RustPython/crates/vm/src/vm/mod.rs at e36962b92ad8b09de306e1684e413d837f5067a9 · youknowone/RustPython

Repository navigation

History

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638

639

640

641

642

643

644

645

646

647

648

649

650

651

652

653

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671

672

673

674

675

676

677

678

679

680

681

682

683

684

685

686

687

688

689

690

691

692

693

694

695

696

697

698

699

700

701

702

703

704

705

706

707

708

709

710

711

712

713

714

715

716

717

718

719

720

721

722

723

724

725

726

727

728

729

730

731

732

733

734

735

736

737

738

739

740

741

742

743

744

745

746

747

748

749

750

751

752

753

754

755

756

757

758

759

760

761

762

763

764

765

766

767

768

769

770

771

772

773

774

775

776

777

778

779

780

781

782

783

784

785

786

787

788

789

790

791

792

793

794

795

796

797

798

799

800

801

802

803

804

805

806

807

808

809

810

811

812

813

814

815

816

817

818

819

820

821

822

823

824

825

826

827

828

829

830

831

832

833

834

835

836

837

838

839

840

841

842

843

844

845

846

847

848

849

850

851

852

853

854

855

856

857

858

859

860

861

862

863

864

865

866

867

868

869

870

871

872

873

874

875

876

877

878

879

880

881

882

883

884

885

886

887

888

889

890

891

892

893

894

895

896

897

898

899

900

901

902

903

904

905

906

907

908

909

910

911

912

913

914

915

916

917

918

919

920

921

922

923

924

925

926

927

928

929

930

931

932

933

934

935

936

937

938

939

940

941

942

943

944

945

946

947

948

949

950

951

952

953

954

955

956

957

958

959

960

961

962

963

964

965

966

967

968

969

970

971

972

973

974

975

976

977

978

979

980

981

982

983

984

985

986

987

988

989

990

991

992

993

994

995

996

997

998

999

1000

//! Implement virtual machine to run instructions.

//!

//! See also:

//! <https://github.com/ProgVal/pythonvm-rust/blob/master/src/processor/mod.rs>

#[cfg(feature = "rustpython-compiler")]

mod compile;

mod context;

mod interpreter;

mod method;

#[cfg(feature = "rustpython-compiler")]

mod python_run;

mod setting;

pub mod thread;

mod vm_new;

mod vm_object;

mod vm_ops;

use crate::{

AsObject, Py, PyObject, PyObjectRef, PyPayload, PyRef, PyResult,

builtins::{

self, PyBaseExceptionRef, PyDict, PyDictRef, PyInt, PyList, PyModule, PyStr, PyStrInterned,

PyStrRef, PyTypeRef, PyUtf8Str, PyUtf8StrInterned, PyWeak,

code::PyCode,

dict::{PyDictItems, PyDictKeys, PyDictValues},

pystr::AsPyStr,

tuple::PyTuple,

codecs::CodecsRegistry,

common::{hash::HashSecret, lock::PyMutex, rc::PyRc},

convert::ToPyObject,

exceptions::types::PyBaseException,

frame::{ExecutionResult, Frame, FrameRef},

frozen::FrozenModule,

function::{ArgMapping, FuncArgs, PySetterValue},

import,

protocol::PyIterIter,

scope::Scope,

signal, stdlib,

warn::WarningsState,

};

use alloc::{borrow::Cow, collections::BTreeMap};

use core::{

cell::{Cell, OnceCell, RefCell},

ptr::NonNull,

sync::atomic::{AtomicBool, AtomicU64, Ordering},

};

use crossbeam_utils::atomic::AtomicCell;

#[cfg(unix)]

use nix::{

sys::signal::{SaFlags, SigAction, SigSet, Signal::SIGINT, kill, sigaction},

unistd::getpid,

};

use std::{

collections::{HashMap, HashSet},

ffi::{OsStr, OsString},

};

pub use context::Context;

pub use interpreter::{Interpreter, InterpreterBuilder};

pub(crate) use method::PyMethod;

pub use setting::{CheckHashPycsMode, Paths, PyConfig, Settings};

pub const MAX_MEMORY_SIZE: usize = isize::MAX as usize;

// Objects are live when they are on stack, or referenced by a name (for now)

/// Top level container of a python virtual machine. In theory you could

/// create more instances of this struct and have them operate fully isolated.

