Files
hakorune/src/runner/selfhost.rs

368 lines
19 KiB
Rust
Raw Normal View History

/*!
* Runner selfhost helpers Ny compiler pipeline (Ny -> JSON v0)
*
* Transitional shim: provides a stable entrypoint from callers, while the
* heavy implementation currently lives in modes/common.rs. Next step will
* migrate the full implementation here.
*/
use super::*;
use nyash_rust::{mir::MirCompiler, parser::NyashParser};
use std::{fs, process};
impl NyashRunner {
/// Selfhost (Ny -> JSON v0) pipeline: EXE/VM/Python フォールバック含む
pub(crate) fn try_run_selfhost_pipeline(&self, filename: &str) -> bool {
use std::io::Write;
// Read input source
let code = match fs::read_to_string(filename) {
Ok(c) => c,
Err(e) => {
eprintln!("[ny-compiler] read error: {}", e);
return false;
}
};
// Optional Phase-15: strip `using` lines and register modules (same policy as execute_nyash_file)
let mut code_ref: std::borrow::Cow<'_, str> = std::borrow::Cow::Borrowed(&code);
if crate::config::env::enable_using() {
match crate::runner::modes::common_util::resolve::strip_using_and_register(self, &code, filename) {
Ok(s) => { code_ref = std::borrow::Cow::Owned(s); }
Err(e) => { eprintln!("[ny-compiler] {}", e); return false; }
}
}
// Write to tmp/ny_parser_input.ny (as expected by Ny parser v0), unless forced to reuse existing tmp
let use_tmp_only = crate::config::env::ny_compiler_use_tmp_only();
let tmp_dir = std::path::Path::new("tmp");
if let Err(e) = std::fs::create_dir_all(tmp_dir) {
eprintln!("[ny-compiler] mkdir tmp failed: {}", e);
return false;
}
// Optional macro preexpand path for selfhost
// Default: auto when macro engine is enabled (safe: PyVM only)
// Gate: NYASH_MACRO_SELFHOST_PRE_EXPAND={1|auto|0}
{
let preenv = std::env::var("NYASH_MACRO_SELFHOST_PRE_EXPAND")
.ok()
.or_else(|| if crate::r#macro::enabled() { Some("auto".to_string()) } else { None });
let do_pre = match preenv.as_deref() {
Some("1") => true,
Some("auto") => crate::r#macro::enabled() && crate::config::env::vm_use_py(),
_ => false,
};
if do_pre && crate::r#macro::enabled() {
crate::cli_v!("[ny-compiler] selfhost macro pre-expand: engaging (mode={:?})", preenv);
match NyashParser::parse_from_string(code_ref.as_ref()) {
Ok(ast0) => {
let ast = crate::r#macro::maybe_expand_and_dump(&ast0, false);
// Compile to MIR and execute (respect VM/PyVM policy similar to vm mode)
let mut mir_compiler = MirCompiler::with_options(true);
match mir_compiler.compile(ast) {
Ok(result) => {
let prefer_pyvm = crate::config::env::vm_use_py();
if prefer_pyvm {
if let Ok(code) = crate::runner::modes::common_util::pyvm::run_pyvm_harness_lib(&result.module, "selfhost-preexpand") {
println!("Result: {}", code);
std::process::exit(code);
} else {
eprintln!("❌ PyVM error (selfhost-preexpand)");
std::process::exit(1);
}
} else {
// For now, only PyVM path is supported in pre-expand mode; fall back otherwise.
crate::cli_v!("[ny-compiler] pre-expand path requires NYASH_VM_USE_PY=1; falling back to default selfhost");
return false;
}
}
Err(e) => {
eprintln!("[ny-compiler] pre-expand compile error: {}", e);
return false;
}
}
}
Err(e) => {
eprintln!("[ny-compiler] pre-expand parse error: {}", e);
return false;
}
}
}
}
let tmp_path = tmp_dir.join("ny_parser_input.ny");
if !use_tmp_only {
match std::fs::File::create(&tmp_path) {
Ok(mut f) => {
if let Err(e) = f.write_all(code_ref.as_bytes()) {
eprintln!("[ny-compiler] write tmp failed: {}", e);
return false;
}
}
Err(e) => {
eprintln!("[ny-compiler] open tmp failed: {}", e);
return false;
}
}
}
// Preferred: run Ny selfhost compiler program (apps/selfhost-compiler/compiler.nyash)
// This avoids inline embedding pitfalls and supports Stage-3 gating via args.
