Possible Solution?

[BenutzerEinsZweiDrei]

As far AS i understand it..

Apparently Google established a redirect Website so requests wont work anymore ..

So First you need to use Selenium Driver?

then apparently Google keeps switching the data-snf css selectors randomly

As far AS i understood this ist not on per User Basis

so what i guess what you truly need is

"a googlesearch-server Module for Automatic Google pattern matching"

this could Run on a Raspberry Pi as API Server

• keeps do the Same Google search
• Filters Out typical headlines and urls
• analyses html pattern around
• adjusts the html pattern for beautifulsoup/selenium

[BenutzerEinsZweiDrei]

here this is a minimal proof of concept using undetected-chromedriver and minimal-server

https://github.com/BenutzerEinsZweiDrei/uc_googlesearch_minimal_server_client/

-> headless mode was detected so running on server with head

now it only needs -> automatic pattern recognition <-

-> the server needs to run continuous tests on changing google patterns in html and automatically detect css results

for example:

• google "python"
• find "https://www.python.org" url
• extract surrounding html pattern
• find title, desc, url
• pass new css markers

[Nv7-GitHub]

Does this solution work?

[BenutzerEinsZweiDrei]

@Nv7-GitHub

chromedriver was detected immediatly.

chromedriver-undetected works perfectly.

if you turn it into headless mode google has bot detection.

i wasnt able to bypass bot detection in headless mode yet.

in the other issue some people said playwright is also detected.

but one could still try playwright-stealth

an easy workaround for now could be to have a client and server

where the server opens a browser and the client gets the results

this is a dirty workaround just for convienience
until someone figures out a headless solution

but it works, yes.