« first day (2064 days earlier)      last day (2770 days later) » 

12:00 AM
?
 
Oh, ha.
I personally rather like this.
@El'endiaStarman Thanks
 
So what makes Haskell a countable noun, and Erlang uncountable?
 
@Downgoat Use Google Custom Search or Lucene
@feersum That's.... a good question....
@wizzwizz4 What's your question?
 
12:19 AM
@1000000000 I'm pretty sure that would just be print (" (650) 543–4800") :P
 
"Oh shoot my challenge is trivial. Well, time to ban numbers from source code I guess."
13
 
@DJMcMayhem That would be one way, another would be to ask for a program to get Facebook's contact number from whatever page its on (if they have one). There are many ways of doing it
 
@quartata erlang is your favorite? it looks kinda ugly to me
 
Just like there are many ways of solving a code golf challenge
 
My nephew just challenged me to a Pokemon battle. Apparently he thought he'd finally beat me if he brought three Arceus with him :/
 
12:29 AM
@Geobits which incarnation?
 
Normal, Ground, and Dragon types.
 
@ConorO'Brien Then you haven't seen good Erlang code
 
@quartata show me good erlang code?
@Geobits I meant, the TCG, the DS,...
 
Oh, 3DS. ORAS
 
12:31 AM
ah, cool
 
It was a random challenge, too. Just a text out of nowhere, saying he had it this time.
It's been months since we fought last lol.
 
you can't be beaten, of course
 
-module(rpn).
-export([rpn/1]).

rpn(Code) when is_list(Code) ->
  rpn(string:tokens(Code, " "), []).

rpn([], Stack) ->
  io:format("Result: ~B~n", [hd(Stack)]);

rpn([Token|Tail], Stack) ->
  rpn(Tail, token_execute(Token, Stack)).

token_execute("+", [Arg1,Arg2|Tail]) ->
  [Arg1 + Arg2|Tail];

token_execute("-", [Arg1,Arg2|Tail]) ->
  [Arg2 - Arg1|Tail];

token_execute("*", [Arg1,Arg2|Tail]) ->
  [Arg1 * Arg2|Tail];

token_execute("/", [Arg1,Arg2|Tail]) ->
  [Arg2 div Arg1|Tail];

token_execute("_", [Arg1|Tail]) ->
-module(dict_server).
-export([main/0]).

main() ->
  {ok, File} = file:open("/usr/share/dict/words", [read, raw]),
  {ok, Socket} = gen_tcp:listen(13372, [binary, {active, false}]),
  spawn(fun() -> start_server(Socket, File, 99171) end).

start_server(Socket, File, Length) ->
  {ok, Connection} = gen_tcp:accept(Socket),
  spawn(fun() -> start_server(Socket, File, Length) end),
  handle_connection(Connection, File, Length).

handle_connection(Connection, File, Length) ->
  io:format("Connection received~n"),
 
Oh sure I can. Just not by him ;)
 
Those were bigger than I had anticipated. Sorry.
 
12:35 AM
shit i started a PPCG influx on five on github
 
@quartata how is this used?
 
@ConorO'Brien Like a normal module
rpn:rpn(string)
 
I mean called
oh
>_<
derp
 
The second example is missing another module
 
@Geobits are you the resident sensei for both pokemon and magic:tg then?
 
12:37 AM
I do what I can. Kids gotta learn somewhere.
 
@quartata that's pretty neat and succinct.
 
Gives a good flavor for the power of pattern matching I thought
 
@Geobits what is your opinion on Yu-gi-oh?
 
hey an O answer
that's the first I've seen in a while
 
12:38 AM
The second is a rudimentary demo of Erlang's concurrency and internet things
 
@quartata is "_" a blank?
 
@ConorO'Brien Never played it and the show seemed dumb. Dunno other than that.
 
@ConorO'Brien "_" is low minus
 
_ is blank
 
12:39 AM
I see
@Geobits show is rather dumb (though 5Ds was pretty decent), but the card game is great
 
The only card game I really got into was MTG. Even pokemon I've only played a few times with my kid.
 
so MTG is good?
 
