Add modularize to file_packager.py

fix python format

apply more ruff fixes

fix pipeline

revert file to the most actual version and apply changes

add modularize to options class

add utilization of EXPORT_NAME with modularize option

simplify declaration of the modularize function

arg documentation

address @sbc100 review

Fix codegen new line formatting

Add file_packager tests and changelog mention

fix some stylistic issues

gichange file_packager modularize test

simplyfi tests

fix test

rename to export_es6

async function promise behaviour

cleanup test

address @sbc100 review

simplify the test

rename promise handlers

introduce createRequire in es6 mode

alternative way to import require

Author: lkwinta

ChangeLog.md

@@ -41,6 +41,8 @@ See docs/process.md for more on how version tagging works.
 - emcc will now error if `MINIMAL_RUNTIME_STREAMING_WASM_COMPILATION` or
   `MINIMAL_RUNTIME_STREAMING_WASM_INSTANTIATION` are used with `SINGLE_FILE`.
   These are fundamentally incompatible but were previously ignored. (#24849)
+- `--export-es6` flag was added to `file_packager.py` available when run 
+  standalone, to enable ES6 imports of generated JavaScript code (#24737)
 
 4.0.12 - 08/01/25
 -----------------

test/test_other.py

@@ -3919,6 +3919,11 @@ def test_file_packager_returns_error_if_target_equal_to_jsoutput(self):
     err = self.expect_fail([FILE_PACKAGER, 'test.data', '--js-output=test.data'])
     self.assertContained(MESSAGE, err)
 
+  def test_file_packager_returns_error_if_emcc_and_export_es6(self):
+    MESSAGE = 'error: Can\'t use --export-es6 option together with --from-emcc since the code should be embedded within emcc\'s code'
+    err = self.expect_fail([FILE_PACKAGER, 'test.data', '--export-es6', '--from-emcc'])
+    self.assertContained(MESSAGE, err)
+
   def test_file_packager_embed(self):
     create_file('data.txt', 'hello data')
 
@@ -3945,6 +3950,43 @@ def test_file_packager_embed(self):
     output = self.run_js('a.out.js')
     self.assertContained('hello data', output)
 
+  def test_file_packager_export_es6(self):
+    create_file('smth.txt', 'hello data')
+    self.run_process([FILE_PACKAGER, 'test.data', '--export-es6', '--preload', 'smth.txt', '--js-output=dataFileLoader.js'])
+
+    create_file('test.c', '''
+    #include <stdio.h>
+    #include <emscripten.h>
+
+    EMSCRIPTEN_KEEPALIVE int test_fun() {
+      FILE* f = fopen("smth.txt", "r");
+      char buf[64] = {0};
+      int rtn = fread(buf, 1, 64, f);
+      buf[rtn] = '\\0';
+      fclose(f);
+      printf("%s\\n", buf);
+      return 0;
+    }
+    ''')
+    self.run_process([EMCC, 'test.c', '-sFORCE_FILESYSTEM', '-sMODULARIZE', '-sEXPORT_ES6', '-o', 'moduleFile.js'])
+
+    create_file('run.js', '''
+    import loadDataFile from './dataFileLoader.js'
+    import {default as loadModule} from './moduleFile.js'
+
+    loadModule().then((module) => {
+      loadDataFile(module)
+        .catch((cause) => Promise.reject(cause))
+        .then(() => {
+          module._test_fun();
+        }
+      );
+    });
+    ''')
+
+    output = self.run_js('run.js')
+    self.assertContained('hello data', output)
+
   @crossplatform
   def test_file_packager_depfile(self):
     create_file('data1.txt', 'data1')

tools/file_packager.py

@@ -21,7 +21,7 @@
 
 Usage:
 
-  file_packager TARGET [--preload A [B..]] [--embed C [D..]] [--exclude E [F..]] [--js-output=OUTPUT.js] [--no-force] [--use-preload-cache] [--indexedDB-name=EM_PRELOAD_CACHE] [--separate-metadata] [--lz4] [--use-preload-plugins] [--no-node] [--help]
+  file_packager TARGET [--preload A [B..]] [--embed C [D..]] [--exclude E [F..]] [--js-output=OUTPUT.js] [--no-force] [--use-preload-cache] [--indexedDB-name=EM_PRELOAD_CACHE] [--separate-metadata] [--lz4] [--use-preload-plugins] [--no-node] [--export-es6] [--help]
 
   --preload  ,
   --embed    See emcc --help for more details on those options.
@@ -41,6 +41,8 @@
 
   --export-name=EXPORT_NAME Use custom export name (default is `Module`)
 
+  --export-es6 Wrap generated code inside ES6 exported function
+
   --no-force Don't create output if no valid input file is specified.
 
