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
//! Streaming data processors for continuous data flows
//!
//! This module provides utilities for processing continuous data streams efficiently:
//!
//! - Stream processing with minimal memory overhead
//! - Pipeline-based data processing for complex transformations
//! - Backpressure handling for rate mismatches
//! - Buffer management for smooth data flow
//! - Fault tolerance with resume capabilities
use crate::error::{CoreError, ErrorContext, ErrorLocation};
use crate::memory_efficient::chunked::{ChunkedArray, ChunkingStrategy};
use crate::memory_efficient::prefetch::PrefetchConfig;
use ::ndarray::{ArrayBase, Dimension, OwnedRepr, RemoveAxis};
use std::collections::{BTreeMap, VecDeque};
use std::sync::{Arc, Condvar, Mutex, RwLock};
use std::thread::{self, JoinHandle};
use std::time::{Duration, Instant};
/// Type alias for the processing function
type ProcessFn<T, U> = Arc<dyn Fn(Vec<T>) -> Result<Vec<U>, CoreError> + Send + Sync>;
/// Stream processing mode
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum StreamMode {
/// Process data as it comes, with no buffering
Immediate,
/// Buffer data up to a certain size before processing
Buffered,
/// Adaptive processing based on system load and data rate
Adaptive,
/// Use a sliding window of data for processing
SlidingWindow,
}
/// Input source for a data stream
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum StreamSource {
/// File input (memory mapped)
File,
/// Network socket input
Network,
/// Real-time sensor data
Sensor,
/// Generated data (simulation, etc.)
Generated,
/// Another stream processor
Stream,
}
/// Stream processor state
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum StreamState {
/// Stream is initialized but not started
Initialized,
/// Stream is currently running
Running,
/// Stream is paused (can be resumed)
Paused,
/// Stream has completed
Completed,
/// Stream has encountered an error
Error,
}
/// Stream processor configuration
#[derive(Debug, Clone)]
pub struct StreamConfig {
/// Processing mode
pub mode: StreamMode,
/// Buffer size in elements
pub buffersize: usize,
/// Maximum batch size for processing
pub max_batch_size: usize,
/// Minimum batch size for processing
pub min_batch_size: usize,
/// Chunk size for chunked processing
pub chunk_size: usize,
/// Whether to use parallel processing
pub parallel: bool,
/// Number of worker threads for parallel processing
pub workers: Option<usize>,
/// Maximum processing rate (items per second, 0 for unlimited)
pub rate_limit: usize,
/// Timeout for waiting for data (milliseconds, 0 for none)
pub timeout_ms: u64,
/// Whether to enable prefetching
pub enable_prefetch: bool,
/// Prefetch configuration
pub prefetch_config: Option<PrefetchConfig>,
/// Whether to enable backpressure handling
pub enable_backpressure: bool,
/// Window size for sliding window mode
pub windowsize: usize,
/// Window stride for sliding window mode
pub window_stride: usize,
}
impl Default for StreamConfig {
fn default() -> Self {
Self {
mode: StreamMode::Buffered,
buffersize: 1024 * 1024, // 1M elements
max_batch_size: 65536, // 64K elements
min_batch_size: 1024, // 1K elements
chunk_size: 1024, // 1K elements
parallel: true,
workers: None,
rate_limit: 0,
timeout_ms: 1000,
enable_prefetch: true,
prefetch_config: None,
enable_backpressure: true,
windowsize: 1024,
window_stride: 256,
}
}
}
/// Builder for stream processor configuration
#[derive(Debug, Clone, Default)]
pub struct StreamConfigBuilder {
config: StreamConfig,
}
impl StreamConfigBuilder {
/// Create a new stream configuration builder with default values
pub fn new() -> Self {
Self::default()
}
/// Set the processing mode
pub const fn mode(mut self, mode: StreamMode) -> Self {
self.config.mode = mode;
self
}
/// Set the buffer size
pub const fn buffersize(mut self, size: usize) -> Self {
self.config.buffersize = size;
self
}
/// Set the maximum batch size
pub const fn max_batch_size(mut self, size: usize) -> Self {
self.config.max_batch_size = size;
self
}
/// Set the minimum batch size
pub const fn min_batch_size(mut self, size: usize) -> Self {
self.config.min_batch_size = size;
self
}
/// Set the chunk size
pub const fn chunk_size(mut self, size: usize) -> Self {
self.config.chunk_size = size;
self
}
/// Enable or disable parallel processing
pub const fn parallel(mut self, enable: bool) -> Self {
self.config.parallel = enable;
self
}
/// Set the number of worker threads
pub const fn workers(mut self, workers: Option<usize>) -> Self {
self.config.workers = workers;
self
}
/// Set the rate limit
pub const fn rate_limit(mut self, limit: usize) -> Self {
self.config.rate_limit = limit;
self
}
/// Set the timeout
pub const fn timeout_ms(mut self, timeout: u64) -> Self {
self.config.timeout_ms = timeout;
self
}
/// Enable or disable prefetching
pub const fn enable_prefetch(mut self, enable: bool) -> Self {
self.config.enable_prefetch = enable;
self
}
/// Set the prefetch configuration
pub const fn prefetch_config(mut self, config: Option<PrefetchConfig>) -> Self {
self.config.prefetch_config = config;
self
}
/// Enable or disable backpressure handling