{
use crate::runner::modes::common_util::selfhost::{child, json};
let exe = std::env::current_exe()
.unwrap_or_else(|_| std::path::PathBuf::from("target/release/nyash"));
let parser_prog = std::path::Path::new("apps/selfhost-compiler/compiler.nyash");
if parser_prog.exists() {
// Build extra args forwarded to child program
let mut extra: Vec<&str> = Vec::new();
if crate::config::env::ny_compiler_min_json() {
extra.extend(["--", "--min-json"]);
}
extra.extend(["--", "--read-tmp"]);
if crate::config::env::ny_compiler_stage3() {
extra.extend(["--", "--stage3"]);
}
let timeout_ms: u64 = crate::config::env::ny_compiler_timeout_ms();
if let Some(line) = child::run_ny_program_capture_json(
&exe,
parser_prog,
timeout_ms,
&extra,
&["NYASH_USE_NY_COMPILER", "NYASH_CLI_VERBOSE"],
&[("NYASH_JSON_ONLY", "1")],
) {
match json::parse_json_v0_line(&line) {
Ok(module) => {
super::json_v0_bridge::maybe_dump_mir(&module);
let emit_only = crate::config::env::ny_compiler_emit_only();
if emit_only {
return false;
}
// Prefer PyVM path when requested
if crate::config::env::vm_use_py() {
if let Some(code) = crate::runner::modes::common_util::selfhost::json::run_pyvm_module(&module, "selfhost") {
println!("Result: {}", code);
std::process::exit(code);
}
}
self.execute_mir_module(&module);
return true;
}
Err(e) => {
eprintln!("[ny-compiler] json parse error (child): {}", e);
}
}
}
}
}
// Python MVP-first: prefer the lightweight harness to produce JSON v0 (unless skipped)
if std::env::var("NYASH_NY_COMPILER_SKIP_PY").ok().as_deref() != Some("1") {
if let Ok(py3) = which::which("python3") {
let py = std::path::Path::new("tools/ny_parser_mvp.py");
if py.exists() {
let mut cmd = std::process::Command::new(&py3);
cmd.arg(py).arg(&tmp_path);
let timeout_ms: u64 = std::env::var("NYASH_NY_COMPILER_TIMEOUT_MS")
.ok()
.and_then(|s| s.parse().ok())
.unwrap_or(2000);
let out = match super::modes::common_util::io::spawn_with_timeout(cmd, timeout_ms) {
Ok(o) => o,
Err(e) => { eprintln!("[ny-compiler] python harness failed: {}", e); return false; }
};
if !out.timed_out {
if let Ok(s) = String::from_utf8(out.stdout) {
if let Some(line) = crate::runner::modes::common_util::selfhost::json::first_json_v0_line(&s) {
match super::json_v0_bridge::parse_json_v0_to_module(&line) {
Ok(module) => {
super::json_v0_bridge::maybe_dump_mir(&module);
let emit_only =
std::env::var("NYASH_NY_COMPILER_EMIT_ONLY")
.unwrap_or_else(|_| "1".to_string())
== "1";
if emit_only {
return false;
}
// Prefer PyVM for selfhost pipeline (parity reference)
if std::env::var("NYASH_VM_USE_PY").ok().as_deref() == Some("1") {
let code = match crate::runner::modes::common_util::pyvm::run_pyvm_harness(&module, "selfhost-py") {
Ok(c) => c,
Err(e) => { eprintln!("❌ PyVM error: {}", e); 1 }
};
println!("Result: {}", code);
std::process::exit(code);
📚 docs: Record field declaration design discussion in papers ## Summary Documented the "init block vs fields-at-top" design discussion as a valuable example of AI-human collaboration in language design. ## Changes ### Paper G (AI Collaboration) - Added field-declaration-design.md documenting the entire discussion flow - Showcased how complex init block proposal evolved to simple "fields at top" rule - Demonstrates AI's tendency toward complexity vs human intuition for simplicity ### Paper H (AI Practical Patterns) - Added Pattern #17: "Gradual Refinement Pattern" (段階的洗練型) - Documents the process: Complex AI proposal → Detailed analysis → Human insight → Convergence - Field declaration design as a typical example ### Paper K (Explosive Incidents) - Added Incident #046: "init block vs fields-at-top incident" - Updated total count to 46 incidents - Shows how a single human comment redirected entire design approach ## Design Decision After analysis, decided that BoxIndex should remain a compiler-internal structure, not a core Box: - Core Boxes: User-instantiable runtime values (String, Integer, Array, Map) - Compiler internals: BoxIndex for name resolution (compile-time only) - Clear separation of concerns between language features and compiler tools ## Philosophy This discussion exemplifies key principles: - The best design needs no explanation - Constraints provide clarity, not limitation - "Everything is Box" doesn't mean "compiler internals are Boxes" - AI tends toward theoretical completeness; humans toward practical simplicity 🐱 Sometimes the simplest answer is right in front of us\!
