« first day (2123 days earlier)      last day (2723 days later) » 

3:00 AM
ಠ_ಠ
 
@DrMcMoylex I'll try that
@quartata Well what else would you call it? :P
 
That's what I'm trying to decide...
 
You should put under jelly's specialty "everything"
 
"Declarative" is the best I can think of honestly but it's imperative in the way it handles input
 
On a more serious note, MATL's specialty is matrices, and osabie's is base conversion
 
3:03 AM
Wait, I need to think about this.
 
@quartata IIRC Blender is, by default, set to 1 BU = 1 meter. There's a setting for it somewhere though. Possibly in the world settings.
 
Would changing "See docs" to just "Docs" in the quick-ref column be acceptable? It's currently making a bunch of rows double-high for basically no reason
 
Are we going for paradigm specifically?
Or something more specific?
 
In the "Type" column?
 
Yes
 
3:06 AM
I was trying to go for paradigm mainly
 
I suppose V is imperative
 
Does stack-based qualify as imperative?
 
Which would make it the only one
 
It's OK to have its own category, it's pretty unique in the world of golfing langs
 
@ETHproductions Nitpick: postfix/prefix are not paradigms. Pyth is imperative
 
3:09 AM
The proper paradigm is concatenative
for stack based
 
Technically V is postfix
 
Yes, maybe paradigm isn't the right word...
I guess I'm describing languages' memory models mainly
 
Rather than trying to perfectly categorize everything, you could go with multiple paradigms
 
Wait, no, that's only stack-based... infix and postfix are not memory models
@DrMcMoylex How so?
 
Like paradigm and syntactical style
 
3:12 AM
@ETHproductions Like "pyth: imperative, prefix notation"
 
Yes, notation is the word I'm looking for
 
"MATL: imperative, postfix notation, stack-based"
 
But aren't practically all of these languages imperative?
 
Jelly is function-level
Brachylog is logic
All of the stack based ones are concatenative
Pyth is procedural
There are a lot of diverse paradigms.
 
Ah, procedural, that's the word I was looking for
@quartata Yeah, but a lot of languages are several of them
 
3:16 AM
True Jelly is partially value level but that's only because of the register
 
So basically, you have notation, memory model and paradigm, all separately
 
Mst of these categories are from a theoretical standpoint mutually exclusive but in practice even if Scheme can have side effects we still call it functional not procedural
There is usually one dominant paradigm
 
@ETHproductions 2 suggestions for your list, goruby (see e.g. github.com/ruby/ruby/blob/trunk/golf_prelude.rb ) and Rebmu ( rebmu.hostilefork.com )
 
Oh hey Rebmu
 
I think the main thing I originally wanted to get across is that the instructions in CJam are run from left-to-right, and the instructions in Pyth form a tree, and the instructions in Jelly form chains, and the instructions in V are... um... well, that's my problem
But it might be good to include more detailed information
 
3:21 AM
hi guys
im on the graduation userscript
love it
 
Hello Oiler V, glad you like the userscript :-)
 
By the way the only difference between 2sable and 05ab1e is how it takes input. Not worth including 2sable imo
 
@ETHproductions left to right
 
It was created just so that we wouldn't need an extra byte for a switch
 
Oh, huh
That's a little over-the-top, but whatever, we're code-golfers after all
 
3:24 AM
I feel like I barely deserve credit for V. I'll I did was shorten some common idioms, Bram did much more than I did
 
@DrMcMoylex But it's quite different from CJam in how it executes those instructions, I believe
So that's what the problem is... there's not really a single attribute that best describes each language
 
The biggest difference is the memory model. 2D with a cursor. Other than that, commands are interpreted very straightforward, character by character
 
@ETHproductions another suggestion: Burlesque github.com/FMNSSun/Burlesque
 
Wait, I don't have Burlesque? Oh, right, I added Brachylog
 
But you're right, the more complex stuff uses macros (kinda like functions) instead of ifs and loops
The cool part is that every macro is stored in a register, so you can execute it, or treat it as a string at the same time
 
3:32 AM
@Downgoat pls check tf2 room
 
Unfortunately, it's time for me to go
 
4:11 AM
@ConorO'Brien
 
Hey all
 
Hi
 
anyone wanna test github.com/wonderlang ?
I need ideas and stuff
 
So I was listening to my own soundcloud page, when I realized it's been four months since I've uploaded anything.
 
saem; my youtube is basically dead right now
anyone play brawlhalla?
 
