Replace the webpack+acorn transform with a Babel plugin

This commit converts the pdfjsdev-loader transform into a Babel plugin,
to skip a AST->string->AST round-trip.

Before this commit, the webpack build process was:
1. Babel parses the code
2. Babel transforms the AST
3. Babel generates the code
4. Acorn parses the code
5. pdfjsdev-loader transforms the AST
6. @javascript-obfuscator/escodegen generates the code
7. Webpack parses the file
8. Webpack concatenates the files

After this commit, it is reduced to:
1. Babel parses the code
2. Babel transforms the AST
3. babel-plugin-pdfjs-preprocessor transforms the AST
4. Babel generates the code
5. Webpack parses the file
6. Webpack concatenates the files

This change improves the build time by ~25% (tested on MacBook Air M2):
- `gulp lib`: 3.4s to 2.6s
- `gulp dist`: 36s to 29s
- `gulp generic`: 5.5s to 4.0s
- `gulp mozcentral`: 4.7s to 3.2s

The new Babel plugin doesn't support the `saveComments` option of
pdfjsdev-loader, and it just always discards comments. Even though
pdfjsdev-loader supported multiple values for that option, it was
effectively ignored due to `acorn` dropping comments by default.

external/builder/babel-plugin-pdfjs-preprocessor.mjs

@@ -0,0 +1,238 @@
+import { types as t, transformSync } from "@babel/core";
+import fs from "fs";
+import { join as joinPaths } from "path";
+import vm from "vm";
+
+const ROOT_PREFIX = "$ROOT/";
+const PDFJS_PREPROCESSOR_NAME = "PDFJSDev";
+
+function isPDFJSPreprocessor(obj) {
+  return obj.type === "Identifier" && obj.name === PDFJS_PREPROCESSOR_NAME;
+}
+
+function evalWithDefines(code, defines) {
+  if (!code || !code.trim()) {
+    throw new Error("No JavaScript expression given");
+  }
+  return vm.runInNewContext(code, defines, { displayErrors: false });
+}
+
+function handlePreprocessorAction(ctx, actionName, args, path) {
+  try {
+    const arg = args[0];
+    switch (actionName) {
+      case "test":
+        if (!t.isStringLiteral(arg)) {
+          throw new Error("No code for testing is given");
+        }
+        return !!evalWithDefines(arg.value, ctx.defines);
+      case "eval":
+        if (!t.isStringLiteral(arg)) {
+          throw new Error("No code for eval is given");
+        }
+        const result = evalWithDefines(arg.value, ctx.defines);
+        if (
+          typeof result === "boolean" ||
+          typeof result === "string" ||
+          typeof result === "number" ||
+          typeof result === "object"
+        ) {
+          return result;
+        }
+        break;
+      case "json":
+        if (!t.isStringLiteral(arg)) {
+          throw new Error("Path to JSON is not provided");
+        }
+        let jsonPath = arg.value;
+        if (jsonPath.indexOf(ROOT_PREFIX) === 0) {
+          jsonPath = joinPaths(
+            ctx.rootPath,
+            jsonPath.substring(ROOT_PREFIX.length)
+          );
+        }
+        return JSON.parse(fs.readFileSync(jsonPath, "utf8"));
+    }
+    throw new Error("Unsupported action");
+  } catch (e) {
+    throw path.buildCodeFrameError(
+      "Could not process " +
+        PDFJS_PREPROCESSOR_NAME +
+        "." +
+        actionName +
+        ": " +
+        e.message
+    );
+  }
+}
+
+function babelPluginPDFJSPreprocessor(babel, ctx) {
+  return {
+    name: "babel-plugin-pdfjs-preprocessor",
+    manipulateOptions({ parserOpts }) {
+      parserOpts.attachComment = false;
+    },
+    visitor: {
+      "ExportNamedDeclaration|ImportDeclaration": ({ node }) => {
+        if (node.source && ctx.map && ctx.map[node.source.value]) {
+          const newValue = ctx.map[node.source.value];
+          node.source.value = newValue;
+        }
+      },
+      "IfStatement|ConditionalExpression": {
+        exit(path) {
+          const { node } = path;
+          if (t.isBooleanLiteral(node.test)) {
+            // if (true) stmt1; => stmt1
+            // if (false) stmt1; else stmt2; => stmt2
+            path.replaceWith(
+              node.test.value === true
+                ? node.consequent
+                : node.alternate || t.emptyStatement()
+            );
+          }
+        },
+      },
+      UnaryExpression(path) {
+        const { node } = path;
+        if (node.operator === "typeof" && isPDFJSPreprocessor(node.argument)) {
+          // typeof PDFJSDev => 'object'
+          path.replaceWith(t.stringLiteral("object"));
+          return;
+        }
