« first day (2693 days earlier)      last day (2150 days later) » 

12:02 AM
o_O
CMP What should ↓⍵ do in RAD? It can't be like Dyalog here, as rank isn't a thing
 
IDK try it.
 
I don't know what it should do, that's the problem.
@CatWizard, why is your bio just "Ł"?
 
@Zacharý Yes
 
12:17 AM
Bye. Gotta sleep.
 
@CatWizard Why tho?
 
Oh I see.
I missed the why
 
;-;
@CatWizard Well, is there even a reason?
 
Yes.
 
What is it?
 
12:19 AM
OH yes.
I thought it would be a good idea to have a bio
 
Very funny, I'm ŁOŁing right now.
@CatWizard Why is your bio just Ł?
 
ł is pronounced like W
So you would be wowing
 
Only for Polish tho'
 
Yeah.
And only after soviet occupation
 
IIRC, some old dialects of Polish have it as a velarized lateral (dark l).
 
12:22 AM
Yes before soviet occupation ended
 
You Polish?
 
I clean
8
 
No I just have a Ł in my bio
 
@LeakyNun What?
 
nothing
 
12:24 AM
perhaps he missed the capital P
 
OH! Missed that.
@CatWizard Pourquoi?!
 
I don't speak polish
 
@CatWizard "Pourquoi" French, like Spanish "Por qué"
 
Oh I don't speak that either
I took spanish in high scrool
 
cough normie
 
12:26 AM
Sorry
 
But seriously though, why Ł over, say, Ð?
 
Hey now, don't bring me into this
;P
 
I can't type Ð
∂ÎÎŒ„´‰ˇÁ¨ˆØØÒÔÓ˝ÏÎÍŸ˛Ç◊ı˜Âµ˜∫√ç≈Ωåß∂ƒ©˙∆˚¬πøˆ¨¥†®´∑œ¡™£¢∞§¶•ªººººº‌​ºººº–“‘æ…≥≤¬…ææææ«« ... See can't do it
 
YET YOU CAN TYPE Ł?
 
Yes
 
12:30 AM
That's one of the FEW symbols I cannot type via my international keyboard
What keyboard layout are you using?
 
QWERTY
 
._.
 
I use speach to text for Ł
 
What do you use for it then?!
 
Use speach to text for?
I use it to type Ł
I'm not sure what question you are asking
 
12:33 AM
What do you say for Ł?
 
"Ł"
 
Insert Kannada-face here
 
( ͡° ͜ʖ ͡°)
 
Kannada, not Lenny. Look of disapproval.
 
I can't type that one
 
12:46 AM
TIL I can type a bunch of random shapes, arrows, and accented letters directly from my keyboard
 
International keyboard layout is best.
 
-2
Q: Decompose in polynomial time if possible

l4m2Given a number being the product of two primes, decompose it into the two primes in polynomial time if possible(currently whether or not is unknown). Proof is required. Shortest code win.

 
1:23 AM
The next OEIS sequence in the OEIS answer-chaining challenge actually looks more solvable than some of the other crazy BS I've pulled off there... IDK why I gave up on it lol. I thought it was a lot harder than it actually is.
 
it's still going on?
 
no
last answer is on march 6
 
oh :c
 
yes :c user202729 linked to one that I completely did not understand until about 3 minutes ago when I looked at it and my brain decided to work
the problem is I don't have programming languages left
I spent way too much time on that challenge lol (26 answers with at least like 3 of them being one-week last minute answers). Then again NieDzejkob has 41 lol
 
@HyperNeutrino yes you do
 
1:31 AM
well ok none that I'm fluent in
 
:| learn one in a week then
 
lol
is there a challenge asking "determine if a sequence is arithmetic/geometric/neither"
 
@HyperNeutrino not even kidding that's what i do when i'm bored
 
TBH I could it's just that I didn't really care enough anymore
 
(well, actually less than a week) (this way you'd have time to write the submission too)
 
Anonymous
1:32 AM
@HyperNeutrino codegolf.stackexchange.com/q/68745/45941 is basically that
 
1
Q: When should I give hints for a challenge?

l4m2If a challenge is difficult, when is it suitable to hint? (I don't think early is good, it breaks other's thought)

 
Has there been an exact polynomial fit question?
 
