Merge base64Decode.js and base64Utils.js into library_base64.js

The optimized base64Decode was only being using by minimal runtime.
After this change it used in the regular runtime too which should speed
up decoding.

.eslintrc.yml

@@ -42,8 +42,6 @@ ignorePatterns:
   - "src/settings_internal.js"
   - "src/arrayUtils.js"
   - "src/deterministic.js"
-  - "src/base64Utils.js"
-  - "src/base64Decode.js"
   - "src/proxyWorker.js"
   - "src/proxyClient.js"
   - "src/IDBStore.js"

src/library_base64.js

@@ -0,0 +1,57 @@
+/**
+ * @license
+ * Copyright 2020 The Emscripten Authors
+ * SPDX-License-Identifier: MIT
+ */
+
+addToLibrary({
+  // Decodes a _known valid_ base64 string (without validation) and returns it as a new Uint8Array.
+  // Benchmarked to be around 5x faster compared to a simple
+  // "Uint8Array.from(atob(b64), c => c.charCodeAt(0))" (TODO: perhaps use this form in -Oz builds?)
+  $base64Decode__postset: `
+  // Precreate a reverse lookup table from chars
+  // "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/" back to
+  // bytes to make decoding fast.
+  for (var base64ReverseLookup = new Uint8Array(123/*'z'+1*/), i = 25; i >= 0; --i) {
+    base64ReverseLookup[48+i] = 52+i; // '0-9'
+    base64ReverseLookup[65+i] = i; // 'A-Z'
+    base64ReverseLookup[97+i] = 26+i; // 'a-z'
+  }
+  base64ReverseLookup[43] = 62; // '+'
+  base64ReverseLookup[47] = 63; // '/'
+`,
+  $base64Decode__docs: '/** @noinline */',
+  $base64Decode: (b64) => {
+#if ENVIRONMENT_MAY_BE_NODE
+    if (ENVIRONMENT_IS_NODE) {
+      var buf = Buffer.from(b64, 'base64');
+      return new Uint8Array(buf.buffer, buf.byteOffset, buf.byteLength);
+    }
+#endif
+
+#if ASSERTIONS
+    assert(b64.length % 4 == 0);
+#endif
+    var b1, b2, i = 0, j = 0, bLength = b64.length;
+    var output = new Uint8Array((bLength*3>>2) - (b64[bLength-2] == '=') - (b64[bLength-1] == '='));
+    for (; i < bLength; i += 4, j += 3) {
+      b1 = base64ReverseLookup[b64.charCodeAt(i+1)];
+      b2 = base64ReverseLookup[b64.charCodeAt(i+2)];
+      output[j] = base64ReverseLookup[b64.charCodeAt(i)] << 2 | b1 >> 4;
+      output[j+1] = b1 << 4 | b2 >> 2;
+      output[j+2] = b2 << 6 | base64ReverseLookup[b64.charCodeAt(i+3)];
+    }
+    return output;
+  },
+
+  // If filename is a base64 data URI, parses and returns data (Buffer on node,
+  // Uint8Array otherwise). If filename is not a base64 data URI, returns
+  // undefined.
+  $tryParseAsDataURI: (filename) => {
+    if (isDataURI(filename)) {
+      return base64Decode(filename.slice(dataURIPrefix.length));
+    }
+  },
+
+  $intArrayFromBase64: 'intArrayFromBase64',
+});

src/modules.js

@@ -71,6 +71,10 @@ globalThis.LibraryManager = {
       libraries.push('library_memoryprofiler.js');
     }
 
+    if (SUPPORT_BASE64_EMBEDDING) {
+      libraries.push('library_base64.js');
+    }
+
     if (AUTODEBUG) {
       libraries.push('library_autodebug.js');
     }
@@ -416,11 +420,6 @@ function exportRuntime() {
     runtimeElements.push('checkStackCookie');
   }
 
-  if (SUPPORT_BASE64_EMBEDDING) {
-    runtimeElements.push('intArrayFromBase64');
-    runtimeElements.push('tryParseAsDataURI');
-  }
-
   if (RETAIN_COMPILER_SETTINGS) {
     runtimeElements.push('getCompilerSetting')
   }

src/postamble_minimal.js

@@ -89,6 +89,10 @@ function initRuntime(wasmExports) {
 
 // Initialize wasm (asynchronous)
 
+#if SINGLE_FILE && WASM == 1 && !WASM2JS
+Module['wasm'] = base64Decode('<<< WASM_BINARY_DATA >>>');
+#endif
+
 var imports = {
 #if MINIFY_WASM_IMPORTED_MODULES
   'a': wasmImports,

src/preamble.js

@@ -44,10 +44,6 @@ if (typeof WebAssembly != 'object') {
 #include "runtime_asan.js"
 #endif
 
