Add initial integration with wasm-bindgen

.circleci/config.yml

@@ -72,6 +72,7 @@ commands:
             curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
             export PATH=${HOME}/.cargo/bin:${PATH}
             rustup target add wasm32-unknown-emscripten
+            cargo install wasm-bindgen-cli
             echo "export PATH=\"\$HOME/.cargo/bin:\$PATH\"" >> $BASH_ENV
   install-node-version:
     description: "install a specific version of node"

emcc.py

@@ -27,7 +27,6 @@
 import shutil
 import sys
 import tarfile
-from dataclasses import dataclass
 from enum import Enum, auto, unique
 
 # This assert needs to happen early, before any too-recent python syntax is used.
@@ -46,7 +45,7 @@
   system_libs,
   utils,
 )
-from tools.cmdline import CLANG_FLAGS_WITH_ARGS, options
+from tools.cmdline import CLANG_FLAGS_WITH_ARGS, LinkFlag, options
 from tools.response_file import substitute_response_files
 from tools.settings import COMPILE_TIME_SETTINGS, default_setting, settings, user_settings
 from tools.shared import DEBUG, DYLIB_EXTENSIONS, in_temp
@@ -96,20 +95,6 @@ class Mode(Enum):
   COMPILE_AND_LINK = auto()
 
 
-@dataclass
-class LinkFlag:
-  """Used to represent a linker flag.
-
-  The flag value is stored along with a bool that distinguishes input
-  files from non-files.
-
-  A list of these is returned by separate_linker_flags.
-  """
-
-  value: str
-  is_file: int
-
-
 class EmccState:
   def __init__(self, args):
     self.mode = Mode.COMPILE_AND_LINK
@@ -578,7 +563,7 @@ def compile_source_file(input_file):
       # Default to assuming the inputs are object files and pass them to the linker
       pass
 
-  return [f.value for f in linker_args]
+  return linker_args
 
 
 if __name__ == '__main__':

site/source/docs/tools_reference/settings_reference.rst

@@ -3288,6 +3288,15 @@ Example use ``-sSIGNATURE_CONVERSIONS=someFunction:_p,anotherFunction:p``
 
 Default value: []
 
+.. _wasm_bindgen:
+
+WASM_BINDGEN
+============
+
+Run wasm-bindgen and integrate the rust-exported symbols into the rest of Emscripten's JS output.
+
+Default value: 0
+
 .. _source_phase_imports:
 
 SOURCE_PHASE_IMPORTS

src/settings.js

@@ -2167,6 +2167,10 @@ var LEGACY_RUNTIME = false;
 // [link]
 var SIGNATURE_CONVERSIONS = [];
 
+// Run wasm-bindgen and integrate the rust-exported symbols into the rest of Emscripten's JS output.
+// [link]
+var WASM_BINDGEN = 0;
+
 // Experimental support for wasm source phase imports.
 // This is only currently implemented in the pre-release/nightly version of
 // node, and not yet supported by browsers.

test/rust/basics/.cargo/config.toml

@@ -0,0 +1,2 @@
+[build]
+target = "wasm32-unknown-emscripten"

test/rust/bindgen_integration/.cargo/config.toml

@@ -0,0 +1,7 @@
+[build]
+target = "wasm32-unknown-emscripten"
+rustflags = [
+    "-Cllvm-args=-enable-emscripten-cxx-exceptions=0",
+    "-Cpanic=abort",
+    "-Crelocation-model=static",
+]

test/rust/bindgen_integration/Cargo.toml

@@ -0,0 +1,8 @@
+[package]
+name = "bindgen_integration"
+edition = "2021"
+
+[lib]
+crate-type = ["staticlib"]
+
+[dependencies]

test/rust/bindgen_integration/src/lib.rs

@@ -0,0 +1,6 @@
+use wasm_bindgen::prelude::*;
+
+#[wasm_bindgen]
+pub fn rs_add(a: i32, b: i32) -> i32 {
+    return a + b;
+}

test/test_other.py

@@ -14830,6 +14830,22 @@ def test_rust_integration_basics(self):
     }''')
     self.do_runf('main.cpp', 'Hello from rust!', cflags=[lib])
 
+  @requires_rust
+  def test_wasm_bindgen_integration(self):
+    copytree(test_file('rust/bindgen_integration'), '.')
+    self.run_process(['cargo', 'add', 'wasm-bindgen'])
+    self.run_process(['cargo', 'build'])
+    lib = 'target/wasm32-unknown-emscripten/debug/libbindgen_integration.a'
+    self.assertExists(lib)
+
+    create_file('empty.c', '')
+    create_file('post.js', '''
+      Module.onRuntimeInitialized = () => out(Module.rs_add(17, 25));
+    ''')
+    emcc_args = [lib, '-sWASM_BINDGEN', '--post-js=post.js', '-lexports.js']
+
+    self.do_runf('empty.c', '42', cflags=emcc_args)
+
   def test_relative_em_cache(self):
     with env_modify({'EM_CACHE': 'foo'}):
       self.assert_fail([EMCC, '-c', test_file('hello_world.c')], 'emcc: error: environment variable EM_CACHE must be an absolute path: foo')

tools/building.py

@@ -277,12 +277,33 @@ def lld_flags_for_executable(external_symbols):
   return cmd
 
