« first day (3979 days earlier)      last day (879 days later) » 

12:00 AM
@user Oh I see, nvm
 
@user MM on the right side of your picture section, FF in the tight triangle on the separate 2nd graph
 
I wonder how you get people to anonymously do these kinds of surveys. There was apparently one about the level of friendship people at an office had where you rated someone from 1 (just a random guy) to 5 (good friend), idk how you'd pull that off
 
My other question is, was this all self-reported, or did they ask people about other people's relationships?
 
Oh yeah, that would definitely have messed things up
 
0
Q: AoCG2021 Day 21: Blinking through the forest

BubblerPart of Advent of Code Golf 2021 event. See the linked meta post for details. Related to AoC2020 Day 3. In the way to vacation, you're traveling through a forest on an airplane. For some biological and geological reasons, the trees in this forest grow only at the exact integer coordinates on a g...

 
12:02 AM
I'll ask my teacher for the original paper tomorrow
 
@user I mean, I'd probably answer that survey (if it was run by someone outside and not somebody at my office).
 
Wait, it's here
@DLosc I personally would be very afraid about my name being leaked tho
 
I don't think I care that much about my coworkers' opinions of me. :P Maybe that's because I don't consider any of them close friends.
 
Oof
 
If one of the options were "I can't stand them, I wish they'd get a job somewhere else," then yeah, I'd be more hesitant :P
 
12:04 AM
1 message moved to ­Trash
 
@NewPosts down?
 
@user Yeah, I haven't clicked with anybody here the way I did at my previous job. COVID probably has something to do with it, but so do the office culture and the personalities.
 
> Adolescents frequently approached interviewers wearing yellow Add
Health buttons and asked when they would be invited to participate in
the study.15 In all, 90% of the students on the school roster participated
in the in-school survey, and over the course of the interview period, 83%
of all students in the school completed in-home interviews.
Huh, apparently people are more willing to do this kind of thing than I thought
 
Interesting
 
Intere... yeah, what Bubbler said
 
12:07 AM
started my copy of NPSP @RedwolfPrograms
 
@DLosc :( Hopefully COVID'll go away soon (as long as another variant doesn't pop up)
 
Hm, you can get the thought bubble on any site just by changing your parent user
(As in, it's not the room, it's the user)
 
I ain't no one's parent :P
How's Winter/Summer Bash going for everyone?
 
I have one normal hat
and six secret ones
 
Ooh
 
12:14 AM
Wait no two normal, repcapped yseterday
 
@user 1 + 2 for me
 
3?
 
1 normal, 2 secret
 
I have ten days left to get 414 rep
 
Oh
 
12:16 AM
Seems like posting lots of answers doesn't do much to get you hats
 
Or at least, want to - I want to get 20k before the end of this year
 
Maybe I should try posting a question
 
@DLosc That'll give you monologue if someone answers, dialogue if you comment under someone's answer, trialogue if they edit it and quadalogue if ???
 
I see how the theme works, but Monologue is an odd name for a hat that you get if someone else answers your question.
@DLosc Also not pictured: child nodes
 
12:30 AM
I would be very concerned if they were pictured
 
@grandBagel \K will remove the match.
 
12:56 AM
@user not sure how to feel about having just downloaded a file called "teensex.pdf"
 
Why'd you download it lol
 
Ugh I swear NPSP goes down every Monday
Hang on, gotta go reboot the server
 
No one likes Mondays :(
 
(It's not a coincidence btw; Mondays are when it does a weekly reboot to apply security updates)
 
@user firefox pdf handling is buggy
that's why lol
 
1:07 AM
Ah
It's not really downloading if it goes into a temp folder while you open it right in the browser, right? :P
 
you are loading it down from "the cloud" to your local storage so i would consider that downloading
 
Not if you stand upside down while downloading it, because then it'd be uploading 😎
That's why I have the Thought Bubble hat :P
wtf is this lol
The revision history is pretty sus
 
