Move base64 decoding into JS library. NFC (#21441)

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.

Author: sbc100

eslint.config.mjs

@@ -55,7 +55,6 @@ export default [{
     'src/growableHeap.js',
     'src/emrun_prejs.js',
     'src/deterministic.js',
-    'src/base64Decode.js',
     'src/proxyWorker.js',
     'src/proxyClient.js',
     'src/IDBStore.js',

src/base64Decode.js

@@ -0,0 +1,66 @@
+/**
+ * @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.length);
+    }
+#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;
+  },
+
+  // Prefix of data URIs emitted by SINGLE_FILE and related options.
+  // Double quotes here needed, othersise jsifier will not include any in the
+  // output.
+  $dataURIPrefix: "'data:application/octet-stream;base64,'",
+
+  /**
+   * Indicates whether filename is a base64 data URI.
+   */
+  $isDataURI: (filename) => filename.startsWith(dataURIPrefix),
+
+  // 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.
+  $tryParseAsDataURI__deps: ['$base64Decode', '$isDataURI', '$dataURIPrefix'],
+  $tryParseAsDataURI: (filename) => {
+    if (isDataURI(filename)) {
+      return base64Decode(filename.slice(dataURIPrefix.length));
+    }
+  },
+});

src/lib/libbase64.js

@@ -102,6 +102,10 @@ export const LibraryManager = {
       libraries.push('libmemoryprofiler.js');
     }
 
+    if (SUPPORT_BASE64_EMBEDDING) {
+      libraries.push('libbase64.js');
+    }
+
     if (AUTODEBUG) {
       libraries.push('libautodebug.js');
     }

src/modules.mjs

@@ -82,6 +82,10 @@ function initRuntime(wasmExports) {
 
 // Initialize wasm (asynchronous)
 
+#if SINGLE_FILE && WASM == 1 && !WASM2JS
+Module['wasm'] = base64Decode('<<< WASM_BINARY_DATA >>>');
+#endif
+
 // In non-fastcomp non-asm.js builds, grab wasm exports to outer scope
 // for emscripten_get_exported_function() to be able to access them.
 #if LibraryManager.has('libexports.js')
@@ -106,10 +110,6 @@ var imports = {
 #endif // MINIFY_WASM_IMPORTED_MODULES
 };
 
-#if SINGLE_FILE && WASM == 1 && !WASM2JS
-Module['wasm'] = base64Decode('<<< WASM_BINARY_DATA >>>');
-#endif
-
 #if MINIMAL_RUNTIME_STREAMING_WASM_INSTANTIATION
 // https://caniuse.com/#feat=wasm and https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/WebAssembly/instantiateStreaming
 // Firefox 52 added Wasm support, but only Firefox 58 added instantiateStreaming.

src/postamble_minimal.js

@@ -17,10 +17,6 @@
 #include "growableHeap.js"
 #endif
 
-#if SUPPORT_BASE64_EMBEDDING
-#include "base64Decode.js"
-#endif
-
 #if USE_ASAN
 #include "runtime_asan.js"
 #endif

src/runtime_shared.js

@@ -1,6 +1,6 @@
 {
-  "a.html": 12449,
-  "a.html.gz": 6808,
-  "total": 12449,
-  "total_gz": 6808
+  "a.html": 12438,
+  "a.html.gz": 6795,
+  "total": 12438,
+  "total_gz": 6795
 }

test/code_size/random_printf_wasm.json

@@ -1938,7 +1938,12 @@ def phase_emscript(in_wasm, wasm_target, js_syms, base_metadata):
 
   # No need to support base64 embedding in wasm2js mode since
   # the module is already in JS format.
-  settings.SUPPORT_BASE64_EMBEDDING = settings.SINGLE_FILE and not settings.WASM2JS
+  if settings.SINGLE_FILE and not settings.WASM2JS:
+    settings.SUPPORT_BASE64_EMBEDDING = 1
+    if settings.MINIMAL_RUNTIME:
+      settings.DEFAULT_LIBRARY_FUNCS_TO_INCLUDE.append('$base64Decode')
+    else:
+      settings.DEFAULT_LIBRARY_FUNCS_TO_INCLUDE.append('$tryParseAsDataURI')
 
   if shared.SKIP_SUBPROCS:
     return