pub const fn enable_backpressure(mut self, enable: bool) -> Self {
self.config.enable_backpressure = enable;
self
}
/// Set the window size for sliding window mode
pub const fn windowsize(mut self, size: usize) -> Self {
self.config.windowsize = size;
self
}
/// Set the window stride for sliding window mode
pub const fn window_stride(mut self, stride: usize) -> Self {
self.config.window_stride = stride;
self
}
/// Build the configuration
pub fn build(self) -> StreamConfig {
self.config
}
}
/// Stream processor statistics
#[derive(Debug, Clone)]
pub struct StreamStats {
/// Number of items processed
pub processed_items: usize,
/// Number of batches processed
pub processed_batches: usize,
/// Average batch size
pub avg_batch_size: f64,
/// Average processing time per batch (milliseconds)
pub avg_batch_time_ms: f64,
/// Average throughput (items per second)
pub avg_throughput: f64,
/// Stream uptime in seconds
pub uptime_seconds: f64,
/// Number of times backpressure was applied
pub backpressure_count: usize,
/// Buffer high water mark (maximum fill level)
pub buffer_high_water_mark: usize,
/// Error count
pub error_count: usize,
/// Last error message
pub lasterror: Option<String>,
}
impl Default for StreamStats {
fn default() -> Self {
Self {
processed_items: 0,
processed_batches: 0,
avg_batch_size: 0.0,
avg_batch_time_ms: 0.0,
avg_throughput: 0.0,
uptime_seconds: 0.0,
backpressure_count: 0,
buffer_high_water_mark: 0,
error_count: 0,
lasterror: None,
}
}
}
/// Stream input buffer for data queuing
#[derive(Debug)]
struct StreamBuffer<T: Clone + Send + 'static> {
/// Buffer data queue
data: VecDeque<T>,
/// Maximum buffer size
maxsize: usize,
/// Mutex for buffer access
mutex: Mutex<()>,
/// Condition variable for buffer synchronization
condvar: Condvar,
/// Whether the stream is closed
closed: bool,
}
impl<T: Clone + Send + 'static> StreamBuffer<T> {
/// Create a new stream buffer
fn new(maxsize: usize) -> Self {
Self {
data: VecDeque::with_capacity(maxsize),
maxsize,
mutex: Mutex::new(()),
condvar: Condvar::new(),
closed: false,
}
}
/// Add an item to the buffer
fn push(&mut self, item: T) -> Result<(), CoreError> {
let mut guard = self.mutex.lock().expect("Failed to acquire lock");
// Check if the buffer is closed
if self.closed {
return Err(CoreError::StreamError(
ErrorContext::new("Stream is closed".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Wait until there's space in the buffer
while self.data.len() >= self.maxsize {
guard = self
.condvar
.wait(guard)
.expect("Condition variable wait failed");
// Check if the buffer was closed while waiting
if self.closed {
return Err(CoreError::StreamError(
ErrorContext::new("Stream is closed".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
}
// Add the item to the buffer
self.data.push_back(item);
// Notify any waiting consumers
self.condvar.notify_one();
Ok(())
}
/// Add multiple items to the buffer
fn push_batch(&mut self, items: Vec<T>) -> Result<(), CoreError> {
let mut guard = self.mutex.lock().expect("Failed to acquire lock");
// Check if the buffer is closed
if self.closed {
return Err(CoreError::StreamError(
ErrorContext::new("Stream is closed".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Wait until there's space in the buffer
while self.data.len() + items.len() > self.maxsize {
guard = self
.condvar
.wait(guard)
.expect("Condition variable wait failed");
// Check if the buffer was closed while waiting
if self.closed {
return Err(CoreError::StreamError(
ErrorContext::new("Stream is closed".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
}
// Add the items to the buffer
self.data.extend(items);
// Notify any waiting consumers
self.condvar.notify_one();
Ok(())
}
/// Get a batch of items from the buffer
fn pop_batch(&mut self, max_batch_size: usize, timeoutms: u64) -> Result<Vec<T>, CoreError> {
let mut guard = self.mutex.lock().expect("Failed to acquire lock");
// Wait until there are items in the buffer
if self.data.is_empty() && !self.closed {
if timeoutms > 0 {
let timeout = Duration::from_millis(timeoutms);
let result = self.condvar.wait_timeout(guard, timeout);
match result {
Ok((g, timeout_result)) => {
#[allow(unused_assignments)]
{
guard = g;
}
// Check if the timeout occurred
if timeout_result.timed_out() && self.data.is_empty() {
return Err(CoreError::TimeoutError(
ErrorContext::new("Timeout waiting for data".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
}
Err(_) => {
return Err(CoreError::StreamError(
ErrorContext::new("Error waiting for data".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
}
} else {
// No timeout, wait indefinitely
#[allow(unused_assignments)]
{
guard = self
.condvar
.wait(guard)
.expect("Condition variable wait failed");
}
}
}
// Check if the buffer is closed and empty
if self.data.is_empty() && self.closed {
return Err(CoreError::EndOfStream(
ErrorContext::new("End of stream".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Get the items (up to max_batch_size)
let batch_size = std::cmp::min(max_batch_size, self.data.len());
let mut batch = Vec::with_capacity(batch_size);
for _ in 0..batch_size {
if let Some(item) = self.data.pop_front() {