@JungHwanMin Only problematic for questions which need clarification (which includes not sandboxed ones)
Reposting of starred messages (long time ago) is likely to be starred?
Mego has posted that 3 times total :/
 
Anonymous
1:49 AM
@user202729 4 - one time didn't onebox
 
Anonymous
The secret is to post funny stuff when it's relevant
 
Of course because the starboard doesn't show surrounding context...
 
ok so my friend complains about my code lines being too long but this is something from his code:
 
@HyperNeutrino What is your screen resolution?
 
return f.read() % ("".join(featured_row %(problem_map[p]['id'],problem_map[p]['title'],problem_map[p]['ctst'],", ".join(problem_map[p]['tags']),sum(map(int,problem_map[p]['pts'].split("/"))),problem_map[p]['id']) for p in sorted(getFeaturedProblems(),key=lambda x:(sum(map(int,problem_map[x]['pts'].split("/"))),problem_map[x]['id']))),"".join(row %(problem_map[p]['id'],problem_map[p]['title'],problem_map[p]['ctst'],", ".join(problem_map[p]['tags']),sum(map(int,problem_map[p]['pts'].split("/"))),problem_map[p]['id']) for p in sorted(getProblems(),key=lambda x:(sum(map(int,problem_map[x]['pts'
 
1:52 AM
@HyperNeutrino wtf
 
Anonymous
That's horrifying
 
Anonymous
I start busting kneecaps at 120
 
even I CAN'T understand it
 
Nevermind.
 
please delete
@HyperNeutrino i can. but it isn't nice
 
1:52 AM
well true if I tried I could but ew
 
my lines are 200 max, and that's only special exceptions
i usually try to stay under 80
@HyperNeutrino oh god
what's with that spacing
 
Next time use named string parameter (or something like that)
 
Anonymous
I stay under 80 so PyCharm doesn't criticize me
 
and like idk why he hates putting spaces around operators. and then when I use a double-list-comp he blames me for "code golfing". like half the time I don't even use double-list-comps in code golf
 
@Mego I don't usually use Python so I'm fine :P Charcoal's lines should all be under 80 though
 
1:53 AM
@ASCII-only yeah idk he hates operator spacing but is also inconsistent. and like the rest of the code for our server is all spaced nicely because I wrote it and I actually bothered to make the code look decent for this project
 
@HyperNeutrino Just a coding convention.
 
@HyperNeutrino no look. featured_row %(
 
oh
uh ew.
 
... inconsistent.
 
Don't worry guys! I need to make some changes to that line so it will be even longer!
 
Anonymous
1:54 AM
Obviously he's calling the function named %
 
@HyperNeutrino In that case add some comments.
 
@HyperNeutrino just use a templating engine
 
Anonymous
Split that up into multiple lines, add some well-named variables, and comment it
 
I think that's recent code because I ran this entire thing through a prettifier some time ago
 
pls just use jinja2
 
Anonymous
1:56 AM
CMC: given some Python code, convert it into functionally-equivalent code that follows PEP 8
 
nvm the prettifier doesn't fix that
 
I pretty-space Python but not C++.
 
@ASCII-only I actually considered that but it's too late to go back. Over the summer if I get bored at home or during my free time at summer camp I might rewrite the whole core regardless of whether he likes it or not
I might even do it in 2 weeks because I only have 2 exams
 
@HyperNeutrino 0/20 2 weeks is too long
 
@ASCII-only fine I'll just not study and work on this I guess
(TBH that's what's going to happen anyway ¯\_(ツ)_/¯)
 
1:58 AM
@HyperNeutrino wait. do you mean start working on this after your second exam is finished
 
that's probably a better idea isn't it >_>
 
yes <_<
 
2:23 AM
Has there been no PPCG podcast episode 2? :(
 
@Downgoat nope
 
Anonymous
We all got busy with other stuff (jobs, life, etc)
 
We should do it.
I'm never busy.
 