 
-def lld_flags(args):
+def get_wasm_bindgen_exported_symbols(input_files):
+  nm_args = [LLVM_NM, '--defined-only', '--extern-only', '--format=just-symbols',
+             '--print-file-name', '--quiet']
+  nm_args += input_files
+
+  result = check_call(nm_args, stdout=PIPE)
+  symbols = []
+  for line in result.stdout.splitlines():
+    _, symbol = line.split()
+    # Skip mangled (non-C) symbols
+    if symbol.startswith(('_Z', '_R', 'anon.')):
+      continue
+    symbols.append(symbol)
+
+  return symbols
+
+
+def lld_flags(args, linker_inputs=None):
   # lld doesn't currently support --start-group/--end-group since the
   # semantics are more like the windows linker where there is no need for
   # grouping.
   args = [a for a in args if a not in {'--start-group', '--end-group'}]
 
+  if settings.WASM_BINDGEN:
+    exported_symbols = get_wasm_bindgen_exported_symbols(linker_inputs)
+    args.extend(f'--export={e}' for e in exported_symbols)
+
   # Emscripten currently expects linkable output (SIDE_MODULE/MAIN_MODULE) to
   # include all archive contents.
   if settings.LINKABLE and (settings.FAKE_DYLIBS or not settings.SIDE_MODULE):
@@ -316,7 +337,7 @@ def lld_flags(args):
   return args
 
 
-def link_lld(args, target, external_symbols=None):
+def link_lld(args, target, external_symbols=None, linker_inputs=None):
   # runs lld to link things.
   if not os.path.exists(WASM_LD):
     exit_with_error('linker binary not found in LLVM directory: %s', WASM_LD)
@@ -325,7 +346,7 @@ def link_lld(args, target, external_symbols=None):
   # normal linker flags that are used when building and executable
   if '--relocatable' not in args and '-r' not in args:
     cmd += lld_flags_for_executable(external_symbols)
-  cmd += lld_flags(args)
+  cmd += lld_flags(args, linker_inputs)
   cmd = get_command_with_possible_response_file(cmd)
   check_call(cmd)
 
@@ -1264,6 +1285,32 @@ def run_wasm_opt(infile, outfile=None, args=[], **kwargs):  # noqa
   return run_binaryen_command('wasm-opt', infile, outfile, args=args, **kwargs)
 
 
+def run_wasm_bindgen(infile):
+  bindgen_out_dir = os.path.join(get_emscripten_temp_dir(), 'bindgen_out')
+
+  wasm_bindgen_bin = shutil.which('wasm-bindgen')
+  if not wasm_bindgen_bin:
+    exit_with_error('wasm-bindgen executable not found in $PATH')
+  cmd = [
+      wasm_bindgen_bin,
+      infile,
+      '--keep-lld-exports',
+      '--keep-debug',
+      '--out-dir',
+      bindgen_out_dir,
+  ]
+  check_call(cmd)
+
+  # Don't try to predict the .wasm filename that wasm-bindgen outputs. Instead
+  # just grab the .wasm file itself.
+  all_output_files = os.listdir(bindgen_out_dir)
+  new_wasm_file = [x for x in all_output_files if x.endswith('.wasm')][0]
+
+  shutil.copyfile(os.path.join(bindgen_out_dir, new_wasm_file), infile)
+
+  return os.path.join(bindgen_out_dir, 'library_bindgen.js')
+
+
 intermediate_counter = 0
 
 

tools/cmdline.py

@@ -9,6 +9,7 @@
 import re
 import shlex
 import sys
+from dataclasses import dataclass
 from enum import Enum, auto, unique
 from subprocess import PIPE
 
@@ -863,3 +864,17 @@ def parse_arguments(args):
   apply_user_settings()
 
   return newargs
+
+
+@dataclass
+class LinkFlag:
+  """Used to represent a linker flag.
+
+  The flag value is stored along with a bool that distinguishes input
+  files from non-files.
+
+  A list of these is returned by separate_linker_flags.
+  """
+
+  value: str
+  is_file: int

tools/emscripten.py

@@ -681,7 +681,7 @@ def create_tsd_exported_runtime_methods(metadata):
   return utils.read_file(in_temp(f'{file}.d.ts'))
 