4:24 AM
Is there npm for Android?
 
I'm not sure
you may be able to do so if you somehow get a terminal
I have an ssh app on my iphone, so I can access node from there
 
yeah if you have node on your host computer, things should work out fine
 
@ETHproductions Thanks for including Pip! One correction: Charcoal uses prefix operators, not infix.
 
4:39 AM
Also: I just surpassed 10k rep! \o/
7
 
Welcome to the club! Would you like a cigar or some caviar?
 
Erm... caviar, s'il vous plait: I think the resale value is higher. :P
 
4:54 AM
for (i = 0; i < 3; i += 1) {
    dictionary[i] = i;
}
Seriously, who does this?
 
@Dennis JS
 
JS or not, what's wrong with [0, 1, 2]?
 
Maybe they are testing how good GCC optimization is
 
@Dennis Two ideas: 1) dictionary has more than three values in it, just those three happen to equal their index; or 2) somebody's used to programming in QBasic. ;)
 
@Downgoat I doubt that's the case here. There are also large if-else blocks or switch statement that can be reduced to a single case by setting the proper variable.
 
5:04 AM
Incidentally, TIL that QBasic arrays are 0-indexed by default--unlike literally everything else in the language, which is 1-indexed. DIM array(5) gives you an array of six elements, indices 0 through 5. O_O
It's like they didn't want to deal with an off-by-one case in calculating the memory addresses (makes sense particularly when you consider multi-dimensional arrays), but they also knew people would do DIM array(5): FOR i = 1 TO 5: INPUT array(i): NEXT and expect it to work. So they compromised and did both.
I guess it does follow the principle of least surprise... but it also makes every dimension of every array one slot bigger than necessary, using up extra memory. Probably not a big deal except for large multi-dimensional arrays.
 
Yeah, a while ago at work I was tasked with rewriting some old VB code in C#. The code was pretty crappy, but I remember the indexing threw me off many many times
 
5:48 AM
@Adnan
In 05AB1E
How would I remove only the first occurrence of a char in a string
 
6:03 AM
?
 
@Oliver There is no command for that
 
If there's not a builtin for it, it's probably not worth doing.
 
@Oliver You can do ©¡¬s¦®ý«, but it's a bit long
@feersum true
 
6:34 AM
I told you I could write a five-language polyglot comment-free: codegolf.stackexchange.com/a/100600/62131
 
@ais523 That is very impressive
 
6:52 AM
I was hoping to make it more polyglotty, either in terms of shared code or in terms of extra languages, but it's hard
 
26
Q: Shortest code that return SIGSEGV

AryaShortest code that returns Segmentation Fault (SIGSEGV) in any programming language.

Jeff? :O
 
Quick question for you web wizards:
var mFile = event.target.files[0];
var params = {o1: mFile, o2: "foo" };
console.log(params);
$.post("request.php",params,function(data) {
	document.getElementsByTagName("body")[0].innerHTML+= data;
 });
why does jquery throw an exception when I pass params like that?
If I do $.post(..., {key: "val"}, ...) it works
nvm got it
jQuery params can only be strings aparently
 
7:07 AM
what was the issue?
ah, right
looks like your answer came in just before my question
 
it's hard to remember how many underscores you need to spell it
 
Educationperfect isn't even trying anymore
 
7:29 AM
Can someone with xxd and a c compiler test this?
@DestructibleWatermelon pterodactyl
 
8:06 AM
@betseg it works
 
@Angs thx
 
I had to add main, of course
 
I know, single functions are allowed
Yay I got for that answer. My second gold badge!
 
8:31 AM
@Mego not sure if you're getting a notification anyway, but I've added a few challenges to your decision problem list.
 
8:46 AM
@betseg it was wednesday
 
Anonymous
@MartinEnder Yep, I get the notification
 
Anonymous
I'll work through them soon
 
9:15 AM
Chat mega challenge: self interpreter in Turtlèd
 
Is Turtléd Turing-complete?
 
