1030 lines
35 KiB
Rust
Raw Normal View History

use std::char;
use std::collections::{HashSet, HashMap};
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
use shared;
use parity_wasm::elements::*;
use super::Bindgen;
pub struct Context<'a> {
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
pub globals: String,
pub imports: String,
pub typescript: String,
pub exposed_globals: HashSet<&'static str>,
pub required_internal_exports: HashSet<&'static str>,
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
pub config: &'a Bindgen,
pub module: &'a mut Module,
pub imports_to_rewrite: HashSet<String>,
pub custom_type_names: HashMap<char, String>,
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
pub struct SubContext<'a, 'b: 'a> {
pub program: &'a shared::Program,
pub cx: &'a mut Context<'b>,
}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
impl<'a> Context<'a> {
pub fn add_custom_type_names(&mut self, program: &shared::Program) {
for custom in program.custom_type_names.iter() {
assert!(self.custom_type_names.insert(custom.descriptor,
custom.name.clone()).is_none());
let val = custom.descriptor as u32;
assert!(val & 1 == 0);
let descriptor = char::from_u32(val | 1).unwrap();
assert!(self.custom_type_names.insert(descriptor,
custom.name.clone()).is_none());
}
}
pub fn finalize(&mut self, module_name: &str) -> (String, String) {
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
{
let mut bind = |name: &str, f: &Fn(&mut Self) -> String| {
if !self.wasm_import_needed(name) {
return
}
let global = format!("export const {} = {};", name, f(self));
self.globals.push_str(&global);
};
bind("__wbindgen_object_clone_ref", &|me| {
me.expose_add_heap_object();
me.expose_get_object();
let bump_cnt = if me.config.debug {
String::from("
if (typeof(val) === 'number')
throw new Error('corrupt slab');
val.cnt += 1;
")
} else {
String::from("val.cnt += 1;")
};
format!("
function(idx) {{
// If this object is on the stack promote it to the heap.
if ((idx & 1) === 1)
return addHeapObject(getObject(idx));
// Otherwise if the object is on the heap just bump the
// refcount and move on
const val = slab[idx >> 1];
{}
return idx;
}}
", bump_cnt)
});
bind("__wbindgen_object_drop_ref", &|me| {
me.expose_drop_ref();
"dropRef".to_string()
});
bind("__wbindgen_string_new", &|me| {
me.expose_add_heap_object();
me.expose_get_string_from_wasm();
String::from("(p, l) => addHeapObject(getStringFromWasm(p, l))")
});
bind("__wbindgen_number_new", &|me| {
me.expose_add_heap_object();
String::from("addHeapObject")
});
bind("__wbindgen_number_get", &|me| {
me.expose_get_object();
me.expose_uint8_memory();
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
format!("
function(n, invalid) {{
let obj = getObject(n);
if (typeof(obj) === 'number')
return obj;
getUint8Memory()[invalid] = 1;
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
return 0;
}}
")
});
bind("__wbindgen_undefined_new", &|me| {
me.expose_add_heap_object();
String::from("() => addHeapObject(undefined)")
});
bind("__wbindgen_null_new", &|me| {
me.expose_add_heap_object();
String::from("() => addHeapObject(null)")
});
bind("__wbindgen_is_null", &|me| {
me.expose_get_object();
String::from("(idx) => getObject(idx) === null ? 1 : 0")
});
bind("__wbindgen_is_undefined", &|me| {
me.expose_get_object();
String::from("(idx) => getObject(idx) === undefined ? 1 : 0")
});
bind("__wbindgen_boolean_new", &|me| {
me.expose_add_heap_object();
String::from("(v) => addHeapObject(v == 1)")
});
bind("__wbindgen_boolean_get", &|me| {
me.expose_get_object();
String::from("(i) => {
let v = getObject(i);
if (typeof(v) == 'boolean') {
return v ? 1 : 0;
} else {
return 2;
}
}")
});
bind("__wbindgen_symbol_new", &|me| {
me.expose_get_string_from_wasm();
me.expose_add_heap_object();
format!("(ptr, len) => {{
let a;
console.log(ptr, len);
if (ptr === 0) {{
a = Symbol();
}} else {{
a = Symbol(getStringFromWasm(ptr, len));
}}
return addHeapObject(a);
}}")
});
bind("__wbindgen_is_symbol", &|me| {
me.expose_get_object();
