« first day (1604 days earlier)      last day (2202 days later) » 

12:03 AM
[ SmokeDetector | MS ] Pattern-matching website in body: ESP8266 WiFi Module to on/off LED? by manish pal on stackoverflow.com
fp- by Makyen
 
 
1 hour later…
1:17 AM
[ SmokeDetector | MS ] Offensive body detected: Transparent Conductive Paint by Jake L. on electronics.SE
 
@SmokeDetector f
 
1:44 AM
@ArtOfCode Then the solution is to not allow this change to allow custom reasons. If we operate on the assumption that users are stupid and will not use it properly, then we shouldn't include the custom report reason feature. It is impossible to handle 'multiple reasons' without making an exception for the type of delimiter accepted; if we are going with the proposed method then that's surrounding the reason in " quotes, and then only accepting that.
oops
clicked the wrong button >.<
Any additional quotes used inside must then be ignored and dropped, or should cause an error case. (That is, in the split on the quotes, if it ends up as len(splitdata)>=3 then FAIL: Invalid syntax)
not disagreeing with you on the 'users are stupid' statement, but the core issue is implementation to prevent the stupidity, and that is nontrivial, so we either need to punish users with a failure message for their being stupid, or decline the feature being implemented entirely.
in other news, Keybase just died on me
 
Why not just assume that anything which doesn't parse as an SE URL is the reason?
 
oops I broke Smokey
 
@Makyen way harder to determine if it's even in the proper URL structure
 
Restart: API quota is 11168.
 
1:54 AM
@ThomasWard For most things they can be initially differentiated by just splitting on spaces, ` `, and looking at the length of the "word". SE URLs will have some minimum length which is much > typical words in a reason. Yes, that means that the reason needs to not start with a bare URL, but if a user wants that then they can just put it in quotes, or some other way of preventing it from being parsed as an SE URL (IME, the URL parser is not forgiving).
 
there's no logic to determine if it's a URL, and we have to do it on an argument-by-argument test.
that means even greater complexity
the way it was proposed, the custom reason would be in the format with a custom report of: !!/report url1 [url2 url3 ...] ["Custom Reason"]
with the [ ] indicating optional items, and the quotes surrounding the custom reason
while in theory we could set it up to split on spaces, which it is doing anyways for splitting the URLs out in the existing Smokey, it's not as simple to split out reasons
this is something that'll probably need discussed out, but 10PM at night isn't when I want to have this discussion. I was simply responding to Art's ping :P
 
@ThomasWard Which is reasonable, but then you started talking about wanting " in reasons, which could be handled with a simple regex. I'm really not seeing the problem (i.e. what spectrum of possible inputs) you're attempting to address which are not easily differentiated (or just rejected).
 
@Makyen not how I wrote the logic
@quartata and others indicateed in here "What if the user puts quotes in their reason?"
and then pointed fingers at potential Charcoal group members who might be on the radar for that
I personally would rather hard-enforce the syntax I just wrote out
which is far easier to do code-wise and doesn't need as much logic (and I can adjust it to only accept one reason, and silently drop any other quoted items, though that could introduce other issues).
no, go away VMware, I don't want you I want PyCharm >.<
(stupid alt+tab is busted)
 
Why? Just assume that everything matched by / "(.*)"\s*$/ is your reason. Having quote characters in the reason is something which you should/could easily handle. It doesn't change that it still must start and end with a quote.
 
2:06 AM
@SmokeDetector f
 
@Makyen I presume that's the entire regex you're searching with then
@Makyen that would work great to extract the reason, but...
given the arguments string of: foo.bar/baz http://evildoers.net https://something.evil "Chaotic Evil "Darkness" Is Coming"
I still have to split on ` "` to get the URLs as a splittable list
though your regex does capture the entire second reason, that won't pass through kindly I bet to MS
 
you can give split a limit
so that it won't try to split the spaces in the reason
e.g. string.split(' "', 1)
 
Or just match out the reason, then substitute out the same regex and split the remaining string.
 