///

/// To construct this, please refer to the [`Interpreter`]

pub struct VirtualMachine {

pub builtins: PyRef<PyModule>,

pub sys_module: PyRef<PyModule>,

pub ctx: PyRc<Context>,

pub frames: RefCell<Vec<FramePtr>>,

pub wasm_id: Option<String>,

exceptions: RefCell<ExceptionStack>,

pub import_func: PyObjectRef,

pub(crate) importlib: PyObjectRef,

pub profile_func: RefCell<PyObjectRef>,

pub trace_func: RefCell<PyObjectRef>,

pub use_tracing: Cell<bool>,

pub recursion_limit: Cell<usize>,

pub(crate) signal_handlers: OnceCell<Box<RefCell<[Option<PyObjectRef>; signal::NSIG]>>>,

pub(crate) signal_rx: Option<signal::UserSignalReceiver>,

pub repr_guards: RefCell<HashSet<usize>>,

pub state: PyRc<PyGlobalState>,

pub initialized: bool,

recursion_depth: Cell<usize>,

/// C stack soft limit for detecting stack overflow (like c_stack_soft_limit)

c_stack_soft_limit: Cell<usize>,

/// Async generator firstiter hook (per-thread, set via sys.set_asyncgen_hooks)

pub async_gen_firstiter: RefCell<Option<PyObjectRef>>,

/// Async generator finalizer hook (per-thread, set via sys.set_asyncgen_hooks)

pub async_gen_finalizer: RefCell<Option<PyObjectRef>>,

/// Current running asyncio event loop for this thread

pub asyncio_running_loop: RefCell<Option<PyObjectRef>>,

/// Current running asyncio task for this thread

pub asyncio_running_task: RefCell<Option<PyObjectRef>>,

}

/// Non-owning frame pointer for the frames stack.

/// The pointed-to frame is kept alive by the caller of with_frame_exc/resume_gen_frame.

#[derive(Copy, Clone)]

pub struct FramePtr(NonNull<Py<Frame>>);

impl FramePtr {

/// # Safety

/// The pointed-to frame must still be alive.

pub unsafe fn as_ref(&self) -> &Py<Frame> {

unsafe { self.0.as_ref() }

}

// SAFETY: FramePtr is only stored in the VM's frames Vec while the corresponding

// FrameRef is alive on the call stack. The Vec is always empty when the VM moves between threads.

unsafe impl Send for FramePtr {}

#[derive(Debug, Default)]

struct ExceptionStack {

stack: Vec<Option<PyBaseExceptionRef>>,

}

pub struct PyGlobalState {

pub config: PyConfig,

pub module_defs: BTreeMap<&'static str, &'static builtins::PyModuleDef>,

pub frozen: HashMap<&'static str, FrozenModule, ahash::RandomState>,

pub stacksize: AtomicCell<usize>,

pub thread_count: AtomicCell<usize>,

pub hash_secret: HashSecret,

pub atexit_funcs: PyMutex<Vec<(PyObjectRef, FuncArgs)>>,

pub codec_registry: CodecsRegistry,

pub finalizing: AtomicBool,

pub warnings: WarningsState,

pub override_frozen_modules: AtomicCell<isize>,

pub before_forkers: PyMutex<Vec<PyObjectRef>>,

pub after_forkers_child: PyMutex<Vec<PyObjectRef>>,

pub after_forkers_parent: PyMutex<Vec<PyObjectRef>>,

pub int_max_str_digits: AtomicCell<usize>,

pub switch_interval: AtomicCell<f64>,

/// Global trace function for all threads (set by sys._settraceallthreads)

pub global_trace_func: PyMutex<Option<PyObjectRef>>,

/// Global profile function for all threads (set by sys._setprofileallthreads)

pub global_profile_func: PyMutex<Option<PyObjectRef>>,

/// Main thread identifier (pthread_self on Unix)

#[cfg(feature = "threading")]

pub main_thread_ident: AtomicCell<u64>,

/// Registry of all threads' slots for sys._current_frames() and sys._current_exceptions()

#[cfg(feature = "threading")]

pub thread_frames: parking_lot::Mutex<HashMap<u64, stdlib::thread::CurrentFrameSlot>>,

/// Registry of all ThreadHandles for fork cleanup

#[cfg(feature = "threading")]

pub thread_handles: parking_lot::Mutex<Vec<stdlib::thread::HandleEntry>>,

/// Registry for non-daemon threads that need to be joined at shutdown

#[cfg(feature = "threading")]

pub shutdown_handles: parking_lot::Mutex<Vec<stdlib::thread::ShutdownEntry>>,

/// sys.monitoring state (tool names, events, callbacks)

pub monitoring: PyMutex<stdlib::sys::monitoring::MonitoringState>,

/// Fast-path mask: OR of all tools' events. 0 means no monitoring overhead.

pub monitoring_events: stdlib::sys::monitoring::MonitoringEventsMask,

/// Incremented on every monitoring state change. Code objects compare their

/// local version against this to decide whether re-instrumentation is needed.

pub instrumentation_version: AtomicU64,

}

pub fn process_hash_secret_seed() -> u32 {

use std::sync::OnceLock;

static SEED: OnceLock<u32> = OnceLock::new();

// os_random is expensive, but this is only ever called once

*SEED.get_or_init(|| u32::from_ne_bytes(rustpython_common::rand::os_random()))

}

impl VirtualMachine {

/// Check whether the current thread is the main thread.

/// Mirrors `_Py_ThreadCanHandleSignals`.