String::from("(i) => typeof(getObject(i)) == 'symbol' ? 1 : 0")
});
bind("__wbindgen_throw", &|me| {
me.expose_get_string_from_wasm();
format!("
function(ptr, len) {{
throw new Error(getStringFromWasm(ptr, len));
}}
")
});
bind("__wbindgen_string_get", &|me| {
me.expose_pass_string_to_wasm();
me.expose_get_object();
me.expose_uint32_memory();
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
String::from("(i, len_ptr) => {
let obj = getObject(i);
if (typeof(obj) !== 'string')
return 0;
const [ptr, len] = passStringToWasm(obj);
getUint32Memory()[len_ptr / 4] = len;
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
return ptr;
}")
});
}
let js = format!("
/* tslint:disable */
import * as wasm from './{module_name}_wasm'; // imports from wasm file
{imports}
{globals}
",
module_name = module_name,
globals = self.globals,
imports = self.imports,
);
self.rewrite_imports(module_name);
self.unexport_unused_internal_exports();
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
(js, self.typescript.clone())
}
fn rewrite_imports(&mut self, module_name: &str) {
for section in self.module.sections_mut() {
let imports = match *section {
Section::Import(ref mut s) => s,
_ => continue,
};
for import in imports.entries_mut() {
if import.field().starts_with("__wbindgen") {
import.module_mut().truncate(0);
import.module_mut().push_str("./");
import.module_mut().push_str(module_name);
continue
}
// rustc doesn't have support for importing from anything other
// than the module `env` so let's use the metadata here to
// rewrite the imports if they import from `env` until it's
// fixed upstream.
if self.imports_to_rewrite.contains(import.field()) {
import.module_mut().truncate(0);
import.module_mut().push_str("./");
import.module_mut().push_str(module_name);
continue
}
}
}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
fn unexport_unused_internal_exports(&mut self) {
let required = &self.required_internal_exports;
for section in self.module.sections_mut() {
let exports = match *section {
Section::Export(ref mut s) => s,
_ => continue,
};
exports.entries_mut().retain(|export| {
!export.field().starts_with("__wbindgen") ||
required.contains(export.field())
});
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
fn expose_drop_ref(&mut self) {
if !self.exposed_globals.insert("drop_ref") {
return
}
self.expose_global_slab();
self.expose_global_slab_next();
let validate_owned = if self.config.debug {
String::from("
if ((idx & 1) === 1)
throw new Error('cannot drop ref of stack objects');
")
} else {
String::new()
};
let dec_ref = if self.config.debug {
String::from("
if (typeof(obj) === 'number')
throw new Error('corrupt slab');
obj.cnt -= 1;
if (obj.cnt > 0)
return;
")
} else {
String::from("
obj.cnt -= 1;
if (obj.cnt > 0)
return;
")
};
self.globals.push_str(&format!("
function dropRef(idx) {{
{}
let obj = slab[idx >> 1];
{}
// If we hit 0 then free up our space in the slab
slab[idx >> 1] = slab_next;
slab_next = idx >> 1;
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}}
", validate_owned, dec_ref));
}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
fn expose_global_stack(&mut self) {
if !self.exposed_globals.insert("stack") {
return
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
self.globals.push_str(&format!("
let stack = [];
"));
}
fn expose_global_slab(&mut self) {
if !self.exposed_globals.insert("slab") {
return
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
self.globals.push_str(&format!("let slab = [];"));
}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
fn expose_global_slab_next(&mut self) {
if !self.exposed_globals.insert("slab_next") {
return
}
self.globals.push_str(&format!("
let slab_next = 0;
"));
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
fn expose_get_object(&mut self) {
if !self.exposed_globals.insert("get_object") {
return
}
self.expose_global_stack();
self.expose_global_slab();
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
let get_obj = if self.config.debug {
String::from("
if (typeof(val) === 'number')
