tomoaki/hakorune
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor(joinir): Modularize function_scope_capture (1588→4 modules)

Browse Source 
Break down monolithic function_scope_capture.rs into 4 focused Box-First modules:

**File structure**:
- mod.rs (57 lines): Public API and module organization
- types.rs (110 lines): CapturedVar, CapturedEnv type definitions
- helpers.rs (411 lines): 11 helper functions for AST analysis
- analyzers.rs (1015 lines): Core analysis functions and FunctionScopeCaptureAnalyzer

**Separation of concerns**:
- types.rs: Pure data structures
- helpers.rs: Reusable utility functions (find_stmt_index, ast_matches, etc.)
- analyzers.rs: High-level analysis logic and orchestration
- mod.rs: Clean public API

**Benefits**:
- Single Responsibility Principle: Each module has one clear purpose
- Maintainability: ~400 line modules instead of 1588 line monolith
- Testability: 12 tests organized by module
- Box-First design: Clear separation with well-defined boundaries

**Verification**:
- 937/937 tests PASS (no regression)
- cargo build --release succeeds
- All docs and comments preserved
- No functional changes

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
nyash-codex
2025-12-12 03:42:57 +09:00
parent 879d3ee08e
commit a763651162
4 changed files with 584 additions and 579 deletions
Download Patch File Download Diff File Expand all files Collapse all files

585
src/mir/loop_pattern_detection/function_scope_capture.rs → src/mir/loop_pattern_detection/function_scope_capture/analyzers.rs
View File

@ -1,98 +1,12 @@
//! Phase 200-A: Function scope capture infrastructure //! Core analysis functions for function scope capture
//!
//! This module provides types for capturing function-scoped variables
//! that are effectively immutable within a loop context.
//!
//! # Example
//!
//! For a function like JsonParser._atoi():
//!
//! ```nyash
//! method _atoi(s, pos, len) {
//! local digits = "0123456789" // <-- Captured variable
//! local value = 0
//! loop(pos < len) {
//! local ch = s.charAt(pos)
//! local digit = digits.indexOf(ch) // Uses captured 'digits'
//! if (digit < 0) { break }
//! value = value * 10 + digit
//! pos = pos + 1
//! }
//! return value
//! }
//! ```
//!
//! Here, `digits` is:
//! - Declared in function scope (before the loop)
//! - Never reassigned (effectively immutable)
//! - Referenced in loop body (digits.indexOf(ch))
//!
//! Phase 200-A creates the infrastructure to capture such variables.
//! Phase 200-B will implement the actual detection logic.
use crate::ast::ASTNode; use crate::ast::ASTNode;
use crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape; use crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape;
use crate::mir::ValueId; use crate::mir::ValueId;
use std::collections::BTreeSet; use std::collections::BTreeSet;
/// A variable captured from function scope for use in loop conditions/body. use super::helpers::*;
/// use super::types::{CapturedEnv, CapturedVar};
/// Example: `local digits = "0123456789"` in JsonParser._atoi()
///
/// # Invariants
///
/// - `name`: Variable name as it appears in the source code
/// - `host_id`: MIR ValueId of the original definition in the host function
/// - `is_immutable`: True if the variable is never reassigned in the function
#[derive(Debug, Clone)]
pub struct CapturedVar {
/// Variable name (e.g., "digits", "table")
pub name: String,
/// MIR ValueId of the original definition in the host function
pub host_id: ValueId,
/// Whether this variable is never reassigned in the function
///
/// Phase 200-B will implement assignment analysis to determine this.
/// For now, this is always set to true as a conservative default.
pub is_immutable: bool,
}
/// Environment containing function-scoped captured variables.
///
/// Phase 200-A: Type definition only, not yet integrated with ConditionEnv.
/// Phase 200-B: Will be populated by FunctionScopeCaptureAnalyzer and
/// integrated into ConditionEnv via ConditionEnvBuilder v2.
#[derive(Debug, Clone, Default)]
pub struct CapturedEnv {
/// List of captured variables
pub vars: Vec<CapturedVar>,
}
impl CapturedEnv {
/// Create a new empty environment
pub fn new() -> Self {
Self { vars: Vec::new() }
}
/// Check if the environment is empty
pub fn is_empty(&self) -> bool {
self.vars.is_empty()
}
/// Add a captured variable to the environment
pub fn add_var(&mut self, var: CapturedVar) {
self.vars.push(var);
}
/// Look up a captured variable by name
///
/// Returns `Some(&CapturedVar)` if found, `None` otherwise.
pub fn get(&self, name: &str) -> Option<&CapturedVar> {
self.vars.iter().find(|v| v.name == name)
}
}
/// Analyzes function-scoped variables that can be safely captured for loop conditions/body. /// Analyzes function-scoped variables that can be safely captured for loop conditions/body.
/// ///
@ -443,461 +357,17 @@ pub(crate) fn analyze_captured_vars_v2(
env env
} }
/// Find the index of a loop statement in the function body
///
/// Returns Some(index) if found, None otherwise.
#[allow(dead_code)]
fn find_stmt_index(fn_body: &[ASTNode], loop_ast: &ASTNode) -> Option<usize> {
// Compare by pointer address (same AST node instance)
fn_body
.iter()
.position(|stmt| std::ptr::eq(stmt as *const ASTNode, loop_ast as *const ASTNode))
}
/// Phase 200-C: Find loop index by structure matching (condition + body comparison)
///
/// Instead of pointer comparison, compare the loop structure.
