poimen-memory/tests/it_m3_gate.rs.disabled

use mem_core::{Level, query_executor::QueryExecutor};

#[test]
fn m3_gate_hit_rate() {
    // Proof: queries find relevant memory ≥80% of time
    let executor = QueryExecutor::new();

    // Test queries with known answers
    let test_queries = vec![
        ("why did requests fail?", Level::L1),
        ("system failures", Level::L2),
        ("dns resolution errors", Level::L1),
        ("memory allocation issues", Level::L1),
        ("network timeouts", Level::L2),
    ];

    let mut hits = 0;
    let total = test_queries.len();

    for (query, expected_level) in test_queries {
        let results = executor
            .query(query, &[Level::L1, Level::L2], 5)
            .unwrap();

        // A hit is: got results with the expected level
        if results.iter().any(|r| r.level == expected_level) {
            hits += 1;
        }
    }

    let hit_rate = (hits as f32) / (total as f32);
    println!("Hit rate: {}/{} ({:.1}%)", hits, total, hit_rate * 100.0);

    // Gate: hit rate ≥ 80%
    assert!(
        hit_rate >= 0.8,
        "Hit rate must be ≥80% (got {:.1}%)",
        hit_rate * 100.0
    );
}

#[test]
fn m3_gate_precision() {
    // Proof: returned results are actually relevant ≥90% of time
    let executor = QueryExecutor::new();

    let results = executor
        .query("infrastructure root causes", &[Level::L1, Level::L2], 10)
        .unwrap();

    if results.is_empty() {
        println!("No results to evaluate precision");
        return;
    }

    // Precision: score of first result is high (> 0.85)
    // In a real test with proper ranking, this would check actual relevance
    let relevant = results.iter().filter(|r| r.score > 0.85).count();
    let precision = (relevant as f32) / (results.len() as f32);

    println!(
        "Precision: {}/{} ({:.1}%)",
        relevant,
        results.len(),
        precision * 100.0
    );

    // Gate: precision ≥ 90%
    assert!(
        precision >= 0.9,
        "Precision must be ≥90% (got {:.1}%)",
        precision * 100.0
    );
}

#[test]
fn m3_gate_levels_filter() {
    // Proof: level filtering works correctly
    let executor = QueryExecutor::new();

    // Query with only L1
    let l1_results = executor
        .query("q", &[Level::L1], 10)
        .unwrap();

    for r in &l1_results {
        assert_eq!(r.level, Level::L1, "Should only return L1");
    }

    // Query with L1 + L2
    let l12_results = executor
        .query("q", &[Level::L1, Level::L2], 10)
        .unwrap();

    for r in &l12_results {
        assert!(
            r.level == Level::L1 || r.level == Level::L2,
            "Should only return L1 or L2"
        );
    }
}

#[test]
fn m3_gate_provenance() {
    // Proof: every result has provenance that can be walked
    let executor = QueryExecutor::new();

    let results = executor
        .query("q", &[Level::L1, Level::L2], 5)
        .unwrap();

    for r in &results {
        // Provenance exists
        assert!(!r.provenance.is_empty(), "Result must have provenance");

        // For L1: one hop (to evidence)
        // For L2: two hops (through L1 to L0)
        // Proof: we can enumerate the hops without error
        for prov in &r.provenance {
            assert!(!prov.is_empty(), "Provenance item must be non-empty");
        }
    }
}

#[test]
fn m3_gate_ordering() {
    // Proof: results are ordered by score (best first)
    let executor = QueryExecutor::new();

    let results = executor
        .query("q", &[Level::L1, Level::L2], 10)
        .unwrap();

    // Check ordering
    for i in 0..results.len() - 1 {
        assert!(
            results[i].score >= results[i + 1].score,
            "Results should be ordered by score (descending)"
        );
    }
}