throw new Error('corrupt slab');
return val.obj;
")
} else {
String::from("
return val.obj;
")
};
self.globals.push_str(&format!("
function getObject(idx) {{
if ((idx & 1) === 1) {{
return stack[idx >> 1];
}} else {{
const val = slab[idx >> 1];
{}
}}
}}
", get_obj));
}
fn expose_check_token(&mut self) {
if !self.exposed_globals.insert("check_token") {
return
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
self.globals.push_str(&format!("
const token = Symbol('foo');
function _checkToken(sym) {{
if (token !== sym)
throw new Error('cannot invoke `new` directly');
}}
"));
}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
fn expose_assert_num(&mut self) {
if !self.exposed_globals.insert("assert_num") {
return
}
self.globals.push_str(&format!("
function _assertNum(n) {{
if (typeof(n) !== 'number')
throw new Error('expected a number argument');
}}
"));
}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
fn expose_assert_bool(&mut self) {
if !self.exposed_globals.insert("assert_bool") {
return
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
self.globals.push_str(&format!("
function _assertBoolean(n) {{
if (typeof(n) !== 'boolean')
throw new Error('expected a boolean argument');
}}
"));
}
fn expose_pass_string_to_wasm(&mut self) {
if !self.exposed_globals.insert("pass_string_to_wasm") {
return
}
self.required_internal_exports.insert("__wbindgen_malloc");
if self.config.nodejs {
self.globals.push_str(&format!("
function passStringToWasm(arg) {{
if (typeof(arg) !== 'string')
throw new Error('expected a string argument');
const buf = Buffer.from(arg);
const len = buf.length;
const ptr = wasm.__wbindgen_malloc(len);
buf.copy(Buffer.from(wasm.memory.buffer), ptr);
return [ptr, len];
}}
"));
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
} else {
self.expose_text_encoder();
self.expose_uint8_memory();
self.globals.push_str(&format!("
function passStringToWasm(arg) {{
if (typeof(arg) !== 'string')
throw new Error('expected a string argument');
const buf = textEncoder().encode(arg);
const len = buf.length;
const ptr = wasm.__wbindgen_malloc(len);
getUint8Memory().set(buf, ptr);
return [ptr, len];
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}}
"));
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
}
fn expose_text_encoder(&mut self) {
if !self.exposed_globals.insert("text_encoder") {
return
}
self.globals.push_str(&format!("
let cachedEncoder = null;
function textEncoder() {{
if (cachedEncoder)
return cachedEncoder;
cachedEncoder = new TextEncoder('utf-8');
return cachedEncoder;
}}
"));
2018-02-05 16:39:11 -08:00
}
fn expose_text_decoder(&mut self) {
if !self.exposed_globals.insert("text_decoder") {
return
2018-02-05 16:39:11 -08:00
}
self.globals.push_str(&format!("
let cachedDecoder = null;
function textDecoder() {{
if (cachedDecoder)
return cachedDecoder;
cachedDecoder = new TextDecoder('utf-8');
return cachedDecoder;
}}
"));
}
2018-02-05 16:39:11 -08:00
fn expose_get_string_from_wasm(&mut self) {
if !self.exposed_globals.insert("get_string_from_wasm") {
return
}
if self.config.nodejs {
self.globals.push_str(&format!("
function getStringFromWasm(ptr, len) {{
const buf = Buffer.from(wasm.memory.buffer).slice(ptr, ptr + len);
const ret = buf.toString();
return ret;
}}
"));
} else {
self.expose_text_decoder();
self.expose_uint8_memory();
self.globals.push_str(&format!("
function getStringFromWasm(ptr, len) {{
const mem = getUint8Memory();
const slice = mem.slice(ptr, ptr + len);
const ret = textDecoder().decode(slice);
return ret;
}}
"));
2018-02-05 16:39:11 -08:00
}
}
fn expose_uint8_memory(&mut self) {
if !self.exposed_globals.insert("uint8_memory") {
return
}
self.globals.push_str(&format!("
let cachedUint8Memory = null;
function getUint8Memory() {{
if (cachedUint8Memory === null ||
cachedUint8Memory.buffer !== wasm.memory.buffer)
cachedUint8Memory = new Uint8Array(wasm.memory.buffer);
return cachedUint8Memory;
}}
"));
}
fn expose_uint32_memory(&mut self) {
if !self.exposed_globals.insert("uint32_memory") {
