From 18373626b243a2d6db63905a2fb9d126dcd59e61 Mon Sep 17 00:00:00 2001
From: Han Xiao <han.xiao@jina.ai>
Date: Mon, 15 Apr 2024 19:27:40 -0700
Subject: [PATCH] fix: catch parse error

---
 backend/functions/src/services/puppeteer.ts | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/backend/functions/src/services/puppeteer.ts b/backend/functions/src/services/puppeteer.ts
index 176876d..c888a92 100644
--- a/backend/functions/src/services/puppeteer.ts
+++ b/backend/functions/src/services/puppeteer.ts
@@ -118,12 +118,21 @@ export class PuppeteerControl extends AsyncService {
         preparations.push(page.evaluateOnNewDocument(READABILITY_JS));
         preparations.push(page.evaluateOnNewDocument(`
 function giveSnapshot() {
+    let parsedContent;
+    try {
+        // Attempt to parse the cloned document
+        parsedContent = new Readability(document.cloneNode(true)).parse();
+    } catch (error) {
+        // If an error occurs, log it and set parsedContent to undefined
+        parsedContent = undefined;
+    }
+
     return {
         title: document.title,
         href: document.location.href,
         html: document.documentElement.outerHTML,
         text: document.body.innerText,
-        parsed: new Readability(document.cloneNode(true)).parse(),
+        parsed: parsedContent
     };
 }
 `));