Well TIL you can't use normal SSDs on a server
Which I have been doing for the last four years
 
wait why not
 
Apparently they aren't compatible with (mumble) firmware (mumble mumble) RAID (mumble) SATA backplane
Probably intentional, since the server SSDs are like 5 thousand dollars
Oh nice, I found a drive that's maybe-sort-of-compatible, and it's only $1k. And a whole 240 GB of storage, too!
Yeah, nah
 
1:25 AM
estion, I noticed a lot of similar questions w
^ Copied from MCR
 
to get manual key hat
 
Hang on, I left an old HDD in my server, and according to the manual you can't mix SSDs and HDDs. I'm going to shut it off, take out the HDD, and see if that fixes any of the issues.
If it catches on fire I blame my cat
Okay, about to boot it up. Got a fire extinguisher on hand :p
 
can i learn regex?
 
Okay so apparently you can't just pull a drive out of a machine that expects it to be there
So I gotta figure out how to make the server forget about it
 
1:33 AM
Use one of those lights that makes people forget things like in the movies
I forgot what it's called though.
 
a\kb will match a[b] a[b] ba but i want [a]b [a]b ba, what is the regex code alt?
 
@Fmbalbuena How about a(?=b) (regex101)
I haven't seen \K being used in any code, better avoid it
 
why?
 
Wait hang on...not only did the server not do that weird RAID foreign config thing, a blinking orange light on one of the drives that's been there for like three years is gone all of a sudden
Wait no it's there now
 
@Bubbler what if a(?=b)c
wait, a(?=b).c but matches [abc]
 
1:39 AM
I think I broke everything lol
 
@RedwolfPrograms what do you mean?
 
@RedwolfPrograms have you tried turning it off and on again?
 
The server seems to expect there to be a /dev/sdc, and because I puiled one of the drives there is not one
 
@DLosc neat
 
@Fmbalbuena That's "match a, then expect that the next char is b, but actually match c" so it matches nothing
 
1:42 AM
didn't think that would work for some reason
 
@Bubbler oh understood
 
@Fmbalbuena If you want "match a, expect b but ignore it, and then match c", you need to use groups like (a)b(c) and extract those groups 1 and 2 in your program.
 
@Bubbler no, [a]b[c]
 
I mean, the regex you want is (a)b(c)
 
@Bubbler maybe a(?=bc)|b\Kc? sorry for \K
 
1:47 AM
A regex cannot match two disconnected chunks.
 
@Bubbler No, i want to match [a]b[c]
 
I'm saying you can't.
 
@Bubbler but checking the left?
 
Checking the left is done with (?<=x)y
but the x must be of constant length
 
1:49 AM
ok, [^#]\K#[^\n]*|^#[^\n]* will match the comments?
 
Okay wtf kind of idiot was I, adding an entry to /etc/fstab referring to a device as sdc1 instead of its UUID
As if there would never be a drive added or removed from the machine
 
@Fmbalbuena Why do you need [^#]\K there?
 
@Bubbler because will match the previous "#".
 
You can just do "match # and then everything that is not a newline", which is simply #[^\n]*
@Fmbalbuena No, regex matches from left to right and the matches don't overlap, so you don't need to worry about it
 
@Bubbler I don't understand
 
1:53 AM
@Bubbler you can pseudo-lookbehind with pcre \K iirc
 
If the code has abc#def#ghi\njkl then the regex #[^\n]* matches the first # and then everything until newline, that is #def#ghi
So you just successfully matched a line comment
 
note that re (the default/canonical regex module) in python does NOT have this, but the regex module does
 
@Bubbler hmmm why i think that.
@Bubbler Test 10/56: You must allow upper case letters. The case insensitive modifier is recommended.
i can't think that without many "|"
 
@Fmbalbuena what flavor is this?
that is not a flavor of regex
you don't have to delete your messages every time, btw
 
@Fmbalbuena What language are you using?
 
