workers: add support for data: URLs

PR-URL: #34584 Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Bradley Farias <[email protected]> Reviewed-By: Jan Krems <[email protected]>

Author: aduh95

doc/api/worker_threads.md

@@ -621,6 +621,13 @@ if (isMainThread){
 <!-- YAML
 added: v10.5.0
 changes:
+ - version: REPLACEME
+ pr-url: https://github.com/nodejs/node/pull/34584
+ description: The `filename` parameter can be a WHATWG `URL` object using
+ `data:` protocol.
+ - version: REPLACEME
+ pr-url: https://github.com/nodejs/node/pull/34394
+ description: The `trackUnmanagedFds` option was set to `true` by default.
  - version:
  - v14.6.0
  pr-url: https://github.com/nodejs/node/pull/34303
@@ -647,7 +654,9 @@ changes:
 *`filename`{string|URL} The path to the Worker’s main script or module. Must
  be either an absolute path or a relative path (i.e. relative to the
  current working directory) starting with `./` or `../`, or a WHATWG `URL`
- object using `file:` protocol.
+ object using `file:` or `data:` protocol.
+ When using a [`data:` URL][], the data is interpreted based on MIME type using
+ the [ECMAScript module loader][].
  If `options.eval` is `true`, this is a string containing JavaScript code
  rather than a path.
 *`options`{Object}
@@ -893,6 +902,7 @@ active handle in the event system. If the worker is already `unref()`ed calling
 [`AsyncResource`]: async_hooks.html#async_hooks_class_asyncresource
 [`Buffer`]: buffer.html
 [`Buffer.allocUnsafe()`]: buffer.html#buffer_static_method_buffer_allocunsafe_size
+[ECMAScript module loader]: esm.html#esm_data_imports
 [`ERR_MISSING_MESSAGE_PORT_IN_TRANSFER_LIST`]: errors.html#errors_err_missing_message_port_in_transfer_list
 [`ERR_WORKER_NOT_RUNNING`]: errors.html#ERR_WORKER_NOT_RUNNING
 [`EventEmitter`]: events.html
@@ -944,3 +954,4 @@ active handle in the event system. If the worker is already `unref()`ed calling
 [child processes]: child_process.html
 [contextified]: vm.html#vm_what_does_it_mean_to_contextify_an_object
 [v8.serdes]: v8.html#v8_serialization_api
+[`data:` URL]: https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/Data_URIs

lib/internal/errors.js

@@ -1451,6 +1451,9 @@ E('ERR_WORKER_PATH', (filename) =>
 (filename.startsWith('file://') ?
 ' Wrap file:// URLs with `new URL`.' : ''
 )+
+(filename.startsWith('data:text/javascript') ?
+' Wrap data: URLs with `new URL`.' : ''
+)+
 ` Received "${filename}"`,
 TypeError);
 E('ERR_WORKER_UNSERIALIZABLE_ERROR',

lib/internal/main/worker_thread.js

@@ -149,7 +149,7 @@ port.on('message', (message) =>{
 debug(`[${threadId}] starts worker script ${filename} `+
 `(eval = ${eval}) at cwd = ${process.cwd()}`);
 port.postMessage({type: UP_AND_RUNNING});
-if(doEval){
+if(doEval==='classic'){
 const{ evalScript }=require('internal/process/execution');
 constname='[worker eval]';
 // This is necessary for CJS module compilation.
@@ -161,6 +161,11 @@ port.on('message', (message) =>{
 });
 process.argv.splice(1,0,name);
 evalScript(name,filename);
+}elseif(doEval==='module'){
+const{ evalModule }=require('internal/process/execution');
+evalModule(filename).catch((e)=>{
+workerOnGlobalUncaughtException(e,true);
+});
 }else{
 // script filename
 // runMain here might be monkey-patched by users in --require.

lib/internal/process/execution.js

@@ -45,7 +45,7 @@ function evalModule(source, print){
 const{ log }=require('internal/console/global');
 const{ loadESM }=require('internal/process/esm_loader');
 const{ handleMainPromise }=require('internal/modules/run_main');
-handleMainPromise(loadESM(async(loader)=>{
+returnhandleMainPromise(loadESM(async(loader)=>{
 const{ result }=awaitloader.eval(source);
 if(print){
 log(result);

lib/internal/worker.js

@@ -4,6 +4,7 @@
 
 const{
  ArrayIsArray,
+ JSONStringify,
  MathMax,
  ObjectCreate,
  ObjectEntries,
@@ -100,7 +101,7 @@ class Worker extends EventEmitter{
 argv=options.argv.map(String);
 }
 
-leturl;
+leturl,doEval;
 if(options.eval){
 if(typeoffilename!=='string'){
 thrownewERR_INVALID_ARG_VALUE(
@@ -110,7 +111,13 @@ class Worker extends EventEmitter{
 );
 }
 url=null;
+doEval='classic';
+}elseif(isURLInstance(filename)&&filename.protocol==='data:'){
+url=null;
+doEval='module';
+filename=`import ${JSONStringify(`${filename}`)}`;
 }else{
+doEval=false;
 if(isURLInstance(filename)){
 url=filename;
 filename=fileURLToPath(filename);
@@ -201,7 +208,7 @@ class Worker extends EventEmitter{
  argv,
 type: messageTypes.LOAD_SCRIPT,
  filename,
-doEval: !!options.eval,
+ doEval,
 cwdCounter: cwdCounter||workerIo.sharedCwdCounter,
 workerData: options.workerData,
 publicPort: port2,

test/parallel/test-worker-data-url.js

@@ -0,0 +1,25 @@
+'use strict';
+
+constcommon=require('../common');
+const{ Worker }=require('worker_threads');
+constassert=require('assert');
+
+newWorker(newURL('data:text/javascript,'))
+.on('error',common.mustNotCall(()=>{}));
+newWorker(newURL('data:text/javascript,export{}'))
+.on('error',common.mustNotCall(()=>{}));
+
+newWorker(newURL('data:text/plain,'))
+.on('error',common.mustCall(()=>{}));
+newWorker(newURL('data:text/javascript,module.exports={}'))
+.on('error',common.mustCall(()=>{}));
+
+newWorker(newURL('data:text/javascript,await Promise.resolve()'))
+.on('error',common.mustNotCall(()=>{}));
+newWorker(newURL('data:text/javascript,await Promise.reject()'))
+.on('error',common.mustCall(()=>{}));
+newWorker(newURL('data:text/javascript,await new Promise(()=>{})'))
+.on(
+'exit',
+common.mustCall((exitCode)=>{assert.strictEqual(exitCode,13);})
+);

test/parallel/test-worker-unsupported-path.js

@@ -33,6 +33,10 @@ const{Worker } = require('worker_threads');
 ()=>{newWorker('file:///file_url');},
 /Wrapfile:\/\/URLswith`newURL`/
 );
+assert.throws(
+()=>{newWorker('data:text/javascript,');},
+/Wrapdata:URLswith`newURL`/
+);
 assert.throws(
 ()=>{newWorker('relative_no_dot');},
 // eslint-disable-next-line node-core/no-unescaped-regexp-dot
@@ -47,6 +51,4 @@ const{Worker } = require('worker_threads');
 };
 assert.throws(()=>{newWorker(newURL('https://www.url.com'));},
 expectedErr);
-assert.throws(()=>{newWorker(newURL('data:application/javascript,'));},
-expectedErr);
 }