error::AlreadyPrunedToSection(1))
            ));
            assert!(journal.last(1).await.unwrap().is_some());

            journal.destroy().await.unwrap();
        });
    }

    #[test_traced]
    fn test_get_many_empty() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg).await.unwrap();
            journal.append(0, &test_digest(0)).await.unwrap();
            assert_eq!(journal.section_len(0).unwrap(), 1);

            let mut buf = [];
            let (items, hits) = journal.get_many(0, &[], &mut buf).await.unwrap();
            assert!(items.is_empty());
            assert_eq!(hits, 0);

            journal.destroy().await.unwrap();
        });
    }

    #[test_traced]
    fn test_get_many_single_section() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg).await.unwrap();

            for i in 0..5 {
                journal.append(0, &test_digest(i)).await.unwrap();
            }
            assert_eq!(journal.section_len(0).unwrap(), 5);

            // Read all 5 items in one call. The reusable buffer is intentionally oversized:
            // get_many slices it to the exact length the batch needs.
            let chunk = Journal::<deterministic::Context, Digest>::CHUNK_SIZE;
            let mut buf = vec![0u8; 6 * chunk];
            let (items, _) = journal
                .get_many(0, &[0, 1, 2, 3, 4], &mut buf)
                .await
                .unwrap();

            for (i, item) in items.iter().enumerate() {
                assert_eq!(*item, test_digest(i as u64));
            }

            journal.destroy().await.unwrap();
        });
    }

    #[test_traced]
    fn test_get_many_subset() {
        // Read a sparse subset of positions.
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg).await.unwrap();

            for i in 0..10 {
                journal.append(0, &test_digest(i)).await.unwrap();
            }
            assert_eq!(journal.section_len(0).unwrap(), 10);

            let chunk = Journal::<deterministic::Context, Digest>::CHUNK_SIZE;
            let positions = [1, 4, 7, 9];
            let mut buf = vec![0u8; positions.len() * chunk];
            let (items, _) = journal.get_many(0, &positions, &mut buf).await.unwrap();

            for (i, &pos) in positions.iter().enumerate() {
                assert_eq!(items[i], test_digest(pos));
            }

            journal.destroy().await.unwrap();
        });
    }

    #[test_traced]
    fn test_get_many_bad_section() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let journal = Journal::<_, Digest>::init(context.child("storage"), cfg)
                .await
                .unwrap();

            let mut buf = vec![0u8; 64];
            let err = journal.get_many(99, &[0], &mut buf).await.unwrap_err();
            assert!(matches!(err, Error::SectionOutOfRange(99)));

            journal.destroy().await.unwrap();
        });
    }

    #[test_traced]
    fn test_get_many_matches_get() {
        // Verify batch read matches individual reads.
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg).await.unwrap();

            for i in 0..8 {
                journal.append(0, &test_digest(i)).await.unwrap();
            }
            assert_eq!(journal.section_len(0).unwrap(), 8);
            journal.sync_all().await.unwrap();

            let chunk = Journal::<deterministic::Context, Digest>::CHUNK_SIZE;
            let positions: Vec<u64> = (0..8).collect();
            let mut buf = vec![0u8; positions.len() * chunk];
            let (batch, _) = journal.get_many(0, &positions, &mut buf).await.unwrap();

            for pos in &positions {
                let single = journal.get(0, *pos).await.unwrap();
                assert_eq!(batch[*pos as usize], single);
            }

            journal.destroy().await.unwrap();
        });
    }

    #[test_traced]
    fn test_segmented_fixed_prune_waits_for_in_flight_start_sync() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let pending = PendingSyncs::default();
            let context = DelayedSyncContext {
                inner: context,
                pending: pending.clone(),
            };
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg)
                .await
                .expect("failed to init");

            journal
                .append(1, &test_digest(0))
                .await
                .expect("failed to append");
            let handle = journal.start_sync(1).await.expect("failed to start sync");
            assert!(!pending.lock().is_empty());

            let started = Arc::new(AtomicUsize::new(0));
            let completed = Arc::new(AtomicUsize::new(0));
            let started_clone = started.clone();
            let completed_clone = completed.clone();
            let waiter = context.inner.child("prune").spawn(|_| async move {
                started_clone.fetch_add(1, Ordering::Relaxed);
                assert!(journal.prune(2).await.expect("failed to prune"));
                completed_clone.fetch_add(1, Ordering::Relaxed);
                journal
            });

            while started.load(Ordering::Relaxed) == 0 {
                commonware_runtime::reschedule().await;
            }
            commonware_runtime::reschedule().await;
            assert_eq!(
                completed.load(Ordering::Relaxed),
                0,
                "prune must wait for in-flight syncs on pruned sections"
            );

            release_pending_syncs(&pending);
            handle
                .await
                .expect("sync handle should complete despite pruning");
            while completed.load(Ordering::Relaxed) == 0 {
                commonware_runtime::reschedule().await;
            }
            let journal = waiter.await.expect("prune task failed");
            assert_eq!(journal.oldest_section(), None);
        });
    }