#[allow(dead_code)]

pub(crate) fn is_main_thread(&self) -> bool {

#[cfg(feature = "threading")]

{

crate::stdlib::thread::get_ident() == self.state.main_thread_ident.load()

}

#[cfg(not(feature = "threading"))]

{

true

}

/// Create a new `VirtualMachine` structure.

pub(crate) fn new(ctx: PyRc<Context>, state: PyRc<PyGlobalState>) -> Self {

flame_guard!("new VirtualMachine");

// make a new module without access to the vm; doesn't

// set __spec__, __loader__, etc. attributes

let new_module = |def| {

PyRef::new_ref(

PyModule::from_def(def),

ctx.types.module_type.to_owned(),

Some(ctx.new_dict()),

)

};

// Hard-core modules:

let builtins = new_module(stdlib::builtins::module_def(&ctx));

let sys_module = new_module(stdlib::sys::module_def(&ctx));

let import_func = ctx.none();

let importlib = ctx.none();

let profile_func = RefCell::new(ctx.none());

let trace_func = RefCell::new(ctx.none());

let signal_handlers = OnceCell::from(signal::new_signal_handlers());

let vm = Self {

builtins,

sys_module,

ctx,

frames: RefCell::new(vec![]),

wasm_id: None,

exceptions: RefCell::default(),

import_func,

importlib,

profile_func,

trace_func,

use_tracing: Cell::new(false),

recursion_limit: Cell::new(if cfg!(debug_assertions) { 256 } else { 1000 }),

signal_handlers,

signal_rx: None,

repr_guards: RefCell::default(),

state,

initialized: false,

recursion_depth: Cell::new(0),

c_stack_soft_limit: Cell::new(Self::calculate_c_stack_soft_limit()),

async_gen_firstiter: RefCell::new(None),

async_gen_finalizer: RefCell::new(None),

asyncio_running_loop: RefCell::new(None),

asyncio_running_task: RefCell::new(None),

};

if vm.state.hash_secret.hash_str("")

!= vm

.ctx

.interned_str("")

.expect("empty str must be interned")

.hash(&vm)

{

panic!("Interpreters in same process must share the hash seed");

}

vm.builtins.init_dict(

vm.ctx.intern_str("builtins"),

Some(vm.ctx.intern_str(stdlib::builtins::DOC.unwrap()).to_owned()),

&vm,

);

vm.sys_module.init_dict(

vm.ctx.intern_str("sys"),

Some(vm.ctx.intern_str(stdlib::sys::DOC.unwrap()).to_owned()),

&vm,

);

// let name = vm.sys_module.get_attr("__name__", &vm).unwrap();

}

/// set up the encodings search function

/// init_importlib must be called before this call

#[cfg(feature = "encodings")]

fn import_encodings(&mut self) -> PyResult<()> {

self.import("encodings", 0).map_err(|import_err| {

let rustpythonpath_env = std::env::var("RUSTPYTHONPATH").ok();

let pythonpath_env = std::env::var("PYTHONPATH").ok();

let env_set = rustpythonpath_env.as_ref().is_some() || pythonpath_env.as_ref().is_some();

let path_contains_env = self.state.config.paths.module_search_paths.iter().any(|s| {

Some(s.as_str()) == rustpythonpath_env.as_deref() || Some(s.as_str()) == pythonpath_env.as_deref()

});

let guide_message = if cfg!(feature = "freeze-stdlib") {

"`rustpython_pylib` may not be set while using `freeze-stdlib` feature. Try using `rustpython::InterpreterBuilder::init_stdlib` or manually call `builder.add_frozen_modules(rustpython_pylib::FROZEN_STDLIB)` in `rustpython_vm::Interpreter::builder()`."

} else if !env_set {

"Neither RUSTPYTHONPATH nor PYTHONPATH is set. Try setting one of them to the stdlib directory."

} else if path_contains_env {

"RUSTPYTHONPATH or PYTHONPATH is set, but it doesn't contain the encodings library. If you are customizing the RustPython vm/interpreter, try adding the stdlib directory to the path. If you are developing the RustPython interpreter, it might be a bug during development."

} else {

"RUSTPYTHONPATH or PYTHONPATH is set, but it wasn't loaded to `PyConfig::paths::module_search_paths`. If you are going to customize the RustPython vm/interpreter, those environment variables are not loaded in the Settings struct by default. Please try creating a customized instance of the Settings struct. If you are developing the RustPython interpreter, it might be a bug during development."