@uoɥʇʎPʎzɐɹC your PR for five broke stuff, should probably nuke/fix
 
I think so, but I might be biased.
 
bah, we're all biased. some biases are preferred more than others :P
 
12:42 AM
iirc rainbolt also plays
 
> ConcurrentSkipListSet
Ah Java
You couldn't have just called it SkipList could you
 
Hell no. Then you wouldn't need to be amped up on coffee to find it.
 
psh, who needs coffee to program
 
If you're using Java and not drinking coffee, you're not going to get past the Verbosity Finger Fatigue.
 
if you drink java while using Java does that make you a super Java programmer?
 
12:55 AM
Nah, just normal :P
 
I know, it's bitterly disappointing :/
 
If I drink a JavaScript, I choke a little on the pieces of paper
 
@ConorO'Brien What if you drink a Python
 
You shouldn't be writing your javascript out on paper. Even a notepad user would say that's not a proper environment.
 
12:56 AM
@Geobits But... I used to use Paper
 
@ASCII-only well, it is a certain type of danger noodle.
 
@ConorO'Brien So does that mean you'll die or does that mean you'll choke
 
@ASCII-only the second, then the first
 
What about eating Ruby/Cheddar/Jelly/Bubblegum/Crayon
 
Eat the second and third, chew the fourth, don't bother with the others.
 
1:07 AM
@ASCII-only (ouch gemstones hurt)/(yum!)/(eugh)/(nom)/(dies)
 
D: crayons are that poisonous?
 
@ConorO'Brien crayons are non toxic
 
which sissy crayons are you using
 
Um, I hope you don't work for crayola
Tomorrows headline, "Kindergaten class dies for eating "manly" crayons"
6
Help Wanted
Esolang creators please add your language to this IDE
 
dude
I have many languages
how does one do this?
 
1:12 AM
@ConorO'Brien if there is one or two, which you really like, and you have a minute to add it in, it shouldnt be to o hard. What you do is provide one jar called LANGUAGE.jar which accepts a file name as a CLA
 
I see
 
And you can also optionally procide some docs, and some syntax highlighting
 
@RohanJhunjhunwala not many people write langs in java
mostly python
 
the langs have to be in java?
 
@ConorO'Brien you can daisy chain the interpreter I guess. I need a better way to support more languages of interpreters
If you write a jar that just does exec("inerpreteter "+args[0]); thats enough
 
1:16 AM
@anyone halp it's better to do a lookup table at runtime instead of precomputing and adding to startup speed right?
 
yeah I think so
 
@ASCII-only depends...
 
@RohanJhunjhunwala depends on?
 
@ASCII-only depends on the size of the lookup table, and the computational cost of generating the look up table on the fly
 
It's base conversion
 
1:22 AM
@ASCII-only that's probably cheap enough that it does not matter
 
in PPCG Minecraft Server, 1 min ago, by Helka Homba
Admin note: I am no longer interested in managing the PPCG MC server. I'll continue to pay the hosting fee until the money runs out but after that it's ok with me if the server quietly dies or its ownership gets transferred. Ping me if you'd be interested in running it.
 
@ConorO'Brien are you interested in adding any languages to the IDE?
 
@RohanJhunjhunwala somewhat, but I'm busy with hw rn
@HelkaHomba why is that?
 
@ConorO'Brien oh, ok. If you do get a minute, I'll accept any PR, or give collaboration rights.
 
1:26 AM
@ConorO'Brien I just never play on it anymore and managing it is a bit of a hassle. It makes more sense for someone who really uses it to run it.
 
ah, I see
 
23 hours ago, by Downgoat
@Geobits Basically I need to make a thing where the user searches for an item and the most relevant item shows up. I have no idea how it should work. A) How should I store the description of the item, a paragraph? an array of keywords? something else? B) how do I effectively do a search? How do I weigh different words based on semantic meaning/importance?
I am not trying to search the web
 
Neither of those were?
Google Custom Search is limited to one site
Lucene is a search framework for general text
 
I need to do this in JS. Lucene is Java?
And still Google Custom search still searched the web or webpages which is not what i want
thanks for suggestion though
 
Your backend is Node JS I take it?
I'll see if there's something for it
 
1:39 AM
@quartata there is no backend >_>
(at least for searching)
 
Well I'm not sure how that will work unless you plan on sending every page to the browser to search
So... backend time?
Would be fairly easy to slam a servlet together and use Glassfish
(or Tomcat if you're a scrub)
 