/// This is useful when the loop AST is constructed dynamically.
fn find_loop_index_by_structure(
fn_body: &[ASTNode],
target_condition: &ASTNode,
target_body: &[ASTNode],
) -> Option<usize> {
for (idx, stmt) in fn_body.iter().enumerate() {
if let ASTNode::Loop {
condition, body, ..
} = stmt
{
// Compare condition and body by structure
if ast_matches(condition, target_condition) && body_matches(body, target_body) {
return Some(idx);
}
}
}
None
}
/// Simple structural AST comparison
///
/// Uses Debug string comparison as a heuristic. This is not perfect but
/// works well enough for finding loops by structure.
fn ast_matches(a: &ASTNode, b: &ASTNode) -> bool {
format!("{:?}", a) == format!("{:?}", b)
}
/// Compare two body slices by structure
fn body_matches(a: &[ASTNode], b: &[ASTNode]) -> bool {
if a.len() != b.len() {
return false;
}
a.iter().zip(b.iter()).all(|(x, y)| ast_matches(x, y))
}
/// Collect local variable declarations from statements
///
/// Returns Vec<(name, init_expr)> for each variable declared with `local`.
fn collect_local_declarations(stmts: &[ASTNode]) -> Vec<(String, Option<Box<ASTNode>>)> {
let mut locals = Vec::new();
for stmt in stmts {
if let ASTNode::Local {
variables,
initial_values,
..
} = stmt
{
// Local declaration can have multiple variables (e.g., local a, b, c)
for (i, name) in variables.iter().enumerate() {
let init_expr = initial_values.get(i).and_then(|opt| opt.clone());
locals.push((name.clone(), init_expr));
}
}
}
locals
}
/// Check if expression is a safe constant (string/integer literal)
///
/// Phase 200-B: Only string and integer literals are allowed.
/// Future: May expand to include other safe constant patterns.
fn is_safe_const_init(expr: &Option<Box<ASTNode>>) -> bool {
match expr {
Some(boxed) => match boxed.as_ref() {
ASTNode::Literal { value, .. } => matches!(
value,
crate::ast::LiteralValue::String(_) | crate::ast::LiteralValue::Integer(_)
),
_ => false,
},
None => false,
}
}
/// Check if variable is reassigned anywhere in function body
///
/// Walks the entire function body AST to detect any assignments to the variable.
/// Returns true if the variable is reassigned (excluding the initial local declaration).
fn is_reassigned_in_fn(fn_body: &[ASTNode], name: &str) -> bool {
fn check_node(node: &ASTNode, name: &str) -> bool {
match node {
// Assignment to this variable
ASTNode::Assignment { target, value, .. } => {
// Check if target is the variable we're looking for
let is_target_match = match target.as_ref() {
ASTNode::Variable { name: var_name, .. } => var_name == name,
ASTNode::FieldAccess { .. } | ASTNode::Index { .. } => {
// Field access or index assignment doesn't count as reassignment
false
}
_ => false,
};
is_target_match || check_node(value, name)
}
// Grouped assignment expression: (x = expr)
ASTNode::GroupedAssignmentExpr { lhs, rhs, .. } => lhs == name || check_node(rhs, name),
// Recursive cases
ASTNode::If {
condition,
then_body,
else_body,
..
} => {
check_node(condition, name)
|| then_body.iter().any(|n| check_node(n, name))
|| else_body
.as_ref()
.map_or(false, |body| body.iter().any(|n| check_node(n, name)))
}
ASTNode::Loop {
condition, body, ..
} => check_node(condition, name) || body.iter().any(|n| check_node(n, name)),
ASTNode::While {
condition, body, ..
} => check_node(condition, name) || body.iter().any(|n| check_node(n, name)),
ASTNode::TryCatch {
try_body,
catch_clauses,
finally_body,
..
} => {
try_body.iter().any(|n| check_node(n, name))
|| catch_clauses
.iter()
.any(|clause| clause.body.iter().any(|n| check_node(n, name)))
|| finally_body
.as_ref()
.map_or(false, |body| body.iter().any(|n| check_node(n, name)))
}
ASTNode::UnaryOp { operand, .. } => check_node(operand, name),
ASTNode::BinaryOp { left, right, .. } => {
check_node(left, name) || check_node(right, name)
}
ASTNode::MethodCall {
object, arguments, ..
} => check_node(object, name) || arguments.iter().any(|arg| check_node(arg, name)),
ASTNode::FunctionCall { arguments, .. } => {
arguments.iter().any(|arg| check_node(arg, name))
}
ASTNode::FieldAccess { object, .. } => check_node(object, name),
ASTNode::Index { target, index, .. } => {
check_node(target, name) || check_node(index, name)
}
ASTNode::Return { value, .. } => value.as_ref().map_or(false, |v| check_node(v, name)),
ASTNode::Local { .. } => {
// Local declarations are not reassignments
false
}
_ => false,
}
}
fn_body.iter().any(|stmt| check_node(stmt, name))
}
/// Check if variable is referenced in loop condition or body
///
/// Returns true if the variable name appears anywhere in the loop AST.
