AILang/crates/ailang-core/tests/hash_pin.rs

//! Hash-pin integration tests, relocated from `ailang-core/src/hash.rs`
//! `#[cfg(test)] mod tests` to a `tests/*` integration crate in iter
//! form-a.1 Task 5. The relocation removes the production-source file's
//! ad-hoc `examples/<stem>.ail.json` reads in favour of loading the
//! `.ail` siblings via `ailang_surface::load_module`, mirroring the
//! Form-A authoring discipline of the form-a-default-authoring
//! milestone. Hash assertions are unchanged: post-form-a the canonical
//! JSON-AST is still the hashable representation, derived in-process
//! from `.ail` by the surface loader.
//!
//! The `examples/<stem>.ail.json` fixtures that backed the original
//! tests are deleted in Task 8; the `.ail` siblings rendered in Task 2
//! are the authored form going forward. The four schema-stability pins
//! (iter13a / iter19b / iter22b1 / ct4) plus the two structural pins
//! (hash_is_stable / hash_changes_with_content / classdef-empty-optionals
//! / forall-without-constraints) all keep their assertion shape.

use ailang_core::ast::*;
use ailang_core::canonical;
use ailang_core::hash::def_hash;

fn examples_dir() -> std::path::PathBuf {
    let manifest_dir = std::path::PathBuf::from(env!("CARGO_MANIFEST_DIR"));
    manifest_dir.parent().unwrap().parent().unwrap().join("examples")
}

fn sample_fn() -> Def {
    Def::Fn(FnDef {
        name: "add".into(),
        ty: Type::Fn {
            params: vec![Type::int(), Type::int()],
            ret: Box::new(Type::int()),
            effects: vec![],
            param_modes: vec![],
            ret_mode: ParamMode::Implicit,
        },
        params: vec!["a".into(), "b".into()],
        body: Term::App {
            callee: Box::new(Term::Var { name: "+".into() }),
            args: vec![
                Term::Var { name: "a".into() },
                Term::Var { name: "b".into() },
            ],
            tail: false,
        },
        suppress: vec![],
        doc: None,
    export: None,
    })
}

#[test]
fn hash_is_stable() {
    let h1 = def_hash(&sample_fn());
    let h2 = def_hash(&sample_fn());
    assert_eq!(h1, h2);
    assert_eq!(h1.len(), 16);
}

#[test]
fn hash_changes_with_content() {
    let mut def = sample_fn();
    let h1 = def_hash(&def);
    if let Def::Fn(ref mut f) = def {
        f.name = "mul".into();
    }
    let h2 = def_hash(&def);
    assert_ne!(h1, h2);
}

/// Iter 13a regression: adding `vars` to TypeDef and `args` to
/// `Type::Con` must NOT change canonical-JSON hashes of any pre-13a
/// definition. Recorded hashes were captured from on-disk modules
/// before the schema extension; if this fires, a
/// `skip_serializing_if` is missing or wrong.
#[test]
fn iter13a_schema_extension_preserves_pre_13a_hashes() {
    let examples = examples_dir();

    let sum_mod = ailang_surface::load_module(&examples.join("sum.ail"))
        .expect("examples/sum.ail loads");
    let sum_def = sum_mod.defs.iter().find(|d| d.name() == "sum").unwrap();
    assert_eq!(def_hash(sum_def), "db33f57cb329935e");

    let list_mod = ailang_surface::load_module(&examples.join("list.ail"))
        .expect("examples/list.ail loads");
    let int_list_def = list_mod.defs.iter().find(|d| d.name() == "IntList").unwrap();
    assert_eq!(def_hash(int_list_def), "b082192bd0c99202");
}