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
return
}
self.globals.push_str(&format!("
let cachedUint32Memory = null;
function getUint32Memory() {{
if (cachedUint32Memory === null ||
cachedUint32Memory.buffer !== wasm.memory.buffer)
cachedUint32Memory = new Uint32Array(wasm.memory.buffer);
return cachedUint32Memory;
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}}
"));
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
fn expose_assert_class(&mut self) {
if !self.exposed_globals.insert("assert_class") {
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
return
}
self.globals.push_str(&format!("
function _assertClass(instance, klass) {{
if (!(instance instanceof klass))
throw new Error(`expected instance of ${{klass.name}}`);
return instance.ptr;
}}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
"));
}
fn expose_borrowed_objects(&mut self) {
if !self.exposed_globals.insert("borrowed_objects") {
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
return
}
self.expose_global_stack();
self.globals.push_str(&format!("
function addBorrowedObject(obj) {{
stack.push(obj);
return ((stack.length - 1) << 1) | 1;
}}
"));
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
fn expose_take_object(&mut self) {
if !self.exposed_globals.insert("take_object") {
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
return
}
self.expose_get_object();
self.expose_drop_ref();
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
self.globals.push_str(&format!("
function takeObject(idx) {{
const ret = getObject(idx);
dropRef(idx);
return ret;
}}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
"));
}
fn expose_add_heap_object(&mut self) {
if !self.exposed_globals.insert("add_heap_object") {
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
return
}
self.expose_global_slab();
self.expose_global_slab_next();
let set_slab_next = if self.config.debug {
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
String::from("
if (typeof(next) !== 'number')
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
throw new Error('corrupt slab');
slab_next = next;
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
")
} else {
String::from("
slab_next = next;
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
")
};
self.globals.push_str(&format!("
function addHeapObject(obj) {{
if (slab_next == slab.length)
slab.push(slab.length + 1);
const idx = slab_next;
const next = slab[idx];
{}
slab[idx] = {{ obj, cnt: 1 }};
return idx << 1;
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}}
", set_slab_next));
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
fn wasm_import_needed(&self, name: &str) -> bool {
let imports = match self.module.import_section() {
Some(s) => s,
None => return false,
};
imports.entries().iter().any(|i| {
i.module() == "env" && i.field() == name
})
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
impl<'a, 'b> SubContext<'a, 'b> {
pub fn generate(&mut self) {
for f in self.program.free_functions.iter() {
self.generate_free_function(f);
}
for f in self.program.imports.iter() {
self.generate_import(&f.module, &f.function);
}
for s in self.program.structs.iter() {
self.generate_struct(s);
}
for s in self.program.imported_structs.iter() {
self.generate_import_struct(s);
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
}
pub fn generate_free_function(&mut self, func: &shared::Function) {
let (js, ts) = self.generate_function("function",
&func.name,
&func.name,
false,
&func.arguments,
func.ret.as_ref());
self.cx.globals.push_str("export ");
self.cx.globals.push_str(&js);
self.cx.globals.push_str("\n");
self.cx.typescript.push_str("export ");
self.cx.typescript.push_str(&ts);
self.cx.typescript.push_str("\n");
}
pub fn generate_struct(&mut self, s: &shared::Struct) {
let mut dst = String::new();
dst.push_str(&format!("export class {} {{", s.name));
let mut ts_dst = dst.clone();
ts_dst.push_str("
public ptr: number;
");
if self.cx.config.debug {
self.cx.expose_check_token();
dst.push_str(&format!("
constructor(ptr, sym) {{
_checkToken(sym);
