hakorune/src/mir/join_ir/lowering/inline_boundary.rs

//! Phase 188-Impl-3: JoinInlineBoundary - Boundary information for JoinIR inlining
//!
//! This module defines the boundary between JoinIR fragments and the host MIR function.
//! It enables clean separation of concerns:
//!
//! - **Box A**: JoinIR Frontend (doesn't know about host ValueIds)
//! - **Box B**: Join→MIR Bridge (converts to MIR using local ValueIds)
//! - **Box C**: JoinInlineBoundary (stores boundary info - THIS FILE)
//! - **Box D**: JoinMirInlineMerger (injects Copy instructions at boundary)
//!
//! ## Design Philosophy
//!
//! The JoinIR lowerer should work with **JoinIR-side ValueIds** allocated via
//! `JoinValueSpace` (Param: 100-999, Local: 1000+) without knowing anything about
//! the host function's ValueId space. This ensures:
//!
//! 1. **Modularity**: JoinIR lowerers are pure transformers
//! 2. **Reusability**: Same lowerer can be used in different contexts
//! 3. **Testability**: JoinIR can be tested independently
//! 4. **Correctness**: SSA properties are maintained via explicit Copy instructions
//!
//! ## Example
//!
//! For `loop(i < 3) { print(i); i = i + 1 }`:
//!
//! ```text
//! Host Function:
//!   ValueId(4) = Const 0        // i = 0 in host
//!
//! JoinIR Fragment:
//!   ValueId(100)  = param       // i_param (JoinIR Param region)
//!   ValueId(1000) = Const 3
//!   ValueId(1001) = Compare ...
//!
//! Boundary:
//!   join_inputs:  [ValueId(100)] // JoinIR's param slot (Param region)
//!   host_inputs:  [ValueId(4)]  // Host's `i` variable
//!
//! Merged MIR (with Copy injection):
//!   entry:
//!     ValueId(100) = Copy ValueId(4)  // Connect host→JoinIR
//!     ValueId(101) = Const 3
//!     ...
//! ```

use super::carrier_info::{CarrierRole, ExitReconnectMode};
use crate::mir::join_ir::JoinFuncId;
use crate::mir::ValueId;
use std::collections::BTreeSet;

/// Explicit binding between JoinIR exit value and host variable
///
/// This structure formalizes the connection between a JoinIR exit PHI value
/// and the host variable it should update. This eliminates implicit assumptions
/// about which variable a ValueId represents.
///
/// # Pattern 3 Example
///
/// For `loop(i < 3) { sum = sum + i; i = i + 1 }`:
///
/// ```text
/// LoopExitBinding {
///     carrier_name: "sum",
///     join_exit_value: ValueId(18),  // k_exit's return value (JoinIR-local)
///     host_slot: ValueId(5),          // variable_map["sum"] in host
/// }
/// ```
///
/// # Multi-Carrier Support (Pattern 4+)
///
/// Multiple carriers can be represented as a vector:
///
/// ```text
/// vec![
///     LoopExitBinding { carrier_name: "sum", join_exit_value: ValueId(18), host_slot: ValueId(5), role: LoopState },
///     LoopExitBinding { carrier_name: "count", join_exit_value: ValueId(19), host_slot: ValueId(6), role: LoopState },
/// ]
/// ```
#[derive(Debug, Clone)]
pub struct LoopExitBinding {
    /// Carrier variable name (e.g., "sum", "count", "is_digit_pos")
    ///
    /// This is the variable name in the host's variable_map that should
    /// receive the exit value.
    pub carrier_name: String,

    /// JoinIR-side ValueId from k_exit (or exit parameter)
    ///
    /// This is the **JoinIR-local** ValueId that represents the exit value.
    /// It will be remapped when merged into the host function.
    pub join_exit_value: ValueId,

    /// Host-side variable_map slot to reconnect
    ///
    /// This is the host function's ValueId for the variable that should be
    /// updated with the exit PHI result.
    pub host_slot: ValueId,

    /// Phase 227: Role of this carrier (LoopState or ConditionOnly)
    ///
    /// Determines whether this carrier should participate in exit PHI:
    /// - LoopState: Needs exit PHI (value used after loop)
    /// - ConditionOnly: No exit PHI (only used in loop condition)
    pub role: CarrierRole,
}