batch.push(item);
} else {
break;
}
}
// Notify any waiting producers
self.condvar.notify_one();
Ok(batch)
}
/// Get the number of items in the buffer
fn len(&self) -> usize {
let _guard = self.mutex.lock().expect("Failed to acquire lock");
self.data.len()
}
/// Check if the buffer is empty
fn is_empty(&self) -> bool {
let _guard = self.mutex.lock().expect("Failed to acquire lock");
self.data.is_empty()
}
/// Close the buffer
fn close(&mut self) {
let _guard = self.mutex.lock().expect("Failed to acquire lock");
self.closed = true;
self.condvar.notify_all();
}
/// Check if the buffer is closed
#[allow(dead_code)]
fn is_closed(&self) -> bool {
let _guard = self.mutex.lock().expect("Failed to acquire lock");
self.closed
}
/// Clear the buffer
fn clear(&mut self) {
let _guard = self.mutex.lock().expect("Failed to acquire lock");
self.data.clear();
self.condvar.notify_all();
}
}
/// Stream processor for continuous data flows
pub struct StreamProcessor<T: Clone + Send + 'static, U: Clone + Send + 'static> {
/// Configuration for the stream processor
config: StreamConfig,
/// Input buffer
input_buffer: Arc<Mutex<StreamBuffer<T>>>,
/// Processing function
processfn: ProcessFn<T, U>,
/// Output buffer
output_buffer: Arc<Mutex<StreamBuffer<U>>>,
/// Current state of the stream processor
state: Arc<RwLock<StreamState>>,
/// Statistics for the stream processor
stats: Arc<RwLock<StreamStats>>,
/// Worker thread handle
worker_thread: Option<JoinHandle<()>>,
/// Start time of the stream processor
start_time: Arc<RwLock<Option<Instant>>>,
}
impl<T, U> std::fmt::Debug for StreamProcessor<T, U>
where
T: Clone + Send + 'static,
U: Clone + Send + 'static,
{
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.debug_struct("StreamProcessor")
.field("config", &self.config)
.field("state", &self.state)
.field("stats", &self.stats)
.field("worker_thread", &self.worker_thread.is_some())
.field("start_time", &self.start_time)
.finish_non_exhaustive()
}
}
impl<T: Clone + Send + 'static, U: Clone + Send + 'static> StreamProcessor<T, U> {
/// Create a new stream processor
pub fn new<F>(config: StreamConfig, processfn: F) -> Self
where
F: Fn(Vec<T>) -> Result<Vec<U>, CoreError> + Send + Sync + 'static,
{
let input_buffer = Arc::new(Mutex::new(StreamBuffer::new(config.buffersize)));
let output_buffer = Arc::new(Mutex::new(StreamBuffer::new(config.buffersize)));
Self {
config,
input_buffer,
processfn: Arc::new(processfn),
output_buffer,
state: Arc::new(RwLock::new(StreamState::Initialized)),
stats: Arc::new(RwLock::new(StreamStats::default())),
worker_thread: None,
start_time: Arc::new(RwLock::new(None)),
}
}
/// Start the stream processor
pub fn start(&mut self) -> Result<(), CoreError> {
let mut state = self.state.write().expect("Failed to acquire write lock");
// Check if the stream is already running
if *state == StreamState::Running {
return Err(CoreError::StreamError(
ErrorContext::new("Stream already running".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Update state
*state = StreamState::Running;
// Set start time
let mut start_time = self
.start_time
.write()
.expect("Failed to acquire write lock");
*start_time = Some(Instant::now());
// Create worker thread
let input_buffer = self.input_buffer.clone();
let output_buffer = self.output_buffer.clone();
let processfn = self.processfn.clone();
let config = self.config.clone();
let state = self.state.clone();
let stats = self.stats.clone();
let start_time_clone = self.start_time.clone();
let worker = thread::spawn(move || {
Self::worker_loop(
input_buffer,
output_buffer,
processfn,
config,
state,
stats,
start_time_clone,
);
});
self.worker_thread = Some(worker);
Ok(())
}
/// Worker loop for processing data
fn worker_loop(
input_buffer: Arc<Mutex<StreamBuffer<T>>>,
output_buffer: Arc<Mutex<StreamBuffer<U>>>,
processfn: ProcessFn<T, U>,
config: StreamConfig,
state: Arc<RwLock<StreamState>>,
stats: Arc<RwLock<StreamStats>>,
start_time: Arc<RwLock<Option<Instant>>>,
) {
// Setup rate limiting if needed
let rate_limit = config.rate_limit;
let mut last_batch_time = Instant::now();
let mut batch_window = VecDeque::new();
// Processing loop
loop {
// Check if we should continue
{
let current_state = state.read().expect("Failed to acquire read lock");
if *current_state != StreamState::Running {
break;
}
}
// Rate limiting
if rate_limit > 0 {
// Calculate the minimum _time per batch
let min_time_per_batch =
Duration::from_secs_f64(config.min_batch_size as f64 / rate_limit as f64);
// Wait if necessary
let elapsed = last_batch_time.elapsed();
if elapsed < min_time_per_batch {
thread::sleep(min_time_per_batch - elapsed);
}
}
// Determine batch size based on the mode
let batch_size = match config.mode {
StreamMode::Immediate => 1,
StreamMode::Buffered => config.max_batch_size,
StreamMode::Adaptive => {
// Simple adaptive batch sizing based on processing time
let stats_guard = stats.read().expect("Failed to acquire read lock");
let avg_time = stats_guard.avg_batch_time_ms;
if avg_time < 10.0 {
// Processing is fast, use larger batches
config.max_batch_size
} else if avg_time < 50.0 {
// Medium processing time, use medium batches
