lix 0.17.1

Embeddable version control for apps and AI agents.
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

use std::hint::black_box;
use std::time::{Duration, Instant};

use super::{ObserveQueryEvaluation, ObserveQueryState, ObserveSharedContent};
use crate::{ExecuteResult, Value};

fn sized_blob_result(size: usize, changed: bool) -> ExecuteResult {
    let mut bytes = vec![0; size];
    if changed {
        bytes[size / 2] = 1;
    }
    ExecuteResult::from_rows(
        vec!["data".to_string()],
        vec![vec![Value::Blob(bytes.into())]],
    )
}

#[test]
#[ignore = "manual performance probe"]
fn observe_result_equivalence_baseline_performance_probe() {
    run_observe_result_equivalence_performance_probe(false);
}

#[test]
#[ignore = "manual performance probe"]
fn observe_result_equivalence_shared_performance_probe() {
    run_observe_result_equivalence_performance_probe(true);
}

fn run_observe_result_equivalence_performance_probe(share_comparison: bool) {
    let runtime = tokio::runtime::Builder::new_current_thread()
        .enable_all()
        .build()
        .expect("build performance probe runtime");
    let mode = if share_comparison {
        "candidate"
    } else {
        "baseline"
    };
    println!("size_mib\tobservers\tworkload\tmode\tmedian_ms\tp10_ms\tp90_ms\titerations");
    for size_mib in [1_usize, 10] {
        for observers in [1_usize, 2, 4, 16] {
            for changed in [false, true] {
                let state = ObserveQueryState::new();
                let initial = runtime
                    .block_on(
                        state.evaluate(1, share_comparison && observers > 1, || async {
                            Ok(sized_blob_result(size_mib * 1024 * 1024, false))
                        }),
                    )
                    .expect("seed performance probe result");
                let mut last_rows = vec![initial.rows; observers];
                let mut last_content = vec![initial.shared_content; observers];
                let mut generation = 1_u64;

                for _ in 0..3 {
                    run_probe_iteration(
                        &runtime,
                        &state,
                        &mut generation,
                        size_mib * 1024 * 1024,
                        changed,
                        share_comparison,
                        &mut last_rows,
                        &mut last_content,
                    );
                }
                let calibration_start = Instant::now();
                for _ in 0..3 {
                    run_probe_iteration(
                        &runtime,
                        &state,
                        &mut generation,
                        size_mib * 1024 * 1024,
                        changed,
                        share_comparison,
                        &mut last_rows,
                        &mut last_content,
                    );
                }
                let per_iteration = calibration_start.elapsed() / 3;
                let iterations = u32::try_from(
                    (Duration::from_millis(30).as_nanos() / per_iteration.as_nanos().max(1))
                        .clamp(2, 100),
                )
                .expect("bounded performance probe iteration count should fit u32");

                let mut samples = Vec::with_capacity(21);
                for _ in 0..21 {
                    let start = Instant::now();
                    for _ in 0..iterations {
                        run_probe_iteration(
                            &runtime,
                            &state,
                            &mut generation,
                            size_mib * 1024 * 1024,
                            changed,
                            share_comparison,
                            &mut last_rows,
                            &mut last_content,
                        );
                    }
                    samples.push(start.elapsed().as_secs_f64() * 1000.0 / f64::from(iterations));
                }
                samples.sort_by(f64::total_cmp);
                println!(
                    "{size_mib}\t{observers}\t{}\t{mode}\t{:.3}\t{:.3}\t{:.3}\t{iterations}",
                    if changed { "changed" } else { "unchanged" },
                    samples[10],
                    samples[2],
                    samples[18]
                );
            }
        }
    }
}

fn run_probe_iteration(
    runtime: &tokio::runtime::Runtime,
    state: &ObserveQueryState,
    generation: &mut u64,
    size: usize,
    changed: bool,
    share_comparison: bool,
    last_rows: &mut [ExecuteResult],
    last_content: &mut [Option<ObserveSharedContent>],
) {
    *generation += 1;
    let variant = changed && (*generation).is_multiple_of(2);
    let evaluation = runtime
        .block_on(state.evaluate(
            *generation,
            share_comparison && last_rows.len() > 1,
            || async move { Ok(sized_blob_result(size, variant)) },
        ))
        .expect("performance probe evaluation");
    for (previous_rows, previous_content) in last_rows.iter_mut().zip(last_content.iter_mut()) {
        let rows_changed = observed_rows_changed(previous_rows, *previous_content, &evaluation);
        *previous_content = evaluation.shared_content;
        if rows_changed {
            *previous_rows = evaluation.rows.clone();
        }
        black_box(rows_changed);
    }
}

fn observed_rows_changed(
    previous_rows: &ExecuteResult,
    previous_content: Option<ObserveSharedContent>,
    evaluation: &ObserveQueryEvaluation,
) -> bool {
    evaluation.rows_changed_since(Some(previous_rows), previous_content)
}