2025-09-16 14:57:05 +09:00
}
self.execute_mir_module(&module);
return true;
}
Err(e) => {
eprintln!("[ny-compiler] json parse error: {}", e);
return false;
📚 docs: Record field declaration design discussion in papers ## Summary Documented the "init block vs fields-at-top" design discussion as a valuable example of AI-human collaboration in language design. ## Changes ### Paper G (AI Collaboration) - Added field-declaration-design.md documenting the entire discussion flow - Showcased how complex init block proposal evolved to simple "fields at top" rule - Demonstrates AI's tendency toward complexity vs human intuition for simplicity ### Paper H (AI Practical Patterns) - Added Pattern #17: "Gradual Refinement Pattern" (段階的洗練型) - Documents the process: Complex AI proposal → Detailed analysis → Human insight → Convergence - Field declaration design as a typical example ### Paper K (Explosive Incidents) - Added Incident #046: "init block vs fields-at-top incident" - Updated total count to 46 incidents - Shows how a single human comment redirected entire design approach ## Design Decision After analysis, decided that BoxIndex should remain a compiler-internal structure, not a core Box: - Core Boxes: User-instantiable runtime values (String, Integer, Array, Map) - Compiler internals: BoxIndex for name resolution (compile-time only) - Clear separation of concerns between language features and compiler tools ## Philosophy This discussion exemplifies key principles: - The best design needs no explanation - Constraints provide clarity, not limitation - "Everything is Box" doesn't mean "compiler internals are Boxes" - AI tends toward theoretical completeness; humans toward practical simplicity 🐱 Sometimes the simplest answer is right in front of us\!
2025-09-16 14:57:05 +09:00
}
}
}
}
}
}
}
}
// EXE-first: if requested, try external parser EXE (nyash_compiler)
if std::env::var("NYASH_USE_NY_COMPILER_EXE").ok().as_deref() == Some("1") {
// Resolve parser EXE path
let exe_path = if let Ok(p) = std::env::var("NYASH_NY_COMPILER_EXE_PATH") {
std::path::PathBuf::from(p)
} else {
let mut p = std::path::PathBuf::from("dist/nyash_compiler");
#[cfg(windows)]
{
p.push("nyash_compiler.exe");
}
#[cfg(not(windows))]
{
p.push("nyash_compiler");
}
if !p.exists() {
// Try PATH
if let Ok(w) = which::which("nyash_compiler") {
w
} else {
p
}
} else {
p
}
};
if exe_path.exists() {
let timeout_ms: u64 = std::env::var("NYASH_NY_COMPILER_TIMEOUT_MS")
.ok()
.and_then(|s| s.parse().ok())
.unwrap_or(2000);
if let Some(module) = super::modes::common_util::selfhost_exe::exe_try_parse_json_v0(filename, timeout_ms) {
super::json_v0_bridge::maybe_dump_mir(&module);
let emit_only = std::env::var("NYASH_NY_COMPILER_EMIT_ONLY")
.unwrap_or_else(|_| "1".to_string())
== "1";
if emit_only { return false; }
// Prefer PyVM when requested (reference semantics)
if std::env::var("NYASH_VM_USE_PY").ok().as_deref() == Some("1") {
if let Ok(py3) = which::which("python3") {
let runner = std::path::Path::new("tools/pyvm_runner.py");
if runner.exists() {
let tmp_dir = std::path::Path::new("tmp");
let _ = std::fs::create_dir_all(tmp_dir);
let mir_json_path = tmp_dir.join("nyash_pyvm_mir.json");
if let Err(e) = crate::runner::mir_json_emit::emit_mir_json_for_harness_bin(&module, &mir_json_path) {
eprintln!("❌ PyVM MIR JSON emit error: {}", e);
process::exit(1);
}
crate::cli_v!("[Bridge] using PyVM (selfhost) → {}", mir_json_path.display());
let entry = if module.functions.contains_key("Main.main") { "Main.main" }
else if module.functions.contains_key("main") { "main" } else { "Main.main" };
let status = std::process::Command::new(py3)
.args(["tools/pyvm_runner.py", "--in", &mir_json_path.display().to_string(), "--entry", entry])
