@SmokeDetector fp- user's profile indicates affiliation with webshopapps, but can't find any affiliation with the product itself. Also, it's the OP's only visible answer and it's a decade old.
@tripleee Pardon my ignorance, but what exactly does this do? Is this checking to see if those emoji characters are treated as a word?
@JeffSchaller Fair enough. The user answered blatant spam as their very first post on the site, and that blatant spam had previously been answered by another known spammer, so I had a very bad feeling about it. I granted in my initial report that it might just be innocent behavior from a very confused new user. Apparently, the folks who gave feedback on the report thought so, too. And now the user has gone on to post other answers, which makes it look less like they are part of the spam wave.
@Spevacus But how did @blackgreen manage to request the same expression be added to the blacklist twice, when they only posted a single chat message?
Yup, I think the additional context that is available to us proves that there's no need to watch that username. That context obviously wasn't available to me at the time, and the user was caught up in a network of spammers stretching across multiple sites, so I definitely erred on the side of a possible FP in reporting and watching them.
@blackgreen @HenryEcker For what it's worth, as a mod, I prefer SIM. It doesn't do as much as Advanced Flagging does, but what it does do is more useful and appropriate for moderators. I never got into using Advanced Flagging as a non-mod, though, so I guess that's a large part of why I don't prefer it. With SIM, you can easily check whether a particular post has been reported (and, if so, get a link to the MS report), and you can also now report it right there from the page, with no risk of a flag automatically being raised that would nuke the post.
Dang, Makyen literally posted exactly that same suggestion a few messages later.
@cigien Currently, as far as I know, posts should not be reported here for being AI-generated, nor should such reports be given TP feedback. (Unless, of course, they are actually spam or abusive independent of the fact that they're AI-generated. Plenty of spammers have been using ChatGPT to generate innocuous-looking content into which they include a spam link. That should definitely be reported, marked TP, and red-flagged, but not because it's ChatGPT—only because it's spam.)
@Mast I see. I don't even routinely read the SOBotics transcript, so I would not have seen that. How did you find those?
@CodyGray That pattern looks like it's already caught by Potentially bad keyword in username Append -force to the command word(s) if you really want to add the pattern you provided.
I convinced myself it was a risk worth taking, but that was before I noticed that there was, in fact, a domain to watch, so perhaps I should have reassessed that judgment in light of that fact.
@Mast There are enough of them such that when there's a page that feels like it should be paginated but doesn't indicate it, I go ahead and try adding the page parameter.
@CodyGray That pattern looks like it's already caught by Blacklisted website in answer and Blacklisted website in body Append -force to the command word(s) if you really want to add the pattern you provided.
@CodyGray That pattern looks like it's already caught by Bad keyword in username Append -force to the command word(s) if you really want to add the pattern you provided.
[ SmokeDetector | MS ] Bad ip for hostname in answer, bad ns for domain in answer, bad keyword in answer, blacklisted website in answer, potentially bad keyword in answer (361): How to say "make money"? by Amit Baria on chinese.SE
@CodyGray That pattern looks like it's already caught by Potentially bad keyword in body and Potentially bad keyword in answer Append -force to the command word(s) if you really want to add the pattern you provided.
@CodyGray that sounds interesting; though I'm not too fond of chat-ops. I like advanced flagging's ability to report to smoke detector while viewing the post. Makyen's message hinted at SIM being able to do that but with chat commands. Disclaimer: I haven't looked into the documentation yet
SIM does it while viewing the post. It's just that, behind the scenes, it posts a chat message into this chat room on your behalf. You never actually have to visit chat.
@CodyGray That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body Append -force to the command word(s) if you really want to add the pattern you provided.
@CodyGray That pattern may be a North American number. If it is, please use a format which starts with an optional 1 followed by possible separator text and has the main number in the format 999-554-3735 where - could be a single alpha character or any [\W_]*+. Alternately, you can add the comment (?#IS NorAm) to the end of the pattern to force also using the alternate normalized form, or (?#NO NorAm) if it's not a North American phone number and it's incorrectly recognized as one. Perhaps try !!/watch-number 999-554-3735. Append -force to the command word(s) if you really want to add the pattern you provided.
@CodyGray That pattern may be a North American number. If it is, please use a format which starts with an optional 1 followed by possible separator text and has the main number in the format 811-395-0449 where - could be a single alpha character or any [\W_]*+. Alternately, you can add the comment (?#IS NorAm) to the end of the pattern to force also using the alternate normalized form, or (?#NO NorAm) if it's not a North American phone number and it's incorrectly recognized as one. Perhaps try !!/watch-number 811-395-0449. Append -force to the command word(s) if you really want to add the pattern you provided.
@CodyGray I don't think the negative lookahead at the end will ever match because there's a Label that's required in the beginning. Did you mean to make the Label part optional as well? I suspect you meant Shrink[\W_]*+Sleeve(?:[\W_]*+(?:Label|Applicator|Machine))?+(?!\.in(?<=shrinksleeveapplicator\.in))
For things like this I often have to guess at what it is that someone really wants to be detecting, as they might easily be looking at something more than or something other than what I think the source text is (or just want something different than what I see). So, there are typically assumptions contained in the regex which may not be accurate. So, I'm usually just suggesting with the expectation that what I'm suggesting may need to be further modified.
@Makyen Ah, that's nice. Took me a while to parse, but it looks better that what I had. (It catches multiple occurrences of machine, label, etc, at least).
[ SmokeDetector | MS ] Bad ip for hostname in body, bad ns for domain in body, bad keyword in body, bad keyword in link text in body, blacklisted website in body, +1 more (458): what is LoadRunner? by priya krish on softwarerecs.SE
@CodyGray yeah, you'll need Python 3 and the regex third-party package, probably in a virtual environment
if you already have the Smoke Detector sources, the README has instructions for how to set up a virtualenv with the packages Smokey depends on
for anything more complex you really need to use Smokey itself because the anchoring of patterns is not entirely obvious or trivial
for a really quick check the regular re library which ships with Python itself is often enough but then you have to know or guess in which aspects regex differs from it
there's a number of online IDEs, repl.it lets you install third-party libraries with full abandon ... but yeah, probably not worth the effort just for this
there's also now the pyscript initiative which lets you run Python in the browser on top of web assembly, but I would not trust that to be 100% reliable yet
> Please be advised that PyScript is very alpha and under heavy development. There are many known issues, from usability to loading times, and you should expect things to change often. We encourage people to play and explore with PyScript, but at this time we do not recommend using it for production.
[ SmokeDetector | MS ] Link at beginning of body, mostly non-latin title, mostly punctuation marks in title, potentially bad keyword in body, several emoji in title (247): 🏦💸🦿🪙🌐💷🎉🏛️👻⛽🧭🤖🧲🔐🔐🏆💰🌏💴🧲⚓🔥🥰🏧🌏⚓🧭🎁🌏💎💎🏦 by BartasCNK on meta.SE (@Ollie @Ethan)
