tephra 0.3.2

A DCB-compliant, immutable event store with global ordering.
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
//! End-to-end tests of the write coordinator through its public API: concurrency,
//! conflict resolution, shutdown, backpressure, and durability across reopen.

use std::collections::HashSet;
use std::sync::{Arc, Barrier};
use std::thread;

use smallvec::SmallVec;
use tempfile::TempDir;

use tephra::Position;
use tephra::event::{Event, EventType, Tag, Tags};
use tephra::log::set::{SegmentConfig, SegmentSet};
use tephra::query::{AppendCondition, Query, QueryItem};
use tephra::writer::{AppendError, ConflictSite, WriteCoordinator, WriteHandle, WriterConfig};

const SEG_SIZE: usize = 1 << 20;

fn config() -> WriterConfig {
    WriterConfig {
        queue_capacity: 256,
        max_batch_records: 256,
        max_batch_bytes: 256 * 1024,
        tips_window: 1_000_000,
        verify_tips: true,
        ..WriterConfig::default()
    }
}

fn open(dir: &TempDir) -> SegmentSet {
    SegmentSet::open(dir.path(), SegmentConfig::new(SEG_SIZE)).unwrap()
}

fn tags(items: &[&str]) -> Tags {
    Tags::new(
        items
            .iter()
            .map(|s| Tag::new(*s).unwrap())
            .collect::<SmallVec<[Tag; 4]>>(),
    )
    .unwrap()
}

fn event(ty: &str, tag_strs: &[&str]) -> Event {
    Event::new(&EventType::new(ty).unwrap(), &tags(tag_strs), b"data").unwrap()
}

/// Reads a query through the handle and collects just the matching positions.
fn read_positions(handle: &WriteHandle, query: &Query, after: Position) -> Vec<Position> {
    let mut reads = handle.read(query, after, None);
    let mut out = Vec::new();
    while let Some(item) = reads.next() {
        out.push(item.expect("read failed").position);
    }
    out
}

fn unique_guard(tag: &str) -> AppendCondition {
    AppendCondition::new(Query::item(QueryItem::with_tags(tags(&[tag]))))
}

#[test]
fn concurrent_appends_are_dense_and_unique() {
    let dir = TempDir::new().unwrap();
    let (coord, handle) = WriteCoordinator::start(open(&dir), config()).unwrap();

    const THREADS: u64 = 8;
    const PER_THREAD: u64 = 250;

    let mut joins = Vec::new();
    for t in 0..THREADS {
        let h: WriteHandle = handle.clone();
        joins.push(thread::spawn(move || {
            let mut positions = Vec::new();
            for i in 0..PER_THREAD {
                let ev = event("Appended", &[&format!("thread:{t}"), &format!("seq:{i}")]);
                let range = h.append(vec![ev], None).unwrap();
                assert_eq!(range.first, range.last);
                positions.push(range.first.get());
            }
            positions
        }));
    }

    let mut all: Vec<u64> = joins.into_iter().flat_map(|j| j.join().unwrap()).collect();
    all.sort_unstable();

    let total = THREADS * PER_THREAD;
    // Dense 1..=total, no gaps, no duplicates.
    assert_eq!(all.len() as u64, total);
    assert_eq!(all.first().copied(), Some(1));
    assert_eq!(all.last().copied(), Some(total));
    assert_eq!(all.iter().collect::<HashSet<_>>().len() as u64, total);
    for (i, p) in all.iter().enumerate() {
        assert_eq!(*p, i as u64 + 1);
    }

    drop(handle);
    let set = coord.shutdown();
    assert_eq!(set.last_position(), Position::new(total));
}

#[test]
fn overlapping_uniqueness_guard_lets_exactly_one_win() {
    // Many threads race to append the same unique tag under a uniqueness guard. Whether
    // they land in the same drain (one SameBatch conflict) or different drains (one
    // Durable conflict), exactly one may win.
    let dir = TempDir::new().unwrap();
    let (coord, handle) = WriteCoordinator::start(open(&dir), config()).unwrap();

    const THREADS: usize = 16;
    let barrier = Arc::new(Barrier::new(THREADS));
    let mut joins = Vec::new();
    for _ in 0..THREADS {
        let h = handle.clone();
        let b = barrier.clone();
        joins.push(thread::spawn(move || {
            b.wait();
            h.append(
                vec![event("Reserved", &["unique:x"])],
                Some(unique_guard("unique:x")),
            )
        }));
    }

    let mut wins = 0;
    let mut conflicts = 0;
    for j in joins {
        match j.join().unwrap() {
            Ok(_) => wins += 1,
            Err(AppendError::Conflict { .. }) => conflicts += 1,
            Err(other) => panic!("unexpected error: {other:?}"),
        }
    }
    assert_eq!(wins, 1, "exactly one writer should win");
    assert_eq!(conflicts, THREADS - 1);

    drop(handle);
    let set = coord.shutdown();
    assert_eq!(set.last_position(), Position::new(1));
}