2:03 AM
it just makes the backscroll harder for people to read since they're missing the context
 
@Bubbler idk
maybe JS
 
is this a site that you're doing example problems on? or is this for a class or something
 
Are you writing your regex enclosed with slashes?
Like /someregex/?
 
@Bubbler yes
@Riker regex101 9th quiz
 
that could be either sed or js probably bubbler but i've seen people use it informally for just pcre
regex101 will certainly have a thing to let you choose what flavor or tell you what flavor
ok, just checked the quiz main page - "The quiz is evaluated using the PCRE2 engine" so PCRE/perl
 
2:05 AM
regex101 allows to choose "flags" at the right end of the regex box. One of them is i, called "ignore case"
Turn it on and you should be good to go
 
@Riker I like fruit flavor
:P
 
In real code it is written as /regex/i btw
 
so in a perl regex the format is (by default at least iirc) /match/flags, so the regex you're using is in between the slashes and you have the flags right after
flags are things like i, g, m
i means 'ignore case' (so a matches both a and A), g is 'global' so it searches for more than just the first match, and m is "multiline", so it can make a match across multiple lines (otherwise it will stop at a newline)
 
Test 17/56: .example.co is not a valid domain.
 
in this case if you want an email address you'll probably be using something like [a-zA-Z], or [a-z] and just globally ignore case (probably fine for an email)
 
2:08 AM
regex code is /[\w\.]+@[\w\.]+\.com/i
 
at risk of sounding like a jerk here - have you tried figuring out why it doesn't work yourself? it doesn't really help you learn too much if somebody tells you why it broke every time
your issue is that it's matching a dot right after the @, which is obviously not gonna fly for a real email - what in your regex matches the stuff after the @? how can you make it not match a dot immediately, but still allow dots later?
 
[\w]+@[\w\.]+\.com
 
now it doesn't match a dot in the first part of the email, which is no good (firstname.lastname@gmail.com)
 
i don't understand. another task
\s{4,} Test 10/14: What if there's more than one occurrence?
 
2:24 AM
@Fmbalbuena You don't learn anything if you skip tasks like that
 
2:52 AM
my quarter rep is 4 while my month is 90
hmmm
 
oh, hey, discord added timeout feature (lets you temp block a user from sending messages, joining voice/stage, etc)
 
yep
 
@hyper-neutrino sounds cool
 
yeah. unfortunatelys granularity on the client-side interface isn't great (after 1 day it's 1 week) and it's hard locked at 28 days max through the API
so it's still better to just have a mute role and make a bot do it
 
3:14 AM
sad
 
Wait I though JWST was launching on the 22nd...
(It's the 24th, apparently)
 
correct
there's still 3 days and 8 hours lol
 
Maybe I should repurpose my CGCC election countdown to be a JWST launch countdown
Although idk who's to say it won't get delayed another decade last minute
 
that site is
cursed
 
3:29 AM
yeah what the hell
 
You can right click the background and set it to picture-in-picture to curse everything else on your screen too
 
oh god
 
Or just right click and disable looping, that'll cause it to be a still image with a rocket
 
or you could just click the stop video button like a normal person
 
3:30 AM
You can double the speed lol
Right click, show controls, and then change speed
 
- background video is disorientingly fast-moving and too prominent for a background especially one that is animated
- stop video button is really small and not obvious, and you cannot resume the video
- the large box changes color when you hover over it despite not being interactable like the buttons are
- the contrast is _barely_ enough
- the page extends to the right and has a horizontal scroll bar but the content does not fill so there's just a random gap if you scroll right
 
also the fact that it loops so quickly doesn't help
 
Double speed rocket go brrr
Okay wow that is shockingly high quality for a gif (ignoring the 10fps)
 
> russian rocket upside down
 
3:35 AM
Well yeah that's the best type
 
3:47 AM
the new log4shell thingy looks tempting
 
is there a third one now?
 
i don’t think so
but the first exploit’s patch isn’t auto distributed
so my computer don’t have patch i think
gonna test it on minecraft lan server >:)
 
Is there a "code that prints itself" challenge already?
 