2:54 AM
Ugh, I need a new SSL certificate
 
3:30 AM
Help how can I disable the (weird mode) in [uim] which is activated by pressing <shift> <space>
(only in some applications)
 
@HyperNeutrino BLECH
 
@user202729 Wow! you just what has been causing my problem in firefox recently. I have no solution, of course you can press <shift> <space> to revert
 
@H.PWiz of course (but that's annoying)
 
agreed. @ASCII-only the underlined text I was getting in firefox was caused by uim with shift+space
 
Which linux distro are you using?
 
3:44 AM
Arch
 
Why don't everything just organized in SE style. Much easier to read.
 
4:03 AM
I think I solved it.
 
@H.PWiz so it was bascially an IME
 
Before that... help how can I undo some configurations that I've done on uim
@H.PWiz For the dialog way, run uim-pref-gtk. Go to "Global key binding 1" tab. Try removing the occurences of <Shift>space there to see if the problem disappears.
The "non-dialog" way is to modify ~/.uim.d/customs/custom-global-keys1.scm
Because there are too many occurences of <Shift>space I don't know which one is "correct". I (accidentally) removed them all, so I can't test.\
 
@user202729 :|
@H.PWiz why use UIM at all though (or at least change to Super+space and Shift+super+space like is default in windows)
 
@ASCII-only I had used it previously because it allowed me to update keyboard configs without restarting x. I don't use it now, and have just removed it
 
4:13 AM
@H.PWiz ah. makes sense i guess
(does xcompose actually not work with unicode)
 
Although it didn't fix the problem...
 
@H.PWiz problem?
 
Shift+space causes weird mode in some applications
I should restart x and then check
Problem fixed!
 
 
4 hours later…
7:58 AM
I can't find any "ruling" about this but if I have input that is the output at the same time (like an array), after the change is done as expected, can I throw an exception?
 
8:19 AM
> If it causes an error that aborts the entire program, or it throws an exception (even if it could be caught!), then I think this should not be allowed.
So, you can't.
(unless it's in a REPL)
 
9:07 AM
My colleague Marshall expects that a simple APL translation of the your Python solution would be a lot faster since the main cost should just be moving list data around, and Dyalog has a lot less data to move because it has bit
Booleans.
 
@Adám :| what's with that random newline
 
@ASCII-only Cut and paste error which happened to line up with the chat input box so I didn't notice.
 
9:27 AM
0
A: Sandbox for Proposed Challenges

BgrWorkerDead Frog Walking code-golfarray-manipulation Introduction Jonny wants to play Frogger. However, he's not very good. In fact, he will only try to move forward, and only after the platforms have moved. Find out if Jonny's frog manages to reach the end of the path, or if it dies and where. Cha...

 
@Adám I see
 
@LeakyNun Would you try?
 
@Adám there isn't really much point though
 
maybe later... speed isn't a big concern
 
performance is only a secondary tiebreaker
and i'd imagine the arrays aren't that big
maybe 10kb, max
 
9:30 AM
@ASCII-only Well, a direct translation should have exactly as many steps as the current one, so it it would win on speed.
OK, never mind that the author of both solutions is the same, but still, competition is competition.
 
@Adám it's still only a tiebreaker :p
and really why not just use forth/julia/rust/whatever the fastest language is instead :P
 
@ASCII-only Because they all have to move the data around, and not many languages use bit Booleans.
 
@Adám you can implement bit booleans though (and C++ has them builtin IIRC?)
@Adám or even handwritten asm :P
 
9
A: Why is a boolean 1 byte and not 1 bit of size?

sukruThe easiest answer is; it's because the CPU addresses memory in bytes and not in bits, and bitwise operations are very slow. However it's possible to use bit-size allocation in C++. There's std::vector specialization for bit vectors, and also structs taking bit sized entries.

 
@Adám exactly
actually it would be even faster if you calculated (even roughly is fine) the threshold between bitwise operations and copying being the bottleneck
 
10:07 AM
0
Q: Pair and sort from two integer lists

user1589188You are given two arbitrary integer lists, e.g. [-10, 99, -6, 43], [100, 32, 44]. You are to pick one integer from each list and sum them up, then sort the best pairs ordered by the absolute distance to 100 in ascending order. Paired up integers from each list shall not be repeated in the output....