(config.max_batch_size + config.min_batch_size) / 2
} else {
// Slow processing, use smaller batches
config.min_batch_size
}
}
StreamMode::SlidingWindow => config.windowsize,
};
// Get a batch of data from the input buffer
let input_batch = match input_buffer
.lock()
.expect("Failed to acquire lock")
.pop_batch(batch_size, config.timeout_ms)
{
Ok(batch) => batch,
Err(err) => {
match err {
CoreError::EndOfStream(_) => {
// End of stream, update state and exit
let mut current_state =
state.write().expect("Failed to acquire write lock");
*current_state = StreamState::Completed;
break;
}
CoreError::TimeoutError(_) => {
// Timeout, continue
continue;
}
_ => {
// Other error, update stats and continue
let mut stats_guard =
stats.write().expect("Failed to acquire write lock");
stats_guard.error_count += 1;
stats_guard.lasterror = Some(err.to_string());
continue;
}
}
}
};
// Check if the batch is empty
if input_batch.is_empty() {
continue;
}
// For sliding window mode, manage the window
let process_input = if config.mode == StreamMode::SlidingWindow {
if batch_window.len() < config.windowsize {
// Still filling the initial window
batch_window.extend(input_batch);
if batch_window.len() < config.windowsize {
// Not enough data for a full window yet
continue;
}
// We now have a full window
batch_window.make_contiguous().to_vec()
} else {
// Slide the window
let stride = std::cmp::min(config.window_stride, input_batch.len());
// Remove old elements
for _ in 0..stride {
batch_window.pop_front();
}
// Add new elements
batch_window.extend(input_batch);
// Return the window for processing
batch_window.make_contiguous().to_vec()
}
} else {
// For non-window modes, just use the batch directly
input_batch
};
// Process the batch
let process_result = {
let batch_start_time = Instant::now();
let result = processfn(process_input.clone());
// Update processing statistics
let mut stats_guard = stats.write().expect("Failed to acquire write lock");
stats_guard.processed_batches += 1;
stats_guard.processed_items += process_input.len();
// Update average batch size
let total_items = stats_guard.processed_items;
let total_batches = stats_guard.processed_batches;
stats_guard.avg_batch_size = total_items as f64 / total_batches as f64;
// Update processing time
let batch_time = batch_start_time.elapsed().as_millis() as f64;
stats_guard.avg_batch_time_ms =
(stats_guard.avg_batch_time_ms * (total_batches - 1) as f64 + batch_time)
/ total_batches as f64;
// Update throughput
if let Some(start) = *start_time.read().expect("Failed to acquire read lock") {
let uptime_seconds = start.elapsed().as_secs_f64();
stats_guard.uptime_seconds = uptime_seconds;
stats_guard.avg_throughput = total_items as f64 / uptime_seconds;
}
// Update buffer statistics
let buffer_len = input_buffer.lock().expect("Failed to acquire lock").len();
if buffer_len > stats_guard.buffer_high_water_mark {
stats_guard.buffer_high_water_mark = buffer_len;
}
result
};
// Handle the processing result
match process_result {
Ok(output_batch) => {
// Send the output to the output _buffer
if !output_batch.is_empty() {
match output_buffer
.lock()
.expect("Failed to acquire lock")
.push_batch(output_batch)
{
Ok(_) => {}
Err(err) => {
// Error sending output, update stats
let mut stats_guard =
stats.write().expect("Failed to acquire write lock");
stats_guard.error_count += 1;
stats_guard.lasterror = Some(err.to_string());
}
}
}
}
Err(err) => {
// Processing error, update stats
let mut stats_guard = stats.write().expect("Failed to acquire write lock");
stats_guard.error_count += 1;
stats_guard.lasterror = Some(err.to_string());
}
}
// Update rate limiting info
last_batch_time = Instant::now();
}
}
/// Stop the stream processor
pub fn stop(&mut self) -> Result<(), CoreError> {
let mut state = self.state.write().expect("Failed to acquire write lock");
// Check if the stream is running
if *state != StreamState::Running {
return Err(CoreError::StreamError(
ErrorContext::new("Stream not running".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Update state
*state = StreamState::Paused;
// Close the input buffer
self.input_buffer
.lock()
.expect("Failed to acquire lock")
.close();
// Wait for the worker thread to finish
if let Some(worker) = self.worker_thread.take() {
match worker.join() {
Ok(_) => {}
Err(_) => {
return Err(CoreError::StreamError(
ErrorContext::new("Error joining worker thread".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
}
}
Ok(())
}
/// Push data to the stream processor
pub fn push(&self, data: T) -> Result<(), CoreError> {
// Check if the stream is running
let state = self.state.read().expect("Failed to acquire read lock");
if *state != StreamState::Running {
return Err(CoreError::StreamError(
ErrorContext::new("Stream not running".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Push data to the input buffer
self.input_buffer
.lock()
.expect("Failed to acquire lock")
.push(data)
}
/// Push a batch of data to the stream processor
pub fn push_batch(&self, data: Vec<T>) -> Result<(), CoreError> {
// Check if the stream is running
let state = self.state.read().expect("Failed to acquire read lock");
if *state != StreamState::Running {
