Aura/crates/aura-engine/tests/random_sweep_e2e.rs

//! End-to-end coverage for the random param-sweep axis (C12.1):
//! `RandomSpace` + `ParamRange` driven through the **public** `sweep` surface a
//! downstream researcher actually writes (the worked author example).
//!
//! The in-module unit tests in `sweep.rs` reach into crate internals
//! (`bootstrap_with_cells`, `sweep_with_threads`, `SplitMix64`); these tests use
//! only the exported API, so they pin the properties a real consumer observes:
//! a `SweepFamily` of `RunReport`s, the typed `SweepError` gate, and the
//! `named_params` view. The blueprint is reconstructed here (the crate-private
//! `test_fixtures` harness is unreachable from an integration test) through the
//! public `Composite` builder + `aura-std` nodes, so the test exercises the same
//! published surface the worked example does.

use std::sync::mpsc;

use aura_core::{Cell, Firing, ScalarKind, Timestamp};
use aura_engine::{
    f64_field, summarize, sweep, BlueprintNode, Composite, Edge, OutField, ParamRange, ParamSpec,
    RandomSpace, Role, RunManifest, RunReport, Scalar, Space, SweepError, Target, VecSource,
};
use aura_std::{Bias, Recorder, SimBroker, Sma, Sub};

/// Seven synthetic F64 ticks (mirrors the crate-private `synthetic_prices`):
/// short enough that small SMA windows warm up, so every run yields finite,
/// non-degenerate metrics. Deterministic input fixture.
fn synthetic_prices() -> Vec<(Timestamp, Scalar)> {
    [
        (1_i64, 1.0000_f64),
        (2, 1.0010),
        (3, 1.0030),
        (4, 1.0060),
        (5, 1.0040),
        (6, 1.0010),
        (7, 0.9990),
    ]
    .iter()
    .map(|&(t, p)| (Timestamp(t), Scalar::f64(p)))
    .collect()
}

/// The SMA-cross signal-quality harness built through the PUBLIC builder API:
/// `[fast: I64, slow: I64, scale: F64]` param-space, ending in an equity sink and
/// an exposure sink. Returns the blueprint plus its two recording receivers (a
/// fresh channel pair per build, so each swept point runs disjointly — C1).
#[allow(clippy::type_complexity)]
fn sma_cross_harness() -> (
    Composite,
    mpsc::Receiver<(Timestamp, Vec<Scalar>)>,
    mpsc::Receiver<(Timestamp, Vec<Scalar>)>,
) {
    let (tx_eq, rx_eq) = mpsc::channel();
    let (tx_ex, rx_ex) = mpsc::channel();
    let sma_cross = Composite::new(
        "sma_cross",
        vec![
            Sma::builder().named("fast").into(),
            Sma::builder().named("slow").into(),
            Sub::builder().into(),
        ],
        vec![
            Edge { from: 0, to: 2, slot: 0, from_field: 0 },
            Edge { from: 1, to: 2, slot: 1, from_field: 0 },
        ],
        vec![Role {
            name: "price".into(),
            targets: vec![Target { node: 0, slot: 0 }, Target { node: 1, slot: 0 }],
            source: None,
        }],
        vec![OutField { node: 2, field: 0, name: "out".into() }],
    );
    let bp = Composite::new(
        "root",
        vec![
            BlueprintNode::Composite(sma_cross),
            Bias::builder().into(),
            SimBroker::builder(0.0001).into(),
            Recorder::builder(vec![ScalarKind::F64], Firing::Any, tx_eq).into(),
            Recorder::builder(vec![ScalarKind::F64], Firing::Any, tx_ex).into(),
        ],
        vec![
            Edge { from: 0, to: 1, slot: 0, from_field: 0 }, // composite out -> Bias
            Edge { from: 1, to: 2, slot: 0, from_field: 0 }, // exposure -> broker slot 0
            Edge { from: 2, to: 3, slot: 0, from_field: 0 }, // equity -> sink
            Edge { from: 1, to: 4, slot: 0, from_field: 0 }, // exposure -> sink
        ],
        vec![Role {
            name: "src".into(),
            targets: vec![
                Target { node: 0, slot: 0 }, // price -> sma_cross role 0
                Target { node: 2, slot: 1 }, // price -> SimBroker price slot
            ],
            source: Some(ScalarKind::F64),
        }],
        vec![], // root ends in sinks
    );
    (bp, rx_eq, rx_ex)
}

/// Build + bootstrap + run + summarize one swept point into a `RunReport`, using
/// only the public surface. A fresh harness (fresh sink channels) per point keeps
/// the runs disjoint (C1). The manifest is a fixed minimal fixture — only the
/// metrics carry the run, so determinism makes this reproduce a point exactly.
fn run_point(point: &[Cell]) -> RunReport {
    let (bp, rx_eq, rx_ex) = sma_cross_harness();
    let mut h = bp
        .bootstrap_with_cells(point)
        .expect("RandomSpace-drawn points are pre-validated against the param-space");
    h.run(vec![Box::new(VecSource::new(synthetic_prices()))]);
    let equity = f64_field(&rx_eq.try_iter().collect::<Vec<_>>(), 0);
    let exposure = f64_field(&rx_ex.try_iter().collect::<Vec<_>>(), 0);
    RunReport {
        manifest: RunManifest {
            commit: "random-sweep-e2e".to_string(),
            params: Vec::new(),
            defaults: Vec::new(),
            window: (Timestamp(0), Timestamp(0)),
            seed: 0,
            broker: "test".to_string(),
            selection: None,
            instrument: None,
            topology_hash: None,
            project: None,
        },
        metrics: summarize(&equity, &exposure),
    }
}