 
-def create_tsd(metadata, embind_tsd):
+def create_tsd(metadata, embind_tsd, bindgen_tsd):
   out = '// TypeScript bindings for emscripten-generated code.  Automatically generated at compile time.\n'
   if settings.EXPORTED_RUNTIME_METHODS:
     out += create_tsd_exported_runtime_methods(metadata)
@@ -713,6 +713,10 @@ def create_tsd(metadata, embind_tsd):
   # Add in embind definitions.
   if embind_tsd:
     export_interfaces += ' & EmbindModule'
+  if settings.WASM_BINDGEN and bindgen_tsd:
+    for file_path in bindgen_tsd:
+      out += utils.read_file(file_path)
+    export_interfaces += ' & BindgenModule'
   out += f'export type MainModule = {export_interfaces};\n'
   if settings.MODULARIZE:
     return_type = 'MainModule'

tools/link.py

@@ -4,6 +4,7 @@
 # found in the LICENSE file.
 
 import base64
+import glob
 import json
 import logging
 import os
@@ -41,7 +42,7 @@
   settings,
   user_settings,
 )
-from .shared import DEBUG, DYLIB_EXTENSIONS, do_replace, in_temp
+from .shared import DEBUG, DYLIB_EXTENSIONS, do_replace, get_emscripten_temp_dir, in_temp
 from .toolchain_profiler import ToolchainProfiler
 from .utils import (
   WINDOWS,
@@ -1837,7 +1838,7 @@ def phase_calculate_system_libraries(options):
 
 
 @ToolchainProfiler.profile_block('link')
-def phase_link(linker_args, wasm_target, js_syms):
+def phase_link(linker_args, linker_inputs, wasm_target, js_syms):
   logger.debug(f'linking: {linker_args}')
 
   # Make a final pass over settings.EXPORTED_FUNCTIONS to remove any
@@ -1859,11 +1860,12 @@ def phase_link(linker_args, wasm_target, js_syms):
     # TODO(sbc): Remove this double execution of wasm-ld if we ever find a way to
     # distinguish EMSCRIPTEN_KEEPALIVE exports from `--export-dynamic` exports.
     settings.LINKABLE = False
-    building.link_lld(linker_args, wasm_target, external_symbols=js_syms)
+    building.link_lld(linker_args, wasm_target, external_symbols=js_syms,
+                      linker_inputs=linker_inputs)
     settings.LINKABLE = True
     rtn = extract_metadata.extract_metadata(wasm_target)
 
-  building.link_lld(linker_args, wasm_target, external_symbols=js_syms)
+  building.link_lld(linker_args, wasm_target, external_symbols=js_syms, linker_inputs=linker_inputs)
   return rtn
 
 
@@ -1885,6 +1887,10 @@ def phase_post_link(options, in_wasm, wasm_target, target, js_syms, base_metadat
 
   settings.TARGET_JS_NAME = os.path.basename(js_target)
 
+  if settings.WASM_BINDGEN:
+    bindgen_jslib = building.run_wasm_bindgen(in_wasm)
+    settings.JS_LIBRARIES.append(bindgen_jslib)
+
   metadata = phase_emscript(in_wasm, wasm_target, js_syms, base_metadata)
 
   if settings.EMBIND_AOT:
@@ -2016,7 +2022,12 @@ def phase_emit_tsd(options, wasm_target, js_target, js_syms, metadata):
   embind_tsd = ''
   if settings.EMBIND:
     embind_tsd = run_embind_gen(options, wasm_target, js_syms, {'EMBIND_AOT': False})
-  all_tsd = emscripten.create_tsd(metadata, embind_tsd)
+  bindgen_ts_files = []
+  if settings.WASM_BINDGEN:
+    bindgen_ts_files = glob.glob(get_emscripten_temp_dir() + '/bindgen_out/*.d.ts')
+    # This list comprehension then filters out any files that end with .wasm.d.ts.
+    bindgen_ts_files = [file for file in bindgen_ts_files if not file.endswith('.wasm.d.ts')]
+  all_tsd = emscripten.create_tsd(metadata, embind_tsd, bindgen_ts_files)
   out_file = os.path.join(os.path.dirname(js_target), filename)
   write_file(out_file, all_tsd)
 
@@ -3066,6 +3077,9 @@ def run(options, linker_args):
   # We have now passed the compile phase, allow reading/writing of all settings.
   settings.limit_settings(None)
 
+  linker_inputs = [f.value for f in linker_args if f.is_file]
+  linker_args = [f.value for f in linker_args]
+
   if settings.RUNTIME_LINKED_LIBS:
     linker_args += settings.RUNTIME_LINKED_LIBS
 
@@ -3129,7 +3143,7 @@ def add_js_deps(sym):
       settings.ASYNCIFY_IMPORTS_EXCEPT_JS_LIBS = settings.ASYNCIFY_IMPORTS[:]
       settings.ASYNCIFY_IMPORTS += ['*.' + x for x in js_info['asyncFuncs']]
 
-  base_metadata = phase_link(linker_args, wasm_target, js_syms)
+  base_metadata = phase_link(linker_args, linker_inputs, wasm_target, js_syms)
 
   # Special handling for when the user passed '-Wl,--version'.  In this case the linker
   # does not create the output file, but just prints its version and exits with 0.