#[allow(dead_code)]
fn is_used_in_loop(loop_ast: &ASTNode, name: &str) -> bool {
fn check_usage(node: &ASTNode, name: &str) -> bool {
match node {
ASTNode::Variable { name: var_name, .. } => var_name == name,
ASTNode::Loop {
condition, body, ..
} => check_usage(condition, name) || body.iter().any(|n| check_usage(n, name)),
ASTNode::If {
condition,
then_body,
else_body,
..
} => {
check_usage(condition, name)
|| then_body.iter().any(|n| check_usage(n, name))
|| else_body
.as_ref()
.map_or(false, |body| body.iter().any(|n| check_usage(n, name)))
}
ASTNode::Assignment { target, value, .. } => {
check_usage(target, name) || check_usage(value, name)
}
ASTNode::UnaryOp { operand, .. } => check_usage(operand, name),
ASTNode::BinaryOp { left, right, .. } => {
check_usage(left, name) || check_usage(right, name)
}
ASTNode::MethodCall {
object, arguments, ..
} => check_usage(object, name) || arguments.iter().any(|arg| check_usage(arg, name)),
ASTNode::FunctionCall { arguments, .. } => {
arguments.iter().any(|arg| check_usage(arg, name))
}
ASTNode::FieldAccess { object, .. } => check_usage(object, name),
ASTNode::Index { target, index, .. } => {
check_usage(target, name) || check_usage(index, name)
}
ASTNode::Return { value, .. } => value.as_ref().map_or(false, |v| check_usage(v, name)),
ASTNode::Local { initial_values, .. } => initial_values
.iter()
.any(|opt| opt.as_ref().map_or(false, |init| check_usage(init, name))),
_ => false,
}
}
check_usage(loop_ast, name)
}
/// Phase 200-C: Check if variable is used in loop condition or body (separate parts)
///
/// This is used by analyze_captured_vars_v2 when condition and body are passed separately.
fn is_used_in_loop_parts(condition: &ASTNode, body: &[ASTNode], name: &str) -> bool {
fn check_usage(node: &ASTNode, name: &str) -> bool {
match node {
ASTNode::Variable { name: var_name, .. } => var_name == name,
ASTNode::Loop {
condition, body, ..
} => check_usage(condition, name) || body.iter().any(|n| check_usage(n, name)),
ASTNode::If {
condition,
then_body,
else_body,
..
} => {
check_usage(condition, name)
|| then_body.iter().any(|n| check_usage(n, name))
|| else_body
.as_ref()
.map_or(false, |body| body.iter().any(|n| check_usage(n, name)))
}
ASTNode::Assignment { target, value, .. } => {
check_usage(target, name) || check_usage(value, name)
}
ASTNode::UnaryOp { operand, .. } => check_usage(operand, name),
ASTNode::BinaryOp { left, right, .. } => {
check_usage(left, name) || check_usage(right, name)
}
ASTNode::MethodCall {
object, arguments, ..
} => check_usage(object, name) || arguments.iter().any(|arg| check_usage(arg, name)),
ASTNode::FunctionCall { arguments, .. } => {
arguments.iter().any(|arg| check_usage(arg, name))
}
ASTNode::FieldAccess { object, .. } => check_usage(object, name),
ASTNode::Index { target, index, .. } => {
check_usage(target, name) || check_usage(index, name)
}
ASTNode::Return { value, .. } => value.as_ref().map_or(false, |v| check_usage(v, name)),
ASTNode::Local { initial_values, .. } => initial_values
.iter()
.any(|opt| opt.as_ref().map_or(false, |init| check_usage(init, name))),
_ => false,
}
}
check_usage(condition, name) || body.iter().any(|n| check_usage(n, name))
}
/// Phase 245C: Collect all variable names used in loop condition and body
///
/// Helper for function parameter capture. Returns a set of all variable names
/// that appear in the loop's condition or body.
fn collect_names_in_loop_parts(condition: &ASTNode, body: &[ASTNode]) -> BTreeSet<String> {
fn collect(node: &ASTNode, acc: &mut BTreeSet<String>) {
match node {
ASTNode::Variable { name, .. } => {
acc.insert(name.clone());
}
ASTNode::If {
condition,
then_body,
else_body,
..
} => {
collect(condition, acc);
for stmt in then_body {
collect(stmt, acc);
}
if let Some(else_stmts) = else_body {
for stmt in else_stmts {
collect(stmt, acc);
}
}
}
ASTNode::Assignment { target, value, .. } => {
collect(target, acc);
collect(value, acc);
}
ASTNode::UnaryOp { operand, .. } => {
collect(operand, acc);
}
ASTNode::Return {
value: Some(operand),
..
} => {
collect(operand, acc);
}
ASTNode::BinaryOp { left, right, .. } => {
collect(left, acc);
collect(right, acc);
}
ASTNode::MethodCall {
object, arguments, ..
} => {
collect(object, acc);
for arg in arguments {
collect(arg, acc);
}
}
ASTNode::FunctionCall { arguments, .. } => {
for arg in arguments {
collect(arg, acc);
}
}
ASTNode::Local { initial_values, .. } => {
for init_opt in initial_values {
if let Some(val) = init_opt {
collect(val, acc);
}
}
}
ASTNode::FieldAccess { object, .. } => {
collect(object, acc);
}
ASTNode::Index { target, index, .. } => {
collect(target, acc);
collect(index, acc);
}
ASTNode::Loop {
condition, body, ..
