refactor: name recovery boundaries and compact IR strings

README.md

@@ -55,6 +55,6 @@ type errors carry source spans and diagnostic identities; invalid operands propa
 
 `lower.lower` turns the checked AST and binding table into an owned `@source/ir` module. functions, globals, local slots and instruction-result temporaries have explicit IDs; typed instructions execute in source evaluation order and basic blocks end in jump, conditional-branch, return or diagnostic-trap terminators. conditional arms join only when execution can continue; local slots and temporaries are function-wide, without SSA or phi nodes. literals are parsed and strings decoded during lowering; the module retains no AST, binding-table or source-buffer references. release it with `ir.deinit`; keep and reuse the diagnostic store used during lowering so diagnostic IDs remain valid. the IR is target-independent and is intended to feed later native backends as well as the current interpreter.
 
-`interpreter.run` consumes only IR and returns a value or a diagnostic trap; returned strings borrow the module. heap-backed invocation frames track their current block and resume it after calls or global initialization. parameters are initialized on entry and locals execute in statement order; premature local reads trap. globals initialize lazily on their first read, cache their result for that run, and trap on initialization cycles. integers use signed 64-bit storage with checked arithmetic; floats use 64-bit IEEE arithmetic and comparisons, including NaN behavior. numeric literal overflow and float underflow to zero are diagnosed; representable subnormals are accepted. strings support `\\`, `\"`, `\n`, `\r`, `\t`, `\0` and `\xNN` escapes.
+`interpreter.run` consumes only IR and returns a value or a diagnostic trap. string values carry a module-local `ir.StringId`, not a slice; resolve bytes with `ir.string_bytes(&module, id)` before releasing the module. decoded bytes and their full backing allocations remain module-owned, without copying strings during calls or returns. heap-backed invocation frames track their current block and resume it after calls or global initialization. parameters are initialized on entry and locals execute in statement order; premature local reads trap. globals initialize lazily on their first read, cache their result for that run, and trap on initialization cycles. integers use signed 64-bit storage with checked arithmetic; floats use 64-bit IEEE arithmetic and comparisons, including NaN behavior. numeric literal overflow and float underflow to zero are diagnosed; representable subnormals are accepted. strings support `\\`, `\"`, `\n`, `\r`, `\t`, `\0` and `\xNN` escapes.
 
 recoverable source errors do not stop execution of unaffected code. unused invalid functions, globals and conditional arms retain diagnostic traps; reaching invalid code reports its original diagnostic identity. missing-return traps affect only paths that fall through, not arms that already return or trap. diagnostics include the filename, line and byte column. exit status is `0` for clean execution, `1` for source diagnostics or a runtime trap, and `2` for usage or infrastructure failure. native code generation, loops, assignment and boolean logical operators remain unimplemented.

source/interpreter/interpreter.hon

@@ -35,7 +35,7 @@ Frame :: struct {
 }
 
 #! executes only typed IR; invocation and global state are temporary and released on every exit.
-#! returned strings borrow `module`; runtime source errors are traps, not infrastructure errors.
+#! returned string IDs resolve through `module`; runtime source errors are traps, not infrastructure errors.
 run proc(
     allocator mem.Allocator,
     module @ir.Module,

source/interpreter/interpreter.test.hon

@@ -153,12 +153,18 @@ function_aliases_and_borrowed_string_values_execute test {
     store := diagnostics.init(mem.c_allocator)
     defer diagnostics.deinit(&store)
     text :: "line\n\"quoted\"\x00"
+    owned :: try mem.alloc(u8, module.allocator, text.len)
+    transferred := false
+    defer if (!transferred) mem.free(module.allocator, owned)
+    memcopy!(owned, text)
+    try arraylist.append(&module.strings, ir.StringStorage{ allocation = owned, len = text.len })
+    transferred = true
     target :: try add_function(&module, mem.empty(ir.Type), 0, .string, 1)
     try emit(&module, target, ir.Instruction{
         op = .constant,
         type = .string,
         destination = ir.ValueId(0),
-        constant = ir.Value{ type = .string, data = ir.ValueData{ string = text } },
+        constant = ir.Value{ type = .string, data = ir.ValueData{ string = ir.StringId(0) } },
     })
     return_value(&module, target, ir.ValueId(0))
     initializer :: try add_function(&module, mem.empty(ir.Type), 0, .function, 1)
@@ -173,7 +179,7 @@ function_aliases_and_borrowed_string_values_execute test {
     outcome :: try run(mem.c_allocator, &module, &store)
     try testing.expect(outcome.trap == null)
     try testing.expect_equal(ir.Type.string, outcome.value.type)
-    try testing.expect_equal([]u8, text, outcome.value.data.string)
+    try testing.expect_equal([]u8, text, ir.string_bytes(&module, outcome.value.data.string))
 }
 