/// loop-recur iter 1 regression: adding `Term::Loop` / `Term::Recur`
/// (and `struct LoopBinder`) must NOT change canonical-JSON hashes
/// of any pre-loop-recur definition. The additive variant carries a
/// new `"t"` tag absent from every pre-existing fixture, so the
/// recorded hashes below must stay byte-identical. If this fires,
/// the extension is non-additive (a `skip_serializing_if` is missing
/// or an existing variant's shape drifted).
#[test]
fn loop_recur_schema_extension_preserves_pre_loop_recur_hashes() {
    let examples = examples_dir();

    let sum_mod = ailang_surface::load_module(&examples.join("sum.ail"))
        .expect("examples/sum.ail loads");
    let sum_def = sum_mod.defs.iter().find(|d| d.name() == "sum").unwrap();
    assert_eq!(def_hash(sum_def), "db33f57cb329935e");

    let list_mod = ailang_surface::load_module(&examples.join("list.ail"))
        .expect("examples/list.ail loads");
    let int_list_def = list_mod.defs.iter().find(|d| d.name() == "IntList").unwrap();
    assert_eq!(def_hash(int_list_def), "b082192bd0c99202");
}

/// Iter 19b regression: adding `suppress` to FnDef must NOT change
/// canonical-JSON hashes of any pre-19b fn whose `suppress` is empty.
/// The `skip_serializing_if = "Vec::is_empty"` predicate on the field
/// is what enforces this; if it is wrong, every existing fixture's
/// hash drifts and `ail diff` / `ail manifest` output breaks.
#[test]
fn iter19b_empty_suppress_preserves_pre_19b_hashes() {
    let with_empty_suppress = sample_fn();
    // Mutate the bare sample to set suppress explicitly to a
    // non-empty Vec, then mutate it back to empty: two distinct
    // construction paths that should still hash identically.
    let mut with_explicit_empty = sample_fn();
    if let Def::Fn(ref mut f) = with_explicit_empty {
        f.suppress = vec![];
    }
    assert_eq!(
        def_hash(&with_empty_suppress),
        def_hash(&with_explicit_empty),
        "two FnDefs with empty suppress must hash identically"
    );

    // And: a FnDef with a *non-empty* suppress hashes differently.
    let mut with_suppress = sample_fn();
    if let Def::Fn(ref mut f) = with_suppress {
        f.suppress = vec![Suppress {
            code: "over-strict-mode".into(),
            because: "test reason".into(),
        }];
    }
    assert_ne!(
        def_hash(&with_empty_suppress),
        def_hash(&with_suppress),
        "non-empty suppress must produce a distinct hash"
    );
}

/// Iter 19b: an on-disk pre-19b fixture must still load cleanly
/// (no `suppress` field present in the JSON) and produce its
/// canonical-byte hash unchanged.
#[test]
fn iter19b_schema_extension_preserves_pre_19b_hashes() {
    let examples = examples_dir();

    let sum_mod = ailang_surface::load_module(&examples.join("sum.ail"))
        .expect("examples/sum.ail loads");
    let sum_def = sum_mod.defs.iter().find(|d| d.name() == "sum").unwrap();
    assert_eq!(def_hash(sum_def), "db33f57cb329935e");
}

/// Iter 22b.1 regression: adding `Def::Class` and `Def::Instance` must
/// NOT change canonical-JSON hashes of any pre-22b def. The new variants
/// are alternatives, not field additions — pre-22b fixtures simply do
/// not produce a `Class` / `Instance` `Def`, and the existing arms are
/// unchanged.
#[test]
fn iter22b1_schema_extension_preserves_pre_22b_hashes() {
    let examples = examples_dir();

    let sum_mod = ailang_surface::load_module(&examples.join("sum.ail"))
        .expect("examples/sum.ail loads");
    let sum_def = sum_mod.defs.iter().find(|d| d.name() == "sum").unwrap();
    assert_eq!(def_hash(sum_def), "db33f57cb329935e");

    let list_mod = ailang_surface::load_module(&examples.join("list.ail"))
        .expect("examples/list.ail loads");
    let int_list_def = list_mod
        .defs
        .iter()
        .find(|d| d.name() == "IntList")
        .unwrap();
    assert_eq!(def_hash(int_list_def), "b082192bd0c99202");
}