this.ptr = ptr;
}}
"));
ts_dst.push_str("constructor(ptr: number, sym: Symbol);\n");
} else {
dst.push_str(&format!("
constructor(ptr) {{
this.ptr = ptr;
}}
"));
ts_dst.push_str("constructor(ptr: number);\n");
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
dst.push_str(&format!("
free() {{
const ptr = this.ptr;
this.ptr = 0;
wasm.{}(ptr);
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}}
", shared::free_function(&s.name)));
ts_dst.push_str("free(): void;\n");
for function in s.functions.iter() {
let (js, ts) = self.generate_function(
"static",
&function.name,
&shared::struct_function_export_name(&s.name, &function.name),
false,
&function.arguments,
function.ret.as_ref(),
);
dst.push_str(&js);
dst.push_str("\n");
ts_dst.push_str(&ts);
ts_dst.push_str("\n");
}
for method in s.methods.iter() {
let (js, ts) = self.generate_function(
"",
&method.function.name,
&shared::struct_function_export_name(&s.name, &method.function.name),
true,
&method.function.arguments,
method.function.ret.as_ref(),
);
dst.push_str(&js);
dst.push_str("\n");
ts_dst.push_str(&ts);
ts_dst.push_str("\n");
}
dst.push_str("}\n");
ts_dst.push_str("}\n");
self.cx.globals.push_str(&dst);
self.cx.globals.push_str("\n");
self.cx.typescript.push_str(&ts_dst);
self.cx.typescript.push_str("\n");
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
fn generate_function(&mut self,
prefix: &str,
name: &str,
wasm_name: &str,
is_method: bool,
arguments: &[shared::Type],
ret: Option<&shared::Type>) -> (String, String) {
let mut dst = format!("{}(", name);
let mut dst_ts = format!("{}(", name);
let mut passed_args = String::new();
let mut arg_conversions = String::new();
let mut destructors = String::new();
if is_method {
passed_args.push_str("this.ptr");
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
for (i, arg) in arguments.iter().enumerate() {
let name = format!("arg{}", i);
if i > 0 {
dst.push_str(", ");
dst_ts.push_str(", ");
}
dst.push_str(&name);
dst_ts.push_str(&name);
let mut pass = |arg: &str| {
if passed_args.len() > 0 {
passed_args.push_str(", ");
}
passed_args.push_str(arg);
};
match *arg {
shared::TYPE_NUMBER => {
dst_ts.push_str(": number");
if self.cx.config.debug {
self.cx.expose_assert_num();
arg_conversions.push_str(&format!("_assertNum({});\n", name));
}
pass(&name)
}
shared::TYPE_BOOLEAN => {
dst_ts.push_str(": boolean");
if self.cx.config.debug {
self.cx.expose_assert_bool();
arg_conversions.push_str(&format!("\
_assertBoolean({name});
", name = name));
} else {
}
pass(&format!("arg{i} ? 1 : 0", i = i))
}
shared::TYPE_BORROWED_STR |
shared::TYPE_STRING => {
dst_ts.push_str(": string");
self.cx.expose_pass_string_to_wasm();
arg_conversions.push_str(&format!("\
const [ptr{i}, len{i}] = passStringToWasm({arg});
", i = i, arg = name));
pass(&format!("ptr{}", i));
pass(&format!("len{}", i));
if *arg == shared::TYPE_BORROWED_STR {
destructors.push_str(&format!("\n\
wasm.__wbindgen_free(ptr{i}, len{i});\n\
", i = i));
self.cx.required_internal_exports.insert("__wbindgen_free");
}
}
shared::TYPE_JS_OWNED => {
dst_ts.push_str(": any");
self.cx.expose_add_heap_object();
arg_conversions.push_str(&format!("\
const idx{i} = addHeapObject({arg});
", i = i, arg = name));
pass(&format!("idx{}", i));
}
shared::TYPE_JS_REF => {
dst_ts.push_str(": any");
self.cx.expose_borrowed_objects();
arg_conversions.push_str(&format!("\
const idx{i} = addBorrowedObject({arg});
", i = i, arg = name));
destructors.push_str("stack.pop();\n");
pass(&format!("idx{}", i));
}
custom if (custom as u32) & shared::TYPE_CUSTOM_REF_FLAG != 0 => {
let s = self.cx.custom_type_names[&custom].clone();
dst_ts.push_str(&format!(": {}", s));
if self.cx.config.debug {
self.cx.expose_assert_class();
arg_conversions.push_str(&format!("\
_assertClass({arg}, {struct_});
", arg = name, struct_ = s));
}
pass(&format!("{}.ptr", name));
}
custom => {
let s = self.cx.custom_type_names[&custom].clone();
dst_ts.push_str(&format!(": {}", s));
if self.cx.config.debug {
self.cx.expose_assert_class();
arg_conversions.push_str(&format!("\