/// A `RandomSpace` over the harness's `[fast, slow, scale]` param-space: integer
/// windows drawn from the 7-tick fixture's proven domain (so SMAs warm up) and a
/// continuous scale. The integer ranges are exact single points so the family
/// stays small but every kind/arm is exercised.
fn sma_cross_random(count: usize, seed: u64) -> RandomSpace {
    let space = sma_cross_harness().0.param_space();
    RandomSpace::new(
        &space,
        vec![
            ParamRange::i64(2, 3),      // fast  in [2, 3]
            ParamRange::i64(4, 5),      // slow  in [4, 5]
            ParamRange::f64(0.25, 1.5), // scale in [0.25, 1.5)
        ],
        count,
        seed,
    )
    .expect("ranges match the sample param-space kinds")
}

/// Property: a `RandomSpace` sweep is fully seed-determined end-to-end — the same
/// `(ranges, count, seed)` driven through the public `sweep` produces a
/// bit-identical `SweepFamily` of `RunReport`s, run after run (C1). The whole
/// pipeline (seeded draw -> bootstrap -> run -> summarize) reproduces, observed at
/// the published JSON boundary, not the internal `points()`.
#[test]
fn random_sweep_is_reproducible_at_the_report_boundary() {
    let render = |seed: u64| -> Vec<String> {
        sweep(&sma_cross_random(8, seed), run_point)
            .points
            .iter()
            .map(|p| p.report.to_json())
            .collect()
    };
    let a = render(0xC0FFEE);
    let b = render(0xC0FFEE);
    assert_eq!(a.len(), 8, "count points were swept");
    assert_eq!(a, b, "same (ranges, count, seed) => bit-identical family (C1)");
}

/// Property: a different seed produces a different family — the sweep genuinely
/// samples the seed, it does not collapse to a constant set of points. Observed
/// via the public `named_params` coordinate view, never an internal field.
#[test]
fn random_sweep_seed_changes_the_family() {
    let coords = |seed: u64| -> Vec<Vec<Scalar>> {
        let family = sweep(&sma_cross_random(8, seed), run_point);
        (0..family.points.len())
            .map(|i| family.named_params(i).into_iter().map(|(_, v)| v).collect())
            .collect()
    };
    assert_ne!(
        coords(1),
        coords(2),
        "different seeds => different swept coordinate sets",
    );
}

/// Property: every coordinate the sweep actually ran on lies inside its declared
/// `ParamRange` — the I64 slots inclusive `[lo, hi]`, the F64 slot half-open
/// `[lo, hi)`. A regression that let a draw escape its range would silently run
/// the strategy out of its declared domain; this pins the bound at the observable
/// `named_params` view of the family that was run.
#[test]
fn swept_points_stay_inside_their_declared_ranges() {
    let family = sweep(&sma_cross_random(200, 0xABCDEF), run_point);
    assert_eq!(family.points.len(), 200);
    for i in 0..family.points.len() {
        let named = family.named_params(i);
        let fast = named[0].1.as_i64();
        let slow = named[1].1.as_i64();
        let scale = named[2].1.as_f64();
        assert!((2..=3).contains(&fast), "fast in [2,3] inclusive, got {fast}");
        assert!((4..=5).contains(&slow), "slow in [4,5] inclusive, got {slow}");
        assert!((0.25..1.5).contains(&scale), "scale in [0.25,1.5), got {scale}");
        // and the run that consumed this in-range point produced a finite metric
        assert!(family.points[i].report.metrics.total_pips.is_finite());
    }
}

/// Property: the typed validation gate rejects a non-numeric param slot BEFORE
/// any run. A `Bool` slot cannot carry a continuous range (it is degenerate), so
/// `RandomSpace::new` returns the public `SweepError::NonNumericRange` value — an
/// observable typed error at the published API, not a panic and not a swept run.
#[test]
fn bool_slot_is_rejected_as_non_numeric_before_any_run() {
    let space = vec![ParamSpec { name: "flag".into(), kind: ScalarKind::Bool }];
    let err = RandomSpace::new(&space, vec![ParamRange::i64(0, 1)], 10, 0)
        .expect_err("a Bool slot is not range-sampleable");
    assert_eq!(err, SweepError::NonNumericRange { slot: 0, kind: ScalarKind::Bool });
}

/// Property: a `count == 0` `RandomSpace` is a valid, explicit empty family (not
/// the "accidental collapse" an empty grid axis would be) — `sweep` over it
/// returns an empty `SweepFamily` while still carrying the param-space schema, so
/// a downstream `named_params` consumer sees a well-formed empty result.
#[test]
fn zero_count_sweep_is_a_well_formed_empty_family() {
    let space = sma_cross_harness().0.param_space();
    let rs = sma_cross_random(0, 0);
    assert!(rs.is_empty(), "count == 0 is the explicit empty space");
    let family = sweep(&rs, run_point);
    assert!(family.points.is_empty(), "no points swept");
    assert_eq!(family.space, space, "the empty family still carries the schema");
}

/// Property: `GridSpace` and `RandomSpace` are interchangeable through the `Space`
/// trait `sweep` is generic over — the same generic helper drives either
/// enumeration. This is the trait abstraction the cut introduced, observed via the
/// public `Space::param_specs`, kept from regressing back to a `GridSpace`-only
/// `sweep` signature.
#[test]
fn random_space_is_driven_through_the_space_trait() {
    fn schema_len<S: Space>(s: &S) -> usize {
        s.param_specs().len()
    }
    let rs = sma_cross_random(4, 7);
    assert_eq!(schema_len(&rs), 3, "the [fast, slow, scale] schema reaches the trait surface");
    // and the generic `sweep` accepts it by value of the same bound
    let family = sweep(&rs, run_point);
    assert_eq!(family.points.len(), 4);
}