} => {
collect(condition, acc);
for stmt in body {
collect(stmt, acc);
}
}
_ => {}
}
}
let mut acc = BTreeSet::new();
collect(condition, &mut acc);
for stmt in body {
collect(stmt, &mut acc);
}
acc
}
#[cfg(test)] #[cfg(test)]
mod tests { mod tests {
use super::*; use super::*;
use crate::ast::{ASTNode, BinaryOperator, LiteralValue, Span};
#[test] use crate::mir::BasicBlockId;
fn test_captured_env_empty() { use std::collections::{BTreeMap, BTreeSet};
let env = CapturedEnv::new();
assert!(env.is_empty());
assert!(env.get("digits").is_none());
}
#[test]
fn test_captured_env_add_and_get() {
let mut env = CapturedEnv::new();
env.add_var(CapturedVar {
name: "digits".to_string(),
host_id: ValueId(42),
is_immutable: true,
});
assert!(!env.is_empty());
let var = env.get("digits").unwrap();
assert_eq!(var.name, "digits");
assert_eq!(var.host_id, ValueId(42));
assert!(var.is_immutable);
}
#[test]
fn test_captured_env_multiple_vars() {
let mut env = CapturedEnv::new();
env.add_var(CapturedVar {
name: "digits".to_string(),
host_id: ValueId(42),
is_immutable: true,
});
env.add_var(CapturedVar {
name: "table".to_string(),
host_id: ValueId(100),
is_immutable: true,
});
assert_eq!(env.vars.len(), 2);
assert!(env.get("digits").is_some());
assert!(env.get("table").is_some());
assert!(env.get("nonexistent").is_none());
}
// Phase 200-B: Capture analysis tests // Phase 200-B: Capture analysis tests
#[test] #[test]
fn test_capture_simple_digits() { fn test_capture_simple_digits() {
use crate::ast::{ASTNode, LiteralValue, Span};
// Build AST for: // Build AST for:
// local digits = "0123456789" // local digits = "0123456789"
// loop(i < 10) { // loop(i < 10) {
@ -949,9 +419,6 @@ mod tests {
let fn_body = vec![digits_decl, loop_node.clone()]; let fn_body = vec![digits_decl, loop_node.clone()];
use crate::mir::BasicBlockId;
use std::collections::{BTreeMap, BTreeSet};
let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape { let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape {
header: BasicBlockId(0), header: BasicBlockId(0),
body: BasicBlockId(1), body: BasicBlockId(1),
@ -978,8 +445,6 @@ mod tests {
#[test] #[test]
fn test_capture_reassigned_rejected() { fn test_capture_reassigned_rejected() {
use crate::ast::{ASTNode, LiteralValue, Span};
// Build AST for: // Build AST for:
// local digits = "0123456789" // local digits = "0123456789"
// digits = "abc" // reassignment // digits = "abc" // reassignment
@ -1041,9 +506,6 @@ mod tests {
let fn_body = vec![digits_decl, reassignment, loop_node.clone()]; let fn_body = vec![digits_decl, reassignment, loop_node.clone()];
use crate::mir::BasicBlockId;
use std::collections::{BTreeMap, BTreeSet};
let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape { let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape {
header: BasicBlockId(0), header: BasicBlockId(0),
body: BasicBlockId(1), body: BasicBlockId(1),
@ -1065,8 +527,6 @@ mod tests {
#[test] #[test]
fn test_capture_after_loop_rejected() { fn test_capture_after_loop_rejected() {
use crate::ast::{ASTNode, LiteralValue, Span};
// Build AST for: // Build AST for:
// loop(i < 10) { } // loop(i < 10) { }
// local digits = "0123456789" // defined AFTER loop // local digits = "0123456789" // defined AFTER loop
@ -1099,9 +559,6 @@ mod tests {
let fn_body = vec![loop_node.clone(), digits_decl]; let fn_body = vec![loop_node.clone(), digits_decl];
use crate::mir::BasicBlockId;
use std::collections::{BTreeMap, BTreeSet};
let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape { let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape {
header: BasicBlockId(0), header: BasicBlockId(0),
body: BasicBlockId(1), body: BasicBlockId(1),
@ -1123,8 +580,6 @@ mod tests {
#[test] #[test]
fn test_capture_method_call_init_rejected() { fn test_capture_method_call_init_rejected() {
use crate::ast::{ASTNode, LiteralValue, Span};
// Build AST for: // Build AST for:
// local result = someBox.getValue() // MethodCall init // local result = someBox.getValue() // MethodCall init
// loop(i < 10) { // loop(i < 10) {
@ -1178,9 +633,6 @@ mod tests {
let fn_body = vec![result_decl, loop_node.clone()]; let fn_body = vec![result_decl, loop_node.clone()];
use crate::mir::BasicBlockId;
use std::collections::{BTreeMap, BTreeSet};
let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape { let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape {
header: BasicBlockId(0), header: BasicBlockId(0),
body: BasicBlockId(1), body: BasicBlockId(1),
@ -1202,8 +654,6 @@ mod tests {
#[test] #[test]
fn test_capture_unused_in_loop_rejected() { fn test_capture_unused_in_loop_rejected() {
use crate::ast::{ASTNode, LiteralValue, Span};
// Build AST for: // Build AST for:
// local digits = "0123456789" // not used in loop // local digits = "0123456789" // not used in loop