 
@ASCII-only My rule is "the array size is the largest power of 2 such that you can handle in time". If you can handle ~10kb in less than time limit then I will try more than that...
If you can make a linear solution it's going to be larger than that.
 
@user202729 yeah but that's only tiebreaker i guess that's a good point
but it's not like two solutions can get the exact same number of operations
hmm. @Cowsquack @Mr.Xcoder do you have any idea how to get 48 bytes for smile more
 
10:23 AM
well idk python so no
 
@Cowsquack 49 bytes is easy though: you should only need one 2-digit number in your source. and one set of syntax parens
 
wait I just remembered, variables became open
 
@Cowsquack open?
 
post-mortem
 
variables?
 
10:27 AM
on anagol
 
ah.
looks that's surprisingly few answers
 
tails beat my sed answer (by 2 byets), I want to see how
 
i wonder how the python 3 smile more is only 51
i ported the py2 49 byter and it became 58
 
0
Q: Embiggen your input

ScootsThis is a bit similar to this dust covered entry but I'm hoping my spin on it makes it unique enough. Not been able to find anything dissuading me from posting this but there is quite the sea out there. Anyway! The challenge: Your code receives a string of characters; It converts this into an A...

 
@Cowsquack similar approach, but clever way of doing the same s/.*/&=&/ when /?/
 
11:11 AM
@ASCII-only should Funcsational be strongly typed?
 
0
A: Sandbox for Proposed Challenges

Jo KingBytes vs Characters popularity-contest? tips For almost every code-golf challenge on the site, the code size is measured in bytes rather than chars, in order to prevent people from taking shortcuts like stuffing a large amount of data into a unicode string. However, other sites, such as code-gol...

 
@Mayube stringly typed it's not practical so no
 
not practical?
 
not a practical language
imo string typing makes it too practical :P
 
Oh right, I thought you meant strongly typed isn't practical
 
11:33 AM
So if the kilogram is the accepted unit of mass in SI
What we call a "gram" should be a "millikilogram"?
 
:| i'm bad at golfing in C
@betseg kinda, but not really
@betseg we'd redefine 1 gram as 1 kg instead probably
 
@ASCII-only what are you trying to do
 
@betseg anagol. smile more
 
Wut
 
0
Q: Large memory support for bottom language

l4m2Assuming an algorithm requires an array of 1080 bytes of RAM(maybe not realistic, but theoretically the algorithm work). For languages like Python, [0]*(10**80) just theoretically work (ignoring optimizing here), but for C, currently there's only 64-bit machine, and avaliable RAM far not enough. ...

 
12:01 PM
Hard
 
12:29 PM
gah wtf js
somehow this outputs true
if (regex.test(symbol) === false) console.log(regex.test(symbol));
 
@Mayube Who says regex.test must consistently return the same value? ...
Or even symbol...
 
given the same regex and the same input, the output should be consistent.. it's regex
 
@Mayube MCVE?
@ASCII-only What's the spec... no spec? Just some example I/O? Looks easy.
 
well here's the thing, given the regex I'm using, and the symbol I'm using, regex.test(symbol) should and is always returning true, except in that if statement
/^[a-zA-Z0-9\-_]+\s*\(?$/g.test('f('); consistently returns true
 
ah, but that's a new regex every time
 
12:33 PM
how so?
 
The regex object is reconstructed every time.
 
if you write regex = /./g; console.log(regex.test(".")); console.log(regex.test(".")); then you get true and false
 
@Mayube There, I know it's it.
 
@Neil why?
 
12:35 PM
ah
 
@Mayube it's for looping when you want more than just the main match results (for which ".".match(/./g) suffices)
I facepalmed myself recently when I realised I'd triggered that behaviour
(I was trying to block certain strings but if you tested the same string twice then the block was ignored)
 
I see yeah, that makes sense
I guess I've just developed the habit of always putting g in my regexes, not sure why
thanks guys!
 
