Fix script parsing to ignore closing tags in comments

JSDOMParser.js

@@ -919,14 +919,59 @@
     },
 
     readScript: function (node) {
-      var index = this.html.indexOf("</script>", this.currentChar);
-      if (index === -1) {
-        index = this.html.length;
+      while (this.currentChar < this.html.length) {
+        var c = this.nextChar();
+        var nextC = this.peekNext();
+        if (c === "<") {
+          if (nextC === "!" || nextC === "?") {
+            // We're still before the ! or ? that is starting this comment:
+            this.currentChar++;
+            node.appendChild(this.discardNextComment());
+            continue;
+          }
+          if (nextC === "/" && this.html.substr(this.currentChar, 8 /*"/script>".length */).toLowerCase() == "/script>") {
+            // Go back before the '<' so we find the end tag.
+            this.currentChar--;
+            // Done with this script tag, the caller will close:
+            return;
+          }
+        }
+        // Either c wasn't a '<' or it was but we couldn't find either a comment
+        // or a closing script tag, so we should just parse as text until the next one
+        // comes along:
+
+        var haveTextNode = node.lastChild && node.lastChild.nodeType === Node.TEXT_NODE;
+        var textNode = haveTextNode ? node.lastChild : new Text();
+        var n = this.html.indexOf("<", this.currentChar);
+        // Decrement this to include the current character *afterwards* so we don't get stuck
+        // looking for the same < all the time.
+        this.currentChar--;
+        if (n === -1) {
+          textNode.textContent += this.html.substring(this.currentChar, this.html.length);
+          this.currentChar = this.html.length;
+        } else {
+          textNode.textContent += this.html.substring(this.currentChar, n);
+          this.currentChar = n;
+        }
+        if (!haveTextNode)
+          node.appendChild(textNode);
       }
-      var txt = new Text();
-      txt.textContent = this.html.substring(this.currentChar, index === -1 ? this.html.length : index);
-      node.appendChild(txt);
-      this.currentChar = index;
+    },
+
+    discardNextComment: function() {
+      if (this.match("--")) {
+        this.discardTo("-->");
+      } else {
+        var c = this.nextChar();
+        while (c !== ">") {
+          if (c === undefined)
+            return null;
+          if (c === '"' || c === "'")
+            this.readString(c);
+          c = this.nextChar();
+        }
+      }
+      return new Comment();
     },
 
 
@@ -964,20 +1009,9 @@
       // them away in readChildren()). So just returning an empty Comment node
       // here is sufficient.
       if (c === "!" || c === "?") {
+        // We're still before the ! or ? that is starting this comment:
         this.currentChar++;
-        if (this.match("--")) {
-          this.discardTo("-->");
-        } else {
-          var c = this.nextChar();
-          while (c !== ">") {
-            if (c === undefined)
-              return null;
-            if (c === '"' || c === "'")
-              this.readString(c);
-            c = this.nextChar();
-          }
-        }
-        return new Comment();
+        return this.discardNextComment();
       }
 
       // If we're reading a closing tag, return null. This means we've reached

test/test-jsdomparser.js

@@ -208,3 +208,42 @@ describe("Test JSDOM functionality", function() {
     }
   });
 });
+
+
+describe("Script parsing", function() {
+  it("should strip ?-based comments within script tags", function() {
+    var html = '<script><?Silly test <img src="test"></script>';
+    var doc = new JSDOMParser().parse(html);
+    expect(doc.firstChild.tagName).eql("SCRIPT");
+    expect(doc.firstChild.textContent).eql("");
+    expect(doc.firstChild.children.length).eql(0);
+    expect(doc.firstChild.childNodes.length).eql(1);
+  });
+
+  it("should strip !-based comments within script tags", function() {
+    var html = '<script><!--Silly test > <script src="foo.js"></script>--></script>';
+    var doc = new JSDOMParser().parse(html);
+    expect(doc.firstChild.tagName).eql("SCRIPT");
+    expect(doc.firstChild.textContent).eql("");
+    expect(doc.firstChild.children.length).eql(0);
+    expect(doc.firstChild.childNodes.length).eql(1);
+  });
+
+  it("should strip any other nodes within script tags", function() {
+    var html = "<script><div>Hello, I'm not really in a </div></script>";
+    var doc = new JSDOMParser().parse(html);
+    expect(doc.firstChild.tagName).eql("SCRIPT");
+    expect(doc.firstChild.textContent).eql("<div>Hello, I'm not really in a </div>");
+    expect(doc.firstChild.children.length).eql(0);
+    expect(doc.firstChild.childNodes.length).eql(1);
+  });
+
+  it("should not be confused by partial closing tags", function() {
+    var html = "<script>var x = '<script>Hi<' + '/script>';</script>";
+    var doc = new JSDOMParser().parse(html);
+    expect(doc.firstChild.tagName).eql("SCRIPT");
+    expect(doc.firstChild.textContent).eql("var x = '<script>Hi<' + '/script>';");
+    expect(doc.firstChild.children.length).eql(0);
+    expect(doc.firstChild.childNodes.length).eql(1);
+  });
+});

test/test-pages/comment-inside-script-parsing/expected-metadata.json

@@ -0,0 +1,5 @@
+{
+  "title": "Test script parsing",
+  "byline": null,
+  "excerpt": "Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod\n        tempor incididunt ut labore et dolore magna aliqua."
+}

test/test-pages/comment-inside-script-parsing/expected.html

@@ -0,0 +1,19 @@
+<div id="readability-page-1" class="page">
+    <div>
+        <p>Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod
+            tempor incididunt ut labore et dolore magna aliqua.</p>
+        <p>Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi
+            ut aliquip ex ea commodo consequat.</p>
+        <p>Duis aute irure dolor in reprehenderit in voluptate velit esse cillum
+            dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non
+            proident, sunt in culpa qui officia deserunt mollit anim id est laborum.</p>
+    </div>
+    <div>
+        <p>Tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
+            quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo
+            consequat.</p>
+        <p>Duis aute irure dolor in reprehenderit in voluptate velit esse cillum
+            dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non
+            proident, sunt in culpa qui officia deserunt mollit anim id est laborum.</p>
+    </div>
+</div>

test/test-pages/comment-inside-script-parsing/source.html

@@ -0,0 +1,34 @@
+<html>
+  <head><title>Test script parsing</title></head>
+<body>
+  <script>
+    <!--
+    Silly test
+    <script src="foo.js"></script>
+    -->
+  </script>
+  <article>
+    <h1>Lorem</h1>
+    <div>
+      <p>Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod
+        tempor incididunt ut labore et dolore magna aliqua.</p>
+      <p>Ut enim ad minim veniam,
+      quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo
+      consequat.</p>
+      <p>Duis aute irure dolor in reprehenderit in voluptate velit esse
+      cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non
+      proident, sunt in culpa qui officia deserunt mollit anim id est laborum.</p>
+    </div>
+    <h2>Foo</h2>
+    <div>
+      <p>Tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
+      quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo
+      consequat.</p>
+      <p>Duis aute irure dolor in reprehenderit in voluptate velit esse
+      cillum dolore eu fugiat nulla pariatur.
+      Excepteur sint occaecat cupidatat non
+      proident, sunt in culpa qui officia deserunt mollit anim id est laborum.</p>
+    </div>
+  </article>
+</body>
+</html>