It's called quine
We have a whole tag for it, for the quine and all the variants (code that prints some manipulation of itself)
 
4:05 AM
@Niko I was doing that the other day lol, it's fun
 
> 404 - image not found!
 
I'd recommend using the standalone server rather than LAN, since it has a console. With singleplayer or LAN you'd have to have the debug log open if you wanted to be able to try things like ${java:version} before doing a real jndi/ldap attack
@pxeger I'd upload to imgur but it's too large
It's just the rocket at double speed and an awful framerate, you're not missing much
 
0
Q: Get the sum of the binary numbers of unicode code points of a string

12944qwertyYeah... it's confusing. Let me explain it a little better: With a string, get the unicode code points of each letter Let's use "Hello World!". The code points would be [72, 101, 108, 108, 111, 32, 87, 111, 114, 108, 100, 33] Of each digit of the code points, get their binary format Get the b...

 
@hyper-neutrino Redwolf's instance is up, yours can go down now
 
4:11 AM
NPSP 2.0 is dropping tomorrow, with full support for multi-instance stuff, so this shouldn't be a problem for long
 
How do NP/SP deal with being message ratelimited?
 
Although I should probably do a four week beta testing period in another room
 
(That should never happen, but might)
 
•́ ‿ ,•̀ I don't know how to think about a challenge.
 
@emanresuA Currently they wait 10s between every message so ratelimiting should never occur, but in NPSP 2.0 it'll just wait the ratelimit plus one second
I don't remember if NPSP 1.0 is robust enough to handle being ratelimited, or if it'll just silently drop the post
 
4:15 AM
alright, duplicate npsp is down
 
@RedwolfPrograms are you gonna publish the source code?
 
Yes
NPSP 1.0 is such a hideous mess I'd be embarassed to, but I'm making a more generic chatbot library for NPSP 2.0, and building it off that
So the library will be published, and NPSP itself will also be published
The only awful garbage in NPSP 2.0 so far is the HTTP library I had to custom design, which will explode if you don't handle it very carefully
 
lol
 
Since none of Node's networking libraries handle cookies and redirects properly. Cookies? Sure. Redirects? Sure. Cookies and redirects? That's some advanced stuff right there.
 
@RedwolfPrograms Post it on CR :P
Whcih of these two is dupe of the other?
@RedwolfPrograms Get outgolfed lol
 
4:52 AM
i like ⍺*⎕
 
@emanresuA quite a lot of the answers to the older one use weird facts that rely on the limit of 2
but my question allows any number so they wouldn't work there
 
So... leave them both open?
 
well the older one also has stupid rules like "no builtins" and "answer must be a function, not a full program" (!)
 
Yeah, leave both open. Doesn't look too much like dupes
 
5:53 AM
Just posted a steganography cop
 
6:12 AM
@PyGamer0 Old APL merch used to say ⍺*⎕ – maybe we should explore that for logos?
 
i think it's quite clever :)
 
Sure, and it was back when all computers had identical keyboards, and pressing Shift+a,p,l gave ⍺*⎕ but these days… and surely a lot of people don't use APLKey+p for * rather preferring Shift+8 or similar.
 
@Adám maybe
 
7:18 AM
ok turns out sizeof(x)/sizeof(x[0]) doesnt work on pointers
 
yeah i was going to say - was trying to load up gcc to confirm. sizeof on a pointer will return the size of the pointer
 
Yeah, that's a correct way to do it. There's no way the map function can find out the incoming array's size without explicitly passing the value
 
that's why i'ts standard to have function args take it a length, fwiw - you'll see that a lot in coding challenges on various sites (like AoC, for the input specs) and iirc it's in the standard i/o formats here
 