I'll try. Remind me in 5 hours plz
 
Reduction To bitfuck for non I/o languages, and I have a program that halts if a given CT program halts
not sure entirely though (of the CT program)
@betseg You will definitely need to account for the $ flags and stuff: could make it fairly difficult... perhaps impossible, because moving around the grid isn't reversible
so maybe not possible
hmmm, apart from the flags and whatever, should be good
 
Why isnt it reversible ;_;
It's possible tho, it makes it not very hard, makes it too dastardly hard
 
9:23 AM
nope
 
@betseg Better hash that word out
 
not possible I do not think
hmmm
maybe is possible
but harder than first suspected even
 
@KritixiLithos i can't @mods plz eleven me
 
I just realised it would need a different method
also, one really long chain of ifs..
The thing about the current interpreter is it takes input in any unicode python likes, not just ascii
(<char>@<char>) times 1000 or something more
wait
I'm thinking it isn't possible again
because you can't move off somewhere else to store the string variable
I actual wish I never added the flag things
so self interpreter could be possible
 
you can un-add them?
 
9:36 AM
user image
4
That is one weird language.
 
9:49 AM
Anyone else love that Stack Exchange has to remind us that There are other rooms?
 
Remind how?
 
if you're only in 1 room it says that where the other rooms normally are
 
10:33 AM
@flawr You have the tag "lebesgue-integral", which seems to imply standard 1-dimensional Lebesgue integration, giving the result 0. But that's probably not what you mean, given the stuff in parentheses...
I added a comment.
 
> Ilkka Törmä
 
Yes, I use my real name on math.se.
 
@Zgarb Thanks, but no, it was just the 1D-case of a variation form (in the context of FEM) of a PDE, I'll add taht.
 
10:49 AM
Okay, I don't really know much about that, sorry.
 
Updated with some more context.
@Zgarb Thank you anways=)
@Zgarb Oh duh, I just need to derive the weak form for the 1d casen and it is easy to see what's going on...
I feel so stupid.
 
11:05 AM
@Zgarb Oh, are you from Finland too?
 
(there is no such thing as not enough snow) @EᴀsᴛᴇʀʟʏIʀᴋ
 
@Zgarb A quick googling tells me we might live the same city
 
Haha, what a coincidence=)
 
This is a university city, so it probably has a disproportionate amount of golfers
 
11:35 AM
Hey
 
So you're saying that many university students are golfers?
 
Not necessarily just students - university cities also tend to have more high-tech industry
And don't forget the academic staff / faculty of universities, although you very well could define them as students.
 
I just passed 2000! \o/
 
@Angs There are two people with an internet presence who have that name. I'm the mathematician, not the architect.
 
11:52 AM
@Zgarb That's what I assumed
 
But yeah, fellow Finn here.
 
Hauska tavata
 
Kuin myös
(Ok, let's not clutter the chat with gibberish :P)
 
Does anyone have thoughts on the input in this: meta.codegolf.stackexchange.com/a/10628/60340
I don't want to make everyone read strings into numbers so numbers should be able to be given as numbers
And it would be nice to be able to give the functions as functions, but with anonymous functions someone could encode the whole problem as the input
Or maybe something like eval(functionname) if arbitrary function names are allowed
 
12:15 PM
Maybe you could compute a fixed expression containing all the operations? For example, take array [a b c d e f] as input, compute (((a+b-c)*d)/e)^f.
 
@Zgarb nice idea
 
The expression could also be more complex and include the operations multiple times, if you want.
 
Yeah, I thought maybe at least two of each to encourage writing functions for each
 
12:34 PM
Regarding the showcase challenge:
Can't we tag it with and slightly extend the meaning of that tag (which in its current state is pretty vague imo)?
Make be about making better answers for challenges of the site, which includes: learning how to shorten code in a certain language (which are most tips questions right now), learning how to write faster code for a certain language (for fastest-code challenges), and learning how to use a language in the first place (which would be the showcase).
 
