Do not delete "fake" source map comments from strings (#9960)

Instead of using `convert-source-map`'s `removeComments` method before
parsing the file, we can first parse the file with `@babel/parser` and then
analyze the comments.
This is needed because it is not possible to reliabily detect comments in
JavaScript without fully parsing the file:
thlorenz/convert-source-map#63

lib/third-party-libs.js.flow

@@ -181,7 +181,7 @@ declare module "convert-source-map" {
     fromJSON(str: string): Converter,
     fromBase64(str: string): Converter,
     fromComment(str: string): Converter,
-    fromMapFileComment(str: string): Converter,
+    fromMapFileComment(str: string, dir: string): Converter,
     fromSource(str: string): Converter,
     fromMapFileSource(str: string, dir: string): Converter,
     removeComments(str: string): string,

packages/babel-core/src/transformation/normalize-file.js

@@ -27,6 +27,20 @@ export default function normalizeFile(
 ): File {
   code = `${code || ""}`;
 
+  if (ast) {
+    if (ast.type === "Program") {
+      ast = t.file(ast, [], []);
+    } else if (ast.type !== "File") {
+      throw new Error("AST root must be a Program or File node");
+    }
+    ast = cloneDeep(ast);
+  } else {
+    // The parser's AST types aren't fully compatible with the types generated
+    // by the logic in babel-types.
+    // $FlowFixMe
+    ast = parser(pluginPasses, options, code);
+  }
+
   let inputMap = null;
   if (options.inputSourceMap !== false) {
     // If an explicit object is passed in, it overrides the processing of
@@ -36,54 +50,33 @@ export default function normalizeFile(
     }
 
     if (!inputMap) {
-      try {
-        inputMap = convertSourceMap.fromSource(code);
-
-        if (inputMap) {
-          code = convertSourceMap.removeComments(code);
+      const lastComment = extractComments(INLINE_SOURCEMAP_REGEX, ast);
+      if (lastComment) {
+        try {
+          inputMap = convertSourceMap.fromComment(lastComment);
+        } catch (err) {
+          debug("discarding unknown inline input sourcemap", err);
         }
-      } catch (err) {
-        debug("discarding unknown inline input sourcemap", err);
-        code = convertSourceMap.removeComments(code);
       }
     }
 
     if (!inputMap) {
-      if (typeof options.filename === "string") {
+      const lastComment = extractComments(EXTERNAL_SOURCEMAP_REGEX, ast);
+      if (typeof options.filename === "string" && lastComment) {
         try {
-          inputMap = convertSourceMap.fromMapFileSource(
-            code,
+          inputMap = convertSourceMap.fromMapFileComment(
+            lastComment,
             path.dirname(options.filename),
           );
-
-          if (inputMap) {
-            code = convertSourceMap.removeMapFileComments(code);
-          }
         } catch (err) {
           debug("discarding unknown file input sourcemap", err);
-          code = convertSourceMap.removeMapFileComments(code);
         }
-      } else {
+      } else if (lastComment) {
         debug("discarding un-loadable file input sourcemap");
-        code = convertSourceMap.removeMapFileComments(code);
       }
     }
   }
 
-  if (ast) {
-    if (ast.type === "Program") {
-      ast = t.file(ast, [], []);
-    } else if (ast.type !== "File") {
-      throw new Error("AST root must be a Program or File node");
-    }
-    ast = cloneDeep(ast);
-  } else {
-    // The parser's AST types aren't fully compatible with the types generated
-    // by the logic in babel-types.
-    // $FlowFixMe
-    ast = parser(pluginPasses, options, code);
-  }
-
   return new File(options, {
     code,
     ast,
@@ -156,3 +149,48 @@ function parser(
     throw err;
   }
 }
+
+// These regexps are copied from the convert-source-map package,
+// but without // or /* at the beginning of the comment.
+
+// eslint-disable-next-line max-len
+const INLINE_SOURCEMAP_REGEX = /^[@#]\s+sourceMappingURL=data:(?:application|text)\/json;(?:charset[:=]\S+?;)?base64,(?:.*)$/;
+const EXTERNAL_SOURCEMAP_REGEX = /^[@#][ \t]+sourceMappingURL=([^\s'"`]+?)[ \t]*$/;
+
+function extractCommentsFromList(regex, comments, lastComment) {
+  if (comments) {
+    comments = comments.filter(({ value }) => {
+      if (regex.test(value)) {
+        lastComment = value;
+        return false;
+      }
+      return true;
+    });
+  }
+  return [comments, lastComment];
+}
+
+function extractComments(regex, ast) {
+  let lastComment = null;
+  t.traverseFast(ast, node => {
+    // $FlowIgnore destructuring with expressions is not supported
+    [node.leadingComments, lastComment] = extractCommentsFromList(
+      regex,
+      node.leadingComments,
+      lastComment,
+    );
+    // $FlowIgnore destructuring with expressions is not supported
+    [node.innerComments, lastComment] = extractCommentsFromList(
+      regex,
+      node.innerComments,
+      lastComment,
+    );
+    // $FlowIgnore destructuring with expressions is not supported
+    [node.trailingComments, lastComment] = extractCommentsFromList(
+      regex,
+      node.trailingComments,
+      lastComment,
+    );
+  });
+  return lastComment;
+}

packages/babel-core/test/fixtures/transformation/source-maps/comment-inside-string/options.json

@@ -0,0 +1,3 @@
+{
+  "inputSourceMap": true
+}

packages/babel-types/src/traverse/traverseFast.js

@@ -7,7 +7,7 @@ import { VISITOR_KEYS } from "../definitions";
  */
 export default function traverseFast(
   node: Object,
-  enter: (node: Node, opts?: Object) => void,
+  enter: (node: BabelNode, opts?: Object) => void,
   opts?: Object,
 ): void {
   if (!node) return;