return Err(CoreError::StreamError(
ErrorContext::new("Stream not running".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Push data to the input buffer
self.input_buffer
.lock()
.expect("Failed to acquire lock")
.push_batch(data)
}
/// Pop processed data from the stream processor
pub fn pop(&self) -> Result<U, CoreError> {
// Check if the stream is running or completed
let state = self.state.read().expect("Failed to acquire read lock");
if *state != StreamState::Running && *state != StreamState::Completed {
return Err(CoreError::StreamError(
ErrorContext::new("Stream not running or completed".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Pop data from the output buffer
let result = self
.output_buffer
.lock()
.expect("Failed to acquire lock")
.pop_batch(1, self.config.timeout_ms)?;
if result.is_empty() {
Err(CoreError::TimeoutError(
ErrorContext::new("Timeout waiting for data".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
))
} else {
Ok(result[0].clone())
}
}
/// Pop a batch of processed data from the stream processor
pub fn pop_batch(&self, batchsize: usize) -> Result<Vec<U>, CoreError> {
// Check if the stream is running or completed
let state = self.state.read().expect("Failed to acquire read lock");
if *state != StreamState::Running && *state != StreamState::Completed {
return Err(CoreError::StreamError(
ErrorContext::new("Stream not running or completed".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Pop data from the output buffer
self.output_buffer
.lock()
.expect("Failed to acquire lock")
.pop_batch(batchsize, self.config.timeout_ms)
}
/// Get the current state of the stream processor
pub fn state(&self) -> StreamState {
*self.state.read().expect("Failed to acquire read lock")
}
/// Get the statistics for the stream processor
pub fn stats(&self) -> StreamStats {
self.stats
.read()
.expect("Failed to acquire read lock")
.clone()
}
/// Check if the stream is empty
pub fn is_empty(&self) -> bool {
self.input_buffer
.lock()
.expect("Failed to acquire lock")
.is_empty()
&& self
.output_buffer
.lock()
.expect("Failed to acquire lock")
.is_empty()
}
/// Clear the stream buffers
pub fn clear(&self) -> Result<(), CoreError> {
// Check if the stream is not running
let state = self.state.read().expect("Failed to acquire read lock");
if *state == StreamState::Running {
return Err(CoreError::StreamError(
ErrorContext::new("Cannot clear running stream".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Clear the buffers
self.input_buffer
.lock()
.expect("Failed to acquire lock")
.clear();
self.output_buffer
.lock()
.expect("Failed to acquire lock")
.clear();
Ok(())
}
}
impl<T: Clone + Send + 'static, U: Clone + Send + 'static> Drop for StreamProcessor<T, U> {
fn drop(&mut self) {
// Stop the stream if it's running
if *self.state.read().expect("Failed to acquire read lock") == StreamState::Running {
let _ = self.stop();
}
}
}
/// A stage in a stream processing pipeline
#[derive(Debug)]
pub struct PipelineStage<I: Clone + Send + 'static, O: Clone + Send + 'static> {
/// Name of the stage
pub name: String,
/// Stream processor for this stage
processor: Arc<Mutex<StreamProcessor<I, O>>>,
/// Whether this stage is parallel
pub parallel: bool,
/// Number of parallel instances
pub parallelism: usize,
}
impl<I: Clone + Send + 'static, O: Clone + Send + 'static> PipelineStage<I, O> {
/// Create a new pipeline stage
pub fn new<F>(
name: String,
config: StreamConfig,
processfn: F,
parallel: bool,
parallelism: usize,
) -> Self
where
F: Fn(Vec<I>) -> Result<Vec<O>, CoreError> + Send + Sync + Clone + 'static,
{
let processor = StreamProcessor::new(config, processfn);
Self {
name,
processor: Arc::new(Mutex::new(processor)),
parallel,
parallelism,
}
}
/// Get the processor for this stage
pub fn processor(&self) -> Arc<Mutex<StreamProcessor<I, O>>> {
self.processor.clone()
}
/// Start the stage
pub fn start(&self) -> Result<(), CoreError> {
self.processor
.lock()
.expect("Failed to acquire lock")
.start()
}
/// Stop the stage
pub fn stop(&self) -> Result<(), CoreError> {
self.processor
.lock()
.expect("Failed to acquire lock")
.stop()
}
/// Get the state of the stage
pub fn state(&self) -> StreamState {
self.processor
.lock()
.expect("Failed to acquire lock")
.state()
}
/// Get the statistics for the stage
pub fn stats(&self) -> StreamStats {
self.processor
.lock()
.expect("Failed to acquire lock")
.stats()
}
}
/// Stream processing pipeline
pub struct Pipeline {
/// Name of the pipeline
pub name: String,
/// Stages in the pipeline
stages: Vec<Box<dyn AnyStage>>,
/// Connections between stages
connections: Vec<(usize, usize)>, // (from_stage, to_stage)
/// Worker threads for the pipeline
workers: Vec<JoinHandle<()>>,
/// Pipeline state
state: Arc<RwLock<StreamState>>,
/// Pipeline statistics
#[allow(dead_code)]
stats: Arc<RwLock<PipelineStats>>,
/// Error context for the pipeline
error_context: Arc<RwLock<Option<ErrorContext>>>,
}
/// Pipeline statistics
#[derive(Debug, Clone)]
pub struct PipelineStats {
/// Statistics for each stage
pub stage_stats: BTreeMap<String, StreamStats>,
/// Total items processed
pub total_items: usize,
/// Pipeline uptime in seconds