};

let mut msg = format!(

"RustPython could not import the encodings module. It usually means something went wrong. Please carefully read the following messages and follow the steps.\n\

\n\

{guide_message}");

if !cfg!(feature = "freeze-stdlib") {

msg += "\n\

If you don't have access to a consistent external environment (e.g. targeting wasm, embedding \

rustpython in another application), try enabling the `freeze-stdlib` feature.\n\

If this is intended and you want to exclude the encodings module from your interpreter, please remove the `encodings` feature from `rustpython-vm` crate.";

}

let err = self.new_runtime_error(msg);

err.set___cause__(Some(import_err));

err

})?;

Ok(())

}

fn import_ascii_utf8_encodings(&mut self) -> PyResult<()> {

// Use the Python import machinery (FrozenImporter) so modules get

// proper __spec__ and __loader__ attributes.

self.import("codecs", 0)?;

// Use dotted names when freeze-stdlib is enabled (modules come from Lib/encodings/),

// otherwise use underscored names (modules come from core_modules/).

let (ascii_module_name, utf8_module_name) = if cfg!(feature = "freeze-stdlib") {

("encodings.ascii", "encodings.utf_8")

} else {

("encodings_ascii", "encodings_utf_8")

};

// Register ascii encoding

// __import__("encodings.ascii") returns top-level "encodings", so

// look up the actual submodule in sys.modules.

self.import(ascii_module_name, 0)?;

let sys_modules = self.sys_module.get_attr(identifier!(self, modules), self)?;

let ascii_module = sys_modules.get_item(ascii_module_name, self)?;

let getregentry = ascii_module.get_attr("getregentry", self)?;

let codec_info = getregentry.call((), self)?;

self.state

.codec_registry

.register_manual("ascii", codec_info.try_into_value(self)?)?;

// Register utf-8 encoding (also as "utf8" alias since normalize_encoding_name

// maps "utf-8" → "utf_8" but leaves "utf8" as-is)

self.import(utf8_module_name, 0)?;

let utf8_module = sys_modules.get_item(utf8_module_name, self)?;

let getregentry = utf8_module.get_attr("getregentry", self)?;

let codec_info = getregentry.call((), self)?;

let utf8_codec: crate::codecs::PyCodec = codec_info.try_into_value(self)?;

self.state

.codec_registry

.register_manual("utf-8", utf8_codec.clone())?;

self.state

.codec_registry

.register_manual("utf8", utf8_codec)?;

// Register latin-1 / iso8859-1 aliases needed very early for stdio

// bootstrap (e.g. PYTHONIOENCODING=latin-1).

if cfg!(feature = "freeze-stdlib") {

self.import("encodings.latin_1", 0)?;

let latin1_module = sys_modules.get_item("encodings.latin_1", self)?;

let getregentry = latin1_module.get_attr("getregentry", self)?;

let codec_info = getregentry.call((), self)?;

let latin1_codec: crate::codecs::PyCodec = codec_info.try_into_value(self)?;

for name in ["latin-1", "latin_1", "latin1", "iso8859-1", "iso8859_1"] {

self.state

.codec_registry

.register_manual(name, latin1_codec.clone())?;

}

Ok(())

}

fn initialize(&mut self) {

flame_guard!("init VirtualMachine");

if self.initialized {

panic!("Double Initialize Error");

}

// Initialize main thread ident before any threading operations

#[cfg(feature = "threading")]

stdlib::thread::init_main_thread_ident(self);

stdlib::builtins::init_module(self, &self.builtins);

stdlib::sys::init_module(self, &self.sys_module, &self.builtins);

self.expect_pyresult(

stdlib::sys::set_bootstrap_stderr(self),

"failed to initialize bootstrap stderr",

);

let mut essential_init = || -> PyResult {

import::import_builtin(self, "_typing")?;

#[cfg(all(not(target_arch = "wasm32"), feature = "host_env"))]

import::import_builtin(self, "_signal")?;

#[cfg(any(feature = "parser", feature = "compiler"))]

import::import_builtin(self, "_ast")?;

#[cfg(not(feature = "threading"))]

import::import_frozen(self, "_thread")?;

let importlib = import::init_importlib_base(self)?;

self.import_ascii_utf8_encodings()?;

{

let io = import::import_builtin(self, "_io")?;

// Full stdio: FileIO → BufferedWriter → TextIOWrapper

#[cfg(all(feature = "host_env", feature = "stdio"))]

let make_stdio = |name: &str, fd: i32, write: bool| -> PyResult<PyObjectRef> {

let buffered_stdio = self.state.config.settings.buffered_stdio;

let unbuffered = write && !buffered_stdio;

let buf = crate::stdlib::io::open(

self.ctx.new_int(fd).into(),

Some(if write { "wb" } else { "rb" }),

crate::stdlib::io::OpenArgs {

buffering: if unbuffered { 0 } else { -1 },

closefd: false,

..Default::default()

self,

)?;

let raw = if unbuffered {

buf.clone()

} else {

buf.get_attr("raw", self)?