/me feels bad when people ask SO-suitable questions on TNB with less details.
(next time please provide a MCVE, thanks)
Q: For this sandbox challenge does being a group give it any special property?
 
1:00 PM
@user202729 yeah
@Neil @Mayube and then there's me who's always known about this but never actually found somewhere to use it >_>
 
0
A: Sandbox for Proposed Challenges

user202729Is there a dangling else? Define a statement as follows (Backur-Naur form) (based on Java syntax): statement ::= ";" | "if(true)" <statement> | "if(true)" <statement> "else " <statement> Example valid statements: ; if(true);else ; if(true);else if(true);else; Your task is, given a vali...

 
@NewSandboxedPosts Very fast this time...
 
> Backur-Naur
@user202729 ^
 
Oops typo.
Next time I will only write BNF. Cannot make a mistake.
 
also: this is basically just determining whether there's an else at depth > 0 right
 
1:04 PM
@ASCII-only Not sure. I want to see if it's too easy.
 
or not
 
1:56 PM
@ASCII-only I don't think so. if(true);else{if(true);else;} contains no dangling else.
 
 
1 hour later…
3:08 PM
@user202729 But the grammar does not contain {}
 
@wastl The {} are just for illustration purposes.
Demonstrates how the statement is parsed.
 
3:30 PM
CMP: What should ↓⍵ do in RAD?
 
3:43 PM
selfdestruct. :P (good day to all)
 
"good day" as in entrance, or exit?
 
both. and yet, neither.
 
3:55 PM
@Zacharý 1↓⍵
 
@Adám That definitely seems easy enough to implement!
 
@Zacharý What does ⊃⍵ do?
 
gives either the first element, or it gives the prototype element (0 or ' ') if the length is 0
 
@Zacharý And ↑⍵?
 
@Adám Pads to rectangular.
 
4:02 PM
@Zacharý Maybe ↓⍵ should remove trailing prototypes?
 
I still have two sets of arrows left to use, so I'm sticking with 1↓⍵, as the temporary name I had (Drop1) fits well with that.
Correction, I still have two sets of VERTICAL arrows left to use for that
 
4:26 PM
Oh, Adám, I found a challenge where ⎕IO←2 would save some bytes: codegolf.stackexchange.com/a/166638/55550
 
4:58 PM
@Zacharý Even better if we had as a ⎕IO-independent primitive: …3 gives 0 1 2 3. 2…9 gives 2 3 4 5 6 7 8 9.
 
whew, finally making some progress with funcsational
 
⎕IO←2 (if it wasn't counted in the byte count) would still shave more bytes than would. 1+⍳9 would become ⍳9 rather than 2…10
 
@Zacharý Sure, but ,\ instead of ∘., is surely a more desirable way to save a byte than that.
 
@Adám Yeah, or just a cartesian product builtin would save even more bytes.
 
@Zacharý A Cartesian-f for each built-in?
 
5:19 PM
@Adám No! Just for ,
 
5:37 PM
@Zacharý Isn't that a bit specific? How about ` ∘.{⍺⍵}`?
 
5:48 PM
is it just me or when you load this does the h1 font update later than the body font
 
404
 
downforeveryoneorjustme.com is so slow that I thought it's down for a minute
 
6:12 PM
 
> This website is under heavy load
We're sorry, too many people are accessing this website at the same time. We're working on this problem. Please try again later.
Ironic.
4
 
6:44 PM
@Dennis try with isup.me then ;)
it also saves you from typing yet another 13 bytes :P
 
@EriktheOutgolfer That's what I did. It was down.
 
and, oh, sometimes you need to be a bit patient with everything, isup.me isn't usually as slow as Soaku said it is above
 
7:00 PM
@Adám Whoops, that's what I meant.
 
@Zacharý We really should have a juxtaposition function. Link may be too odd, but {⍺⍵} isn't odd. I'd vote for .
 
@Adám What do you mean by "link"?
 
@Zacharý {(⊂⍺),⊆⍵}.
 
@Dennis ohhh, you actually replied, chat bug hiding replies strikes again
 
will probably be used as the juxtaposition function in RAD (It's in the CP already)
 
7:06 PM
and, ah, the way you said it made it look like you tried Adám's link
 
@EriktheOutgolfer Adám's Iverson's.
 