+        if (node.operator === "!" && t.isBooleanLiteral(node.argument)) {
+          // !true => false,  !false => true
+          path.replaceWith(t.booleanLiteral(!node.argument.value));
+        }
+      },
+      LogicalExpression: {
+        exit(path) {
+          const { node } = path;
+          if (!t.isBooleanLiteral(node.left)) {
+            return;
+          }
+
+          switch (node.operator) {
+            case "&&":
+              // true && expr => expr
+              // false && expr => false
+              path.replaceWith(
+                node.left.value === true ? node.right : node.left
+              );
+              break;
+            case "||":
+              // true || expr => true
+              // false || expr => expr
+              path.replaceWith(
+                node.left.value === true ? node.left : node.right
+              );
+              break;
+          }
+        },
+      },
+      BinaryExpression: {
+        exit(path) {
+          const { node } = path;
+
+          if (["==", "===", "!=", "!=="].includes(node.operator)) {
+            // folding == and != check that can be statically evaluated
+            const { confident, value } = path.evaluate();
+            if (confident) {
+              path.replaceWith(t.booleanLiteral(value));
+            }
+          }
+        },
+      },
+      CallExpression(path) {
+        const { node } = path;
+        if (
+          t.isMemberExpression(node.callee) &&
+          isPDFJSPreprocessor(node.callee.object) &&
+          t.isIdentifier(node.callee.property) &&
+          !node.callee.computed
+        ) {
+          // PDFJSDev.xxxx(arg1, arg2, ...) => transform
+          const action = node.callee.property.name;
+          const result = handlePreprocessorAction(
+            ctx,
+            action,
+            node.arguments,
+            path
+          );
+          path.replaceWith(t.inherits(t.valueToNode(result), path.node));
+        }
+
+        // require('string')
+        if (
+          t.isIdentifier(node.callee, { name: "require" }) &&
+          node.arguments.length === 1 &&
+          t.isStringLiteral(node.arguments[0]) &&
+          ctx.map &&
+          ctx.map[node.arguments[0].value]
+        ) {
+          const requireName = node.arguments[0];
+          requireName.value = requireName.raw = ctx.map[requireName.value];
+        }
+      },
+      BlockStatement: {
+        // Visit node in post-order so that recursive flattening
+        // of blocks works correctly.
+        exit(path) {
+          const { node } = path;
+
+          let subExpressionIndex = 0;
+          while (subExpressionIndex < node.body.length) {
+            switch (node.body[subExpressionIndex].type) {
+              case "EmptyStatement":
+                // Removing empty statements from the blocks.
+                node.body.splice(subExpressionIndex, 1);
+                continue;
+              case "BlockStatement":
+                // Block statements inside a block are flattened
+                // into the parent one.
+                const subChildren = node.body[subExpressionIndex].body;
+                node.body.splice(subExpressionIndex, 1, ...subChildren);
+                subExpressionIndex += Math.max(subChildren.length - 1, 0);
+                continue;
+              case "ReturnStatement":
+              case "ThrowStatement":
+                // Removing dead code after return or throw.
+                node.body.splice(
+                  subExpressionIndex + 1,
+                  node.body.length - subExpressionIndex - 1
+                );
+                break;
+            }
+            subExpressionIndex++;
+          }
+        },
+      },
+      Function: {
+        exit(path) {
+          if (!t.isBlockStatement(path.node.body)) {
+            // Arrow function with expression body
+            return;
+          }
+
+          const { body } = path.node.body;
+          if (
+            body.length > 0 &&
+            t.isReturnStatement(body.at(-1), { argument: null })
+          ) {
+            // Function body ends with return without arg -- removing it.
+            body.pop();
+          }
+        },
+      },
+    },
+  };
+}
+
+function preprocessPDFJSCode(ctx, content) {
+  return transformSync(content, {
+    configFile: false,
+    plugins: [[babelPluginPDFJSPreprocessor, ctx]],
+  }).code;
+}
+
+export { babelPluginPDFJSPreprocessor, preprocessPDFJSCode };