// loop(i < 10) { // loop(i < 10) {
@ -1238,9 +688,6 @@ mod tests {
let fn_body = vec![digits_decl, loop_node.clone()]; let fn_body = vec![digits_decl, loop_node.clone()];
use crate::mir::BasicBlockId;
use std::collections::{BTreeMap, BTreeSet};
let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape { let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape {
header: BasicBlockId(0), header: BasicBlockId(0),
body: BasicBlockId(1), body: BasicBlockId(1),
@ -1264,8 +711,6 @@ mod tests {
#[test] #[test]
fn test_capture_function_param_used_in_condition() { fn test_capture_function_param_used_in_condition() {
use crate::ast::{ASTNode, BinaryOperator, LiteralValue, Span};
// Simulate: fn parse_number(s, p, len) { loop(p < len) { ... } } // Simulate: fn parse_number(s, p, len) { loop(p < len) { ... } }
// Expected: 'len' should be captured (used in condition, not reassigned) // Expected: 'len' should be captured (used in condition, not reassigned)
@ -1302,9 +747,6 @@ mod tests {
span: Span::unknown(), span: Span::unknown(),
}]; }];
use crate::mir::BasicBlockId;
use std::collections::{BTreeMap, BTreeSet};
let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape { let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape {
header: BasicBlockId(0), header: BasicBlockId(0),
body: BasicBlockId(1), body: BasicBlockId(1),
@ -1331,8 +773,6 @@ mod tests {
#[test] #[test]
fn test_capture_function_param_used_in_method_call() { fn test_capture_function_param_used_in_method_call() {
use crate::ast::{ASTNode, BinaryOperator, LiteralValue, Span};
// Simulate: fn parse_number(s, p) { loop(p < s.length()) { ch = s.charAt(p) } } // Simulate: fn parse_number(s, p) { loop(p < s.length()) { ch = s.charAt(p) } }
// Expected: 's' should be captured (used in condition and body, not reassigned) // Expected: 's' should be captured (used in condition and body, not reassigned)
@ -1392,9 +832,6 @@ mod tests {
}, },
]; ];
use crate::mir::BasicBlockId;
use std::collections::{BTreeMap, BTreeSet};
let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape { let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape {
header: BasicBlockId(0), header: BasicBlockId(0),
body: BasicBlockId(1), body: BasicBlockId(1),
@ -1421,8 +858,6 @@ mod tests {
#[test] #[test]
fn test_capture_function_param_reassigned_rejected() { fn test_capture_function_param_reassigned_rejected() {
use crate::ast::{ASTNode, BinaryOperator, LiteralValue, Span};
// Simulate: fn bad_func(x) { x = 5; loop(x < 10) { x = x + 1 } } // Simulate: fn bad_func(x) { x = 5; loop(x < 10) { x = x + 1 } }
// Expected: 'x' should NOT be captured (reassigned in function) // Expected: 'x' should NOT be captured (reassigned in function)
@ -1472,9 +907,6 @@ mod tests {
span: Span::unknown(), span: Span::unknown(),
}]; }];
use crate::mir::BasicBlockId;
use std::collections::{BTreeMap, BTreeSet};
let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape { let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape {
header: BasicBlockId(0), header: BasicBlockId(0),
body: BasicBlockId(1), body: BasicBlockId(1),
@ -1497,8 +929,6 @@ mod tests {
#[test] #[test]
fn test_capture_mixed_locals_and_params() { fn test_capture_mixed_locals_and_params() {
use crate::ast::{ASTNode, BinaryOperator, LiteralValue, Span};
// Simulate: fn parse(s, len) { local digits = "0123"; loop(p < len) { ch = digits.indexOf(...); s.charAt(...) } } // Simulate: fn parse(s, len) { local digits = "0123"; loop(p < len) { ch = digits.indexOf(...); s.charAt(...) } }
// Expected: 'len', 's', and 'digits' should all be captured // Expected: 'len', 's', and 'digits' should all be captured
@ -1560,9 +990,6 @@ mod tests {
span: Span::unknown(), span: Span::unknown(),
}]; }];
use crate::mir::BasicBlockId;
use std::collections::{BTreeMap, BTreeSet};
let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape { let scope = crate::mir::join_ir::lowering::loop_scope_shape::LoopScopeShape {
header: BasicBlockId(0), header: BasicBlockId(0),
body: BasicBlockId(1), body: BasicBlockId(1),

411
src/mir/loop_pattern_detection/function_scope_capture/helpers.rs Normal file
View File

@ -0,0 +1,411 @@
//! Helper functions for AST analysis, reassignment detection, and structural matching
use crate::ast::ASTNode;
use std::collections::BTreeSet;
/// Find the index of a loop statement in the function body
///
/// Returns Some(index) if found, None otherwise.
#[allow(dead_code)]
pub(super) fn find_stmt_index(fn_body: &[ASTNode], loop_ast: &ASTNode) -> Option<usize> {
// Compare by pointer address (same AST node instance)
fn_body
.iter()
.position(|stmt| std::ptr::eq(stmt as *const ASTNode, loop_ast as *const ASTNode))
}
/// Phase 200-C: Find loop index by structure matching (condition + body comparison)
///
/// Instead of pointer comparison, compare the loop structure.