/// Boundary information for inlining a JoinIR fragment into a host function
///
/// This structure captures the "interface" between a JoinIR fragment and the
/// host function, allowing the merger to inject necessary Copy instructions
/// to connect the two SSA value spaces.
///
/// # Design Note
///
/// This is a **pure data structure** with no logic. All transformation logic
/// lives in the merger (merge_joinir_mir_blocks).
#[derive(Debug, Clone)]
pub struct JoinInlineBoundary {
    /// JoinIR-local ValueIds that act as "input slots"
    ///
    /// These are the ValueIds used **inside** the JoinIR fragment to refer
    /// to values that come from the host. They should be in the JoinValueSpace
    /// Param region (100-999). (They are typically allocated sequentially.)
    ///
    /// Example: For a loop variable `i`, JoinIR uses ValueId(100) as the parameter.
    pub join_inputs: Vec<ValueId>,

    /// Host-function ValueIds that provide the input values
    ///
    /// These are the ValueIds from the **host function** that correspond to
    /// the join_inputs. The merger will inject Copy instructions to connect
    /// host_inputs[i] → join_inputs[i].
    ///
    /// Example: If host has `i` as ValueId(4), then host_inputs = [ValueId(4)].
    pub host_inputs: Vec<ValueId>,

    /// JoinIR-local ValueIds that represent outputs (if any)
    ///
    /// For loops that produce values (e.g., loop result), these are the
    /// JoinIR-local ValueIds that should be visible to the host after inlining.
    ///
    /// Phase 188/189 ではまだ利用していないが、将来的な Multi-carrier パターン
    /// (複数の変数を一度に返すループ) のために予約している。
    pub join_outputs: Vec<ValueId>,

    /// Host-function ValueIds that receive the outputs (DEPRECATED)
    ///
    /// **DEPRECATED**: Use `exit_bindings` instead for explicit carrier naming.
    ///
    /// These are the destination ValueIds in the host function that should
    /// receive the values from join_outputs, or (Pattern 3 のような単一
    /// キャリアのケースでは) ループ exit PHI の結果を受け取るホスト側の
    /// SSA スロットを表す。
    ///
    /// Phase 188-Impl-3 までは未使用だったが、Phase 189 で
    /// loop_if_phi.hako の sum のような「ループの出口で更新されるキャリア」の
    /// 再接続に利用する。
    #[deprecated(since = "Phase 190", note = "Use exit_bindings instead")]
    pub host_outputs: Vec<ValueId>,

    /// Explicit exit bindings for loop carriers (Phase 190+)
    ///
    /// Each binding explicitly names which variable is being updated and
    /// where the value comes from. This eliminates ambiguity and prepares
    /// for multi-carrier support.
    ///
    /// For Pattern 3 (single carrier "sum"):
    /// ```
    /// exit_bindings: vec![
    ///     LoopExitBinding {
    ///         carrier_name: "sum",
    ///         join_exit_value: ValueId(18),  // k_exit return value
    ///         host_slot: ValueId(5),          // variable_map["sum"]
    ///     }
    /// ]
    /// ```
    pub exit_bindings: Vec<LoopExitBinding>,

    /// Condition-only input variables (Phase 171+ / Phase 171-fix)
    ///
    /// **DEPRECATED**: Use `condition_bindings` instead (Phase 171-fix).
    ///
    /// These are variables used ONLY in the loop condition, NOT as loop parameters.
    /// They need to be available in JoinIR scope but are not modified by the loop.
    ///
    /// # Example
    ///
    /// For `loop(start < end) { i = i + 1 }`:
    /// - Loop parameter: `i` → goes in `join_inputs`/`host_inputs`
    /// - Condition-only: `start`, `end` → go in `condition_inputs`
    ///
    /// # Format
    ///
    /// Each entry is `(variable_name, host_value_id)`:
    /// ```
    /// condition_inputs: vec![
    ///     ("start".to_string(), ValueId(33)),  // HOST ID for "start"
    ///     ("end".to_string(), ValueId(34)),    // HOST ID for "end"
    /// ]
    /// ```
    ///
    /// The merger will:
    /// 1. Extract unique variable names from condition AST
    /// 2. Look up HOST ValueIds from `builder.variable_map`
    /// 3. Inject Copy instructions for each condition input
    /// 4. Remap JoinIR references to use the copied values
    #[deprecated(since = "Phase 171-fix", note = "Use condition_bindings instead")]
    pub condition_inputs: Vec<(String, ValueId)>,