    #[test_traced]
    fn test_segmented_fixed_destroy_waits_for_in_flight_start_sync() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let pending = PendingSyncs::default();
            let context = DelayedSyncContext {
                inner: context,
                pending: pending.clone(),
            };
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg)
                .await
                .expect("failed to init");

            journal
                .append(1, &test_digest(0))
                .await
                .expect("failed to append");
            let handle = journal.start_sync(1).await.expect("failed to start sync");
            assert!(!pending.lock().is_empty());

            let started = Arc::new(AtomicUsize::new(0));
            let completed = Arc::new(AtomicUsize::new(0));
            let started_clone = started.clone();
            let completed_clone = completed.clone();
            let waiter = context.inner.child("destroy").spawn(|_| async move {
                started_clone.fetch_add(1, Ordering::Relaxed);
                journal.destroy().await.expect("failed to destroy");
                completed_clone.fetch_add(1, Ordering::Relaxed);
            });

            while started.load(Ordering::Relaxed) == 0 {
                commonware_runtime::reschedule().await;
            }
            commonware_runtime::reschedule().await;
            assert_eq!(
                completed.load(Ordering::Relaxed),
                0,
                "destroy must wait for in-flight syncs"
            );

            release_pending_syncs(&pending);
            handle
                .await
                .expect("sync handle should complete despite destruction");
            while completed.load(Ordering::Relaxed) == 0 {
                commonware_runtime::reschedule().await;
            }
            waiter.await.expect("destroy task failed");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_clear_waits_for_in_flight_start_sync() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let pending = PendingSyncs::default();
            let context = DelayedSyncContext {
                inner: context,
                pending: pending.clone(),
            };
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg)
                .await
                .expect("failed to init");

            journal
                .append(1, &test_digest(0))
                .await
                .expect("failed to append");
            let handle = journal.start_sync(1).await.expect("failed to start sync");
            assert!(!pending.lock().is_empty());

            let started = Arc::new(AtomicUsize::new(0));
            let completed = Arc::new(AtomicUsize::new(0));
            let started_clone = started.clone();
            let completed_clone = completed.clone();
            let waiter = context.inner.child("clear").spawn(|_| async move {
                started_clone.fetch_add(1, Ordering::Relaxed);
                journal.clear().await.expect("failed to clear");
                completed_clone.fetch_add(1, Ordering::Relaxed);
                journal
            });

            while started.load(Ordering::Relaxed) == 0 {
                commonware_runtime::reschedule().await;
            }
            commonware_runtime::reschedule().await;
            assert_eq!(
                completed.load(Ordering::Relaxed),
                0,
                "clear must wait for in-flight syncs"
            );

            release_pending_syncs(&pending);
            handle
                .await
                .expect("sync handle should complete despite clearing");
            while completed.load(Ordering::Relaxed) == 0 {
                commonware_runtime::reschedule().await;
            }
            let mut journal = waiter.await.expect("clear task failed");

            // The journal must remain usable after clear.
            assert_eq!(journal.oldest_section(), None);
            let position = journal
                .append(1, &test_digest(1))
                .await
                .expect("failed to append after clear");
            assert_eq!(position, 0);
            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_rewind_waits_for_in_flight_start_sync() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let pending = PendingSyncs::default();
            let context = DelayedSyncContext {
                inner: context,
                pending: pending.clone(),
            };
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg)
                .await
                .expect("failed to init");

            journal
                .append(1, &test_digest(0))
                .await
                .expect("failed to append");
            journal
                .append(2, &test_digest(1))
                .await
                .expect("failed to append");
            let handle = journal.start_sync(2).await.expect("failed to start sync");
            assert!(!pending.lock().is_empty());

            let size = journal.size(1).expect("failed to get size");
            let started = Arc::new(AtomicUsize::new(0));
            let completed = Arc::new(AtomicUsize::new(0));
            let started_clone = started.clone();
            let completed_clone = completed.clone();
            let waiter = context.inner.child("rewind").spawn(move |_| async move {
                started_clone.fetch_add(1, Ordering::Relaxed);
                journal.rewind(1, size).await.expect("failed to rewind");
                completed_clone.fetch_add(1, Ordering::Relaxed);
                journal
            });

            while started.load(Ordering::Relaxed) == 0 {
                commonware_runtime::reschedule().await;
            }
            commonware_runtime::reschedule().await;
            assert_eq!(
                completed.load(Ordering::Relaxed),
                0,
                "rewind must wait for in-flight syncs on removed sections"
            );

            release_pending_syncs(&pending);
            handle
                .await
                .expect("sync handle should complete despite rewind");
            while completed.load(Ordering::Relaxed) == 0 {
                commonware_runtime::reschedule().await;
            }
            let journal = waiter.await.expect("rewind task failed");
            assert_eq!(journal.size(2).expect("failed to get size"), 0);
            journal.destroy().await.expect("failed to destroy");
        });
    }

    #[test_traced]
    fn test_segmented_fixed_prune_surfaces_failed_in_flight_start_sync() {
        let executor = deterministic::Runner::default();
        executor.start(|context| async move {
            let pending = PendingSyncs::default();
            let context = DelayedSyncContext {
                inner: context,
                pending: pending.clone(),
            };
            let cfg = test_cfg(&context);
            let mut journal = Journal::init(context.child("storage"), cfg)
                .await
                .expect("failed to init");

            journal
                .append(1, &test_digest(0))
                .await
                .expect("failed to append");
            let handle = journal.start_sync(1).await.expect("failed to start sync");
            fail_pending_syncs(&pending);

            let err = journal
                .prune(2)
                .await
                .expect_err("prune must surface a failed in-flight sync");
            assert!(matches!(err, Error::Runtime(RError::Io(_))));

            let err = handle.await.expect_err("sync handle should fail");
            assert!(matches!(err, RError::Io(_)));
        });
    }
}