@quartata i wasn't talking about splitting out the reason
i was talking about splitting on ' "' still while regex extracting the reason
which I'm messing with in a test environment now
 
Oh right
yeah you wouldn't need the regex really
 
2:12 AM
actually that would work
>>> args = 'foo.bar/baz evildoers.net something.evil "Chaotic Evil "Darkness" Is Coming"'
>>> argsraw = args.split(' "', 1)
>>> argsraw
['foo.bar/baz evildoers.net something.evil';, 'Chaotic Evil "Darkness" Is Coming"']
I still need the logic with the try/except to capture nonexistent reasons
 
True, a split on ` "` with a limit of 1 would result in the URL args and the reason split.
 
and to trim the trailing "
but it would work
@quartata but this still doesn't fix Art's consideration point of "multiple reasons" which user stupidity might introduce
 
Then multiple reasons can be treated as one.
 
also not Art's concern
standby for example case
>>> args = 'foo.bar/baz evildoers.net something.evil "Spam URL" "Malware Distribution"'
>>> argsraw = args.split(' "', 1)>>> argsraw
['foo.bar/baz evildoers.net something.evil';, 'Spam URL" "Malware Distribution"']
>>> argsraw[1][:-1]
'Spam URL" "Malware Distribution'
in this case I can replace " " with ; but this is the 'rare' case
 
[ SmokeDetector | MS ] Bad keyword in answer, bad keyword in username, username similar to website in answer: Samba fails to add a user entry, how do I fix this? by obat amandel on askubuntu.com
tpu- by Makyen
 
2:17 AM
case in point:
>>> args = 'foo.bar/baz evildoers.net something.evil "Spam URL" "Malware Distribution"'
>>> argsraw = args.split(' "', 1)
>>> argsraw[1][:-1].replace('" "', '; ')
'Spam URL; Malware Distribution'
which would fix Art's concern about malformatting.
 
That case is just not supported. If they want different reasons for different URLs, use multiple !!/report.
 
KILL THE SPAMMER! laughs evilly as the orbital firebombs come forth upon the spammer, called forth by the power of Diamonds.
sorry I've had a chaotic day, bear with my moments of insanity :)
@Makyen I agree, but i'm complaining about Art's complaint :P
the above would 'fix' Art's concerns
but either way it annoys me :P
I'd much rather punish the user who called the function (metaphorically) for invalid syntax by triggering a fail/error case >:D
 
I'm in the why bother camp. If the person wants different reasons, then use a different !!/report. IMO, the single !!/report is for posts that fit the single reason provided, if a reason is provided. If they want multiple reasons, then use a different !!/report. If you really want to support multiple reasons, then the syntax should be !!/report URL1 ["reason1"] [URL2 ["reason2"]] [URL3 ["reason3"]] [URL4 ["reason4"]] [URL5 ["reason5"]].
 
@Makyen preaching to the choir. Art is concerned about users being stupid.
so blame Art for this whole discussion
 
Then they are stupid and all of their reasons get grouped into one and/or they get an error (depending on what they actually did). Just make the documentation clear.
 
2:25 AM
again, I agree. Tell that to @ArtOfCode who believes users are stupid and don't follow the documentation :p
which has happened
 
And it will happen, no matter how much effort is put into trying to anticipate the stupidity.
 
"Breaking from the stated syntax will cause undefined behavior, and one of the esteemed SmokeDetector developers will come and find you in the middle of the night and destroy you torment you for inducing undefined behavior."
I'm kidding of course, but still.
 
[ SmokeDetector | MS ] Bad keyword in body: Express.js not adding into mysql by Priyesh Goswami on stackoverflow.com
 
:-) Although, I'd say that the behavior should be reasonably well defined.
 
sd why
 
2:28 AM
@ThomasWard [:43882450] Body - Position 196-205: BeastMode, Position 243-252: BeastMode
fp- by Makyen
 
@SmokeDetector fp;
 
@ThomasWard Registered question as false positive.
 
oops
meh
@Makyen "Undefined stupidity of grammar syntax" then.
 