};

raw.set_attr("name", self.ctx.new_str(format!("<{name}>")), self)?;

let isatty = self.call_method(&raw, "isatty", ())?.is_true(self)?;

let write_through = !buffered_stdio;

let line_buffering = buffered_stdio && (isatty || fd == 2);

let newline = if cfg!(windows) { None } else { Some("\n") };

let encoding = self.state.config.settings.stdio_encoding.as_deref();

// stderr always uses backslashreplace (ignores stdio_errors)

let errors = if fd == 2 {

Some("backslashreplace")

} else {

self.state.config.settings.stdio_errors.as_deref().or(

if self.state.config.settings.stdio_encoding.is_some() {

Some("strict")

} else {

Some("surrogateescape")

)

};

let stdio = self.call_method(

&io,

"TextIOWrapper",

(

buf,

encoding,

errors,

newline,

line_buffering,

write_through,

)?;

let mode = if write { "w" } else { "r" };

stdio.set_attr("mode", self.ctx.new_str(mode), self)?;

Ok::<_, self::PyBaseExceptionRef>(stdio)

};

// Sandbox stdio: lightweight wrapper using Rust's std::io directly

#[cfg(all(not(feature = "host_env"), feature = "stdio"))]

let make_stdio = |name: &str, fd: i32, write: bool| {

let mode = if write { "w" } else { "r" };

let stdio = stdlib::sys::SandboxStdio {

fd,

name: format!("<{name}>"),

mode: mode.to_owned(),

}

.into_ref(&self.ctx);

Ok(stdio.into())

};

// No stdio: set to None (embedding use case)

#[cfg(not(feature = "stdio"))]

let make_stdio = |_name: &str, _fd: i32, _write: bool| {

Ok(crate::builtins::PyNone.into_pyobject(self))

};

let set_stdio = |name, fd, write| {

let stdio: PyObjectRef = make_stdio(name, fd, write)?;

let dunder_name = self.ctx.intern_str(format!("__{name}__"));

self.sys_module.set_attr(

dunder_name, // e.g. __stdin__

stdio.clone(),

self,

)?;

self.sys_module.set_attr(name, stdio, self)?;

Ok(())

};

set_stdio("stdin", 0, false)?;

set_stdio("stdout", 1, true)?;

set_stdio("stderr", 2, true)?;

let io_open = io.get_attr("open", self)?;

self.builtins.set_attr("open", io_open, self)?;

}

Ok(importlib)

};

let res = essential_init();

let importlib = self.expect_pyresult(res, "essential initialization failed");

#[cfg(feature = "host_env")]

if self.state.config.settings.allow_external_library

&& cfg!(feature = "rustpython-compiler")

&& let Err(e) = import::init_importlib_package(self, importlib)

{

eprintln!(

"importlib initialization failed. This is critical for many complicated packages."

);

self.print_exception(e);

}

#[cfg(not(feature = "host_env"))]

let _ = importlib;

let _expect_stdlib = cfg!(feature = "freeze-stdlib")

|| !self.state.config.paths.module_search_paths.is_empty();

#[cfg(feature = "encodings")]

if _expect_stdlib {

if let Err(e) = self.import_encodings() {

eprintln!(

"encodings initialization failed. Only utf-8 encoding will be supported."

);

self.print_exception(e);

}

} else {

// Here may not be the best place to give general `path_list` advice,

// but bare rustpython_vm::VirtualMachine users skipped proper settings must hit here while properly setup vm never enters here.

eprintln!(

"feature `encodings` is enabled but `paths.module_search_paths` is empty. \

Please add the library path to `settings.path_list`. If you intended to disable the entire standard library (including the `encodings` feature), please also make sure to disable the `encodings` feature.\n\

Tip: You may also want to add `\"\"` to `settings.path_list` in order to enable importing from the current working directory."

);

}

self.initialized = true;

}

/// Set the custom signal channel for the interpreter

pub fn set_user_signal_channel(&mut self, signal_rx: signal::UserSignalReceiver) {

self.signal_rx = Some(signal_rx);

}

/// Execute Python bytecode (`.pyc`) from an in-memory buffer.

///

/// When the RustPython CLI is available, `.pyc` files are normally executed by

/// invoking `rustpython <input>.pyc`. This method provides an alternative for

/// environments where the binary is unavailable or file I/O is restricted

/// (e.g. WASM).

///

/// ## Preparing a `.pyc` file

///

/// First, compile a Python source file into bytecode:

///

/// ```sh

/// # Generate a .pyc file

/// $ rustpython -m py_compile <input>.py

/// ```

///

/// ## Running the bytecode

///

/// Load the resulting `.pyc` file into memory and execute it using the VM:

///

/// ```no_run

/// use rustpython_vm::Interpreter;

/// Interpreter::without_stdlib(Default::default()).enter(|vm| {

/// let bytes = std::fs::read("__pycache__/<input>.rustpython-314.pyc").unwrap();

/// let main_scope = vm.new_scope_with_main().unwrap();

/// vm.run_pyc_bytes(&bytes, main_scope);

/// });

/// ```

pub fn run_pyc_bytes(&self, pyc_bytes: &[u8], scope: Scope) -> PyResult<()> {

let code = PyCode::from_pyc(pyc_bytes, Some("<pyc_bytes>"), None, None, self)?;

self.with_simple_run("<source>", |_module_dict| {

self.run_code_obj(code, scope)?;

Ok(())

})

}

pub fn run_code_obj(&self, code: PyRef<PyCode>, scope: Scope) -> PyResult {

use crate::builtins::{PyFunction, PyModule};

// Create a function object for module code, similar to CPython's PyEval_EvalCode

let func = PyFunction::new(code.clone(), scope.globals.clone(), self)?;

let func_obj = func.into_ref(&self.ctx).into();

// Extract builtins from globals["__builtins__"], like PyEval_EvalCode

let builtins = match scope

.globals

.get_item_opt(identifier!(self, __builtins__), self)?