   --use-preload-cache Stores package in IndexedDB so that subsequent loads don't need to do XHR. Checks package version.
@@ -129,6 +131,7 @@ def __init__(self):
     self.use_preload_plugins = False
     self.support_node = True
     self.wasm64 = False
+    self.export_es6 = False
 
 
 class DataFile:
@@ -362,7 +365,7 @@ def main():  # noqa: C901, PLR0912, PLR0915
   To revalidate these numbers, run `ruff check --select=C901,PLR091`.
   """
   if len(sys.argv) == 1:
-    err('''Usage: file_packager TARGET [--preload A [B..]] [--embed C [D..]] [--exclude E [F..]] [--js-output=OUTPUT.js] [--no-force] [--use-preload-cache] [--indexedDB-name=EM_PRELOAD_CACHE] [--separate-metadata] [--lz4] [--use-preload-plugins] [--no-node] [--help]
+    err('''Usage: file_packager TARGET [--preload A [B..]] [--embed C [D..]] [--exclude E [F..]] [--js-output=OUTPUT.js] [--no-force] [--use-preload-cache] [--indexedDB-name=EM_PRELOAD_CACHE] [--separate-metadata] [--lz4] [--use-preload-plugins] [--no-node] [--export-es6] [--help]
   Try 'file_packager --help' for more details.''')
     return 1
 
@@ -391,6 +394,9 @@ def main():  # noqa: C901, PLR0912, PLR0915
     elif arg == '--no-force':
       options.force = False
       leading = ''
+    elif arg == '--export-es6':
+      options.export_es6 = True
+      leading = ''
     elif arg == '--use-preload-cache':
       options.use_preload_cache = True
       leading = ''
@@ -485,6 +491,11 @@ def main():  # noqa: C901, PLR0912, PLR0915
     diagnostics.error('TARGET should not be the same value of --js-output')
     return 1
 
+  if options.from_emcc and options.export_es6:
+    diagnostics.error('Can\'t use --export-es6 option together with --from-emcc since the code should be embedded '
+        'within emcc\'s code')
+    return 1
+
   walked.append(__file__)
   for file_ in data_files:
     if not should_ignore(file_.srcpath):
@@ -621,20 +632,38 @@ def generate_js(data_target, data_files, metadata):
   if options.from_emcc:
     ret = ''
   else:
-    ret = '''
+    if options.export_es6:
+      ret = 'export default async function loadDataFile(Module) {\n'
+    else:
+      ret = '''
   var Module = typeof %(EXPORT_NAME)s != 'undefined' ? %(EXPORT_NAME)s : {};\n''' % {"EXPORT_NAME": options.export_name}
 
   ret += '''
   Module['expectedDataFileDownloads'] ??= 0;
-  Module['expectedDataFileDownloads']++;
-  (() => {
+  Module['expectedDataFileDownloads']++;'''
+
+  if not options.export_es6:
+    ret += '''
+  (() => {'''
+
+  ret += '''
     // Do not attempt to redownload the virtual filesystem data when in a pthread or a Wasm Worker context.
     var isPthread = typeof ENVIRONMENT_IS_PTHREAD != 'undefined' && ENVIRONMENT_IS_PTHREAD;
     var isWasmWorker = typeof ENVIRONMENT_IS_WASM_WORKER != 'undefined' && ENVIRONMENT_IS_WASM_WORKER;
     if (isPthread || isWasmWorker) return;\n'''
 
   if options.support_node:
     ret += "    var isNode = typeof process === 'object' && typeof process.versions === 'object' && typeof process.versions.node === 'string';\n"
+
+  if options.support_node and options.export_es6:
+        ret += '''if (isNode) {
+    const { createRequire } = await import('module');
+    /** @suppress{duplicate} */
+    var require = createRequire(import.meta.url);
+  }\n'''
+
+  if options.export_es6:
+    ret += 'return new Promise((loadDataResolve, loadDataReject) => {\n'
   ret += '    async function loadPackage(metadata) {\n'
 
   code = '''
@@ -684,6 +713,8 @@ def generate_js(data_target, data_files, metadata):
     create_data = '''// canOwn this data in the filesystem, it is a slide into the heap that will never change
           Module['FS_createDataFile'](this.name, null, byteArray, true, true, true);
           Module['removeRunDependency'](`fp ${that.name}`);'''
+    ready_promise = '''
+          loadDataResolve();'''
 
     if not options.lz4:
       # Data requests - for getting a block of data out of the big archive - have
@@ -710,14 +741,14 @@ def generate_js(data_target, data_files, metadata):
         finish: function(byteArray) {
           var that = this;
           %s
-          this.requests[this.name] = null;
+          this.requests[this.name] = null;%s
         }
       };
 
       var files = metadata['files'];
       for (var i = 0; i < files.length; ++i) {
         new DataRequest(files[i]['start'], files[i]['end'], files[i]['audio'] || 0).open('GET', files[i]['filename']);
-      }\n''' % (create_preloaded if options.use_preload_plugins else create_data)
+      }\n''' % (create_preloaded if options.use_preload_plugins else create_data, ready_promise if options.export_es6 else '')
 
   if options.has_embedded and not options.obj_output:
     diagnostics.warn('--obj-output is recommended when using --embed.  This outputs an object file for linking directly into your application is more efficient than JS encoding')
@@ -958,6 +989,9 @@ def generate_js(data_target, data_files, metadata):
           return contents.buffer;
         }'''.strip()
 
+    reject_promise = '''
+          loadDataReject();'''
+
     ret += '''
       async function fetchRemotePackage(packageName, packageSize) {
         %(node_support_code)s
@@ -1128,7 +1162,14 @@ def generate_js(data_target, data_files, metadata):
     }
     loadPackage(%s);\n''' % json.dumps(metadata)
 
-  ret += '''
+  if options.export_es6:
+    ret += '''
+  });
+}
+// END the loadDataFile function
+'''
+  else:
+    ret += '''
   })();\n'''
 
   return ret