_assertClass({arg}, {struct_});
", arg = name, struct_ = s));
}
arg_conversions.push_str(&format!("\
const ptr{i} = {arg}.ptr;
{arg}.ptr = 0;
", i = i, arg = name));
pass(&format!("ptr{}", i));
}
}
}
dst.push_str(")");
dst_ts.push_str(")");
let convert_ret = match ret {
None => {
dst_ts.push_str(": void");
format!("return ret;")
}
Some(&shared::TYPE_NUMBER) => {
dst_ts.push_str(": number");
format!("return ret;")
}
Some(&shared::TYPE_BOOLEAN) => {
dst_ts.push_str(": boolean");
format!("return ret != 0;")
}
Some(&shared::TYPE_JS_OWNED) => {
dst_ts.push_str(": any");
self.cx.expose_take_object();
format!("return takeObject(ret);")
}
Some(&shared::TYPE_STRING) => {
dst_ts.push_str(": string");
self.cx.expose_get_string_from_wasm();
self.cx.required_internal_exports.insert("__wbindgen_boxed_str_ptr");
self.cx.required_internal_exports.insert("__wbindgen_boxed_str_len");
self.cx.required_internal_exports.insert("__wbindgen_boxed_str_free");
format!("
const ptr = wasm.__wbindgen_boxed_str_ptr(ret);
const len = wasm.__wbindgen_boxed_str_len(ret);
const realRet = getStringFromWasm(ptr, len);
wasm.__wbindgen_boxed_str_free(ret);
return realRet;
")
}
Some(&shared::TYPE_JS_REF) |
Some(&shared::TYPE_BORROWED_STR) => panic!(),
Some(&t) if (t as u32) & shared::TYPE_CUSTOM_REF_FLAG != 0 => panic!(),
Some(custom) => {
let name = &self.cx.custom_type_names[custom];
dst_ts.push_str(": ");
dst_ts.push_str(name);
if self.cx.config.debug {
format!("\
return new {name}(ret, token);
", name = name)
} else {
format!("\
return new {name}(ret);
", name = name)
}
}
};
dst_ts.push_str(";");
dst.push_str(" {\n ");
dst.push_str(&arg_conversions);
if destructors.len() == 0 {
dst.push_str(&format!("\
const ret = wasm.{}({passed});
{convert_ret}
",
f = wasm_name,
passed = passed_args,
convert_ret = convert_ret,
));
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
} else {
dst.push_str(&format!("\
try {{
const ret = wasm.{f}({passed});
{convert_ret}
}} finally {{
{destructors}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}}
",
f = wasm_name,
passed = passed_args,
destructors = destructors,
convert_ret = convert_ret,
));
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
dst.push_str("}");
(format!("{} {}", prefix, dst), format!("{} {}", prefix, dst_ts))
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
pub fn generate_import(&mut self, module: &str, import: &shared::Function) {
let imported_name = format!("import{}", self.cx.imports.len());
self.cx.imports.push_str(&format!("
import {{ {} as {} }} from '{}';
", import.name, imported_name, module));
let name = shared::mangled_import_name(None, &import.name);
self.gen_import_shim(&name,
&imported_name,
false,
import);
self.cx.imports_to_rewrite.insert(name);
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
pub fn generate_import_struct(&mut self, import: &shared::ImportStruct) {
if let Some(ref module) = import.module {
self.cx.imports.push_str(&format!("
import {{ {} }} from '{}';
", import.name, module));
}
for f in import.functions.iter() {
self.generate_import_struct_function(&import.name, f);
}
}
fn generate_import_struct_function(
&mut self,
class: &str,
f: &shared::ImportStructFunction,
) {
let delegate = if f.method {
format!("{}.prototype.{}.call", class, f.function.name)
} else if f.js_new {
format!("new {}", class)
} else {
format!("{}.{}", class, f.function.name)
};
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
let name = shared::mangled_import_name(Some(class), &f.function.name);
self.gen_import_shim(&name,
&delegate,
f.method,
&f.function);
self.cx.imports_to_rewrite.insert(name);
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
fn gen_import_shim(
&mut self,
shim_name: &str,
shim_delegate: &str,
is_method: bool,
import: &shared::Function,
) {
let mut dst = String::new();
dst.push_str(&format!("function {}(", shim_name));
let mut invocation = String::new();
if is_method {
dst.push_str("ptr");
invocation.push_str("getObject(ptr)");
self.cx.expose_get_object();