@Adám what? is isitdownrightnow.com Iverson's?
 
@EriktheOutgolfer Whoa, what a misunderstanding (?) You spoke to Z about my hyperlink exactly as I spoke to him about a link function.
 
@Adám uh... I was talking to, eh, "D", not "Z"
it's just that I didn't want to double-ping him
 
@EriktheOutgolfer Ah, but that was very unclear without a ↰
 
7:21 PM
Jeez. APL dialects vary/varied/changed A LOT. I thought ⎕ML was bad >_<
 
@Zacharý On top of the original APL ("APL1" ≈ APL\360) there were basically two competing extensions: IPSA's and NARS. Dictionary follows IPSA. Dyalog follows NARS.
 
@Adám And IPSA more resembles J, correct?
 
@Zacharý J is more or less ASCIIlliterated IPSA APL plus a few primitives and tacit functions.
 
8:01 PM
But NARS is clearly superior :P
 
The NARS-following ones, of course!
 
> This is a duplicate of #3068, which will be fixed in the next release, likely 5.2.0 which is due out around 2018-07-01.
god damn it apple making stupid freaking paywalls and making every stop working if you don't update immediately ffs
 
hah and remember when you were defending apple a while back in hdf
 
8:19 PM
hdf?
 
@Pavel Base on what? Do you know what the functional (not just glyph choices) differences are?
 
@Cowsquack oh I think apple is fine in concept
but at this point I'm so freaking fed up with their stupid freaking version control
you can't even install xcode easily on anything other htan the latest macOS version
@Zacharý this one other chat we're both in
also @Cowsquack when are you going back to kritixi on PPCG?
 
I like the pilcrow permalinks
 
@Riker this
not just referring to xcode, but many other things in general
 
@Adám That was a joke, I have no idea what's an extension and what's base APL.
 
8:26 PM
@Riker when I finish school
 
@AdmBorkBork that's what the symbol generally means, apart from the fact that you may know it better as == \n (because of you know what)
 
@Riker tbh nh's page looked like a scam to me when I first saw it
 
@Cowsquack I mean you're not wrong for a while it really was
@Cowsquack wait are you not done?
smth smth "updated constnatly" something
 
@EriktheOutgolfer Really? I always understood it to be a paragraph break, not an HTML anchor point.
 
well, it is a paragraph break
however, if it is permalinked, it's usually a permalink :)
 
8:30 PM
@Riker Why not just update your mac
Problem solved
 
2 reasons: one, it errors on install since I have a dualbooted partition atm? no clue but that seems to be teh cause
two, I've been working on a largeish project and I don't want to mess it up by updating (backups yes but still I want a working computer), that ends in 2 days so I'll update then
but yes under any other circumstances I would have updated a while ago
 
I can't stand not having the most up to date software
 
so for example you'd upgrade Ubuntu every 6 months?
 
@EriktheOutgolfer I would not use Ubuntu because Ubuntu has a lot of outdated software in its repos.
But I do upgrade Fedora twice a year, yes.
 
I stick to LTS versions
and I don't upgrade until they get a .1 suffix (Ubuntu)
 
8:41 PM
Fedora has no LTS, every release reaches end of life in a year
I'm considering switching to arch
 
tfw minxomat posts a link
6
 
9:45 PM
0
A: Sandbox for Proposed Challenges

Sherlock9Generate the Recamán Sequence code-golfsequence The newest Numberphile video talks about a sequence created by Bernardo Recamán Santos, a Colombian mathematician. The OEIS entry is here and it is defined as follows: Start with the non-negative integers and with the number 0 at index 0. The it...

 
 
2 hours later…
11:47 PM
0
A: Sandbox for Proposed Challenges

BeefsterASCII Maze Compression (WIP) code-challenge compression maze Challenge Design a compression algorithm specialized for compressing ASCII mazes. You will need to create both a compression algorithm and a decompression algorithm. Your score will be based on the size of your compressed mazes. M...

 

« first day (2693 days earlier)      last day (2150 days later) »