 premature_local_reads_and_global_cycles_report_instruction_spans test {

source/ir/ir.hon

@@ -7,6 +7,7 @@ GlobalId :: distinct u32
 LocalId :: distinct u32
 ValueId :: distinct u32
 BlockId :: distinct u32
+StringId :: distinct u32
 no_value :: maxval!(ValueId)
 
 Type :: enum {
@@ -23,7 +24,7 @@ ValueData :: union {
     integer i64
     real f64
     boolean bool
-    string []u8
+    string StringId
     function FunctionId
 }
 Value :: struct {
@@ -101,13 +102,19 @@ Global :: struct {
     span diagnostics.Span
 }
 
+#! retains the original allocation for freeing while exposing only decoded bytes.
+StringStorage :: struct {
+    allocation []mut u8
+    len usize
+}
+
 #! owns executable storage, including decoded strings; no AST or binding table is retained.
 #! spans and diagnostic IDs refer to the caller's source and diagnostic store.
 Module :: struct {
     allocator mem.Allocator
     functions arraylist.ArrayList(Function)
     globals arraylist.ArrayList(Global)
-    strings arraylist.ArrayList([]mut u8)
+    strings arraylist.ArrayList(StringStorage)
     entry ?FunctionId = null
     entry_diagnostic diagnostics.DiagnosticId = diagnostics.no_diagnostic
 }
@@ -136,6 +143,12 @@ init_block proc(allocator mem.Allocator) Block {
     }
 }
 
+#! resolves a module-local ID to bytes borrowed until module deinit.
+string_bytes proc(module @Module, id StringId) []u8 {
+    string :: module.strings.items[usize(id)]
+    return string.allocation[..string.len]
+}
+
 #! releases all executable storage using the retained allocator.
 deinit proc(module @mut Module) void {
     for (0..(module.functions.items.len)) |i| {
@@ -149,7 +162,9 @@ deinit proc(module @mut Module) void {
         arraylist.deinit(&function.locals)
         arraylist.deinit(&function.blocks)
     }
-    for (module.strings.items) |string| mem.free(module.allocator, string)
+    for (module.strings.items) |string|
+        mem.free(module.allocator, string.allocation)
+
     arraylist.deinit(&module.functions)
     arraylist.deinit(&module.globals)
     arraylist.deinit(&module.strings)

source/lower/lower.hon

@@ -560,8 +560,17 @@ lower_string proc(state @mut State, id ast.NodeId) ir.ValueId ! (mem.AllocError
         trap(state, try diagnostics.add(state.diagnostics, .invalid_escape, span), span)
         return ir.no_value
     }
-    # retain the complete allocation for freeing, but expose only decoded bytes.
+    if (state.module.strings.items.len > usize(maxval!(ir.StringId)))
-    try arraylist.append(&state.module.strings, owned)
+        return .out_of_memory
+
+    string_id :: ir.StringId(state.module.strings.items.len)
+    try arraylist.append(&state.module.strings, ir.StringStorage{
+        allocation = owned,
+        len = length,
+    })
     transferred = true
-    return try constant(state, ir.Value{ type = .string, data = ir.ValueData{ string = owned[..length] } }, span)
+    return try constant(state, ir.Value{
+        type = .string,
+        data = ir.ValueData{ string = string_id },
+    }, span)
 }