I wanted to do a tips on writing challenges, and it got shot down because tips was already too broad (and we didn't want to perpetuate the problem)
so, I'm not sure how much success you'll have with that
 
A tips on writing challenges would actually be very valuable…
 
Weird how just about everyone got excited about EM drives at the same time
 
I came across an interesting paper that gave me an idea for a code challenge, but it's probably way too big/annoying... though I'd mention it here
The paper was by one Paul Tarau, about bijective Gödel numbering
So the challenge is to write a program that maps valid programs in your language of choice to natural number, and back, bijectively
 
bijective what?
 
12:46 PM
and the score would be the number you get if you give it your own program as input
 
what's to stop you from minusing your program's score from the number?
 
Nathan Merril nothing, if you can still make it bijective
 
if(myProgram) return 0 else {bijective from 1 to +inf}
 
I mean, that seems rather easy. If you already have a bijective numbering, then all you need to do perform a constant subtraction, and you'll maintain the bijective numbering
 
the difficult part is the second but once you get that everyone has a score of 0
 
12:49 PM
Fatalize: I guess that would work, but you'd need a +1 on the else
yeah, you could cheat that way. Not sure a contest could be formulated to avoid it
But anyway, mapping programs bijectively to integers is cool. I think mapping anything bijectively to integers is cool :)
 
I think that mapping a pair of integers to an integer is really cool
or even a list of integers
the fact that its even possible blows my mind
 
I mean, Mathematica programs could be mapped bijectively rather easily
Considering that all of them are already just expression trees
I'd wager that a similar thing can be done with Python's ast
 
should be "easy" in all lisps too,
but it's more subtle than it seems :)
It's easy if you allow for invalid programs
 
Yeah, the hard part is getting bijection with trees
But I can think of certain methods that can work
 
I wrote a little program a while ago to map from matrices to java bignums and vice versa
 
1:03 PM
Also, define "valid". Does it just have to be syntactically valid or not have any warnings during execution?
 
the first attempt required factoring, so it was too slow. Then I figured out many other good ways to do it
syntactically valid is more than challenging enough
 
E.g., would print(1/0) be a valid program
 
Okay, give me an hour or two
 
otherwise I'm pretty sure enumerating them is undecidable
or can you enumerate the terminating programs? I'm not steady on this sort of thing
 
1:06 PM
enumerating the terminating programs is just as hard as enumerating non-terminating programs: you have to determine whether or not its terminating
if you've already got a list, then sure
 
1
Q: Convert CSV to Table

Basically Alan TuringThe Challenge Given a CSV input, output a proper unicode table using box characters. Formatting The table will be formatted using the following rules: Column width will be equal to the longest value of that column All table data will be left justified Each table will assume the first csv row...

 
+1 for not saying "ASCII".
 
1:22 PM
this is an argument I end up being involved in a ton, due to working on NetHack
sometimes people use it to mean the actual character set ASCII7, sometimes people use it as a general term referring to text-based output
sometimes people use it to mean a display in which the cells of the display are or are inspired by letters, digits, and punctuation marks
sometimes people require it to be actual literal text on a grid that's monospaced, copy-and-pastable, and all the same font
 
ASCII is the 7-bit codepage. Any other thing is extended ASCII, Unicode or not ASCII (EBCDIC, etc...)
 
I've decided it's best just to make sure you're clear on what people mean and that you don't start a flamewar yourself (although I agree, ASCII is totally a 7-bit character set)
 
Did you know 0x33 is 3?
 
yes
 
γes
 
1:33 PM
in ASCII, 0x3*x*, where x is a decimal digit, represents x
I don't think I ever memorized the entirety of ASCII
 
#justcodegolfthings
 
but you end up memorizing large sections of it over time for one reason or another, and some of them stick in the mind
 
65 and 97
 
Argh, Mathematica symbol names are weird
 
I know the codepoint of nul (¯\_(ツ)_/¯), space, newline, carriage return, esc, del, the digit section and the uppercase alphabet section
 
