« first day (2210 days earlier)      last day (2640 days later) » 

12:02 AM
Oh gosh I've started feeling Phantom Vibrations.
I need to use SE at work less.
 
@wat solution: don't use electron
 
wat
@Downgoat Kek.
 
0
A: Sandbox for Proposed Challenges

ETHproductionsEfficient counting code-golf number When I was a kid, and wanted to count the dollar bills in my life savings, I would count out loud: one, two, three, four, five, six, seven, eight, nine, ten; eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty; ...

 
@wat don't say its name
i've been told that if you say its name three times it replaces every directory in node_modules with itself
 
wat
@quartata Electron Electron Electron?
 
12:15 AM
...
 
ಠ_ಠ
electron is pretty awesome
 
Well, I wouldn't know. I only program via punch card.
 
@Maltysen why a full chrome instance
 
@quartata why not :P
 
@Downgoat ugh
 
@Downgoat Just to make sure you're asking this as a joke right
 
wat
I have now written an app!
Yay!
 
@wat oh that reminds me I forgot to finish the one I was writing in shoes
It almost works just missing editing I think
 
wat
@quartata shoes?
 
wat
12:32 AM
I really really want to work on mine, it's just that I cannot figure out the API
Otherwise I would be working on it right now
 
@wat download is 0.2MB/s, is server like strained?
 
wat
@Downgoat no...
 
O_o
 
wat
was max speed for me when I tested
Where are you located? Like, rough location
 
wat
12:34 AM
Wat.
 
@wat california, i get 120mbps download usually
 
wat
@Downgoat Huh. The server is in San Francisco.
 
120?!
I get 14.
 
@wat i think you should just use nativifier
 
wat
goddammit i placed it in the wrong place. all the other OS versions should be fine.
 
12:35 AM
Australian Internet is 100/10, absolutely perfect.
Smaller numbers are better right? This is code golf.
 
wat
100/10 is good
 
@wat did you fix?
 
wat
@Downgoat yeha repackaging and uploading
 
Hmm, looks like the scroll bar is broken...
That works at least though
 
wat
@quartata Does my app work?
 
12:39 AM
Oh I wasn't trying that. I was chatting from what remains of what I was writing a year ago
It's only 56 lines so I wasn't expecting it to work
 
> > Sees "Add a language to a Polyglot" at the top of active
> > *WHAT DID THEY DO NOW*
 
But I can send and read messages, the scrolling is just busted
 
wat
oic
 
Bog standard BF
Since SMBF was already a thing
 
can someone tell me why this latex doesn't render:
$$\left\langle N,\Sigma, R, S \right\rangle$$
is there some error?
 
$$\left\langle N,\Sigma, R, S \right\rangle$$
@Downgoat Works for me
 
Renders fine on ChatJax
 
Yeah, me too. What LaTeX renderer are you using?
Is this in a document?
 
MathJax
 
 
12:43 AM
complex latex like:
$$
\begin{array}{rl}
A \rightarrow& A a \mid A \\\\
a \rightarrow& \Sigma_n
\end{array}
$$
works fine too which is weird
 
@quartata I think it might be possible to get busy-beaver levels of scores on the challenge as-is; that might make it slightly terser to do so
 
@ais523 Oh yeah, absolutely. The question is will they be crackable
 
at least, if the best score on that doesn't beat Graham's Number, I'll be disappointed
 
assuming you allow the cracking regex to be posted in an abbreviated form because it's too long for SE
 
12:45 AM
Yeah, we've done that in the past.
 
$$
\begin{array}{rl}
A \rightarrow& A a \mid A \\\\
a \rightarrow& \Sigma_n
\end{array}
$$
 
Hmm, should I make a leaderboard script for this?
 
Yeah, renders fine.
 
It's pretty easy to calculate your scores for yourself but
 
I have an idea for a challenge: Build a regex engine (no builtins)
 
12:46 AM
@Pavel we've already done that
spoiler alert haskell won
 
Ooh, where?
 
not big surprise
 
I love how ChatJax looks.
It's a beautiful contrast to the otherwise plain text.
 
Crap, I can't find it. One second.
 
Thanks $\LaTeX$
 
wat
12:47 AM
@ATaco link?
 
@MartinEnder Oh crap, I just noticed this:
72
Q: Cops and Robbers: Regex Golf

Martin Ender Note: This challenge is now closed. Any future cops' submission will not be considered for the accepted answer. This is to ensure that no one can post a very simple regex in the future that only remains uncracked because no one is interested in the challenge any more. The Cops' Challenge Yo...