[ SmokeDetector | MS ] Bad NS for domain in body, bad keyword in body, blacklisted website in body: Loss Decreases recuperation time by Srj wsf on apple.SE
 
either way it's "not really well defined behavior" if you use an evil syntax
 
2:30 AM
tpu- by NobodyNada
 
also I instinictively knew we could limit splits, but it's now firmly kept in my head now >:D
 
3:11 AM
[ SmokeDetector | MS ] Few unique characters in answer, repeating characters in answer, blacklisted user: Linking HTML code with Java Code by Amandeep Singh on stackoverflow.com
 
It is extremely useful
 
sd why
 
[:43882937] Post - Contains 1 unique characters
Post - Repeated character: *aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa*
Blacklisted user - blacklisted for //stackoverflow.com/a/49547742 (https://m.erwaysoftware.com/posts/by-url?url=//stackoverflow.com/a/49547742) by the metasmoke API
 
Serial vandal?
 
tpu- by NobodyNada
 
3:15 AM
@SmokeDetector did you forget the link?
 
Hm?
 
The spam post didn't include a link
 
Oh, right
 
3:28 AM
Happens about once a day: we get a very short post on Drupal with all the keywords but no link.
A malfunctioning bot somewhere.
 
3:40 AM
[ SmokeDetector | MS ] Bad keyword in body, blacklisted website in body, blacklisted user: Can you expect Up to 9% off rs 3 gold for 2018 RuneScape Game Jam Apr.14-Apr.18? by wrwetiuru on askubuntu.com
 
@SmokeDetector tpu-
 
@SmokeDetector tp-
 
tpu by Nisse Engström
tpu- by Makyen
 
4:04 AM
[ SmokeDetector | MS ] Potentially bad keyword in username: Font size variation within siddurim by Rayaz on judaism.SE
tpu- by K.Dᴀᴠɪs
 
Really? All that caught was Rayaz?
!!/watch pus+y
 
> hairy pussy
 
@K.Dᴀᴠɪs Added pus+y to watchlist
 
specific enough for blacklist?
pussy by itself could be a cat
 
Which is why it's a watch not a blacklist
 
4:06 AM
Although there is only 1 match on all of Pets.
 
I'd rather have rare fp's for that term but catch all the tps
 
[ SmokeDetector ] SmokeDetector started at rev 3af0b22 (SmokeDetector: Auto watch of pus+y by K.Dᴀᴠɪs --autopull) (running on teward/Solar Flare)
Restart: API quota is 10362.
 
Looks like I missed a lot earlier today
I saw that keybase was blowing up my phone. I had to turn off notifications lol
Wasn't in the position to get involved earlier
 
API quota rolled over with 10290 requests remaining. Current quota: 19999.
math: 35
stackoverflow: 6
physics: 4
superuser: 3
mathoverflow.net: 2
workplace: 2
askubuntu: 2
health: 1
reverseengineering: 1
scifi: 1
datascience: 1
gaming: 1
drupal: 1
security: 1
networkengineering: 1
stats: 1
french: 1
sqa: 1
bitcoin: 1
expressionengine: 1
anime: 1
raspberrypi: 1
bioinformatics: 1
ell: 1
movies: 1
dba: 1
hsm: 1
[ SmokeDetector | MS ] URL in title, bad keyword in body, blacklisted website in body, blacklisted website in title, pattern-matching website in body, +1 more: vitacleanseblogs.com/vigra-fast-austria/ by Haryqwit on puzzling.SE (@Mithrandir @micsthepick)
tpu- by K.Dᴀᴠɪs
 
4:24 AM
!!/blacklist-keyword rs\W?gold
 
@tripleee Blacklisted rs\W?gold
 
A J
Morning all!
 
CI on b0ec994 succeeded. Message contains 'autopull', pulling...
 
!!/blacklist-keyword keto\W?tone(?:\W?diet)?
 
[ SmokeDetector ] SmokeDetector started at rev b0ec994 (SmokeDetector: Auto blacklist of rs\W?gold by tripleee --autopull) (running on teward/Solar Flare)
Restart: API quota is 19952.
 
4:27 AM
!!/blacklist-keyword keto\W?tone(?:\W?diet)?
 
@tripleee Blacklisted keto\W?tone(?:\W?diet)?
CI on 424078c succeeded. Message contains 'autopull', pulling...
Restart: API quota is 19926.
 
@AJ \o
 
smirk
 
Maybe a negative lookahead there...
 