    /// Phase 171-fix: Condition bindings with explicit JoinIR ValueIds
    ///
    /// Each binding explicitly specifies:
    /// - Variable name
    /// - HOST ValueId (source)
    /// - JoinIR ValueId (destination)
    ///
    /// This replaces `condition_inputs` to ensure proper ValueId separation.
    pub condition_bindings: Vec<super::condition_to_joinir::ConditionBinding>,

    /// Phase 33-14: Expression result ValueId (JoinIR-local)
    ///
    /// If the loop is used as an expression (like `return loop(...)`), this field
    /// contains the JoinIR-local ValueId of k_exit's return value.
    ///
    /// - `Some(ValueId)`: Loop returns a value → k_exit return goes to exit_phi_inputs
    /// - `None`: Loop only updates carriers → no exit_phi_inputs generation
    ///
    /// # Example: joinir_min_loop.hako (expr result pattern)
    ///
    /// ```nyash
    /// loop(i < 3) { if (i >= 2) { break } i = i + 1 }
    /// return i
    /// ```
    ///
    /// Here, `expr_result = Some(i_exit)` because the loop's result is used.
    ///
    /// # Example: trim pattern (carrier-only)
    ///
    /// ```nyash
    /// loop(start < end) { start = start + 1 }
    /// print(start)  // Uses carrier after loop
    /// ```
    ///
    /// Here, `expr_result = None` because the loop doesn't return a value.
    pub expr_result: Option<crate::mir::ValueId>,

    /// Phase 33-16: Loop variable name (for LoopHeaderPhiBuilder)
    ///
    /// The name of the loop control variable (e.g., "i" in `loop(i < 3)`).
    /// Used to track which PHI corresponds to the loop variable.
    pub loop_var_name: Option<String>,

    /// Phase 228: Carrier metadata (for header PHI generation)
    ///
    /// Contains full carrier information including initialization policies.
    /// This allows header PHI generation to handle ConditionOnly carriers
    /// with explicit bool initialization.
    ///
    /// - `Some(CarrierInfo)`: Full carrier metadata available
    /// - `None`: Legacy path (derive carriers from exit_bindings)
    pub carrier_info: Option<super::carrier_info::CarrierInfo>,

    /// Phase 132 P1: Continuation contract (SSOT)
    ///
    /// JoinIR merge must not infer/guess continuation functions. The router/lowerer
    /// must declare continuation JoinFuncIds here.
    ///
    /// Merge may still choose to *skip* a continuation function if it is a pure
    /// exit stub (structural check), but it must never skip based on name/ID alone.
    pub continuation_func_ids: BTreeSet<JoinFuncId>,

    /// Phase 131 P1.5: Exit reconnection mode
    ///
    /// Controls whether exit values are reconnected via PHI generation (Phi)
    /// or direct variable_map update (DirectValue).
    ///
    /// - `Phi` (default): Existing loop patterns use exit PHI generation
    /// - `DirectValue`: Normalized shadow uses direct value wiring
    pub exit_reconnect_mode: ExitReconnectMode,
}

impl JoinInlineBoundary {
    /// Phase 132-R0 Task 1: SSOT for default continuation function IDs
    ///
    /// Returns the default set of continuation functions (k_exit = join_func_2).
    /// This is the single source of truth for continuation function identification.
    ///
    /// # Rationale
    ///
    /// - Router/lowerer must declare continuation functions explicitly
    /// - Merge must NOT infer continuations by name or ID
    /// - This method centralizes the default continuation contract
    ///
    /// # Usage
    ///
    /// Use this method when constructing JoinInlineBoundary objects:
    ///
    /// ```ignore
    /// JoinInlineBoundary {
    ///     // ...
    ///     continuation_func_ids: JoinInlineBoundary::default_continuations(),
    ///     // ...
    /// }
    /// ```
    pub fn default_continuations() -> BTreeSet<JoinFuncId> {
        BTreeSet::from([JoinFuncId::new(2)])
    }