/// This is useful when the loop AST is constructed dynamically.
pub(super) fn find_loop_index_by_structure(
fn_body: &[ASTNode],
target_condition: &ASTNode,
target_body: &[ASTNode],
) -> Option<usize> {
for (idx, stmt) in fn_body.iter().enumerate() {
if let ASTNode::Loop {
condition, body, ..
} = stmt
{
// Compare condition and body by structure
if ast_matches(condition, target_condition) && body_matches(body, target_body) {
return Some(idx);
}
}
}
None
}
/// Simple structural AST comparison
///
/// Uses Debug string comparison as a heuristic. This is not perfect but
/// works well enough for finding loops by structure.
pub(super) fn ast_matches(a: &ASTNode, b: &ASTNode) -> bool {
format!("{:?}", a) == format!("{:?}", b)
}
/// Compare two body slices by structure
pub(super) fn body_matches(a: &[ASTNode], b: &[ASTNode]) -> bool {
if a.len() != b.len() {
return false;
}
a.iter().zip(b.iter()).all(|(x, y)| ast_matches(x, y))
}
/// Collect local variable declarations from statements
///
/// Returns Vec<(name, init_expr)> for each variable declared with `local`.
pub(super) fn collect_local_declarations(
stmts: &[ASTNode],
) -> Vec<(String, Option<Box<ASTNode>>)> {
let mut locals = Vec::new();
for stmt in stmts {
if let ASTNode::Local {
variables,
initial_values,
..
} = stmt
{
// Local declaration can have multiple variables (e.g., local a, b, c)
for (i, name) in variables.iter().enumerate() {
let init_expr = initial_values.get(i).and_then(|opt| opt.clone());
locals.push((name.clone(), init_expr));
}
}
}
locals
}
/// Check if expression is a safe constant (string/integer literal)
///
/// Phase 200-B: Only string and integer literals are allowed.
/// Future: May expand to include other safe constant patterns.
pub(super) fn is_safe_const_init(expr: &Option<Box<ASTNode>>) -> bool {
match expr {
Some(boxed) => match boxed.as_ref() {
ASTNode::Literal { value, .. } => matches!(
value,
crate::ast::LiteralValue::String(_) | crate::ast::LiteralValue::Integer(_)
),
_ => false,
},
None => false,
}
}
/// Check if variable is reassigned anywhere in function body
///
/// Walks the entire function body AST to detect any assignments to the variable.
/// Returns true if the variable is reassigned (excluding the initial local declaration).
pub(super) fn is_reassigned_in_fn(fn_body: &[ASTNode], name: &str) -> bool {
fn check_node(node: &ASTNode, name: &str) -> bool {
match node {
// Assignment to this variable
ASTNode::Assignment { target, value, .. } => {
// Check if target is the variable we're looking for
let is_target_match = match target.as_ref() {
ASTNode::Variable { name: var_name, .. } => var_name == name,
ASTNode::FieldAccess { .. } | ASTNode::Index { .. } => {
// Field access or index assignment doesn't count as reassignment
false
}
_ => false,
};
is_target_match || check_node(value, name)
}
// Grouped assignment expression: (x = expr)
ASTNode::GroupedAssignmentExpr { lhs, rhs, .. } => lhs == name || check_node(rhs, name),
// Recursive cases
ASTNode::If {
condition,
then_body,
else_body,
..
} => {
check_node(condition, name)
|| then_body.iter().any(|n| check_node(n, name))
|| else_body
.as_ref()
.map_or(false, |body| body.iter().any(|n| check_node(n, name)))
}
ASTNode::Loop {
condition, body, ..
} => check_node(condition, name) || body.iter().any(|n| check_node(n, name)),
ASTNode::While {
condition, body, ..
} => check_node(condition, name) || body.iter().any(|n| check_node(n, name)),
ASTNode::TryCatch {
try_body,
catch_clauses,
finally_body,
..
} => {
try_body.iter().any(|n| check_node(n, name))
|| catch_clauses
.iter()
.any(|clause| clause.body.iter().any(|n| check_node(n, name)))
|| finally_body
.as_ref()
.map_or(false, |body| body.iter().any(|n| check_node(n, name)))
}
ASTNode::UnaryOp { operand, .. } => check_node(operand, name),
ASTNode::BinaryOp { left, right, .. } => {
check_node(left, name) || check_node(right, name)
}
ASTNode::MethodCall {
object, arguments, ..
} => check_node(object, name) || arguments.iter().any(|arg| check_node(arg, name)),
ASTNode::FunctionCall { arguments, .. } => {
arguments.iter().any(|arg| check_node(arg, name))
}
ASTNode::FieldAccess { object, .. } => check_node(object, name),
ASTNode::Index { target, index, .. } => {
check_node(target, name) || check_node(index, name)
}
ASTNode::Return { value, .. } => value.as_ref().map_or(false, |v| check_node(v, name)),
ASTNode::Local { .. } => {
// Local declarations are not reassignments
false
}
_ => false,
}
}
fn_body.iter().any(|stmt| check_node(stmt, name))
}
/// Check if variable is referenced in loop condition or body
///
/// Returns true if the variable name appears anywhere in the loop AST.
#[allow(dead_code)]
pub(super) fn is_used_in_loop(loop_ast: &ASTNode, name: &str) -> bool {
fn check_usage(node: &ASTNode, name: &str) -> bool {
match node {
ASTNode::Variable { name: var_name, .. } => var_name == name,
ASTNode::Loop {
condition, body, ..