Yes
Well if it's stdin/stdout then some problems get away with not giving the number of items in advance, and C/C++ users are constantly annoyed by those
 
7:22 AM
so i think C will not have [1, 2, [3, 5, [], 6]] # Python arrays
 
well, you could have an array of pointers
 
do i use a struct + union combination
 
but that's funky
i'm gonna call that one an x-y problem - why do you want a jagged array
 
You might want to use tagged unions
 
@Riker personal reasons well just seeing if i can implement some simple python stuff in C
 
7:24 AM
TBF some code golf problems do require jagged array I/O
 
i think it would be reasonable to request a padded 2d array maybe but it's a moot point - an array of pointers should work for that
i.e. int row1[]; int row2[]; int* arr[]
 
@Riker but what if its nested even more
 
and then store int arrSize[] with sizes of each row
 
array of pointers to pointers to pointers?
 
well. do you have another option? it's C, there's literally no other clean way to store it
assuming you're averse to some malloc wizardry at least
 
7:27 AM
> do you have another option?
structures?
 
i'm going to call you a heathen if you're attempting to use structs for arrays of length n :/ i think that wouldn't expand well at all
 
well how is python implemented in C then?
 
a lot of wizardry, and it probably pads them to an even length i'm guessing
or it just does it directly via malloc and tracks each row's length
i.e. it counts how long the row is at each point, mallocs the right length, and stores how long we've allocated somewhere (that point you've probably got like a 2 element struct - pointer & size? 3 if you include some type flag maybe idk)
cpython is on github iirc?
 
wtf. ok so i have serv aliased to ssh <my server> on my computer
i tried running serv like 8 times and it kept just hanging
so i run the ssh command directly and it works
and now serv works too
i hate computers
 
@Riker yes
 
7:31 AM
actually yeah, that's just an n-ary tree at that point
so yeah, the easiest way is probably a simple struct with a pointer + size, and then an array of structs, rather than trying to store anything in the struct itself. dunno, i've never taken a formal class that used C (esp not a data structures one)
this is probably basic knowledge to anybody who has / works with C professionally
 
so is it possible to write a C answer for vvvv
17
Q: Jam don't add like that

DennisBackground Jelly's arithmetic atoms vectorize automatically. In fact, x + y is well-defined whenever x and y are numbers or ragged arrays of numbers. Jelly's source code implements this behavior using a generic vectorizer, but for this challenge, we'll only consider addition of integers and nest...

cause i am gonna bounty it (no deadline)
 
C is TC so yes /s
lemme take a look. i haven't coded non-really-messed-up-C in a while
 
0
A: List of bounties with no deadline

PyGamer0250 Reputation I will give anyone who writes a working answer in C (not C++), for Jam don't add like that. NOTE: Comment to receive your reputation

^ posted
 
oh god i remember this coming around the first time
wouldn't be surprised if leaky's java answer works well ported
 
uh
 
7:41 AM
@PyGamer0 oh, actually, that's an excellent point. should probably edit your bounty to request a non-string input form otherwise i'm pretty sure kenny's strategy could be ported - take it as a char array or something and just parse it from there
assuming the bounty is because c n-dimensional jagged arrays are clunky
 
ok i edited
 
An almost minimal example of jagged array in C:
2
 
^ proof certain things are crimes (or at least should be) /s
 
@Riker thats just macros
 
0
A: List of bounties with no deadline

PyGamer0250 Reputation I will give anyone who writes a working answer in C (not C++), for Jam don't add like that. Condition The input must be taken as integers, not as char pointers (or strings) NOTE: Comment to receive your reputation

 
7:47 AM
@PyGamer0 look at what they do :)
 
wtf am i looking at
 
@Riker i dont understand that lol
@hyper-neutrino what your eyes are seeing
:P
 
what, nero's macromagic? he was fueled by spite, the lack of desire to actually refactor the code, and too many illadvised SO questions
 
having the black python format plugin for atom is so convenient. not only cuz it makes my code actually follow good practice (for formatting at least) but because when moving code around or editing things i can leave it as messy as i want and just press one shortcut and have it fix it for me
 