.status()
.map_err(|e| format!("spawn pyvm: {}", e))
.unwrap();
let code = status.code().unwrap_or(1);
println!("Result: {}", code);
std::process::exit(code);
}
}
}
self.execute_mir_module(&module);
return true;
} else {
return false;
}
}
}
📚 docs: Record field declaration design discussion in papers ## Summary Documented the "init block vs fields-at-top" design discussion as a valuable example of AI-human collaboration in language design. ## Changes ### Paper G (AI Collaboration) - Added field-declaration-design.md documenting the entire discussion flow - Showcased how complex init block proposal evolved to simple "fields at top" rule - Demonstrates AI's tendency toward complexity vs human intuition for simplicity ### Paper H (AI Practical Patterns) - Added Pattern #17: "Gradual Refinement Pattern" (段階的洗練型) - Documents the process: Complex AI proposal → Detailed analysis → Human insight → Convergence - Field declaration design as a typical example ### Paper K (Explosive Incidents) - Added Incident #046: "init block vs fields-at-top incident" - Updated total count to 46 incidents - Shows how a single human comment redirected entire design approach ## Design Decision After analysis, decided that BoxIndex should remain a compiler-internal structure, not a core Box: - Core Boxes: User-instantiable runtime values (String, Integer, Array, Map) - Compiler internals: BoxIndex for name resolution (compile-time only) - Clear separation of concerns between language features and compiler tools ## Philosophy This discussion exemplifies key principles: - The best design needs no explanation - Constraints provide clarity, not limitation - "Everything is Box" doesn't mean "compiler internals are Boxes" - AI tends toward theoretical completeness; humans toward practical simplicity 🐱 Sometimes the simplest answer is right in front of us\!
2025-09-16 14:57:05 +09:00
// Fallback: inline VM run (embed source into a tiny wrapper that prints JSON)
// This avoids CLI arg forwarding complexity and does not require FileBox.
let mut json_line = String::new();
{
📚 docs: Record field declaration design discussion in papers ## Summary Documented the "init block vs fields-at-top" design discussion as a valuable example of AI-human collaboration in language design. ## Changes ### Paper G (AI Collaboration) - Added field-declaration-design.md documenting the entire discussion flow - Showcased how complex init block proposal evolved to simple "fields at top" rule - Demonstrates AI's tendency toward complexity vs human intuition for simplicity ### Paper H (AI Practical Patterns) - Added Pattern #17: "Gradual Refinement Pattern" (段階的洗練型) - Documents the process: Complex AI proposal → Detailed analysis → Human insight → Convergence - Field declaration design as a typical example ### Paper K (Explosive Incidents) - Added Incident #046: "init block vs fields-at-top incident" - Updated total count to 46 incidents - Shows how a single human comment redirected entire design approach ## Design Decision After analysis, decided that BoxIndex should remain a compiler-internal structure, not a core Box: - Core Boxes: User-instantiable runtime values (String, Integer, Array, Map) - Compiler internals: BoxIndex for name resolution (compile-time only) - Clear separation of concerns between language features and compiler tools ## Philosophy This discussion exemplifies key principles: - The best design needs no explanation - Constraints provide clarity, not limitation - "Everything is Box" doesn't mean "compiler internals are Boxes" - AI tends toward theoretical completeness; humans toward practical simplicity 🐱 Sometimes the simplest answer is right in front of us\!
