disable headless auto form filling

[dogancanbakir]

Closes #876

```console
$ go run . -u http://testphp.vulnweb.com -hl -aff -j -o test.json
...

$ grep -r 'method":"POST"' test.json | grep -o '"raw":.*' | cut -c 1-50
"raw":"POST /search.php?test=query HTTP/1.1\r\nHos
"raw":"POST /search.php?test=query HTTP/1.1\r\nHos
"raw":"POST /userinfo.php HTTP/1.1\r\nHost: testph
"raw":"POST /userinfo.php HTTP/1.1\r\nHost: testph
"raw":"POST /secured/newuser.php HTTP/1.1\r\nHost:
"raw":"POST /secured/newuser.php HTTP/1.1\r\nHost:
"raw":"POST /guestbook.php HTTP/1.1\r\nHost: testp
"raw":"POST /guestbook.php HTTP/1.1\r\nHost: testp

[Mzack9999]

It seems like we should review the proposed solution based upon #876 (comment). Probably form filling should be reconsidered/disabled as a whole for headless navigation as the way we are doing it is totally wrong, as it should happen via headless actions within the page context (ex. click on submit button).

[alban-stourbe-wmx]

Hello,
Do you think it would be a good idea to create a function that automatically fills out forms using the rod package?

Is it possible that for each request launched by page.Navigate, we search through xpaths/selector for forms and content to fill them again with predefined data and click? If this were possible, the request would be hijacked and we wouldn't have to filter requests by tag, and the navigateRequest function would suffice on its own.

I don't realize how hard it would be to add such a feature

[dogancanbakir]

Disabled headless auto form filling for now. I'll create a follow-up ticket to track this.

[Miracles666]

hi @alban-stourbe-wmx Have you successfully solved this problem?

[alban-stourbe-wmx]

hi @alban-stourbe-wmx Have you successfully solved this problem?

Hello @Miracles666, to fix the issue, the automatic-form has been disabled with headless mode. For my case, i detect forms with the option with the standard mode only.

If we really want to use automatic-form with headless mode, a new feature need to be develop to detect form and send data through the web browser.

I hope you understand

[Miracles666]

@alban-stourbe-wmx standard mode is very few things that can be crawled up ，I think this is a very serious bug。

[alban-stourbe-wmx]

is very few things that can be crawled up

That's why if you think a bit, we can use the SDK Katana to crawl your website with the headless and the standard mode together. After you can merge and the two outputs to obtain all informations ;)