} => check_usage(condition, name) || body.iter().any(|n| check_usage(n, name)),
ASTNode::If {
condition,
then_body,
else_body,
..
} => {
check_usage(condition, name)
|| then_body.iter().any(|n| check_usage(n, name))
|| else_body
.as_ref()
.map_or(false, |body| body.iter().any(|n| check_usage(n, name)))
}
ASTNode::Assignment { target, value, .. } => {
check_usage(target, name) || check_usage(value, name)
}
ASTNode::UnaryOp { operand, .. } => check_usage(operand, name),
ASTNode::BinaryOp { left, right, .. } => {
check_usage(left, name) || check_usage(right, name)
}
ASTNode::MethodCall {
object, arguments, ..
} => check_usage(object, name) || arguments.iter().any(|arg| check_usage(arg, name)),
ASTNode::FunctionCall { arguments, .. } => {
arguments.iter().any(|arg| check_usage(arg, name))
}
ASTNode::FieldAccess { object, .. } => check_usage(object, name),
ASTNode::Index { target, index, .. } => {
check_usage(target, name) || check_usage(index, name)
}
ASTNode::Return { value, .. } => value.as_ref().map_or(false, |v| check_usage(v, name)),
ASTNode::Local { initial_values, .. } => initial_values
.iter()
.any(|opt| opt.as_ref().map_or(false, |init| check_usage(init, name))),
_ => false,
}
}
check_usage(loop_ast, name)
}
/// Phase 200-C: Check if variable is used in loop condition or body (separate parts)
///
/// This is used by analyze_captured_vars_v2 when condition and body are passed separately.
pub(super) fn is_used_in_loop_parts(condition: &ASTNode, body: &[ASTNode], name: &str) -> bool {
fn check_usage(node: &ASTNode, name: &str) -> bool {
match node {
ASTNode::Variable { name: var_name, .. } => var_name == name,
ASTNode::Loop {
condition, body, ..
} => check_usage(condition, name) || body.iter().any(|n| check_usage(n, name)),
ASTNode::If {
condition,
then_body,
else_body,
..
} => {
check_usage(condition, name)
|| then_body.iter().any(|n| check_usage(n, name))
|| else_body
.as_ref()
.map_or(false, |body| body.iter().any(|n| check_usage(n, name)))
}
ASTNode::Assignment { target, value, .. } => {
check_usage(target, name) || check_usage(value, name)
}
ASTNode::UnaryOp { operand, .. } => check_usage(operand, name),
ASTNode::BinaryOp { left, right, .. } => {
check_usage(left, name) || check_usage(right, name)
}
ASTNode::MethodCall {
object, arguments, ..
} => check_usage(object, name) || arguments.iter().any(|arg| check_usage(arg, name)),
ASTNode::FunctionCall { arguments, .. } => {
arguments.iter().any(|arg| check_usage(arg, name))
}
ASTNode::FieldAccess { object, .. } => check_usage(object, name),
ASTNode::Index { target, index, .. } => {
check_usage(target, name) || check_usage(index, name)
}
ASTNode::Return { value, .. } => value.as_ref().map_or(false, |v| check_usage(v, name)),
ASTNode::Local { initial_values, .. } => initial_values
.iter()
.any(|opt| opt.as_ref().map_or(false, |init| check_usage(init, name))),
_ => false,
}
}
check_usage(condition, name) || body.iter().any(|n| check_usage(n, name))
}
/// Phase 245C: Collect all variable names used in loop condition and body
///
/// Helper for function parameter capture. Returns a set of all variable names
/// that appear in the loop's condition or body.
pub(super) fn collect_names_in_loop_parts(
condition: &ASTNode,
body: &[ASTNode],
) -> BTreeSet<String> {
fn collect(node: &ASTNode, acc: &mut BTreeSet<String>) {
match node {
ASTNode::Variable { name, .. } => {
acc.insert(name.clone());
}
ASTNode::If {
condition,
then_body,
else_body,
..
} => {
collect(condition, acc);
for stmt in then_body {
collect(stmt, acc);
}
if let Some(else_stmts) = else_body {
for stmt in else_stmts {
collect(stmt, acc);
}
}
}
ASTNode::Assignment { target, value, .. } => {
collect(target, acc);
collect(value, acc);
}
ASTNode::UnaryOp { operand, .. } => {
collect(operand, acc);
}
ASTNode::Return {
value: Some(operand),
..
} => {
collect(operand, acc);
}
ASTNode::BinaryOp { left, right, .. } => {
collect(left, acc);
collect(right, acc);
}
ASTNode::MethodCall {
object, arguments, ..
} => {
collect(object, acc);
for arg in arguments {
collect(arg, acc);
}
}
ASTNode::FunctionCall { arguments, .. } => {
for arg in arguments {
collect(arg, acc);
}
}
ASTNode::Local { initial_values, .. } => {
for init_opt in initial_values {
if let Some(val) = init_opt {
collect(val, acc);
}
}
}
ASTNode::FieldAccess { object, .. } => {
collect(object, acc);
}
ASTNode::Index { target, index, .. } => {
collect(target, acc);
collect(index, acc);
}
ASTNode::Loop {
condition, body, ..