4:33 AM
probably should use a negative lookahead on that word
ditto
or pedo(?:file)?\b
 
the regular spelling is with ph, but it's probably easier to enumerate the non-offensive correctly spelled words
 
meh true. I don't type it out everyday so yeah :)
 
though I really have no idea what pedotrible or pedomorphism mean, do we really need to guard against those?
*tribe
 
CI on 18c364d succeeded. Message contains 'autopull', pulling...
[ SmokeDetector ] SmokeDetector started at rev 18c364d (tripleee: reduce FP hits for pedometer etc --autopull) (running on teward/Solar Flare)
Restart: API quota is 19853.
 
4:40 AM
actually should we trigger on paedo too? or is that too sophisticated for trolls?
 
@tripleee it won't hurt to p[ea]{1,2}
 
that will FP on Padova etc, probably prefer pa?edo
but not altogether convinced this is necessary or useful
 
probably not
 
[ SmokeDetector | MS ] Bad keyword in body, blacklisted website in body, link at end of body: I will clarify what this does? by veradotsan on askubuntu.com
 
I was thinking you used a word boundary, which is why I suggested it
 
4:43 AM
tpu- by K.Dᴀᴠɪs
[ SmokeDetector | MS ] Username similar to website in answer: Comparing documents between two MongoDB collections by Studio 3T on stackoverflow.com
 
clearly not if pedometer triggered
 
in your update...
I didn't check it
But suggested it
 
fp- by tripleee
 
@SmokeDetector disclosure, maybe the question is dubious though
 
[ SmokeDetector | MS ] Username similar to website in answer: Synchronize mongo databases on different servers by Studio 3T on stackoverflow.com
 
4:45 AM
Now it looks like spam, eh?
 
yeah, but Bhargav already deleted the second one and left a comment, and he reviewed the first one and edited it, so I'm not touching this
 
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, bad keyword in username, blacklisted website in body, link at end of body, +1 more: Keto Tone Diet - How To Performe? by Keto Tone Diet on askubuntu.com
tpu- by K.Dᴀᴠɪs
[ SmokeDetector | MS ] URL-only title, bad NS for domain in body, bad NS for domain in title, bad keyword in body, bad keyword in title, +6 more: healthpurelives.com/luna-trim/ by fxrv on askubuntu.com
tpu- by K.Dᴀᴠɪs
 
5:11 AM
[ SmokeDetector | MS ] Bad keyword in body, bad pattern in URL body, blacklisted website in body, pattern-matching product name in body: tablet will permit a person to put off all of the sexual issues by user22616 on astronomy.SE
tpu- by PeterJ
[ SmokeDetector | MS ] URL-only title, bad keyword in body, bad keyword in title, link at end of body: apexatropin.co.za/ by lapperdbee on askubuntu.com
tpu- by NobodyNada
[ SmokeDetector | MS ] URL in title, bad NS for domain in body, bad NS for domain in title, bad keyword in body, bad keyword in title, +5 more: healthpurelives.com/luna-trim/ by fxrv on apple.SE
tpu- by tripleee
 
5:30 AM
!!/blacklist-website apexatropin\.co\.za
 
@tripleee That pattern looks like it's already caught by Bad keyword in answer and Bad keyword in body; append -force if you really want to do that.
[ SmokeDetector | MS ] Bad NS for domain in body, bad keyword in body, bad keyword in title, blacklisted website in body, pattern-matching website in body: Steel RX A On this same note, we didn’t by Fleischer on apple.SE
tpu- by K.Dᴀᴠɪs
 
5:49 AM
[ SmokeDetector | MS ] Bad NS for domain in body, bad keyword in body, bad keyword in title, blacklisted website in body, link following arrow in body, +1 more: Keto Tone This thing is adequately by Bricietjen on graphicdesign.SE
tpu- by K.Dᴀᴠɪs
 
Can unprivileged members get a notification when a spam post is registered?
@TriyugiNarayanMani You can just type !!/notify 11540 hinduism.stackexchange.com but I don't know if it works.
 