2025-09-16 14:57:05 +09:00
// Escape source for embedding as string literal
let mut esc = String::with_capacity(code_ref.len());
for ch in code_ref.chars() {
match ch {
'\\' => esc.push_str("\\\\"),
'"' => esc.push_str("\\\""),
'\n' => esc.push_str("\n"),
'\r' => esc.push_str(""),
_ => esc.push(ch),
}
}
let inline_path = std::path::Path::new("tmp").join("inline_selfhost_emit.nyash");
let inline_code = format!(
"include \"apps/selfhost-compiler/boxes/parser_box.nyash\"\ninclude \"apps/selfhost-compiler/boxes/emitter_box.nyash\"\nstatic box Main {{\n main(args) {{\n local s = \"{}\"\n local p = new ParserBox()\n p.stage3_enable(1)\n local json = p.parse_program2(s)\n local e = new EmitterBox()\n json = e.emit_program(json, \"[]\")\n print(json)\n return 0\n }}\n}}\n",
📚 docs: Record field declaration design discussion in papers ## Summary Documented the "init block vs fields-at-top" design discussion as a valuable example of AI-human collaboration in language design. ## Changes ### Paper G (AI Collaboration) - Added field-declaration-design.md documenting the entire discussion flow - Showcased how complex init block proposal evolved to simple "fields at top" rule - Demonstrates AI's tendency toward complexity vs human intuition for simplicity ### Paper H (AI Practical Patterns) - Added Pattern #17: "Gradual Refinement Pattern" (段階的洗練型) - Documents the process: Complex AI proposal → Detailed analysis → Human insight → Convergence - Field declaration design as a typical example ### Paper K (Explosive Incidents) - Added Incident #046: "init block vs fields-at-top incident" - Updated total count to 46 incidents - Shows how a single human comment redirected entire design approach ## Design Decision After analysis, decided that BoxIndex should remain a compiler-internal structure, not a core Box: - Core Boxes: User-instantiable runtime values (String, Integer, Array, Map) - Compiler internals: BoxIndex for name resolution (compile-time only) - Clear separation of concerns between language features and compiler tools ## Philosophy This discussion exemplifies key principles: - The best design needs no explanation - Constraints provide clarity, not limitation - "Everything is Box" doesn't mean "compiler internals are Boxes" - AI tends toward theoretical completeness; humans toward practical simplicity 🐱 Sometimes the simplest answer is right in front of us\!
2025-09-16 14:57:05 +09:00
esc
);
if let Err(e) = std::fs::write(&inline_path, inline_code) {
eprintln!("[ny-compiler] write inline failed: {}", e);
return false;
}
let exe = std::env::current_exe()
.unwrap_or_else(|_| std::path::PathBuf::from("target/release/nyash"));
let mut cmd = std::process::Command::new(exe);
📚 docs: Record field declaration design discussion in papers ## Summary Documented the "init block vs fields-at-top" design discussion as a valuable example of AI-human collaboration in language design. ## Changes ### Paper G (AI Collaboration) - Added field-declaration-design.md documenting the entire discussion flow - Showcased how complex init block proposal evolved to simple "fields at top" rule - Demonstrates AI's tendency toward complexity vs human intuition for simplicity ### Paper H (AI Practical Patterns) - Added Pattern #17: "Gradual Refinement Pattern" (段階的洗練型) - Documents the process: Complex AI proposal → Detailed analysis → Human insight → Convergence - Field declaration design as a typical example ### Paper K (Explosive Incidents) - Added Incident #046: "init block vs fields-at-top incident" - Updated total count to 46 incidents - Shows how a single human comment redirected entire design approach ## Design Decision After analysis, decided that BoxIndex should remain a compiler-internal structure, not a core Box: - Core Boxes: User-instantiable runtime values (String, Integer, Array, Map) - Compiler internals: BoxIndex for name resolution (compile-time only) - Clear separation of concerns between language features and compiler tools ## Philosophy This discussion exemplifies key principles: - The best design needs no explanation - Constraints provide clarity, not limitation - "Everything is Box" doesn't mean "compiler internals are Boxes" - AI tends toward theoretical completeness; humans toward practical simplicity 🐱 Sometimes the simplest answer is right in front of us\!