    /// Create a new boundary with input mappings only
    ///
    /// This is the common case for loops like Pattern 1 where:
    /// - Inputs: loop variables (e.g., `i` in `loop(i < 3)`)
    /// - Outputs: none (loop returns void/0)
    pub fn new_inputs_only(join_inputs: Vec<ValueId>, host_inputs: Vec<ValueId>) -> Self {
        assert_eq!(
            join_inputs.len(),
            host_inputs.len(),
            "join_inputs and host_inputs must have same length"
        );
        Self {
            join_inputs,
            host_inputs,
            join_outputs: vec![],
            #[allow(deprecated)]
            host_outputs: vec![],
            exit_bindings: vec![],
            #[allow(deprecated)]
            condition_inputs: vec![], // Phase 171: Default to empty (deprecated)
            condition_bindings: vec![], // Phase 171-fix: Default to empty
            expr_result: None,          // Phase 33-14: Default to carrier-only pattern
            loop_var_name: None,        // Phase 33-16
            carrier_info: None,         // Phase 228: Default to None
            continuation_func_ids: Self::default_continuations(),
            exit_reconnect_mode: ExitReconnectMode::default(), // Phase 131 P1.5: Default to Phi
        }
    }

    /// Create a new boundary with both inputs and outputs (DEPRECATED)
    ///
    /// **DEPRECATED**: Use `new_with_exit_bindings` instead.
    ///

    /// Create a new boundary with inputs and **host outputs only** (DEPRECATED)
    ///
    /// **DEPRECATED**: Use `new_with_exit_bindings` instead for explicit carrier naming.
    ///
    /// JoinIR 側の exit 値 (k_exit の引数など) を 1 つの PHI にまとめ、
    /// その PHI 結果をホスト側の変数スロットへ再接続したい場合に使う。
    ///
    /// 典型例: Pattern 3 (loop_if_phi.hako)
    /// - join_inputs  : [i_init, sum_init]
    /// - host_inputs  : [host_i, host_sum]
    /// - host_outputs : [host_sum]  // ループ exit 時に上書きしたい変数
    #[deprecated(since = "Phase 190", note = "Use new_with_exit_bindings instead")]
    pub fn new_with_input_and_host_outputs(
        join_inputs: Vec<ValueId>,
        host_inputs: Vec<ValueId>,
        host_outputs: Vec<ValueId>,
    ) -> Self {
        assert_eq!(
            join_inputs.len(),
            host_inputs.len(),
            "join_inputs and host_inputs must have same length"
        );
        Self {
            join_inputs,
            host_inputs,
            join_outputs: vec![],
            #[allow(deprecated)]
            host_outputs,
            exit_bindings: vec![],
            #[allow(deprecated)]
            condition_inputs: vec![], // Phase 171: Default to empty (deprecated)
            condition_bindings: vec![], // Phase 171-fix: Default to empty
            expr_result: None,          // Phase 33-14
            loop_var_name: None,        // Phase 33-16
            carrier_info: None,         // Phase 228
            continuation_func_ids: Self::default_continuations(),
            exit_reconnect_mode: ExitReconnectMode::default(), // Phase 131 P1.5: Default to Phi
        }
    }

