reCAPTCHA v1 being turned off March 1 2018

(Douglas (@finnatic at @waikato)) #1

From Google this morning:

We announced the reCAPTCHA v1 deprecation in May 2016. Starting in November 2017, a small percentage of reCAPTCHA v1 traffic will begin to show a notice informing users that the old API will soon be retired. Any calls to the v1 API will not work after March 31, 2018.
To ensure continued functionality, you’ll need to update your website to a current version of reCAPTCHA. You can learn more about reCAPTCHA v2, Invisible reCAPTCHA and reCAPTCHA Android API in our Developer’s Guide. The new APIs are simple to implement and will streamline the captcha experience for your users. If you need help, you can engage in the reCAPTCHA Google Developer Group or post to Stack Overflow with the ‘recaptcha’ tag.

(Marcus Fong) #2

Just for reference, reCAPTCHA v2 support was added to Squiz Matrix in version 5.4.2.0, released July 3 this year:

https://matrix.squiz.net/releases/5.4/5.4.2.0#google-recaptcha-version-2-support

(Tbaatar) #3

Hi Marcus,

Is it possible to use reCaptcha v2 on a standard HTML form within Matrix by somehow calling the required keyword using global method?

We use standard HTML to post form submission to AutoPilot and need a way to improve spam preventation.

(Marcus Fong) #4

Hi,

I don’t think I’ve ever tried that, but looking at the code the reCAPTCHA keyword replacements seem to take place within the Form asset. The relevant manual page also says the same:

The following keyword replacements can be used on the Custom Form.

%form_recaptcha% This will display the Google reCAPTCHA fields.

So if by “standard HTML form” you mean a form that doesn’t use the Matrix Custom Form asset, I think you’d have to do it with the standard non-CMS method rather than using Matrix keywords.

(Tbaatar) #5

Hi Marcus,

Thanks for the feedback.
Yes, I’m wanting it to do via non Matrix form and figured it might be best to serve the keys via Matrix method (server side) as opposed on the front end for more secure and bulletproof approach.

Unless it doesn’t make any difference on the outcome the JS should be fine.

Thanks.