 
(Why isn't my Markdown Rendering..?)
 
12:49 AM
Completely different but same name
I'll need a better title.
 
wat
@Downgoat I have updated. Try redownloading.
 
@ATaco markdown does not render for multiline messages
 
@DJMcMayhem Whatever happened to github.com/DJMcMayhem/ThePeoplesPython ? :p
 
wat
@DJMcMayhem In your peoples pythin thing, cna you make "wat" mean "import"?
 
Hmm. Not sure I like this new title but it's the best I can think of for now
 
wat
12:52 AM
I would rather have it be if, but if is taken
 
@HelkaHomba it's best not we talk about those things
them be bad things
 
lang-python:wat string
wat gzinflate
 
@wat And make it so imports have to be ended with a question mark.
wat random?
 
@quartata I think you could get away with renaming the old one; it's more of a reverse regex golf
 
wat
@Pavel yeeeah
 
12:53 AM
given that the cops are trying to set an obfuscated solution to a question of their choice and the robbers are tasked with determining what the question was
actually I'll do that now
or, hmm, the problem is that editing it will bump it to the front page
despite the fact that it doesn't allow new submissions
 
@quartata Yeah, can't find it anywhere. Are you sure you're not thinking of a similar challenge?
 
I don't feel right submitting answers in RProgN for CNR.
 
@Pavel No, I'm almost positive
 
I submit answers in 7 to CNRs all the time
it's a good language for the purpose because hardly anyone can read it
 
RProgN does a lot of things that aren't obvious by reading it.
 
12:54 AM
@ais523 I don't really like regex golf CnR as a title for mine either
 
actually, I nearly always have versions of my 7 programs with plenty of whitespace in
 
I renamed it "Regex Golf Generators" but I don't like that either
 
to make them easier to understand when I come back to them
 
~1aL1{}: prints all numbers between 1 and 26. Now tell me which part means what.
 
Hmmm
I wonder if there's one to write a CFG engine.
 
wat
12:55 AM
@ais523 You only have 7 programs?
I have 8
 
and even with all the whitespace (it doesn't have comments that aren't made of digits), I have problems reading my own 7 code
like, you could write a comment with something like 17…613, but given that the comment could only contain digits (and would need have balanced 76 pairs), I'm not sure there's any point
 
wat
@ais523 make base 8 encoding with the other remaining digits, encode ASCII chars ino that, make a translator program
 
Or, alternatively, make non-octal characters no-ops.
 
@ATaco well it's a stack-based language, so I'm guessing the print is either implicit or comes last, the range endpoints come near the start, and the range operator is in the middle
 
Close.
 
12:58 AM
@ATaco non-octal characters are used as clues the program isn't ASCII; that way I don't need arguments to specify the program encoding
 
Darn.
 
like, if the program is just ASCII octal digits and whitespace, it's treated as ASCII, otherwise as binary-compressed octal
I guess I could write my comments as Whitespace programs?
 
Just hit the third hour of listening to yourepeat.com/watch?v=CxPFyu5AGng
 
~ is a ZSS, which is an RProgN specific thing, 1aL1 is three different numbers, 1, 26 and 1, {}: is a for loop that does nothing, thus, leaving its numbers on the reg stack, which is implicitly printed.
 
I'm guessing this isn't designed primarily as a golfing language?
makes sense
because those things normally have range operators
 
1:03 AM
It has a range operator.
It just doesn't have a small byte count.
Also, correct.
to and both 2 bytes, get all the things between two things inclusively.
I've got an update in the works to add a 1 byte solution to a lot of less golfy functions, but I don't have many 1 bytes left.
 
there's nothing wrong with not being a golfing language :-)
very few of my languages are golfy
in fact, many don't even have input, or an implementation
 
I've occasionally beaten other languages, or tied with some.
 
(and often, output is an optional extension that exists only to make testing easier)
 
I'm not sure, RProgN is just my first born, and I love it.
 
oh, my first esolang was so bad I didn't even post it publicly
(although possibly still better than half of esolangs.org, there's some real dreck there in addition to some real gems)
 
1:08 AM
Well, RProgN isn't my first Esolang.
But before that it's just Brainfuck and Basic Derivatives.
 
CS teachers teach you not to regex so that you don't rely on it
 
@Downgoat that should really mention Perl 6 regexes; they have almost exactly the feature set you mention/use in your blog post
much to the disappointment of people who liked doing other things with regexes in Perl 5
 
@ATaco no they teach us not to use regex because they don't know regex
@ais523 I mentioned it is PCRE but is it really the only flavor with group recursion?
 
