Hi! We’re experiencing a lot of spam that’s not caught by Akismet. The messages usually have (international) phone numbers in them which made us think of using a regex to catch them. After experimenting a bit we found the following expression: (?:\+|00)?\d{1,3}?[-\s]?\d{2,4}[-\s]?\d{2,4}[-\s]?\d{2,4}|\b\d{10}\b We’re not able to enter this though as the admin screen breaks it up into separate chunks: [The image shows a text editor interface with a highlighted box containing a regular expression and various character codes used in the expression. (Captioned by AI)] Any suggestions how we should handle this?

I think it’s picking up on the pipes | when the regex is copy-pasted in. I think if you type it in manually it may ignore them as separators and allow you to do it as one entry. :crossed_fingers: ETA: From a quick test, this seems promising. :+1:

Brilliant, that solved it! :+1:

Não é possível inserir regex de palavras assistidas para capturar números de telefone

Suporte ao Produto

bartv (Bart ) 17 Maio , 2025 07:27 4

Brilhante, isso resolveu!

Tópico		Respostas	Vistas
Increase allowed length of watched words Feature watched-words	5	1059	9 Dezembro , 2020
Using Regex with Watched Words Site Management reference , regex , watched-words , content	1	2124	14 Junho , 2024
Invalid regular expressions in 'Watched Words' makes no watched word work Bug watched-words	2	722	31 Maio , 2021
Watched word regular expression crash (2025) Bug regex , watched-words	6	392	6 Outubro , 2025
How to use Discourse regexes with watched words? Product Support	6	2606	30 Maio , 2019

Não é possível inserir regex de palavras assistidas para capturar números de telefone

Related topics