So as I was going to post a link to an article from google’s blog like so:
our instance of Discourse is giving me the following message:
Any ideas what went wrong, and how to fix it?
So as I was going to post a link to an article from google’s blog like so:
our instance of Discourse is giving me the following message:
Any ideas what went wrong, and how to fix it?
The word password
has the word ass
in it.
… you’re kidding me. right?
Password is a censored word therefor?
Does seem somewhat buggy it would not be enforce word boundaries, @eviltrout can you have a quick look?
Thanks @sam. I would offer to help, but this stuff is beyond my understanding.
Also, we should fix that message. It’s very bad UX.
At the very least, something like .split("|").sort.join(", ")
.
You might want to add this test into your CI server, since it is something that could easily affect a lot of users.
And @zogstrip, though I do agree the way it’s being displayed could be improved, perhaps it’d be better not to show the list at all; if only to avoid helping them find workarounds… and though I’ve found it amusing to read all the weird censored terms, to some people this could result as an unpleasant shocker. Particularly if they didn’t put any of these words in it.
e.g. Imagine we were censoring images, and a list of actual explicit pictures popped up to say “these are censored”. Not bueno.
That’s even better, but requires more work: only show the censored words.
Ah ha. This is the validation I added. Will fix.
Title includes one or more censored words:
Well it actually does Title : tit - having that list pop up is more offensive than some of the words.
If my community of nuns mention spotting a blue tit in the convent gardens … this list will pop up …? lol
Context matters. It’s safe to say some instances would not only allow but require actual body parts to be mentioned (e.g. medical forums).
not sure how this is handled in those scenarios.
This fix needs to be backported to 1.7 @tgxworld so make sure it is tagged as such.
The feature was added post 1.7 so no need to backport
https://github.com/discourse/discourse/commit/ce07da1d8bc12467ceede077ba60123be7a9c02d