1:36 PM
I know newline, escape and EOT
 
@betseg or in hex, 0x41 and 0x61
 
The complete list of allowed characters after the first character:
U+0001 - U+0006, U+0008, U+000E - U+001F, U+0024, U+0041 - U+005A, U+0060 - U+007A, U+0080 - U+009F, U+00A1 - U+00AB, U+00AD - U+00B0, U+00B2 - U+00B6, U+00B8 - U+00D6, U+00D8 - U+00F6, U+00F8 - U+2004, U+2006 - U+2008, U+200B - U+2017, U+201A - U+201B, U+201E - U+2027, U+202A - U+2042, U+2044 - U+205E, U+2061, U+2063 - U+218F, U+219A - U+21A3, U+21A8 - U+21BB, U+21C7 - U+21CA, U+21CD - U+21CF, U+21D6 - U+21E3, U+21E6 - U+21F4, U+21F6 - U+21FF, U+2201, U+2205, U+220A, U+220E, U+2214, U+2217, U+2219, U+221B - U+221C, U+221E - U+2222, U+2224, U+2229 - U+222A, U+222C - U+222D, U+2230 - U+2231,
 
> This message is too long.
Indeed
 
Yeah, codeblock messages need to be multi-line
 
does that correspond to a simple expression when written in terms of Unicode properties?
 
1:38 PM
idk, Mathematica doesn't support them
 
my brain's already trying to golf it :-D
 
How I got the list:
 
there's this fairly obscure feature in recent versions of Perl that lets you do arithmetic on character classes inside regexes
you're not supposed to use it in production yet because they haven't quite finalized how it works
 
Not sure, but I think 0x21 (one after space) is !
 
but it seems like it might produce a nice short solution to that problem
betseg: yes, it is
 
1:39 PM
Quiet[Select[CharacterRange[1, 65535],
  Check[Symbol[# <> "test"]; True, False] &]]
 
@betseg Yes
@ais523 Not IRC here. To ping you need to use @user
 
Basically, it tries to make it into a symbol, and checks if it succeeds
 
after space, ASCII almost follows the shifted digits on a US keyboard, but not quite
@TuxCopter: it wasn't intended as a ping, it was intended as an in-reply-to marker
when there are a ton of conversations going on I tend to drop usernames in to help keep them apart
 
(not memorized these, found on Google) why are those in this order? *+,-./
 
but I can do proper pings if you like
 
1:40 PM
@ais523 SE chat support replies (In the arrow at the left of a message, there is the button reply to this message)
 
4 mins ago, by ais523
@betseg or in hex, 0x41 and 0x61
 
yep, I did one of those earlier, it's just so fiddly
 
Oh permalinks don't show the arrows
 
Hello :)
 
Hey
 
1:46 PM
2321
 
The full list of characters that can be used in a symbol, but not as its first character:
, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9
(The first character is a U+000B)
(don't ask why)
 
I don't even think that's a very widely used control code
which one is it?
 
Vertical tab
 
Tab (according to google) (ninja'd)
 
normal tab is U+0009
 
1:49 PM
So
 
Vertical tab in a variable name... Wow
 
A
<Insert VT here>
 B
is a valid symbol name? :P
 
I seem to remember that Perl had a fairly banal argument about whether vertical tab counted as whitespace
 
Renders like a U+1D160 (eighth note) in Mathematica's font
 
they ended up making a breaking change as a result of it, with tons of warnings because it was a breaking change, but in the belief that people were unlikely to be regexing the things in the first place (and if they were they were probably listing them explicitly)
 
1:51 PM
Yeah, they act pretty weird
 
Math𝅘𝅥𝅮Music𝅘𝅥𝅮𝅘𝅥𝅮𝅘𝅥𝅮
 
lots of squares there
 
You have bad font
 
how does Mathematica interpret a literal U+1D160?
 
@TuxCopter how do I get good font
 
1:52 PM
DejaVu
 
also, I went and installed a font pack a while back which was specifically intended to try to fill in Unicode characters that weren't otherwise easily available
I forget what it was caled
*called
DejaVu has the vast majority of Unicode, though, especially the older characters
 
I use DejaVu and it work well (The only problem is no advocad emoji Q_______Q)
 
ew, tarballs
 
@ΛεγίωνΜάμμαλϠΟΗʹ \\:
ew that's a bad syntax for literals
 
1:54 PM
Escaped, \:01ef is a four-hex-digit character literal
 
even worse, that's writing code in UTF-16
 
ikr
Mathematica doesn't support codepoints above U+FFFF
 
like, I can just about understand it as a format for internal use for backward compatibility reasons
but using it as a user input method?
 