pub uptime_seconds: f64,
/// Overall throughput (items per second)
pub overall_throughput: f64,
/// Bottleneck stage (slowest stage)
pub bottleneck_stage: Option<String>,
/// Bottleneck throughput (items per second)
pub bottleneck_throughput: f64,
}
impl Default for PipelineStats {
fn default() -> Self {
Self {
stage_stats: BTreeMap::new(),
total_items: 0,
uptime_seconds: 0.0,
overall_throughput: 0.0,
bottleneck_stage: None,
bottleneck_throughput: f64::MAX,
}
}
}
/// Trait for pipeline stages of any type
pub trait AnyStage: Send + Sync {
/// Get the name of the stage
fn name(&self) -> &str;
/// Start the stage
fn start(&self) -> Result<(), CoreError>;
/// Stop the stage
fn stop(&self) -> Result<(), CoreError>;
/// Get the state of the stage
fn state(&self) -> StreamState;
/// Get the statistics for the stage
fn stats(&self) -> StreamStats;
/// Check if the stage is empty
fn is_empty(&self) -> bool;
/// Push raw data to the stage
fn push_raw(&self, data: Box<dyn std::any::Any + Send>) -> Result<(), CoreError>;
/// Pop raw data from the stage
fn pop_raw(&self) -> Result<Box<dyn std::any::Any + Send>, CoreError>;
/// Clone the stage into a new Box
fn clone_box_impl(&self) -> Box<dyn AnyStage>;
}
/// Pipeline builder
pub struct PipelineBuilder {
/// Name of the pipeline
name: String,
/// Stages in the pipeline
stages: Vec<Box<dyn AnyStage>>,
/// Connections between stages
connections: Vec<(usize, usize)>, // (from_stage, to_stage)
}
impl PipelineBuilder {
/// Create a new pipeline builder
pub fn new(name: String) -> Self {
Self {
name,
stages: Vec::new(),
connections: Vec::new(),
}
}
/// Add a stage to the pipeline
pub fn add_stage<I, O, F>(
&mut self,
name: String,
config: StreamConfig,
processfn: F,
parallel: bool,
parallelism: usize,
) -> usize
where
I: Clone + Send + 'static,
O: Clone + Send + 'static,
F: Fn(Vec<I>) -> Result<Vec<O>, CoreError> + Send + Sync + Clone + 'static,
{
let stage = PipelineStage::new(name, config, processfn, parallel, parallelism);
let stage_index = self.stages.len();
self.stages.push(Box::new(StageWrapper::new(stage)));
stage_index
}
/// Connect two stages in the pipeline
pub fn connect(&mut self, from_stage: usize, tostage: usize) -> &mut Self {
if from_stage < self.stages.len() && tostage < self.stages.len() {
self.connections.push((from_stage, tostage));
}
self
}
/// Build the pipeline
pub fn build(self) -> Pipeline {
Pipeline {
name: self.name,
stages: self.stages,
connections: self.connections,
workers: Vec::new(),
state: Arc::new(RwLock::new(StreamState::Initialized)),
stats: Arc::new(RwLock::new(PipelineStats::default())),
error_context: Arc::new(RwLock::new(None)),
}
}
}
impl Pipeline {
/// Start the pipeline
pub fn start(&mut self) -> Result<(), CoreError> {
let mut state = self.state.write().expect("Failed to acquire write lock");
// Check if the pipeline is already running
if *state == StreamState::Running {
return Err(CoreError::StreamError(
ErrorContext::new("Pipeline already running".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Start all stages
for stage in &self.stages {
stage.start()?;
}
// Create worker threads for each connection
for (from_stage, to_stage) in &self.connections {
let from_stage = &self.stages[*from_stage];
let to_stage = &self.stages[*to_stage];
let from_stage_clone = from_stage.clone_box();
let to_stage_clone = to_stage.clone_box();
let state_clone = self.state.clone();
let error_context_clone = self.error_context.clone();
let worker = thread::spawn(move || {
Self::connection_worker(
from_stage_clone,
to_stage_clone,
state_clone,
error_context_clone,
);
});
self.workers.push(worker);
}
// Update state
*state = StreamState::Running;
Ok(())
}
/// Worker function for processing data between stages
fn connection_worker(
from_stage: Box<dyn AnyStage>,
to_stage: Box<dyn AnyStage>,
state: Arc<RwLock<StreamState>>,
error_context: Arc<RwLock<Option<ErrorContext>>>,
) {
let mut consecutiveerrors = 0;
let error_threshold = 10; // Maximum number of consecutive errors before giving up
// Processing loop
loop {
// Check if we should continue
{
let current_state = state.read().expect("Failed to acquire read lock");
if *current_state != StreamState::Running {
break;
}
}
// Try to get data from the source stage
match from_stage.pop_raw() {
Ok(data) => {
// Reset error counter
consecutiveerrors = 0;
// Try to push data to the destination stage
if let Err(err) = to_stage.push_raw(data) {
// Handle error
consecutiveerrors += 1;
// Update error context
let mut error_context_guard =
error_context.write().expect("Failed to acquire write lock");
*error_context_guard = Some(
ErrorContext::new(format!(
"Error pushing data from {} to {}: {}",
from_stage.name(),
to_stage.name(),
err
))
.with_location(ErrorLocation::new(file!(), line!())),
);
// Check if we should give up
if consecutiveerrors >= error_threshold {
let mut current_state =
state.write().expect("Failed to acquire write lock");
*current_state = StreamState::Error;
break;
}
// Sleep before retrying
thread::sleep(Duration::from_millis(100));
}
}
Err(err) => {
match err {
CoreError::EndOfStream(_) => {
// End of stream, exit gracefully
break;
}
