RSS Google search returns coded redirects that GPT modules cannot parse

How To
1

I’m trying to build an automation that runs an rss google search and then passes the result to GPT, but the url returned by the rss search is a “coded” redirect that every make.com module I try cannot parse to the original source url.

Here’s an example of the url returned by the rss search module:

https://news.google.com/rss/articles/CBMifEFVX3lxTE4zdjZWWURIeEV4SWVvLWRBUUdXSWs1OGtka2VlVktkZlg4RldyZXczS3l1ZXZkZ1Bta3RQUC0xaEc1Y3FsbjY2YTBIMXNiVXkzUldLU2hDM3B6Q19iOVQyTkZHWG54UjFoWjN1Vk12cXZMMDBCbld4c0hVNzk?oc=5

Plugging this url into any browser redirects immediately to the original source url, but it proves a dead end for any make.com automation I have tried - which have been many variations of the HTTP module, custom javascript, puppeteer, etc.

The simplicity of the redirect in a browser would seem to indicate this should be easy, but my pea-brain is officially stumped. Thanks for any suggestions.

2

Welcome to the Make community!

Duplicate of Scrape Content from Redirected Google News Links - #6 by samliew

3

Thanks for the reply. I’ve spent the day trying to understand how to use and testing in make.com and I’m failing completely. I’m only able to see various root-level domain calls related to cdns, css, typekit, analytics, etc. I’m not getting anything that returns any part of the original source url, just the main doman and the aforementioned plumbing.

Admittedly, the documentation is over my head. Is there a simple way to just extract the source domain including the full page url with this method?

Thanks!