Add test for input containing invalid attributes

mozilla · Oct 21, 2024 · f48f5e3 · f48f5e3
1 parent 8bf0a99
commit f48f5e3
Show file tree

Hide file tree

Showing 6 changed files with 97 additions and 19 deletions.
diff --git a/package-lock.json b/package-lock.json
diff --git a/package.json b/package.json
@@ -35,6 +35,7 @@
     "mocha": "10.1.0",
     "prettier": "^3.3.2",
     "release-it": "17.0.1",
-    "sinon": "14.0.2"
+    "sinon": "14.0.2",
+    "xml-name-validator": "^5.0.0"
   }
 }
diff --git a/test/test-pages/invalid-attributes/expected-metadata.json b/test/test-pages/invalid-attributes/expected-metadata.json
@@ -0,0 +1,9 @@
+{
+  "title": "Lorem Ipsum",
+  "byline": null,
+  "dir": null,
+  "excerpt": "Lorem ipsum dolor sit amet, consectetur adipiscing elit.",
+  "siteName": null,
+  "publishedTime": null,
+  "readerable": false
+}
diff --git a/test/test-pages/invalid-attributes/expected.html b/test/test-pages/invalid-attributes/expected.html
@@ -0,0 +1,7 @@
+<div id="readability-page-1" class="page">
+  <div "="">
+    <p>
+        Lorem ipsum dolor sit amet, consectetur adipiscing elit.
+    </p>
+  </div>
+</div>
diff --git a/test/test-pages/invalid-attributes/source.html b/test/test-pages/invalid-attributes/source.html
@@ -0,0 +1,19 @@
+<!DOCTYPE html>
+<html>
+    <head>
+        <title>Lorem Ipsum</title>
+    </head>
+    <body>
+        <main>
+            <section>
+                <div "="">
+                    <div>
+                        <div>
+                            Lorem ipsum dolor sit amet, consectetur adipiscing elit.
+                        </div>
+                    </div>
+                </div>
+            </section>
+        </main>
+    </body>
+</html>
diff --git a/test/test-readability.js b/test/test-readability.js
@@ -1,6 +1,7 @@
 /* eslint-env node, mocha */
 
 var JSDOM = require("jsdom").JSDOM;
+var xmlNameValidator = require("xml-name-validator").name;
 var chai = require("chai");
 var sinon = require("sinon");
 chai.config.includeStack = true;
@@ -121,14 +122,17 @@ function runTestsWithItems(
 
       function attributesForNode(node) {
         return Array.from(node.attributes)
+          .filter(function(attr) {
+            return xmlNameValidator(attr.name)
+          })
           .map(function (attr) {
             return attr.name + "=" + attr.value;
           })
-          .join(",");
       }
 
       var actualDOM = domGenerationFn(prettyPrint(result.content));
       var expectedDOM = domGenerationFn(prettyPrint(expectedContent));
+
       traverseDOM(
         function (actualNode, expectedNode) {
           if (actualNode && expectedNode) {
@@ -152,21 +156,21 @@ function runTestsWithItems(
               }
               // Compare attributes for element nodes:
             } else if (actualNode.nodeType == 1) {
-              var actualNodeDesc = attributesForNode(actualNode);
-              var expectedNodeDesc = attributesForNode(expectedNode);
+              var actualNodeAttributes = attributesForNode(actualNode);
+              var expectedNodeAttributes = attributesForNode(expectedNode);
               var desc =
                 "node " +
                 nodeStr(actualNode) +
                 " attributes (" +
-                actualNodeDesc +
+                actualNodeAttributes.join(",") +
                 ") should match (" +
-                expectedNodeDesc +
-                ") ";
-              expect(actualNode.attributes.length, desc).eql(
-                expectedNode.attributes.length
+                expectedNodeAttributes.join(",") +
+                ") 1";
+              expect(actualNodeAttributes.length, desc).eql(
+                expectedNodeAttributes.length
               );
-              for (var i = 0; i < actualNode.attributes.length; i++) {
-                var attr = actualNode.attributes[i].name;
+              for (var i = 0; i < actualNodeAttributes.length; i++) {
+                var attr = actualNodeAttributes[i].name;
                 var actualValue = actualNode.getAttribute(attr);
                 var expectedValue = expectedNode.getAttribute(attr);
                 expect(