#[test]
fn explicit_shutdown_returns_a_consistent_set() {
    let dir = TempDir::new().unwrap();
    let (coord, handle) = WriteCoordinator::start(open(&dir), config()).unwrap();

    for i in 0..10 {
        handle
            .append(vec![event("E", &[&format!("k:{i}")])], None)
            .unwrap();
    }
    drop(handle);
    let set = coord.shutdown();
    assert_eq!(set.last_position(), Position::new(10));
    drop(set);

    // Reopening the directory sees exactly the acknowledged writes.
    let reopened = open(&dir);
    assert_eq!(reopened.last_position(), Position::new(10));
}

#[test]
fn drop_based_shutdown_persists_acknowledged_writes() {
    let dir = TempDir::new().unwrap();
    {
        let (coord, handle) = WriteCoordinator::start(open(&dir), config()).unwrap();
        for i in 0..5 {
            handle
                .append(vec![event("E", &[&format!("k:{i}")])], None)
                .unwrap();
        }
        // Drop both: the coordinator's Drop signals shutdown and joins the thread.
        drop(handle);
        drop(coord);
    }
    let reopened = open(&dir);
    assert_eq!(reopened.last_position(), Position::new(5));
}

#[test]
fn tiny_queue_still_serves_all_writes() {
    // queue_capacity = 1 forces the bounded channel to block and unblock repeatedly;
    // every write must still complete, with dense positions.
    let dir = TempDir::new().unwrap();
    let cfg = WriterConfig {
        queue_capacity: 1,
        ..config()
    };
    let (coord, handle) = WriteCoordinator::start(open(&dir), cfg).unwrap();

    let mut joins = Vec::new();
    for t in 0..4 {
        let h = handle.clone();
        joins.push(thread::spawn(move || {
            for i in 0..100 {
                h.append(
                    vec![event("E", &[&format!("t:{t}"), &format!("i:{i}")])],
                    None,
                )
                .unwrap();
            }
        }));
    }
    for j in joins {
        j.join().unwrap();
    }

    drop(handle);
    let set = coord.shutdown();
    assert_eq!(set.last_position(), Position::new(400));
}

#[test]
fn index_search_sees_own_writes_across_rollovers() {
    // Read-your-writes through the index: a query issued right after an append must see
    // that append, including the batch that triggered a rollover (visible in the freshly
    // sealed segment) and everything before it. Tiny segments force many rollovers mid-run.
    let dir = TempDir::new().unwrap();
    let set = SegmentSet::open(dir.path(), SegmentConfig::new(512)).unwrap();
    let cfg = WriterConfig {
        queue_capacity: 64,
        max_batch_records: 64,
        max_batch_bytes: 256,
        tips_window: 1_000_000,
        verify_tips: true,
        ..WriterConfig::default()
    };
    let (coord, handle) = WriteCoordinator::start(set, cfg).unwrap();

    // Every event carries course:c1, so after the i-th append the query must return
    // exactly positions 1..=i.
    let mut expected = Vec::new();
    for i in 0..80u64 {
        let ty = if i % 2 == 0 { "Enrolled" } else { "Renamed" };
        let range = handle
            .append(vec![event(ty, &["course:c1"])], None)
            .unwrap();
        expected.push(range.first);

        let got = read_positions(
            &handle,
            &Query::item(QueryItem::with_tags(tags(&["course:c1"]))),
            Position::ZERO,
        );
        assert_eq!(got, expected, "read-your-writes after append {i}");
    }

    // A type filter and an `after` bound also compose across the sealed segments.
    let enrolled = read_positions(
        &handle,
        &Query::item(QueryItem::of_types(vec![
            EventType::new("Enrolled").unwrap(),
        ])),
        Position::new(40),
    );
    assert!(enrolled.iter().all(|p| p.get() > 40));
    assert!(!enrolled.is_empty());

    drop(handle);
    coord.shutdown();
}

#[test]
fn durable_conflict_detected_through_the_index_across_sealed_segments() {
    // With verify off, the durable arm runs the index existence check alone (no scan
    // cross-check), so this exercises the real production path. Tiny segments
    // force seals, so the guarded tag lives in an early *sealed* segment: the check must
    // find it there, proving the index path works across sealed segments, not just the
    // active tail. This is the `after`-omitted uniqueness guard, the shape the index most improves.
    let dir = TempDir::new().unwrap();
    let set = SegmentSet::open(dir.path(), SegmentConfig::new(512)).unwrap();
    let cfg = WriterConfig {
        queue_capacity: 64,
        max_batch_records: 64,
        max_batch_bytes: 256,
        verify_tips: false,
        ..WriterConfig::default()
    };
    let (coord, handle) = WriteCoordinator::start(set, cfg).unwrap();