@NogShine Yup. It's under the "Commands for Everyone" charcoal-se.org/smokey/Commands.html#commands-for-everyone
 
@K.Dᴀᴠɪs Cool
 
[ SmokeDetector | MS ] URL-only title, bad NS for domain in body, bad NS for domain in title, bad keyword in body, bad keyword in title, +5 more: healthpurelives.com/luna-trim/ by ccqvvssms on askubuntu.com
 
!!/notify 11540 hinduism.stackexchange.com
 
6:03 AM
tpu- by NobodyNada
 
@SmokeDetector tpu-
 
@TriyugiNarayanMani You'll now get pings from me if I report a post on hinduism.stackexchange.com, in room 11540 on chat.stackexchange.com
 
@SmokeDetector great
 
!!/notify 11540 stackoverflow.com
 
6:06 AM
@PrincessLuna You'll now get pings from me if I report a post on stackoverflow.com, in room 11540 on chat.stackexchange.com
[ SmokeDetector | MS ] Bad keyword in body, blacklisted website in body, link at end of body, pattern-matching website in body: Naturalist Coffee Intellectual by Hadis tiews on drupal.SE
tpu- by K.Dᴀᴠɪs
[ SmokeDetector | MS ] URL-only title, bad NS for domain in body, bad NS for domain in title, bad keyword in body, bad keyword in title, +6 more: healthpurelives.com/luna-trim/ by sadadradhwaafakhoury on askubuntu.com
 
@SmokeDetector Why didn't Smokey reply that a user is not privileged?
 
tpu- by K.Dᴀᴠɪs
@NogShine That's not a report.
 
It would only reply with that if a user attempts to use a privileged command, and !!/notify is not one.
 
So, can a user reply Great, thanks etc., ?
10 mins ago, by Triyugi Narayan Mani
@SmokeDetector great
 
Appears that way. He can also use !!/amiprivileged to see if he can provide feedback. Getting privileged for that is not difficult, assuming the requirements are met.
!!/amiprivileged
 
6:17 AM
@K.Dᴀᴠɪs ✓ You are a privileged user.
[ SmokeDetector | MS ] URL-only title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +4 more: www.garciniacambogialean.com/vitax-lean by Dows1939 on askubuntu.com
tpu- by K.Dᴀᴠɪs
[ SmokeDetector | MS ] Bad keyword in title: Tribulus Terrestris For Man Sexual Health by joirekosxs on puzzling.SE (@Mithrandir @micsthepick)
tpu- by PeterJ
 
morning
 
> I cannot believe the information would be passed down if there wasn't some truth to it.
A gullible spammer
 
[ SmokeDetector | MS ] Manually reported answer: Unit tests running phantomJS run much slower on windows than iOS/Linux by stlo on stackoverflow.com (@PrincessLuna)
 
ooops
 
naa- by Makyen
 
6:25 AM
It's well past my bedtime :D
 
[ SmokeDetector | MS ] URL-only title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +1 more: mumybear.com/high-tech-cbd/ by user816344 on askubuntu.com
tpu- by K.Dᴀᴠɪs
 
6:45 AM
[ SmokeDetector | MS ] URL-only title, bad NS for domain in body, bad NS for domain in title, bad keyword in body, bad keyword in title, +6 more: supplementforhelp.com/forskolin-fat-loss-extract by waliyagalli on askubuntu.com
tpu- by K.Dᴀᴠɪs
[ SmokeDetector | MS ] Bad NS for domain in body, bad keyword in body, bad keyword in title, blacklisted website in body, link at end of body, +3 more: Eyeserum Magic is really work by nicoleread on askubuntu.com
tpu- by K.Dᴀᴠɪs
 
!!/blacklist-keyword eye\W?serum\W?magic
 
@tripleee Blacklisted eye\W?serum\W?magic
CI on f3d0c42 succeeded. Message contains 'autopull', pulling...
Restart: API quota is 19039.
[ SmokeDetector | MS ] Bad keyword in answer, username similar to website in answer: Apply English punctuation rules by fluper on codegolf.SE (@Mego)
tpu- by A J
 
A J
!!/watch fluper\.com
 
@AJ You don't have code privileges, but I've created PR#1989 for you.
 
@SmokeDetector chat flagged?
 
A J
6:55 AM
I didn't.
I declined the flag though.
 
[ SmokeDetector | MS ] Link at end of answer, potentially bad keyword in answer: Relation between a Wizard's Previous button and history's Back button by athman8 on ux.SE
 
same here
 
tpu- by Makyen
 
I flagged the smokey report on ppcg as spam, but not the chat entry
 
Yeah, we're talking about the chat message being flagged.
 

« first day (1604 days earlier)      last day (2202 days later) »