/// Iter 22b.1: a ClassDef with no doc, no superclass, and an empty
/// methods list serialises to a stable canonical form. Two construction
/// paths (default-elided optionals vs. JSON without those keys at all)
/// hash bit-identically.
#[test]
fn iter22b1_classdef_empty_optionals_hash_stable() {
    let bare = Def::Class(ClassDef {
        name: "Empty".into(),
        param: "a".into(),
        superclass: None,
        methods: vec![],
        doc: None,
    });

    let json = r#"{"kind":"class","name":"Empty","param":"a","methods":[]}"#;
    let parsed: Def = serde_json::from_str(json).unwrap();

    assert_eq!(
        def_hash(&bare),
        def_hash(&parsed),
        "ClassDef with elided optionals must hash identically to construction with explicit None"
    );
}

/// Iter 22b.2: adding `constraints` to `Type::Forall` must NOT alter
/// canonical-JSON bytes of any pre-22b.2 polymorphic type.
#[test]
fn forall_without_constraints_hashes_bit_identical_to_pre_22b2() {
    let t = Type::Forall {
        vars: vec!["a".into()],
        constraints: vec![],
        body: Box::new(Type::fn_implicit(
            vec![Type::Var { name: "a".into() }],
            Type::Var { name: "a".into() },
            vec![],
        )),
    };
    let bytes = canonical::to_bytes(&t);
    let s = std::str::from_utf8(&bytes).unwrap();
    assert!(
        !s.contains("constraints"),
        "Type::Forall serialised must omit `constraints` when empty; got: {s}"
    );
}

/// Iter ct.4 (canonical-type-names milestone close): pin the
/// canonical-form hashes of the two cross-module fixtures migrated by
/// `ail migrate-canonical-types` in ct.1. These hashes are the
/// post-migration state.
#[test]
fn ct4_migrated_fixtures_have_canonical_form_hashes() {
    let examples = examples_dir();

    let ord_mod = ailang_surface::load_module(&examples.join("ordering_match.ail"))
        .expect("examples/ordering_match.ail loads");
    let main_def = ord_mod.defs.iter().find(|d| d.name() == "main").unwrap();
    // Iter rpe.1 updated the hash: `ordering_match.ail`'s body migrated
    // from `(do io/print_int x)` to `(app print x)` as part of the
    // corpus migration retiring the per-type print effect-ops.
    assert_eq!(
        def_hash(main_def),
        "b65a7f834703ffb4",
        "ordering_match::main canonical hash must match captured post-rpe.1 value"
    );

    let dup_a_mod = ailang_surface::load_module(&examples.join("test_22b1_dup_a.ail"))
        .expect("examples/test_22b1_dup_a.ail loads");
    assert_eq!(dup_a_mod.defs.len(), 1, "test_22b1_dup_a expected to have exactly 1 def (the instance)");
    assert_eq!(
        def_hash(&dup_a_mod.defs[0]),
        "392c247f07de6517",
        "test_22b1_dup_a instance hash drifted; expected post-24.2 captured value"
    );

    let dup_classmod_mod = ailang_surface::load_module(&examples.join("test_22b1_dup_classmod.ail"))
        .expect("examples/test_22b1_dup_classmod.ail loads");
    assert_eq!(dup_classmod_mod.defs.len(), 1, "test_22b1_dup_classmod expected to have exactly 1 def (class TShow)");
    assert_eq!(
        def_hash(&dup_classmod_mod.defs[0]),
        "b8bca96c2d09ed93",
        "test_22b1_dup_classmod class TShow canonical hash; post-24.2 captured value"
    );
}

/// Iter ct.4: re-assert that the canonical-form tightening did NOT
/// change hashes of intra-module-only fixtures.
#[test]
fn ct4_unmigrated_fixtures_remain_bit_identical() {
    let examples = examples_dir();

    let sum_mod = ailang_surface::load_module(&examples.join("sum.ail"))
        .expect("examples/sum.ail loads");
    let sum_def = sum_mod.defs.iter().find(|d| d.name() == "sum").unwrap();
    assert_eq!(def_hash(sum_def), "db33f57cb329935e",
        "sum.sum hash drifted across canonical-form tightening — unexpected");

    let list_mod = ailang_surface::load_module(&examples.join("list.ail"))
        .expect("examples/list.ail loads");
    let int_list_def = list_mod.defs.iter().find(|d| d.name() == "IntList").unwrap();
    assert_eq!(def_hash(int_list_def), "b082192bd0c99202",
        "list.IntList hash drifted across canonical-form tightening — unexpected");
}