    // The guarded tag lands first, then enough filler to seal several later segments.
    handle
        .append(vec![event("Reserved", &["unique:early"])], None)
        .unwrap();
    for i in 0..80u64 {
        handle
            .append(vec![event("Filler", &[&format!("f:{i}")])], None)
            .unwrap();
    }

    // A uniqueness guard (after: 0) on the early tag must see the durable event and conflict.
    let err = handle
        .append(
            vec![event("Reserved", &["unique:early"])],
            Some(unique_guard("unique:early")),
        )
        .unwrap_err();
    assert!(
        matches!(
            err,
            AppendError::Conflict {
                at: ConflictSite::Durable(_)
            }
        ),
        "expected a durable conflict from a sealed segment, got {err:?}"
    );

    // A guard on a tag no event carries succeeds.
    handle
        .append(
            vec![event("Reserved", &["unique:fresh"])],
            Some(unique_guard("unique:fresh")),
        )
        .unwrap();

    drop(handle);
    let set = coord.shutdown();
    assert!(
        set.sealed_len() >= 1,
        "tiny segments should have sealed, so unique:early was in a sealed segment"
    );
}

#[test]
fn append_with_no_events_is_rejected() {
    let dir = TempDir::new().unwrap();
    let (coord, handle) = WriteCoordinator::start(open(&dir), config()).unwrap();
    assert!(matches!(
        handle.append(vec![], None),
        Err(AppendError::Empty)
    ));
    drop(handle);
    coord.shutdown();
}

#[cfg(feature = "async")]
use std::future::Future;

/// Drives a future to completion on the calling thread with a park/unpark waker. The
/// worker replies from its own thread, so its `wake` unparks us. Keeps the async tests
/// free of an executor dependency.
#[cfg(feature = "async")]
fn block_on<F: Future>(fut: F) -> F::Output {
    use std::pin::pin;
    use std::sync::Arc;
    use std::task::{Context, Poll, Wake, Waker};
    use std::thread::{self, Thread};

    struct ThreadWaker(Thread);
    impl Wake for ThreadWaker {
        fn wake(self: Arc<Self>) {
            self.0.unpark();
        }
        fn wake_by_ref(self: &Arc<Self>) {
            self.0.unpark();
        }
    }

    let waker = Waker::from(Arc::new(ThreadWaker(thread::current())));
    let mut cx = Context::from_waker(&waker);
    let mut fut = pin!(fut);
    loop {
        match fut.as_mut().poll(&mut cx) {
            Poll::Ready(out) => return out,
            Poll::Pending => thread::park(),
        }
    }
}

#[cfg(feature = "async")]
#[test]
fn append_async_assigns_dense_positions_and_read_sees_own_writes() {
    let dir = TempDir::new().unwrap();
    let (coord, handle) = WriteCoordinator::start(open(&dir), config()).unwrap();

    block_on(async {
        // Every event carries course:c1, so after the i-th append the read must return
        // exactly positions 1..=i. `read` is synchronous (it runs on this thread over the
        // published snapshot); read-your-writes holds because `append_async` resolves only
        // after the writer has published the watermark.
        let mut expected = Vec::new();
        for i in 0..20u64 {
            let range = handle
                .append_async(vec![event("Enrolled", &["course:c1"])], None)
                .await
                .unwrap();
            assert_eq!(range.first, range.last, "one event, one position");
            assert_eq!(range.first, Position::new(i + 1), "dense positions");
            expected.push(range.first);

            let got = read_positions(
                &handle,
                &Query::item(QueryItem::with_tags(tags(&["course:c1"]))),
                Position::ZERO,
            );
            assert_eq!(got, expected, "read-your-writes after async append {i}");
        }
    });

    drop(handle);
    let set = coord.shutdown();
    assert_eq!(set.last_position(), Position::new(20));
}

#[cfg(feature = "async")]
#[test]
fn append_async_honors_the_uniqueness_guard() {
    let dir = TempDir::new().unwrap();
    let (coord, handle) = WriteCoordinator::start(open(&dir), config()).unwrap();

    block_on(async {
        // Empty events are rejected before ever touching the channel.
        assert!(matches!(
            handle.append_async(vec![], None).await,
            Err(AppendError::Empty)
        ));

        // First guarded write wins; the second, guarded on the same tag, sees the durable
        // conflict.
        handle
            .append_async(
                vec![event("Reserved", &["unique:x"])],
                Some(unique_guard("unique:x")),
            )
            .await
            .unwrap();
        assert!(matches!(
            handle
                .append_async(
                    vec![event("Reserved", &["unique:x"])],
                    Some(unique_guard("unique:x")),
                )
                .await,
            Err(AppendError::Conflict { .. })
        ));
    });

    drop(handle);
    let set = coord.shutdown();
    assert_eq!(set.last_position(), Position::new(1));
}