{

Some(b) => {

if let Some(module) = b.downcast_ref::<PyModule>() {

module.dict().into()

} else {

}

None => self.builtins.dict().into(),

};

let frame =

Frame::new(code, scope, builtins, &[], Some(func_obj), self).into_ref(&self.ctx);

self.run_frame(frame)

}

#[cold]

pub fn run_unraisable(&self, e: PyBaseExceptionRef, msg: Option<String>, object: PyObjectRef) {

// During interpreter finalization, sys.unraisablehook may not be available,

// but we still need to report exceptions (especially from atexit callbacks).

// Write directly to stderr like PyErr_FormatUnraisable.

if self.state.finalizing.load(Ordering::Acquire) {

self.write_unraisable_to_stderr(&e, msg.as_deref(), &object);

return;

}

let sys_module = self.import("sys", 0).unwrap();

let unraisablehook = sys_module.get_attr("unraisablehook", self).unwrap();

let exc_type = e.class().to_owned();

let exc_traceback = e.__traceback__().to_pyobject(self); // TODO: actual traceback

let exc_value = e.into();

let args = stdlib::sys::UnraisableHookArgsData {

exc_type,

exc_value,

exc_traceback,

err_msg: self.new_pyobj(msg),

object,

};

if let Err(e) = unraisablehook.call((args,), self) {

println!("{}", e.as_object().repr(self).unwrap());

}

/// Write unraisable exception to stderr during finalization.

/// Similar to _PyErr_WriteUnraisableDefaultHook in CPython.

fn write_unraisable_to_stderr(

&self,

e: &PyBaseExceptionRef,

msg: Option<&str>,

object: &PyObjectRef,

) {

// Get stderr once and reuse it

let stderr = crate::stdlib::sys::get_stderr(self).ok();

let write_to_stderr = |s: &str, stderr: &Option<PyObjectRef>, vm: &VirtualMachine| {

if let Some(stderr) = stderr {

let _ = vm.call_method(stderr, "write", (s.to_owned(),));

} else {

eprint!("{}", s);

}

};

let msg_str = if let Some(msg) = msg {

format!("{msg}: ")

} else {

"Exception ignored in: ".to_owned()

};

write_to_stderr(&msg_str, &stderr, self);

let repr_result = object.repr(self);

let repr_wtf8 = repr_result

.as_ref()

.map_or("<object repr failed>".as_ref(), |s| s.as_wtf8());

write_to_stderr(&format!("{repr_wtf8}\n"), &stderr, self);

// Write exception type and message

let exc_type_name = e.class().name();

let msg = match e.as_object().str(self) {

Ok(exc_str) if !exc_str.as_wtf8().is_empty() => {

format!("{}: {}\n", exc_type_name, exc_str.as_wtf8())

}

_ => format!("{}\n", exc_type_name),

};

write_to_stderr(&msg, &stderr, self);

// Flush stderr to ensure output is visible

if let Some(ref stderr) = stderr {

let _ = self.call_method(stderr, "flush", ());

}

#[inline(always)]

pub fn run_frame(&self, frame: FrameRef) -> PyResult {

match self.with_frame(frame, |f| f.run(self))? {

ExecutionResult::Return(value) => Ok(value),

_ => panic!("Got unexpected result from function"),

}

/// Run `run` with main scope.

fn with_simple_run(

&self,

path: &str,

run: impl FnOnce(&Py<PyDict>) -> PyResult<()>,

) -> PyResult<()> {

let sys_modules = self.sys_module.get_attr(identifier!(self, modules), self)?;

let main_module = sys_modules.get_item(identifier!(self, __main__), self)?;

let module_dict = main_module.dict().expect("main module must have __dict__");

// Track whether we set __file__ (for cleanup)

let set_file_name = !module_dict.contains_key(identifier!(self, __file__), self);

if set_file_name {

module_dict.set_item(

identifier!(self, __file__),

self.ctx.new_str(path).into(),

self,

)?;

module_dict.set_item(identifier!(self, __cached__), self.ctx.none(), self)?;

}

let result = run(&module_dict);

self.flush_io();

// Cleanup __file__ and __cached__ after execution

if set_file_name {

let _ = module_dict.del_item(identifier!(self, __file__), self);

let _ = module_dict.del_item(identifier!(self, __cached__), self);

}

result

}

/// flush_io

///

/// Flush stdout and stderr. Errors are silently ignored.

fn flush_io(&self) {

if let Ok(stdout) = self.sys_module.get_attr("stdout", self) {

let _ = self.call_method(&stdout, identifier!(self, flush).as_str(), ());

}

if let Ok(stderr) = self.sys_module.get_attr("stderr", self) {

let _ = self.call_method(&stderr, identifier!(self, flush).as_str(), ());

}

/// Clear module references during shutdown.