@TuxCopter help how do I change font in a mobile browser
 
@betseg carefully
 
1:56 PM
Note that when the user just types in the hex literal, the interpreter auto-expands it, so I just escaped it and wrapped it in a ToExpression (i.e., eval) for presentation
 
@TuxCopter Hey that's my joke
 
ಠ_ಠ
 
do mobile browsers accept javascript: URLs? you could probably write a small JavaScript oneliner to do it, although it'd only last until the next page load
 
Yes they (at least mine) do
 
1:59 PM
%C9%A2oogle.com I know this
 
Yep, ɢoogle.com
I tried to go on this URL and got this URL...
 
@TuxCopter So basically http://<blah>.ilovevitaly.com/<blah>?
 
Yes
 
there's a # just after the .com/, so all that stuff is never seen by the server
it's visible from JavaScript but not from HTTP
thus, on a site that isn't designed to give a special meaning to anchors, you can put what you like there and the URL still works
 
0
Q: Handling errors

AngsWhat's great about golfing is not having to deal with errors. Except this time you won't get off so lightly! I need to do some arithmetic with certain limitations, and I wan't to know what goes wrong if anything. Challenge Given a list of signed integer values [n1..n11], give the following resu...

 
2:08 PM
@NewMainPosts 7 minutes, you've outdone you
 
that's new? I'm pretty sure I saw it earlier
was it on the sandbox?
 
Wait a second...
Money is the root of all evil.
Therefore, money squared is evil.
However, time is money.
Therefore, time squared is evil.
That means that Times Square is evil, so you'd better stay away from Manhattan!
 
Hopefully I'm in Corsica
 
@ais523 yes
 
Also, breaking news: According to a dream I had last night, atoms are on the scale of 10^-5 meters!
(don't ask)
 
2:11 PM
wat
 
@ΛεγίωνΜάμμαλϠΟΗʹ ... wtf
 
According to a dream I had, we are all dead
 
@TuxCopter idk, it also had something to do with my wristwatch's water resistance rating
I'm not even making this up
 
I had the best dream about two months ago. I was in an apron, waiting for airplane. Than the airplane that I will get in crashed into the apron.
 
2:20 PM
All your dreams sound great. I just had a nightmare where my homemade kebab meat was raw and had huge chunks of cooked carrots in it so I couldn't keep it together to slice it.
 
I am going to be re-invented soon :)
 
@Angs Sounds like a shish kebab
Source: ate one yesterday (not raw btw)
 
Yesterday, I had a dream where I met a fellow code-golfer in real life, and I would never have guessed him to be a part of PPCG, and it was all happy
 
@KritixiLithos Hmm... that would be what it would looked like if you encountered me, minus the "all happy" part
 
The guy had 40k+ rep
 
2:26 PM
@betseg It was about a döner kebab / gyro kinda thing that I had cooked the day before. It didn't have any carrots at all, which was nice.
 
@KritixiLithos I have 40k+ in my dreams
 
I do have some Turkish friends...
 
hi wise people
which of the answers at codegolf.stackexchange.com/questions/100192/… should I accept?
my favourite or one that is best by some objective criterion?
 
Whichever has the better score.
 
@Dennis then why does a human need to be involved?
 
TIL that a governor in Turkey discharged a doctor in 2014 because he said rice that watered from river Ergene caused cancer and that's why people from Edirne have higher chance of getting cancer.
 
Or none at all
 
I don't mean that as badly as it seems :) I just have never understood if on ppcg you can just accept the one you like the most
 
@Lembik Because we're a competition site on a Q&A network.
13
A: What's the "accepted answer"?

Calvin's HobbiesThe accepted answer is the answer that wins the challenge As defined by the objective scoring criterion the challenge writer gave the challenge when it was created. That's it really. If there are unforeseen circumstances that cause ambiguity on what the winning answer is (say a challenge is ti...

 
for example, in this question xnor basically solved it and everyone else copied him/her
 
2:36 PM
You can always accept none of them, but if you accept one, it should be the winner by the scoring criterion.
 
2:56 PM
accepted the one in Actually :)
because I always do what I am told
I just feel bad for xnor
 
0*Null=0 lol
 

« first day (2123 days earlier)      last day (2723 days later) »