    /// Create a new boundary with explicit exit bindings (Phase 190+)
    ///
    /// This is the recommended constructor for loops with exit carriers.
    /// Each exit binding explicitly names the carrier variable and its
    /// source/destination values.
    ///
    /// # Example: Pattern 3 (single carrier)
    ///
    /// ```ignore
    /// let boundary = JoinInlineBoundary::new_with_exit_bindings(
    ///     vec![ValueId(0), ValueId(1)],     // join_inputs (i, sum init)
    ///     vec![loop_var_id, sum_var_id],    // host_inputs
    ///     vec![
    ///         LoopExitBinding {
    ///             carrier_name: "sum".to_string(),
    ///             join_exit_value: ValueId(18),  // k_exit return value
    ///             host_slot: sum_var_id,          // variable_map["sum"]
    ///         }
    ///     ],
    /// );
    /// ```
    ///
    /// # Example: Pattern 4+ (multiple carriers)
    ///
    /// ```ignore
    /// let boundary = JoinInlineBoundary::new_with_exit_bindings(
    ///     vec![ValueId(0), ValueId(1), ValueId(2)],  // join_inputs
    ///     vec![i_id, sum_id, count_id],              // host_inputs
    ///     vec![
    ///         LoopExitBinding { carrier_name: "sum".to_string(), ... },
    ///         LoopExitBinding { carrier_name: "count".to_string(), ... },
    ///     ],
    /// );
    /// ```
    pub fn new_with_exit_bindings(
        join_inputs: Vec<ValueId>,
        host_inputs: Vec<ValueId>,
        exit_bindings: Vec<LoopExitBinding>,
    ) -> Self {
        assert_eq!(
            join_inputs.len(),
            host_inputs.len(),
            "join_inputs and host_inputs must have same length"
        );
        Self {
            join_inputs,
            host_inputs,
            join_outputs: vec![],
            #[allow(deprecated)]
            host_outputs: vec![],
            exit_bindings,
            #[allow(deprecated)]
            condition_inputs: vec![], // Phase 171: Default to empty (deprecated)
            condition_bindings: vec![], // Phase 171-fix: Default to empty
            expr_result: None,          // Phase 33-14
            loop_var_name: None,        // Phase 33-16
            carrier_info: None,         // Phase 228
            continuation_func_ids: Self::default_continuations(),
            exit_reconnect_mode: ExitReconnectMode::default(), // Phase 131 P1.5: Default to Phi
        }
    }

    /// Create a new boundary with condition inputs (Phase 171+)
    ///
    /// # Arguments
    ///
    /// * `join_inputs` - JoinIR-local ValueIds for loop parameters
    /// * `host_inputs` - HOST ValueIds for loop parameters
    /// * `condition_inputs` - Condition-only variables [(name, host_value_id)]
    ///
    /// # Example
    ///
    /// ```ignore
    /// let boundary = JoinInlineBoundary::new_with_condition_inputs(
    ///     vec![ValueId(0)],           // join_inputs (i)
    ///     vec![ValueId(5)],            // host_inputs (i)
    ///     vec![
    ///         ("start".to_string(), ValueId(33)),
    ///         ("end".to_string(), ValueId(34)),
    ///     ],
    /// );
    /// ```
    pub fn new_with_condition_inputs(
        join_inputs: Vec<ValueId>,
        host_inputs: Vec<ValueId>,
        condition_inputs: Vec<(String, ValueId)>,
    ) -> Self {
        assert_eq!(
            join_inputs.len(),
            host_inputs.len(),
            "join_inputs and host_inputs must have same length"
        );
        Self {
            join_inputs,
            host_inputs,
            join_outputs: vec![],
            #[allow(deprecated)]
            host_outputs: vec![],
            exit_bindings: vec![],
            #[allow(deprecated)]
            condition_inputs,
            condition_bindings: vec![], // Phase 171-fix: Will be populated by new constructor
            expr_result: None,          // Phase 33-14
            loop_var_name: None,        // Phase 33-16
            carrier_info: None,         // Phase 228
            continuation_func_ids: Self::default_continuations(),
            exit_reconnect_mode: ExitReconnectMode::default(), // Phase 131 P1.5: Default to Phi
        }
    }

    /// Create boundary with inputs, exit bindings, AND condition inputs (Phase 171+)
    ///
    /// This is the most complete constructor for loops with carriers and condition variables.
    ///
    /// # Example: Pattern 3 with condition variables
    ///
    /// ```ignore
    /// let boundary = JoinInlineBoundary::new_with_exit_and_condition_inputs(
    ///     vec![ValueId(0), ValueId(1)],     // join_inputs (i, sum)
    ///     vec![ValueId(5), ValueId(10)],    // host_inputs
    ///     vec![
    ///         LoopExitBinding {
    ///             carrier_name: "sum".to_string(),
    ///             join_exit_value: ValueId(18),
    ///             host_slot: ValueId(10),
    ///         }
    ///     ],
    ///     vec![
    ///         ("start".to_string(), ValueId(33)),
    ///         ("end".to_string(), ValueId(34)),
    ///     ],
    /// );
    /// ```
    pub fn new_with_exit_and_condition_inputs(
        join_inputs: Vec<ValueId>,
        host_inputs: Vec<ValueId>,
        exit_bindings: Vec<LoopExitBinding>,
        condition_inputs: Vec<(String, ValueId)>,
    ) -> Self {
        assert_eq!(
            join_inputs.len(),
            host_inputs.len(),
            "join_inputs and host_inputs must have same length"
        );
        Self {
            join_inputs,
            host_inputs,
            join_outputs: vec![],
            #[allow(deprecated)]
            host_outputs: vec![],
            exit_bindings,
            #[allow(deprecated)]
            condition_inputs,
            condition_bindings: vec![], // Phase 171-fix: Will be populated by new constructor
            expr_result: None,          // Phase 33-14
            loop_var_name: None,        // Phase 33-16
            carrier_info: None,         // Phase 228
            continuation_func_ids: Self::default_continuations(),
            exit_reconnect_mode: ExitReconnectMode::default(), // Phase 131 P1.5: Default to Phi
        }
    }