2025-09-16 14:57:05 +09:00
cmd.arg("--backend").arg("vm").arg(&inline_path);
cmd.env_remove("NYASH_USE_NY_COMPILER");
cmd.env_remove("NYASH_CLI_VERBOSE");
cmd.env("NYASH_JSON_ONLY", "1");
let timeout_ms: u64 = std::env::var("NYASH_NY_COMPILER_TIMEOUT_MS")
.ok()
.and_then(|s| s.parse().ok())
.unwrap_or(2000);
let out = match super::modes::common_util::io::spawn_with_timeout(cmd, timeout_ms) {
Ok(o) => o,
Err(e) => { eprintln!("[ny-compiler] spawn inline vm failed: {}", e); return false; }
};
if out.timed_out {
let head = String::from_utf8_lossy(&out.stdout).chars().take(200).collect::<String>();
eprintln!("[ny-compiler] inline timeout after {} ms; stdout(head)='{}'", timeout_ms, head.replace('\n', "\\n"));
}
let stdout = String::from_utf8_lossy(&out.stdout).to_string();
if let Some(line) = crate::runner::modes::common_util::selfhost::json::first_json_v0_line(&stdout) {
json_line = line;
}
}
if json_line.is_empty() {
return false;
}
match super::json_v0_bridge::parse_json_v0_to_module(&json_line) {
Ok(module) => {
super::json_v0_bridge::maybe_dump_mir(&module);
let emit_only = std::env::var("NYASH_NY_COMPILER_EMIT_ONLY")
.unwrap_or_else(|_| "1".to_string())
== "1";
if emit_only {
return false;
}
📚 docs: Record field declaration design discussion in papers ## Summary Documented the "init block vs fields-at-top" design discussion as a valuable example of AI-human collaboration in language design. ## Changes ### Paper G (AI Collaboration) - Added field-declaration-design.md documenting the entire discussion flow - Showcased how complex init block proposal evolved to simple "fields at top" rule - Demonstrates AI's tendency toward complexity vs human intuition for simplicity ### Paper H (AI Practical Patterns) - Added Pattern #17: "Gradual Refinement Pattern" (段階的洗練型) - Documents the process: Complex AI proposal → Detailed analysis → Human insight → Convergence - Field declaration design as a typical example ### Paper K (Explosive Incidents) - Added Incident #046: "init block vs fields-at-top incident" - Updated total count to 46 incidents - Shows how a single human comment redirected entire design approach ## Design Decision After analysis, decided that BoxIndex should remain a compiler-internal structure, not a core Box: - Core Boxes: User-instantiable runtime values (String, Integer, Array, Map) - Compiler internals: BoxIndex for name resolution (compile-time only) - Clear separation of concerns between language features and compiler tools ## Philosophy This discussion exemplifies key principles: - The best design needs no explanation - Constraints provide clarity, not limitation - "Everything is Box" doesn't mean "compiler internals are Boxes" - AI tends toward theoretical completeness; humans toward practical simplicity 🐱 Sometimes the simplest answer is right in front of us\!
2025-09-16 14:57:05 +09:00
// Phase-15 policy: when NYASH_VM_USE_PY=1, prefer PyVM as reference executor
// regardless of BoxCall presence to ensure semantics parity (e.g., PHI merges).
let prefer_pyvm = std::env::var("NYASH_VM_USE_PY").ok().as_deref() == Some("1");
// Backward compatibility: if not preferring PyVM explicitly, still auto-enable when BoxCalls exist.
let needs_pyvm = !prefer_pyvm
&& module.functions.values().any(|f| {
f.blocks.values().any(|bb| {
bb.instructions.iter().any(|inst| {
matches!(inst, crate::mir::MirInstruction::BoxCall { .. })
})
})
});
📚 docs: Record field declaration design discussion in papers ## Summary Documented the "init block vs fields-at-top" design discussion as a valuable example of AI-human collaboration in language design. ## Changes ### Paper G (AI Collaboration) - Added field-declaration-design.md documenting the entire discussion flow - Showcased how complex init block proposal evolved to simple "fields at top" rule - Demonstrates AI's tendency toward complexity vs human intuition for simplicity ### Paper H (AI Practical Patterns) - Added Pattern #17: "Gradual Refinement Pattern" (段階的洗練型) - Documents the process: Complex AI proposal → Detailed analysis → Human insight → Convergence - Field declaration design as a typical example ### Paper K (Explosive Incidents) - Added Incident #046: "init block vs fields-at-top incident" - Updated total count to 46 incidents - Shows how a single human comment redirected entire design approach ## Design Decision After analysis, decided that BoxIndex should remain a compiler-internal structure, not a core Box: - Core Boxes: User-instantiable runtime values (String, Integer, Array, Map) - Compiler internals: BoxIndex for name resolution (compile-time only) - Clear separation of concerns between language features and compiler tools ## Philosophy This discussion exemplifies key principles: - The best design needs no explanation - Constraints provide clarity, not limitation - "Everything is Box" doesn't mean "compiler internals are Boxes" - AI tends toward theoretical completeness; humans toward practical simplicity 🐱 Sometimes the simplest answer is right in front of us\!
2025-09-16 14:57:05 +09:00
if prefer_pyvm || needs_pyvm {
let label = if prefer_pyvm { "selfhost" } else { "selfhost-fallback" };
if let Some(code) = crate::runner::modes::common_util::selfhost::json::run_pyvm_module(&module, label) {
println!("Result: {}", code);
std::process::exit(code);
}
}
self.execute_mir_module(&module);
true
}
Err(e) => {
eprintln!("❌ JSON v0 bridge error: {}", e);
false
}
}
}
}