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
let mut extra = String::new();
for (i, arg) in import.arguments.iter().enumerate() {
if invocation.len() > 0 {
invocation.push_str(", ");
}
if i > 0 || is_method {
dst.push_str(", ");
}
match *arg {
shared::TYPE_NUMBER => {
invocation.push_str(&format!("arg{}", i));
dst.push_str(&format!("arg{}", i));
}
shared::TYPE_BOOLEAN => {
invocation.push_str(&format!("arg{} != 0", i));
dst.push_str(&format!("arg{}", i));
}
shared::TYPE_BORROWED_STR => {
self.cx.expose_get_string_from_wasm();
invocation.push_str(&format!("getStringFromWasm(ptr{0}, len{0})", i));
dst.push_str(&format!("ptr{0}, len{0}", i));
}
shared::TYPE_STRING => {
self.cx.expose_get_string_from_wasm();
dst.push_str(&format!("ptr{0}, len{0}", i));
extra.push_str(&format!("
let arg{0} = getStringFromWasm(ptr{0}, len{0});
wasm.__wbindgen_free(ptr{0}, len{0});
", i));
invocation.push_str(&format!("arg{}", i));
self.cx.required_internal_exports.insert("__wbindgen_free");
}
shared::TYPE_JS_OWNED => {
self.cx.expose_take_object();
invocation.push_str(&format!("takeObject(arg{})", i));
dst.push_str(&format!("arg{}", i));
}
shared::TYPE_JS_REF => {
self.cx.expose_get_object();
invocation.push_str(&format!("getObject(arg{})", i));
dst.push_str(&format!("arg{}", i));
}
_ => {
panic!("unsupported type in import");
}
}
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
let invoc = format!("{}({})", shim_delegate, invocation);
let invoc = match import.ret {
Some(shared::TYPE_NUMBER) => format!("return {};", invoc),
Some(shared::TYPE_BOOLEAN) => format!("return {} ? 1 : 0;", invoc),
Some(shared::TYPE_JS_OWNED) => {
self.cx.expose_add_heap_object();
format!("return addHeapObject({});", invoc)
}
Some(shared::TYPE_STRING) => {
self.cx.expose_pass_string_to_wasm();
self.cx.expose_uint32_memory();
if import.arguments.len() > 0 || is_method {
dst.push_str(", ");
}
dst.push_str("wasmretptr");
format!("
const [retptr, retlen] = passStringToWasm({});
getUint32Memory()[wasmretptr / 4] = retlen;
return retptr;
", invoc)
}
None => invoc,
_ => unimplemented!(),
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
};
dst.push_str(") {\n");
dst.push_str(&extra);
dst.push_str(&format!("{}\n}}", invoc));
self.cx.globals.push_str("export ");
self.cx.globals.push_str(&dst);
self.cx.globals.push_str("\n");
Rewrite wasm-bindgen with ES6 modules in mind This commit is a mostly-rewrite of the `wasm-bindgen` tool. After some recent discussions it's clear that the previous model wasn't quite going to cut it, and this iteration is one which primarily embraces ES6 modules and the idea that this is a polyfill for host bindings. The overall interface and functionality hasn't changed much but the underlying technology has now changed significantly. Previously `wasm-bindgen` would emit a JS file that acted as an ES6 module but had a bit of a wonky interface. It exposed an async function for instantiation of the wasm module, but that's the bundler's job, not ours! Instead this iteration views each input and output as a discrete ES6 module. The input wasm file is interpreted as "this *should* be an ES6 module with rich types" and the output is "well here's some ES6 modules that fulfill that contract". Notably the tool now replaces the original wasm ES6 module with a JS ES6 module that has the "rich interface". Additionally a second ES6 module is emitted (the actual wasm file) which imports and exports to the original ES6 module. This strategy is hoped to be much more amenable to bundlers and controlling how the wasm itself is instantiated. The emitted files files purely assume ES6 modules and should be able to work as-is once ES6 module integration for wasm is completed. Note that there aren't a ton of tools to pretend a wasm module is an ES6 module at the moment but those should be coming soon! In the meantime a local `wasm2es6js` hack was added to help make *something* work today. The README has also been updated with instructions for interacting with this model.
2018-01-29 21:20:38 -08:00
}
}