Avoid splitting node labels in the middle of unicode surrogate pairs. F…

…ixes #11697
GoogleChrome · Nov 23, 2020 · 32afa11 · 32afa11
1 parent 5608cbd
commit 32afa11
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 1 deletion.
diff --git a/lighthouse-core/lib/page-functions.js b/lighthouse-core/lib/page-functions.js
@@ -375,7 +375,7 @@ function getNodeLabel(node) {
     if (str.length <= maxLength) {
       return str;
     }
-    return str.slice(0, maxLength - 1) + '…';
+    return Array.from(str).slice(0, maxLength - 1).join('') + '…';
   }
   const tagName = node.tagName.toLowerCase();
   // html and body content is too broad to be useful, since they contain all page content

diff --git a/lighthouse-core/test/lib/page-functions-test.js b/lighthouse-core/test/lib/page-functions-test.js
@@ -116,6 +116,12 @@ describe('Page Functions', () => {
       assert.equal(pageFunctions.getNodeLabel(el).length, 80);
     });
 
+    it('Truncates long text containing unicode surrogate pairs', () => {
+      const el = dom.createElement('div');
+      el.innerText = Array(78).fill('a').join('') + '💡💡💡';
+      assert.equal(pageFunctions.getNodeLabel(el), Array(78).fill('a').join('') + '💡…');
+    });
+
     it('Uses tag name for html tags', () => {
       const el = dom.createElement('html');
       assert.equal(pageFunctions.getNodeLabel(el), 'html');