} => {
collect(condition, acc);
for stmt in body {
collect(stmt, acc);
}
}
_ => {}
}
}
let mut acc = BTreeSet::new();
collect(condition, &mut acc);
for stmt in body {
collect(stmt, &mut acc);
}
acc
}

57
src/mir/loop_pattern_detection/function_scope_capture/mod.rs Normal file
View File

@ -0,0 +1,57 @@
//! Phase 200-A: Function scope capture infrastructure
//!
//! This module provides types and analysis functions for capturing function-scoped variables
//! that are effectively immutable within a loop context.
//!
//! # Example
//!
//! For a function like JsonParser._atoi():
//!
//! ```nyash
//! method _atoi(s, pos, len) {
//! local digits = "0123456789" // <-- Captured variable
//! local value = 0
//! loop(pos < len) {
//! local ch = s.charAt(pos)
//! local digit = digits.indexOf(ch) // Uses captured 'digits'
//! if (digit < 0) { break }
//! value = value * 10 + digit
//! pos = pos + 1
//! }
//! return value
//! }
//! ```
//!
//! Here, `digits` is:
//! - Declared in function scope (before the loop)
//! - Never reassigned (effectively immutable)
//! - Referenced in loop body (digits.indexOf(ch))
//!
//! Phase 200-A creates the infrastructure to capture such variables.
//! Phase 200-B implements the actual detection logic.
//!
//! # Module Structure
//!
//! This module is organized following the Box-First principle:
//!
//! - `types` - Core type definitions (CapturedVar, CapturedEnv)
//! - `analyzers` - Analysis functions (analyze_captured_vars, analyze_captured_vars_v2)
//! - `helpers` - Helper functions for AST analysis and structural matching
//!
//! # Public API
//!
//! The primary entry points are:
//!
//! - `analyze_captured_vars()` - Main analysis function (uses pointer comparison)
//! - `analyze_captured_vars_v2()` - Alternative using structural matching (Phase 200-C)
//! - `CapturedVar` - Represents a captured variable
//! - `CapturedEnv` - Environment containing all captured variables
// Module declarations
mod analyzers;
mod helpers;
mod types;
// Public re-exports
pub(crate) use analyzers::{analyze_captured_vars, analyze_captured_vars_v2};
pub use types::{CapturedEnv, CapturedVar};

110
src/mir/loop_pattern_detection/function_scope_capture/types.rs Normal file
View File

@ -0,0 +1,110 @@
//! Type definitions for function scope capture
use crate::mir::ValueId;
/// A variable captured from function scope for use in loop conditions/body.
///
/// Example: `local digits = "0123456789"` in JsonParser._atoi()
///
/// # Invariants
///
/// - `name`: Variable name as it appears in the source code
/// - `host_id`: MIR ValueId of the original definition in the host function
/// - `is_immutable`: True if the variable is never reassigned in the function
#[derive(Debug, Clone)]
pub struct CapturedVar {
/// Variable name (e.g., "digits", "table")
pub name: String,
/// MIR ValueId of the original definition in the host function
pub host_id: ValueId,
/// Whether this variable is never reassigned in the function
///
/// Phase 200-B will implement assignment analysis to determine this.
/// For now, this is always set to true as a conservative default.
pub is_immutable: bool,
}
/// Environment containing function-scoped captured variables.
///
/// Phase 200-A: Type definition only, not yet integrated with ConditionEnv.
/// Phase 200-B: Will be populated by FunctionScopeCaptureAnalyzer and
/// integrated into ConditionEnv via ConditionEnvBuilder v2.
#[derive(Debug, Clone, Default)]
pub struct CapturedEnv {
/// List of captured variables
pub vars: Vec<CapturedVar>,
}
impl CapturedEnv {
/// Create a new empty environment
pub fn new() -> Self {
Self { vars: Vec::new() }
}
/// Check if the environment is empty
pub fn is_empty(&self) -> bool {
self.vars.is_empty()
}
/// Add a captured variable to the environment
pub fn add_var(&mut self, var: CapturedVar) {
self.vars.push(var);
}
/// Look up a captured variable by name
///
/// Returns `Some(&CapturedVar)` if found, `None` otherwise.
pub fn get(&self, name: &str) -> Option<&CapturedVar> {
self.vars.iter().find(|v| v.name == name)
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn test_captured_env_empty() {
let env = CapturedEnv::new();
assert!(env.is_empty());
assert!(env.get("digits").is_none());
}
#[test]
fn test_captured_env_add_and_get() {
let mut env = CapturedEnv::new();
env.add_var(CapturedVar {
name: "digits".to_string(),
host_id: ValueId(42),
is_immutable: true,
});
assert!(!env.is_empty());
let var = env.get("digits").unwrap();
assert_eq!(var.name, "digits");
assert_eq!(var.host_id, ValueId(42));
assert!(var.is_immutable);
}
#[test]
fn test_captured_env_multiple_vars() {
let mut env = CapturedEnv::new();
env.add_var(CapturedVar {
name: "digits".to_string(),
host_id: ValueId(42),
is_immutable: true,
});
env.add_var(CapturedVar {
name: "table".to_string(),
host_id: ValueId(100),
is_immutable: true,
});
assert_eq!(env.vars.len(), 2);
assert!(env.get("digits").is_some());
assert!(env.get("table").is_some());
assert!(env.get("nonexistent").is_none());
}
}