@Downgoat group recursion is based into the fundamental syntax of Perl 6 regex, as opposed to PCRE where it's very much an add-on
btw, I've never had to teach people not to use regex when teaching CS
 
1:18 AM
I don't get to use that much Regex.
 
mostly because I've been teaching languages (Java, C) where regex is rarely used
 
Lua Patterns are the only tool I have.
 
0
Q: Shortest Code to Legitimately Slack Off

Challenger5I'm a developer, and I don't feel like doing my work. I know from XKCD that the best excuse for slacking off is that your code's compiling. Because of this, I think I need some code that will compile forever! And because I'm lazy and don't want to have to type much, this has to be done with the s...

 
@ais523 looks like PCRE, and Ruby, and probably more also support group recursion
May not be named but those can be emulated
(?<DEFINE>...>) (?<1>)
 
What do you guys think about a code golf challenge for outputting the first 10 prime numbers in as few bytes as possible [No Literals allowed]. It
It's interesting to me because the prime numbers have no true generation pattern. But there are many patterns that can be found just for the first 3 or 7 or 10 or X.
 
1:25 AM
No literals?
 
Meaning hard coding in values such as "2", "3", char, unicode strings, etc. Depends on language
 
people would just use a prime generation builtin
the hardest part is specifying the 10 without literals involved
 
Yeah that was my one qualm with it is languages with built in generators
 
I don't like these types of challenge because literals feel very I'll defined
 
Literal is a term from C (my native language) I'm not sure how well it translates to other languages
 
1:27 AM
I think that should be able to be checked by a computer
2
 
RProgN doesn't actually have Number literals.
 
I agree wheat
 
wat
@AlbertRenshaw are symbols allowed?
 
It just happens that all numbers are functions which push themselves.
 
Not sure @wat I haven't ever used symbols can't say I know what they are
 
1:31 AM
@wat I don't think that would help since they either are literals or need a string depending on the language
 
wat
noice
 
@wat I can't even begin to guess who that was.
 
@ATaco I'd treat that as a literal; the main defining feature of a literal to me is that it comes as part of a near-infinite family of commands that differs only in the numbers or strings involved
 
I concur
 
Generally, I see literals as being treated as such by the interpreter.
For example, Strings in RProgN do not follow the function definition, and instead are individually processed.
 
1:33 AM
well, something like 123 parses as a single token, doesn't it?
so it's already somewhat special-cased
 
Nope.
In a ZSS, 123 pushes 1, 2, and 3 separately.
 
Sometimes literals might be necessary too. I think in some C IDEs you have to define all variables. So I can't just create an integer and start working with it from 0 or 1, it will instead just grab some random variable from memory. So restricted-source may require exceptions for non-null variables that must be assigned.
 
You can set 4 to be a different function, all that fun.
4 and a work, on the surface, basically the same.
 
I treat literals as literals even when you can assign to them (e.g. Forte, INTERCAL, FORTRAN)
in GolfScript you can assign to basically anything you like, including built-in functions you aren't using
 
is the function a, which pushes the alphabet literal? What about X, which pushes "X"?
 
1:36 AM
@AlbertRenshaw you can just use argc or errno to get a value to start off with
@ATaco I'd say yes and yes, although those don't comply with my previous definition
 
Generally, my argument is that a literal is too difficult to define among esoteric languages, and such, probably too difficult to use in a challenge.
 
in most of the stack-based languages it's easy
with things like BF, though, it's much harder
is [-]+++++ a literal?
 
wat
@AlbertRenshaw Ruby: require 'prime';l,i=Array.new,$$-$$;loop{l<<i if i.prime?;break if l.length > :aaaaaaaaa.length;i+=$$/$$};puts l
 
MmMmmMmMm, That formatting.
 
oh, that's a fairly simple way to find an integer earlier on, but you can't use +=1
 
wat
1:40 AM
I hope that is allowed, that is what I meant by symbols
 
does Ruby have a ++?
 
wat
oh crap
@ais523 no
 
 
Perl has a ton of preinitialized variables
 
wat
@ATaco What.
 
1:40 AM
there's one that's preinitialized to 60, which is very rarely useful in golfing, but is fun to know about
 
Chat Formatting~
Wait, but are Pre-initilized variables Literals by some obscure definition?
Because that's all X and a are.
 
wat
How did you get code blocks to thing?
 