source/lower/lower.test.hon

@@ -91,7 +91,20 @@ owns_decoded_strings_after_source_and_frontend_are_freed test {
     outcome :: try interpreter.run(mem.c_allocator, &module, &diagnostic_store)
     try testing.expect(outcome.trap == null)
     try testing.expect_equal(ir.Type.string, outcome.value.type)
-    try testing.expect(mem.eql(outcome.value.data.string, "a\\\"\n\r\t\x00Az"))
+    try testing.expect(mem.eql(ir.string_bytes(&module, outcome.value.data.string), "a\\\"\n\r\t\x00Az"))
+}
+
+resolves_module_local_string_ids_after_frontend_cleanup test {
+    diagnostic_store := diagnostics.init(mem.c_allocator)
+    defer diagnostics.deinit(&diagnostic_store)
+    module := try compile_owned(
+        "first :: \"not the result\"\nidentity proc(value string) string { return value }\nmain proc() string { return identity(\"second\\n\\x00\") }\n",
+        &diagnostic_store,
+    )
+    defer ir.deinit(&module)
+    outcome :: try interpreter.run(mem.c_allocator, &module, &diagnostic_store)
+    try testing.expect(outcome.trap == null)
+    try testing.expect_equal([]u8, "second\n\x00", ir.string_bytes(&module, outcome.value.data.string))
 }
 
 honors_float_context_and_float_division test {

source/main.hon

@@ -133,19 +133,19 @@ run_file proc(allocator mem.Allocator, runtime io.Io, path [;0]u8) i32 ! DriverE
     module := try compile(allocator, source.items, &diagnostic_store)
     defer ir.deinit(&module)
     outcome :: try interpreter.run(allocator, &module, &diagnostic_store)
-    if (outcome.trap == null) try print_result(io.stdout(runtime), outcome.value)
+    if (outcome.trap == null) try print_result(io.stdout(runtime), &module, outcome.value)
     try report_diagnostics(allocator, io.stderr(runtime), path, source.items, &diagnostic_store, outcome.trap)
     if (outcome.trap != null or diagnostics.count(&diagnostic_store) != 0) return 1
     return 0
 }
 
 @hide:file
-print_result proc(output io.Writer, value ir.Value) void ! io.WriteError {
+print_result proc(output io.Writer, module @ir.Module, value ir.Value) void ! io.WriteError {
     match value.type {
         .int: try io.print("{}\n", output, {value.data.integer})
         .float: try io.print("{d}\n", output, {value.data.real})
         .bool: try io.print("{}\n", output, {value.data.boolean})
-        .string: try io.print("{s}\n", output, {value.data.string})
+        .string: try io.print("{s}\n", output, {ir.string_bytes(module, value.data.string)})
         .void: {}
         else: unreachable
     }

source/parser/parse.hon

@@ -408,16 +408,7 @@ parse_call proc(state @mut State, callee ast.NodeId) ast.NodeId ! ParseError {
     arguments arraylist.ArrayList(ast.NodeId) := arraylist.init(state.allocator)
     defer arraylist.deinit(&arguments)
 
-    while (
+    while (!is_expression_boundary(current(state).kind)) {
-        !next_is(state, .close_paren)
-        and !next_is(state, .newline)
-        and !next_is(state, .close_curly)
-        and !next_is(state, .open_curly)
-        and !next_is(state, .else)
-        and !next_is(state, .return)
-        and !next_is(state, .if)
-        and !next_is(state, .eof)
-    ) {
         argument :: try parse_expr(state)
         try arraylist.append(&arguments, argument)
         if (!next_is(state, .comma)) break
@@ -482,17 +473,7 @@ parse_primary proc(state @mut State) ast.NodeId ! ParseError {
             else
                 try diagnostics.add(state.diagnostics, .unexpected_token, token.span)
 
-            if (
+            if token.kind != .comma and !is_expression_boundary(token.kind) {
-                token.kind != .newline
-                and token.kind != .comma
-                and token.kind != .close_paren
-                and token.kind != .close_curly
-                and token.kind != .open_curly
-                and token.kind != .else
-                and token.kind != .return
-                and token.kind != .if
-                and token.kind != .eof
-            ) {
                 state.next_token += 1
             }
 
@@ -501,6 +482,14 @@ parse_primary proc(state @mut State) ast.NodeId ! ParseError {
     }
 }
 
+@hide:file
+is_expression_boundary proc(kind lexer.TokenKind) bool {
+    return match kind {
+        .close_paren, .newline, .close_curly, .open_curly, .else, .return, .if, .eof: true
+        else: false
+    }
+}
+
 @hide:file
 recover_decl proc(
     state @mut State,