external/builder/fixtures_esprima/blocks-expected.js

@@ -1,10 +1,10 @@
 function test() {
- "test";
- "1";
- "2";
- "3";
- if ("test") {
-  "5";
- }
- "4";
+  "test";
+  "1";
+  "2";
+  "3";
+  if ("test") {
+    "5";
+  }
+  "4";
 }

external/builder/fixtures_esprima/comments-expected.js

@@ -1,17 +1,17 @@
 function f1() {
- "1";
- "2";
+  "1";
+  "2";
 }
 function f2() {
- "1";
- "2";
+  "1";
+  "2";
 }
 function f3() {
- if ("1") {
-  "1";
- }
- "2";
- if ("3") {
-  "4";
- }
+  if ("1") {
+    "1";
+  }
+  "2";
+  if ("3") {
+    "4";
+  }
 }

external/builder/fixtures_esprima/constants-expected.js

@@ -10,6 +10,6 @@ var i = true;
 var j = false;
 var k = false;
 var l = true;
-var m = '1' === true;
+var m = false;
 var n = false;
 var o = true;

external/builder/fixtures_esprima/deadcode-expected.js

@@ -1,13 +1,21 @@
-function f1() {
-}
+function f1() {}
 function f2() {
- return 1;
+  return 1;
 }
 function f3() {
- var i = 0;
- throw "test";
+  var i = 0;
+  throw "test";
 }
 function f4() {
- var i = 0;
+  var i = 0;
+}
+var obj = {
+  method1() {},
+  method2() {}
+};
+class C {
+  method1() {}
+  method2() {}
 }
-
+var arrow1 = () => {};
+var arrow2 = () => {};

external/builder/fixtures_esprima/deadcode.js

@@ -23,3 +23,15 @@ function f4() {
   var j = 0;
 }
 
+var obj = {
+  method1() { return; var i = 0; },
+  method2() { return; },
+};
+
+class C {
+  method1() { return; var i = 0; }
+  method2() { return; }
+}
+
+var arrow1 = () => { return; var i = 0; };
+var arrow2 = () => { return; };

external/builder/fixtures_esprima/evals-expected.js

@@ -3,11 +3,17 @@ var b = true;
 var c = true;
 var d = false;
 var e = true;
-var f = 'text';
+var f = "text";
 var g = {
- "obj": { "i": 1 },
- "j": 2
+  obj: {
+    i: 1
+  },
+  j: 2
+};
+var h = {
+  test: "test"
 };
-var h = { 'test': 'test' };
 var i = '0';
-var j = { "i": 1 };
+var j = {
+  i: 1
+};

external/builder/fixtures_esprima/evals.json

@@ -1 +1 @@
-{ 'test': 'test' }
+{ "test": "test" }

external/builder/fixtures_esprima/ifs-expected.js

@@ -1,17 +1,17 @@
 if ('test') {
- "1";
+  "1";
 }
 {
- "1";
+  "1";
 }
 {
- "1";
+  "1";
 }
 ;
 {
- "2";
+  "2";
 }
 ;
 if ('1') {
- "1";
+  "1";
 }