    /// Phase 171-fix: Create boundary with ConditionBindings (NEW constructor)
    ///
    /// This is the recommended constructor that uses ConditionBindings instead of
    /// the deprecated condition_inputs.
    ///
    /// # Arguments
    ///
    /// * `join_inputs` - JoinIR-local ValueIds for loop parameters
    /// * `host_inputs` - HOST ValueIds for loop parameters
    /// * `condition_bindings` - Explicit HOST ↔ JoinIR mappings for condition variables
    ///
    /// # Example
    ///
    /// ```ignore
    /// let boundary = JoinInlineBoundary::new_with_condition_bindings(
    ///     vec![ValueId(0)],           // join_inputs (loop param i)
    ///     vec![ValueId(5)],            // host_inputs (loop param i)
    ///     vec![
    ///         ConditionBinding {
    ///             name: "start".to_string(),
    ///             host_value: ValueId(33),  // HOST
    ///             join_value: ValueId(1),    // JoinIR
    ///         },
    ///         ConditionBinding {
    ///             name: "end".to_string(),
    ///             host_value: ValueId(34),  // HOST
    ///             join_value: ValueId(2),    // JoinIR
    ///         },
    ///     ],
    /// );
    /// ```
    pub fn new_with_condition_bindings(
        join_inputs: Vec<ValueId>,
        host_inputs: Vec<ValueId>,
        condition_bindings: Vec<super::condition_to_joinir::ConditionBinding>,
    ) -> Self {
        assert_eq!(
            join_inputs.len(),
            host_inputs.len(),
            "join_inputs and host_inputs must have same length"
        );
        Self {
            join_inputs,
            host_inputs,
            join_outputs: vec![],
            #[allow(deprecated)]
            host_outputs: vec![],
            exit_bindings: vec![],
            #[allow(deprecated)]
            condition_inputs: vec![], // Deprecated, use condition_bindings instead
            condition_bindings,
            expr_result: None,   // Phase 33-14
            loop_var_name: None, // Phase 33-16
            carrier_info: None,  // Phase 228
            continuation_func_ids: Self::default_continuations(),
            exit_reconnect_mode: ExitReconnectMode::default(), // Phase 131 P1.5: Default to Phi
        }
    }
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_boundary_inputs_only() {
        let boundary = JoinInlineBoundary::new_inputs_only(
            vec![ValueId(0)], // JoinIR uses ValueId(0) for loop var
            vec![ValueId(4)], // Host has loop var at ValueId(4)
        );

        assert_eq!(boundary.join_inputs.len(), 1);
        assert_eq!(boundary.host_inputs.len(), 1);
        assert_eq!(boundary.join_outputs.len(), 0);
        #[allow(deprecated)]
        {
            assert_eq!(boundary.host_outputs.len(), 0);
            assert_eq!(boundary.condition_inputs.len(), 0); // Phase 171: Deprecated field
        }
        assert_eq!(boundary.condition_bindings.len(), 0); // Phase 171-fix: New field
    }

    #[test]
    #[should_panic(expected = "join_inputs and host_inputs must have same length")]
    fn test_boundary_mismatched_inputs() {
        JoinInlineBoundary::new_inputs_only(vec![ValueId(0), ValueId(1)], vec![ValueId(4)]);
    }
}