tidaldb/tidal/benches/search.rs

#![allow(clippy::unwrap_used)]
//! Criterion benchmarks for the SEARCH query pipeline.
//!
//! Measures end-to-end `db.search()` latency at 10K items to validate the
//! < 50ms target specified in the m5p3 phase acceptance criteria.

use std::{collections::HashMap, time::Duration};

use criterion::{Criterion, black_box, criterion_group, criterion_main};
use tidaldb::{
    TidalDb,
    query::search::Search,
    schema::{DecaySpec, EntityId, EntityKind, SchemaBuilder, TextFieldType, Timestamp, Window},
};

fn search_schema() -> tidaldb::schema::Schema {
    let mut builder = SchemaBuilder::new();
    let _ = builder
        .signal(
            "view",
            EntityKind::Item,
            DecaySpec::Exponential {
                half_life: Duration::from_secs(7 * 24 * 3600),
            },
        )
        .windows(&[Window::TwentyFourHours])
        .velocity(false)
        .add();
    let _ = builder
        .signal(
            "like",
            EntityKind::Item,
            DecaySpec::Exponential {
                half_life: Duration::from_secs(30 * 24 * 3600),
            },
        )
        .windows(&[Window::TwentyFourHours])
        .velocity(false)
        .add();
    builder.text_field("title", TextFieldType::Text);
    builder.text_field("description", TextFieldType::Text);
    builder.text_field("category", TextFieldType::Keyword);
    builder.build().unwrap()
}

/// Build a `TidalDb` with N items indexed for text search.
fn make_db(n: u64) -> TidalDb {
    let db = TidalDb::builder()
        .ephemeral()
        .with_schema(search_schema())
        .open()
        .unwrap();

    let ts = Timestamp::now();
    for i in 0..n {
        let mut meta = HashMap::new();
        // Vary titles to produce realistic IDF scores.
        meta.insert(
            "title".to_string(),
            format!("Rust tutorial {i} async concurrency programming"),
        );
        meta.insert(
            "description".to_string(),
            "Learn Rust with practical examples and real projects.".to_string(),
        );
        let cat = if i % 3 == 0 {
            "programming"
        } else if i % 3 == 1 {
            "systems"
        } else {
            "web"
        };
        meta.insert("category".to_string(), cat.to_string());
        db.write_item_with_metadata(EntityId::new(i), &meta)
            .unwrap();

        // Add some signals to make profile scoring non-trivial.
        if i % 5 == 0 {
            db.signal("view", EntityId::new(i), 1.0, ts).unwrap();
        }
    }

    // Wait for the background text syncer to commit all pending documents
    // (syncer commits every 1_000 items or 2s; 10K items = 10 batch commits).
    // Then reload the reader so the searcher sees all committed documents.
    std::thread::sleep(std::time::Duration::from_millis(500));
    db.reload_text_index().unwrap();
    db
}

/// Benchmark: `db.search()` with a text-only query at 10K items.
///
/// Target: < 50ms.
fn bench_search_text_10k(c: &mut Criterion) {
    let db = make_db(10_000);
    let query = Search::builder()
        .query("Rust async")
        .limit(20)
        .build()
        .unwrap();

    c.bench_function("search_text_10k", |b| {
        b.iter(|| db.search(black_box(&query)).unwrap());
    });
}

/// Benchmark: `db.search()` with a keyword-scoped query at 10K items.
///
/// Target: < 50ms.
fn bench_search_keyword_10k(c: &mut Criterion) {
    let db = make_db(10_000);
    let query = Search::builder()
        .query("category:programming")
        .limit(20)
        .build()
        .unwrap();

    c.bench_function("search_keyword_10k", |b| {
        b.iter(|| db.search(black_box(&query)).unwrap());
    });
}

criterion_group!(
    search_benches,
    bench_search_text_10k,
    bench_search_keyword_10k
);
criterion_main!(search_benches);