[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, body starts with title and ends in url, link following arrow in body, potentially bad ns for domain in body, +1 more: Keto Burn Diet Find recipes with fruits by Cherylalma on graphicdesign.SE
@Makyen That pattern looks like it's already caught by Potentially bad ns for domain in body, Potentially bad keyword in answer, Potentially bad keyword in body, and Potentially bad ns for domain in answer; append -force if you really want to do that.
[ SmokeDetector | MS ] Blacklisted website in body, body starts with title and ends in url, pattern-matching website in body, potentially bad ns for domain in body: he reason is because a by user45789 on rpg.SE
@Makyen That pattern looks like it's already caught by Pattern-matching website in answer, Potentially bad keyword in body, Potentially bad keyword in answer, and Pattern-matching website in body; append -force if you really want to do that.
@Makyen That pattern looks like it's already caught by Pattern-matching website in answer and Pattern-matching website in body; append -force if you really want to do that.
@TetsuyaYamamoto That pattern looks like it's already caught by Pattern-matching website in answer and Pattern-matching website in body; append -force if you really want to do that.
[ SmokeDetector | MS ] Bad keyword in body, blacklisted website in body, link at end of body, pattern-matching product name in body, pattern-matching website in body, +2 more: Hints for effective food regimen by eepv on superuser.com
@Makyen That pattern looks like it's already caught by Pattern-matching website in body and Pattern-matching website in answer; append -force if you really want to do that.
@Makyen Added buymodafinilonline\.net to watchlist
CI on 19fd9c5 succeeded. Message contains 'autopull', pulling...
[ SmokeDetector | MS ] Bad keyword in body, blacklisted website in body, body starts with title and ends in url, link at end of body, pattern-matching product name in body, +3 more: when you exercise by gwpvgwpv on superuser.com
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted website in body, body starts with title and ends in url, link at end of body, +3 more: Keto Ultra Diet by Meimomotesh on superuser.com
[ SmokeDetector | MS ] Blacklisted website in body, body starts with title and ends in url, link at end of body, pattern-matching website in body, potentially bad ns for domain in body: Boost SX Pro Styling requires less by luvardn12 on askubuntu.com
It'd certainly help. At a minimum, just checking against the most recently executed command would fix 95% of the problems we experience (dup by edit & dup by multiple people trying to add the same thing at almost the same time).
@iBug BTW: Given that from time to time SD will use the original version, rather than the newly edited version, we really shouldn't edit commands that we really don't want SD to execute twice. Fortunately, you have GitHub write access, so you can resolve such issues.
Yeah, the only thing I'd want to prevent programmatically, at least at this time, is the whole process of !!/watch and !!/blacklist-* grinding to a halt. As far as I know that only happens when CI sees an exact match, or at least that's the only thing I've seen.
@iBug If that's what really happened (likely, needs testing), then that issue is happening for you right now on a much higher frequency than I've seen before. Previously, it was once in a great while.
links to an external site, and it's either that this user is plagiarizing from that site or there's a serious lack of attribution affiliation disclosure.
It's assured that for blacklist/watch commands that take the source of the message, Smokey will always use the old message source instead of the edited
Fixing this requires changing ChatExchange, however.
Ahhh... That's not what I've experienced. I've routinely edited messages from !!/watch to !!/watch-force, etc. and seen other people do the same on many occasions.
Some parts of the code use content, while some other parts use content_source. E.g., command parsing uses HTML content, but the blacklist function takes the source
So, when you update a message from watch to watch-force, the command processor takes the edited message and decides that it's watch-force, then the HTML is splitted and sent to individual command functions.
Individual function may require that the original message object be provided, like do_blacklist, which takes content_source and parses it on its own (ignoring the argument generated by command dispatcher)
Well, it's certainly possible without a CE tweak, as long as you know the chat server and message ID, you can always fetch the source yourself. The question is if it's worth it.
Personally, I think it's reasonable, particularly given that it's only needed on some commands.
Only some commands use the content_source. It was a relatively recent change to !!/watch and !!/blacklist to start using it, so that extra escapes did not need to be in the text used for the regexes.
At the time that change was made, I got the impression that no other commands were using content_source, but I could be wrong on that.
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted website in body, link at end of body, pattern-matching product name in body, +3 more: you to maintain our muscle agencies by sins on superuser.com
[ SmokeDetector | MS ] Bad keyword in body, blacklisted website in body, link at end of body, pattern-matching product name in body, pattern-matching website in body, +2 more: Jessica Alba Jennifer Aniston by qwpvqwpv on superuser.com
@FrenzyLi That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body; append -force if you really want to do that.
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted website in body, body starts with title and ends in url, link at end of body, +4 more: Keto Ultra Diet by Barry Rjrk on superuser.com
[ SmokeDetector | MS ] Blacklisted website in body, link at end of body, pattern-matching product name in body, pattern-matching website in body, potentially bad ns for domain in body, +1 more: Drink milk contemporary research by wdpv on superuser.com
[ SmokeDetector | MS ] Blacklisted website in body, body starts with title and ends in url, link at end of body, pattern-matching website in body, potentially bad ns for domain in body: keto go assiduity in a in the forthright by Meens on askubuntu.com
[ SmokeDetector | MS ] Blacklisted website in body, link at end of body, pattern-matching website in body, potentially bad ns for domain in body: you have swallowed a good by gapvgapv on superuser.com