@Bubbler I love (and somewhat hate) the fact that I can write that kind of code without looking up more than a page on cppreference.com
 
8:03 AM
it's pretty clean code too :)
 
^
although i dont see free( calls (although i think thats unnecessary for small demonstrations)
 
Yeah, I'm leaking the memory
Doing any kind of manipulation on an Array, including destroying it, requires writing a recursive function that does case analysis based on tag of each Item
 
RIP garbage collection my beloved :(
 
Yeah, if you need to do it in production, then you need to code GC or (at least) refcounting yourself
That is pretty much how Python does things
 
i wouldn't be surprised if they do both
 
8:10 AM
Why intentionally leak memory when you can unintentionally leak memory?
3
 
swag
 
"I leak memory, no one can stop me"
 
@Bubbler except when your computer crashes
 
Btw, as of December 2021, you can cheat at the Jam thing in C: use libk (a library usable in C code that runs K)
 
o_O
 
8:24 AM
Anyone know of the equivalent in other languages?
123
A: Import Python Script Into Another?

jedwardsIt depends on how the code in the first file is structured. If it's just a bunch of functions, like: # first.py def foo(): print("foo") def bar(): print("bar") Then you could import it and use the functions as follows: # second.py import first first.foo() # prints "foo" first.bar() # ...

 
JS modules (or commonjs + require) need explicit exports, so no real equivalent except <script src=...>
 
@Adám What is the exact requirement? Is it sufficient to be able to import another source code in the same directory?
 
Vyxal has no imports
 
@Bubbler Two #! files, and yes, same dir is OK, use one from within the other.
 
Oh, so it is limited to scripted languages.
 
8:28 AM
Yes.
 
@Adám java uses import (unnecessary when in same pkg/directory), C & friends use #include
 
@Riker Neither of those are for #! scripts, correct?
 
I can't think of any scripting language that supports #! and modules, other than python
 
ah, for things that specifically use shebangs? idk
I mean, i use shebangs for my random c scripts sometimes
 
Background: We're soon releasing Dyalog 18.2 with #! support, but there's no way to call one such script from another (other than starting a command processor – and then you can't really "use it").
 
8:31 AM
Vyxal might be getting a module / import / package management system at some point
 
are you wondering whether or not shebang support implies the ability to import another file? i'd say no
 
And I'm pretty sure #!/path/to/vyxal_executable works
 
@emanresuA that should work by definition i think?
 
While one script could print stuff to stdout and the other could capture this, it'd be nice to return proper values and use them in the top-level script.
 
Judging from the docs, BQN can import with •Import, where the target can be a script or module
and a script can return a value
 
8:34 AM
yeah it allows shebangs
 
@Riker ¯\_(ツ)_/¯ Throws a parse error in stuff like Jelly
Vyxal just has # comments
 
oh, makes sense. i was forgetting that the shebang is ignored via comment a lot of the time, not just skipped entirely
 
Yeah, the target language should have a # line comment or specifically ignore the first line if shebang is detected (the latter is planned for Dyalog APL, IIRC)
 
# at the start of a line throws a parse error in jelly anyway
(Because # is a quick with no links to operate on)
 
8:41 AM
Not sure what I'm looking at.
 
JS?
 
Shell based wrapper for JS? Looks useful
 
looks like a package for js
 
Yep
 
 
3 hours later…
11:37 AM
do y'all ever just put off doing high priority stuff on your todo list because it requires too much thinking?
because I do
 
Of course.
 
like there's something I've been putting off for 17 days now
2 weeks in a row now I've said "I'll work on that this week"
but no
I don't
 
My colleagues started complaining that my weekly report's "coming up" list kept having the same items week after week, so now I have to only list things I realistically think I'll get around to.
 

« first day (3979 days earlier)      last day (879 days later) »