CoreError::TimeoutError(_) => {
// Timeout, continue
continue;
}
_ => {
// Other error, increment counter
consecutiveerrors += 1;
// Update error context
let mut error_context_guard =
error_context.write().expect("Failed to acquire write lock");
*error_context_guard = Some(
ErrorContext::new(format!(
"Error popping data from {}: {}",
from_stage.name(),
err
))
.with_location(ErrorLocation::new(file!(), line!())),
);
// Check if we should give up
if consecutiveerrors >= error_threshold {
let mut current_state =
state.write().expect("Failed to acquire write lock");
*current_state = StreamState::Error;
break;
}
// Sleep before retrying
thread::sleep(Duration::from_millis(100));
}
}
}
}
}
}
/// Stop the pipeline
pub fn stop(&mut self) -> Result<(), CoreError> {
let mut state = self.state.write().expect("Failed to acquire write lock");
// Check if the pipeline is running
if *state != StreamState::Running {
return Err(CoreError::StreamError(
ErrorContext::new("Pipeline not running".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
// Update state
*state = StreamState::Paused;
// Stop all stages
for stage in &self.stages {
stage.stop()?;
}
// Wait for worker threads to finish
for worker in self.workers.drain(..) {
match worker.join() {
Ok(_) => {}
Err(_) => {
return Err(CoreError::StreamError(
ErrorContext::new("Error joining worker thread".to_string())
.with_location(ErrorLocation::new(file!(), line!())),
));
}
}
}
Ok(())
}
/// Get the current state of the pipeline
pub fn state(&self) -> StreamState {
*self.state.read().expect("Failed to acquire read lock")
}
/// Get the statistics for the pipeline
pub fn stats(&self) -> PipelineStats {
let mut stats = PipelineStats::default();
// Collect stats from all stages
for stage in &self.stages {
let stage_stats = stage.stats();
let stage_name = stage.name().to_string();
stats
.stage_stats
.insert(stage_name.clone(), stage_stats.clone());
// Update bottleneck stats
let stage_throughput = stage_stats.avg_throughput;
if stage_throughput > 0.0 && stage_throughput < stats.bottleneck_throughput {
stats.bottleneck_throughput = stage_throughput;
stats.bottleneck_stage = Some(stage_name);
}
// Update total items processed (use the final stage's count)
if !self
.connections
.iter()
.any(|(_, to)| *to == self.stages.len() - 1)
{
stats.total_items = stage_stats.processed_items;
}
}
// Calculate overall statistics
let mut max_uptime = 0.0;
for stage_stats in stats.stage_stats.values() {
if stage_stats.uptime_seconds > max_uptime {
max_uptime = stage_stats.uptime_seconds;
}
}
stats.uptime_seconds = max_uptime;
if max_uptime > 0.0 {
stats.overall_throughput = stats.total_items as f64 / max_uptime;
}
stats
}
/// Get the last error from the pipeline
pub fn lasterror(&self) -> Option<ErrorContext> {
self.error_context
.read()
.expect("Failed to acquire read lock")
.clone()
}
/// Check if the pipeline is empty
pub fn is_empty(&self) -> bool {
self.stages.iter().all(|stage| stage.is_empty())
}
/// Get a stage by index
pub fn stage(&self, index: usize) -> Option<&dyn AnyStage> {
self.stages.get(index).map(|s| s.as_ref())
}
/// Get the number of stages in the pipeline
pub fn num_stages(&self) -> usize {
self.stages.len()
}
}
impl Drop for Pipeline {
fn drop(&mut self) {
// Stop the pipeline if it's running
if *self.state.read().expect("Failed to acquire read lock") == StreamState::Running {
let _ = self.stop();
}
}
}
/// Wrapper for pipeline stages to implement AnyStage
struct StageWrapper<I: Clone + Send + 'static, O: Clone + Send + 'static> {
stage: PipelineStage<I, O>,
}
impl<I: Clone + Send + 'static, O: Clone + Send + 'static> StageWrapper<I, O> {
/// Create a new stage wrapper
fn new(stage: PipelineStage<I, O>) -> Self {
Self { stage }
}
}
impl<I: Clone + Send + 'static, O: Clone + Send + 'static> AnyStage for StageWrapper<I, O> {
fn name(&self) -> &str {
&self.stage.name
}
fn start(&self) -> Result<(), CoreError> {
self.stage.start()
}
fn stop(&self) -> Result<(), CoreError> {
self.stage.stop()
}
fn state(&self) -> StreamState {
self.stage.state()
}
fn stats(&self) -> StreamStats {
self.stage.stats()
}
fn clone_box_impl(&self) -> Box<dyn AnyStage> {
Box::new(self.clone())
}
fn is_empty(&self) -> bool {
self.stage
.processor
.lock()
.expect("Failed to acquire lock")
.is_empty()
}
fn push_raw(&self, data: Box<dyn std::any::Any + Send>) -> Result<(), CoreError> {
let input = match data.downcast::<Vec<I>>() {
Ok(input) => *input,
Err(data) => {
// Try to downcast to a single item
match data.downcast::<I>() {
Ok(item) => vec![*item],
Err(_) => {
return Err(CoreError::StreamError(
ErrorContext::new(format!(
"Type mismatch when pushing data to stage {}",
self.name()
))
.with_location(ErrorLocation::new(file!(), line!())),
));
}
}
}
};
self.stage
.processor
.lock()
.expect("Failed to acquire lock")
.push_batch(input)
}
fn pop_raw(&self) -> Result<Box<dyn std::any::Any + Send>, CoreError> {
let output = self
.stage
.processor
.lock()
.expect("Failed to acquire lock")
.pop_batch(100)?;
Ok(Box::new(output))
}
}
impl dyn AnyStage {
/// Clone the stage into a new Box
fn clone_box(&self) -> Box<dyn AnyStage> {
self.clone_box_impl()
}
}