/// Follows the same phased algorithm as pylifecycle.c finalize_modules():

/// no hardcoded module names, reverse import order, only builtins/sys last.

pub fn finalize_modules(&self) {

// Phase 1: Set special sys/builtins attributes to None, restore stdio

self.finalize_modules_delete_special();

// Phase 2: Remove all modules from sys.modules (set values to None),

// and collect weakrefs to modules preserving import order.

// No strong refs are kept — modules freed when their last ref drops.

let module_weakrefs = self.finalize_remove_modules();

// Phase 3: Clear sys.modules dict

self.finalize_clear_modules_dict();

// Phase 4: GC collect — modules removed from sys.modules are freed,

// exposing cycles (e.g., dict ↔ function.__globals__). GC collects

// these and calls __del__ while module dicts are still intact.

crate::gc_state::gc_state().collect_force(2);

// Phase 5: Clear module dicts in reverse import order using 2-pass algorithm.

// Skip builtins and sys — those are cleared last.

self.finalize_clear_module_dicts(&module_weakrefs);

// Phase 6: GC collect — pick up anything freed by dict clearing.

crate::gc_state::gc_state().collect_force(2);

// Phase 7: Clear sys and builtins dicts last

self.finalize_clear_sys_builtins_dict();

}

/// Phase 1: Set special sys attributes to None and restore stdio.

fn finalize_modules_delete_special(&self) {

let none = self.ctx.none();

let sys_dict = self.sys_module.dict();

// Set special sys attributes to None

for attr in &[

"path",

"argv",

"ps1",

"ps2",

"last_exc",

"last_type",

"last_value",

"last_traceback",

"path_importer_cache",

"meta_path",

"path_hooks",

] {

let _ = sys_dict.set_item(*attr, none.clone(), self);

}

// Restore stdin/stdout/stderr from __stdin__/__stdout__/__stderr__

for (std_name, dunder_name) in &[

("stdin", "__stdin__"),

("stdout", "__stdout__"),

("stderr", "__stderr__"),

] {

let restored = sys_dict

.get_item_opt(*dunder_name, self)

.ok()

.flatten()

.unwrap_or_else(|| none.clone());

let _ = sys_dict.set_item(*std_name, restored, self);

}

// builtins._ = None

let _ = self.builtins.dict().set_item("_", none, self);

}

/// Phase 2: Set all sys.modules values to None and collect weakrefs.

/// No strong refs are kept — modules are freed when removed from sys.modules

/// (if nothing else references them), allowing GC to collect their cycles.

fn finalize_remove_modules(&self) -> Vec<(String, PyRef<PyWeak>)> {

let mut module_weakrefs = Vec::new();

let Ok(modules) = self.sys_module.get_attr(identifier!(self, modules), self) else {

return module_weakrefs;

};

let Some(modules_dict) = modules.downcast_ref::<PyDict>() else {

return module_weakrefs;

};

let none = self.ctx.none();

let items: Vec<_> = modules_dict.into_iter().collect();

for (key, value) in items {

let name = key

.downcast_ref::<PyUtf8Str>()

.map(|s| s.as_str().to_owned())

.unwrap_or_default();

// Save weakref to module (for later dict clearing)

if value.downcast_ref::<PyModule>().is_some()

&& let Ok(weak) = value.downgrade(None, self)

{

module_weakrefs.push((name, weak));

}

// Set the value to None in sys.modules

let _ = modules_dict.set_item(&*key, none.clone(), self);

}

module_weakrefs

}

/// Phase 3: Clear sys.modules dict.

fn finalize_clear_modules_dict(&self) {

if let Ok(modules) = self.sys_module.get_attr(identifier!(self, modules), self)

&& let Some(modules_dict) = modules.downcast_ref::<PyDict>()

{

modules_dict.clear();

}

/// Phase 5: Clear module dicts in reverse import order.

/// Skip builtins and sys — those are cleared last in Phase 7.

fn finalize_clear_module_dicts(&self, module_weakrefs: &[(String, PyRef<PyWeak>)]) {

let builtins_dict = self.builtins.dict();

let sys_dict = self.sys_module.dict();

for (_name, weakref) in module_weakrefs.iter().rev() {

let Some(module_obj) = weakref.upgrade() else {

continue;

};

let Some(module) = module_obj.downcast_ref::<PyModule>() else {

continue;

};

let dict = module.dict();

// Skip builtins and sys — they are cleared last

if dict.is(&builtins_dict) || dict.is(&sys_dict) {

continue;

}

Self::module_clear_dict(&dict, self);

}

/// 2-pass module dict clearing (_PyModule_ClearDict algorithm).

/// Pass 1: Set names starting with '_' (except __builtins__) to None.