-#if SUPPORT_BASE64_EMBEDDING || FORCE_FILESYSTEM
-#include "base64Utils.js"
-#endif
-
 // Wasm globals
 
 var wasmMemory;

src/preamble_minimal.js

@@ -51,11 +51,6 @@ if (Module['doWasm2JS']) {
 #endif
 #endif
 
-#if SINGLE_FILE && WASM == 1 && !WASM2JS
-#include "base64Decode.js"
-Module['wasm'] = base64Decode('<<< WASM_BINARY_DATA >>>');
-#endif
-
 var HEAP8, HEAP16, HEAP32, HEAPU8, HEAPU16, HEAPU32, HEAPF32, HEAPF64,
 #if WASM_BIGINT
   HEAP64, HEAPU64,

src/settings_internal.js

@@ -148,7 +148,7 @@ var SOURCE_MAP_BASE = '';
 // this setting will always be true.
 var MEM_INIT_IN_WASM = true;
 
-// If set to 1, src/base64Utils.js will be included in the bundle.
+// If set to 1 then base64 decoding functions will be included in the bundle.
 // This is set internally when needed (SINGLE_FILE)
 var SUPPORT_BASE64_EMBEDDING = false;
 

src/shell.js

@@ -61,6 +61,10 @@ var Module = typeof {{{ EXPORT_NAME }}} != 'undefined' ? {{{ EXPORT_NAME }}} : {
 // See https://caniuse.com/mdn-javascript_builtins_bigint64array
 #include "polyfill/bigint64array.js"
 #endif
+
+#if SUPPORT_BASE64_EMBEDDING && (ENVIRONMENT_MAY_BE_SHELL || (ENVIRONMENT_MAY_BE_NODE && MIN_NODE_VERSION < 160000))
+#include "polyfill/atob.js"
+#endif
 #endif // POLYFILL
 
 #if MODULARIZE

test/code_size/random_printf_wasm.json

@@ -1,6 +1,6 @@
 {
-  "a.html": 12734,
-  "a.html.gz": 6968,
-  "total": 12734,
-  "total_gz": 6968
+  "a.html": 12723,
+  "a.html.gz": 6958,
+  "total": 12723,
+  "total_gz": 6958
 }

test/code_size/random_printf_wasm2js.json

@@ -1,6 +1,6 @@
 {
-  "a.html": 17275,
-  "a.html.gz": 7537,
-  "total": 17275,
-  "total_gz": 7537
+  "a.html": 17371,
+  "a.html.gz": 7580,
+  "total": 17371,
+  "total_gz": 7580
 }

tools/link.py

@@ -193,14 +193,6 @@ def setup_environment_settings():
     exit_with_error('when building with multithreading enabled and a "-sENVIRONMENT=" directive is specified, it must include "worker" as a target! (Try e.g. -sENVIRONMENT=web,worker)')
 
 
-def embed_memfile(options):
-  return (settings.SINGLE_FILE or
-          (settings.WASM2JS and not options.memory_init_file and
-           (not settings.MAIN_MODULE and
-            not settings.SIDE_MODULE and
-            not settings.GENERATE_SOURCE_MAP)))
-
-
 def generate_js_sym_info():
   # Runs the js compiler to generate a list of all symbols available in the JS
   # libraries.  This must be done separately for each linker invocation since the
@@ -1796,6 +1788,15 @@ def get_full_import_name(name):
   if settings.USE_CLOSURE_COMPILER or not settings.MINIFY_WHITESPACE:
     settings.MAYBE_CLOSURE_COMPILER = 1
 
+  if (settings.SINGLE_FILE or (settings.WASM2JS and not options.memory_init_file and
+           (not settings.MAIN_MODULE and
+            not settings.SIDE_MODULE and
+            not settings.GENERATE_SOURCE_MAP))):
+    settings.SUPPORT_BASE64_EMBEDDING = 1
+    settings.DEFAULT_LIBRARY_FUNCS_TO_INCLUDE.append('$base64Decode')
+    if not settings.MINIMAL_RUNTIME:
+      settings.DEFAULT_LIBRARY_FUNCS_TO_INCLUDE.append('$tryParseAsDataURI')
+
   return target, wasm_target
 
 
@@ -1870,8 +1871,6 @@ def phase_emscript(options, in_wasm, wasm_target, js_syms):
   # Emscripten
   logger.debug('emscript')
 
-  settings.SUPPORT_BASE64_EMBEDDING = embed_memfile(options)
-
   if shared.SKIP_SUBPROCS:
     return
 
@@ -2080,7 +2079,7 @@ def phase_final_emitting(options, state, target, wasm_target, memfile):
   elif settings.PROXY_TO_WORKER:
     generate_worker_js(target, js_target, target_basename)
 
-  if embed_memfile(options) and memfile:
+  if memfile and (settings.SINGLE_FILE or (settings.WASM2JS and not options.memory_init_file)):
     delete_file(memfile)
 
   if settings.SPLIT_MODULE:
@@ -2446,7 +2445,7 @@ def generate_traditional_runtime_html(target, options, js_target, target_basenam
   }
 '''
     # add required helper functions such as tryParseAsDataURI
-    for filename in ('arrayUtils.js', 'base64Utils.js', 'URIUtils.js'):
+    for filename in ('arrayUtils.js', 'URIUtils.js'):
       content = shared.read_and_preprocess(utils.path_from_root('src', filename))
       script.inline = content + script.inline