impl<I: Clone + Send + 'static, O: Clone + Send + 'static> Clone for StageWrapper<I, O> {
fn clone(&self) -> Self {
let stage = PipelineStage {
name: self.stage.name.clone(),
processor: self.stage.processor(),
parallel: self.stage.parallel,
parallelism: self.stage.parallelism,
};
Self { stage }
}
}
/// Extensions to the StreamProcessor to enable ndarray processing
impl<A, D> StreamProcessor<ArrayBase<OwnedRepr<A>, D>, ArrayBase<OwnedRepr<A>, D>>
where
A: Clone + Send + Default + 'static,
D: Dimension + Clone + Send + 'static + RemoveAxis,
{
/// Create a new array stream processor
pub fn newarray<F>(config: StreamConfig, processfn: F) -> Self
where
F: Fn(
Vec<ArrayBase<OwnedRepr<A>, D>>,
) -> Result<Vec<ArrayBase<OwnedRepr<A>, D>>, CoreError>
+ Send
+ Sync
+ 'static,
{
Self::new(config, processfn)
}
/// Process arrays chunk-wise
pub fn chunk_wise<F>(config: StreamConfig, chunk_size: usize, processfn: F) -> Self
where
F: Fn(&ArrayBase<OwnedRepr<A>, D>) -> Result<ArrayBase<OwnedRepr<A>, D>, CoreError>
+ Send
+ Sync
+ Clone
+ 'static,
{
let chunking_strategy = ChunkingStrategy::Fixed(chunk_size);
let processfn_clone = processfn.clone();
let chunks_fn = move |arrays: Vec<ArrayBase<OwnedRepr<A>, D>>| -> Result<Vec<ArrayBase<OwnedRepr<A>, D>>, CoreError> {
let mut results = Vec::with_capacity(arrays.len());
for array in arrays {
// Create chunked array
let chunked = ChunkedArray::new(array, chunking_strategy);
// Process each chunk and combine results
let mut chunk_results = Vec::new();
for chunk in chunked.get_chunks() {
let result = processfn_clone(&chunk)?;
chunk_results.push(result);
}
// Combine chunk results by concatenating along the flattened dimension
let combined = if !chunk_results.is_empty() {
// For 1D arrays, concatenate directly
if let Ok(combined_1d) = crate::ndarray::concatenate(
crate::ndarray::Axis(0),
&chunk_results
.iter()
.map(|arr| arr.view())
.collect::<Vec<_>>()
) {
// Try to reshape back to original dimensions
if let Ok(reshaped) = combined_1d.into_dimensionality::<D>() {
reshaped
} else {
// If reshaping fails, use the first chunk as fallback
chunk_results.into_iter().next().expect("Expected at least one result")
}
} else {
// For multi-dimensional arrays, more complex logic would be needed
// For now, concatenate and reshape approach
chunk_results.into_iter().next().expect("Expected at least one result")
}
} else {
return Err(CoreError::ValueError(ErrorContext::new(
"No chunks to process".to_string(),
)));
};
results.push(combined);
}
Ok(results)
};
Self::new(config, chunks_fn)
}
/// Process arrays in parallel
#[cfg(feature = "parallel")]
pub fn parallel<F>(config: StreamConfig, processfn: F) -> Self
where
F: Fn(&ArrayBase<OwnedRepr<A>, D>) -> Result<ArrayBase<OwnedRepr<A>, D>, CoreError>
+ Send
+ Sync
+ Clone
+ 'static,
A: Send + Sync,
{
let workers = config.workers.unwrap_or_else(num_cpus::get);
let parallel_fn = move |arrays: Vec<ArrayBase<OwnedRepr<A>, D>>| -> Result<Vec<ArrayBase<OwnedRepr<A>, D>>, CoreError> {
// Process arrays in parallel using rayon
use crate::parallel_ops::*;
let pool = ThreadPoolBuilder::new()
.num_threads(workers)
.build()
.map_err(|e| CoreError::StreamError(
ErrorContext::new(format!("{e}"))
.with_location(ErrorLocation::new(file!(), line!()))
))?;
let processfn_clone = processfn.clone();
pool.install(|| {
let results: Result<Vec<_>, _> = arrays
.par_iter()
.map(|array| processfn_clone(array))
.collect();
results
})
};
Self::new(config, parallel_fn)
}
}
/// Create a new stream processor with default configuration
#[allow(dead_code)]
pub fn create_stream_processor<T, U, F>(processfn: F) -> StreamProcessor<T, U>
where
T: Clone + Send + 'static,
U: Clone + Send + 'static,
F: Fn(Vec<T>) -> Result<Vec<U>, CoreError> + Send + Sync + 'static,
{
StreamProcessor::new(StreamConfig::default(), processfn)
}
/// Create a new pipeline
#[allow(dead_code)]
pub fn create_pipeline(name: &str) -> PipelineBuilder {
PipelineBuilder::new(name.to_string())
}
/// Extension trait for error handling in stream processing
pub trait StreamError {
/// Convert to a stream error
#[allow(dead_code)]
fn to_streamerror(self, message: &str) -> CoreError;
}
impl<T> StreamError for std::result::Result<T, CoreError> {
fn to_streamerror(self, message: &str) -> CoreError {
match self {
Ok(_) => CoreError::StreamError(
ErrorContext::new(message.to_string())
.with_location(ErrorLocation::new(file!(), line!())),
),
Err(e) => CoreError::StreamError(
ErrorContext::new(format!("{message}, {e}"))
.with_location(ErrorLocation::new(file!(), line!())),
),
}
}
}
/// Extension to CoreError for stream errors
impl CoreError {
/// Create a new end of stream error
pub fn message(message: &str) -> Self {
CoreError::EndOfStream(
ErrorContext::new(message.to_string())
.with_location(ErrorLocation::new(file!(), line!())),
)
}
/// Create a new stream error
pub fn message_2(message: &str) -> Self {
CoreError::StreamError(
ErrorContext::new(message.to_string())
.with_location(ErrorLocation::new(file!(), line!())),
)
}
/// Create a new timeout error
pub fn message_3(message: &str) -> Self {
CoreError::TimeoutError(
ErrorContext::new(message.to_string())
.with_location(ErrorLocation::new(file!(), line!())),
)
}
}