/// Pass 2: Set all remaining names (except __builtins__) to None.

pub(crate) fn module_clear_dict(dict: &Py<PyDict>, vm: &VirtualMachine) {

let none = vm.ctx.none();

// Pass 1: names starting with '_' (except __builtins__)

for (key, value) in dict.into_iter().collect::<Vec<_>>() {

if vm.is_none(&value) {

continue;

}

if let Some(key_str) = key.downcast_ref::<PyStr>() {

let name = key_str.as_wtf8();

if name.starts_with("_") && name != "__builtins__" {

let _ = dict.set_item(key_str, none.clone(), vm);

}

// Pass 2: all remaining (except __builtins__)

for (key, value) in dict.into_iter().collect::<Vec<_>>() {

if vm.is_none(&value) {

continue;

}

if let Some(key_str) = key.downcast_ref::<PyStr>()

&& key_str.as_bytes() != b"__builtins__"

{

let _ = dict.set_item(key_str.as_wtf8(), none.clone(), vm);

}

/// Phase 7: Clear sys and builtins dicts last.

fn finalize_clear_sys_builtins_dict(&self) {

Self::module_clear_dict(&self.sys_module.dict(), self);

Self::module_clear_dict(&self.builtins.dict(), self);

}

pub fn current_recursion_depth(&self) -> usize {

self.recursion_depth.get()

}

/// Stack margin bytes (like _PyOS_STACK_MARGIN_BYTES).

/// 2048 * sizeof(void*) = 16KB for 64-bit.

const STACK_MARGIN_BYTES: usize = 2048 * core::mem::size_of::<usize>();

/// Get the stack boundaries using platform-specific APIs.

/// Returns (base, top) where base is the lowest address and top is the highest.

#[cfg(all(not(miri), windows))]

fn get_stack_bounds() -> (usize, usize) {

use windows_sys::Win32::System::Threading::{

GetCurrentThreadStackLimits, SetThreadStackGuarantee,

};

let mut low: usize = 0;

let mut high: usize = 0;

unsafe {

GetCurrentThreadStackLimits(&mut low as *mut usize, &mut high as *mut usize);

// Add the guaranteed stack space (reserved for exception handling)

let mut guarantee: u32 = 0;

SetThreadStackGuarantee(&mut guarantee);

low += guarantee as usize;

}

(low, high)

}

/// Get stack boundaries on non-Windows platforms.

/// Falls back to estimating based on current stack pointer.

#[cfg(all(not(miri), not(windows)))]

fn get_stack_bounds() -> (usize, usize) {

// Use pthread_attr_getstack on platforms that support it

#[cfg(any(target_os = "linux", target_os = "android"))]

{

use libc::{

pthread_attr_destroy, pthread_attr_getstack, pthread_attr_t, pthread_getattr_np,

pthread_self,

};

let mut attr: pthread_attr_t = unsafe { core::mem::zeroed() };

unsafe {

if pthread_getattr_np(pthread_self(), &mut attr) == 0 {

let mut stack_addr: *mut libc::c_void = core::ptr::null_mut();

let mut stack_size: libc::size_t = 0;

if pthread_attr_getstack(&attr, &mut stack_addr, &mut stack_size) == 0 {

pthread_attr_destroy(&mut attr);

let base = stack_addr as usize;

let top = base + stack_size;

return (base, top);

}

pthread_attr_destroy(&mut attr);

}

#[cfg(target_os = "macos")]

{

use libc::{pthread_get_stackaddr_np, pthread_get_stacksize_np, pthread_self};

unsafe {

let thread = pthread_self();

let stack_top = pthread_get_stackaddr_np(thread) as usize;

let stack_size = pthread_get_stacksize_np(thread);

let stack_base = stack_top - stack_size;

return (stack_base, stack_top);

}

// Fallback: estimate based on current SP and a default stack size

#[allow(unreachable_code)]

{

let current_sp = psm::stack_pointer() as usize;

// Assume 8MB stack, estimate base

let estimated_size = 8 * 1024 * 1024;

let base = current_sp.saturating_sub(estimated_size);

let top = current_sp + 1024 * 1024; // Assume we're not at the very top

(base, top)

}

/// Calculate the C stack soft limit based on actual stack boundaries.

/// soft_limit = base + 2 * margin (for downward-growing stacks)

#[cfg(not(miri))]

fn calculate_c_stack_soft_limit() -> usize {

let (base, _top) = Self::get_stack_bounds();

// Soft limit is 2 margins above the base

base + Self::STACK_MARGIN_BYTES * 2

}

/// Miri doesn't support inline assembly, so disable C stack checking.

#[cfg(miri)]

fn calculate_c_stack_soft_limit() -> usize {

}

/// Check if we're near the C stack limit (like _Py_MakeRecCheck).

/// Returns true only when stack pointer is in the "danger zone" between

View remainder of file in raw view

FilesExpand file tree

mod.rs

Latest commit

History

mod.rs

File metadata and controls