@wat I think you mean wat? Also great Ruby hahaha
 
@ATaco hmm, an interesting point; in this case assigning to them has side effects, which makes them less obviously literals
 
1:42 AM
I'm not sure what the stars do in this chat, I've just been using them like a like button?
 
@AlbertRenshaw they temporarily pin the message to the right hand side of the screen, the more stars the longer it stays there
but mostly it's a sort-of combination of "like" and "approve" and "this needs more visibility"
 
If you see a message, and think "More people should see this message", it's star worthy.
 
Gotcha thanks guys! Alright I'm heading out to dinner, peace
 
wat
@ATaco can you set your scripts' filenames to *.user.js
 
Sure thing.
Done.
 
wat
1:50 AM
thx'
@ATaco Huh.
 
Huh?
 
wat
Your user scripts are not activating.
 
https:// at the tail causes it to not match.
And I've no idea how to make that optional.
 
@DJMcMayhem: a compile time infinite loop is not a duplicate of a runtime infinite loop, it's very rare that code for one will work for the other (my Perl program is possibly a special case but it wouldn't work in other languages, which don't have universal compile time hoists)
 
Well, I do.
I'll fix that.
 
1:55 AM
@ais523 fair enough
 
it's interesting how whether challenges are duplicate depends on the existence of universal constructors
e.g. "do a quine but X" is normally uninteresting due to the existence of universal quine constructors, unless is involved somehow
but "do X but at compile time" is typically completely different
 
"do X but at compile time" = "do X, but use a preprocessor instead of a real language"
 
wat
do x but at compile time?
 
Primality testing at compile time would involve creating a family of classes/types/sth such that type_x::out returns a truthy value iff x is a prime, in constant time.
 
yep
or more likely type<x>::out
as recursive templates are the most common trick for that sort of thing
 
2:02 AM
Alright, Now my userscripts should correctly match all variations of the chat.
 
(regardless of language)
 
It's challenging in C++, impossible in Java (I think), and only a minor complication in Lisp.
Ruby doesn't have a compilation phase. You can create new classes on the fly, but creating a class with a class method that returns a specific value isn't too hard either.
 
wat
method_missing
 
The only real trick is keeping the scope throughout the creation process.
Not even that if said class can be a Proc :-)
def const_cls value
  Class.new do
    define_method :get_value do
      value
    end
  end
end

const_cls(1).new.get_value #=> 1
defs create their own scope for local variables, but blocks don't.
 
wat
2:27 AM
this is cool
 
2:40 AM
@NathanMerrill re 4: do you mean omeone who posted the best submission but a lot of poor ones as well vs someone who posted a single submission
 
wat
ples halp
 
With?
 
wat
programming
 
Gonna need you to narrow it down.
 
with, or without cue?
 
wat
2:48 AM
I need help with Ruby on Rails, specifically creating a check in page for items
 
Good luck.
 
wat
I have an items model and stuff, I have the page design done.
Each item has an "Item ID" which is composed of an "Item ID" and an "Item Instance ID"
The "Item ID" is basically the type of item
the ItemInstance ID is the specific item of that type.
I would like the users to be able to put in the item ID (such as RPI 1) as "RPI1" or "RPI 1" and parse it correctly in both cases.
More info> Item IDs are always 3 characters.
Also, how do I create that?
 
I can't answer this without switching to sarcasm.
 
wat
what do you mean?
 
I mostly write in Lua, Thus, Ruby is seen as inefficent and evil.
 
2:56 AM
It's neither, unless you mean in terms of CPU cycles.
 
@NathanMerrill
I'm going to do 1 once I figure out how to incorporate it into the score. 2 I won't do -- special characters are just a nuisance to both sides and I don't see the point. 3 is so I can have an objective way of stopping bowled self cracks -- otherwise it doesn't really fall under our definition of not competing since it maximizes the cop score and provides a robber win too. 5 is so strong but doable submissions don't end up with 0s while lazy completely uncrackable submissions don't get points
Ugh that's why I couldn't send it. Too long.
 
wat
@ATaco I dislike Lua, its syntax is cryptic and weird. Ruby is mmuch more understandable and concise
 
It took me ten minutes to figure that out since on mobile it doesn't say anything. It just does nothing when you press the button
 
Lua's syntax is remarkably simple, it's just different.
 
@wat Ruby and Lua are almost identical in syntax
 
wat
2:59 AM
@quartata ...no
how do you pass blocks in Lua?
 
My usual response for Lua is that it's disgustingly barebones
 

« first day (2210 days earlier)      last day (2640 days later) »