>_> I think you are overestimating what I'm seraching
I basically have a JSON with about ~50 items, each with a description around a paragraph long
I need to find the item with the most relevant description
 
@Downgoat have an array for words -> maximum matching word count maybe?
 
but typos and all
 
@Downgoat 50 items? You could do that word by word and just bucket counts. I thought you needed "real" search :P
 
1:49 AM
well at the moment around 50 items
who knows it could grow to 200
that's probably not that big of a difference :/
 
Yeah, 200 paragraphs is still basically nothing for this.
It's still going to be hard to determine "relevancy" unless you're working in a very narrow scope though.
(and your users type in sensible queries)
 
pretty sure most people will type "goat goat goat goat iwonderwhatshowsup goat goat", or "asdfjhasdljfahsdkasdfhjagsbdcalhj" :P
 
Well since all 50 items are "goat", that's going to make it hard to determine which one is most relevant ;)
 
Hows everyone doing this evening?
 
1
 
1:59 AM
good good
 
0
 
that's sad
 
Why? Doesn't it mean 'no error'?
 
^ this is true
 
I was thinking 1 was a truthy "fine"
"good" etc.
 
2:01 AM
I figured. That's why I went the other way with it.
 
either way I'm at 1 now, error found, still fine.
 
I'm just sleepy and reading a bit before I decide to check out for the night
 
well, everybody's got to start somewhere...
 
I've just never seen someone post a ~whole~ homework question on SO before lol
 
There was one here recently
 
2:11 AM
I was just talking about formatting questions on SO. people post homework all the time.
 
Ah the formatting is an easy fix once you've posted a few times. i'm still new to posting/answering on SO
 
@confusedandamused Stick around. You'll see it far too often :/
 
So first time for everything I suppose :)
 
@SetBigO the google nlp is exactly what this is for
 
True enough. I didn't want to recommend something like that without using it myself first though.
 
2:20 AM
also, your name
 
Hey, where'd Geobits go?!
 
are we doing anagrams again or something?
 
@Maltysen Anagrams were old... now they're just beating a dead horse
(did i use metaphor right? :3)
 
halp anyone here familiar with llvm?
 
kinda
I know how to tail call optimize llvm
besides that no
Pls halp. how to say "où est-ce que elle est" in correct frenhc
 
2:27 AM
@confusedandamused and now it's gone...
 
@Linus like magic lol
So what happens if you answer a question - and the OP never marks it as the answer (or marks any answer as the correct answer for that matter)
 
nothing
 
I figured as such
 
somebody else may upvote you of course
 
halp what's the difference between LLVM's APInt and GMP's gmp_class
 
2:55 AM
code challenge idea: given an arbitrary, connected graph, find the minimum needed number of disjoint complete graphs that cover the original graph
 
cover nodes or cover edges?
 
nodes
aka, you have a group of friends, and only some of them like each other. You need to put them into groups such that everybody likes each other in their respective groups
 
so for a square A-B-C-D-A. You can use A-B and C-D (without covering B-C and D-A), but not the complete graph of all four.
 
right. B-C and D-A would also be a valid solution
but A-B,C,D wouldn't be valid because it has 3 graphs instead of the minimal 2
 
But we're not counting A-C,B-D as cover ... so the edges play some role
 
3:01 AM
right
oh, you remove edges
actually, that gives a different metric
you need to remove as few edges to construct a set of disjoint completely connected graphs
 
@NathanMerrill [...] find the minimum needed number of disjoint complete sub-graphs that cover the original graph
 
right
0
Q: Algorithm complexity for minimum number of clique in a graph

flatronkaI have written an algorithm which solves the minimum number of clique in a graph. I have tested my backtracking algorithm, but I couldn't calculate the worst case time complexity, I have tried a lot of times. I know that this problem is an NP hard problem, but I think is it possible to give a wo...

 
I've made some great progress on static code analysis for my language Element.
To be more specific, I'm trying to make my analyzer as flexible as possible, so adding support for Element is merely a matter of defining the commands.
 
is this related to the "combination game" we were talking about way back when?
if not, I'm not sure what Element is
 
3:14 AM
spec?
 
Simple language, notable for having two stacks and a map as its main global data structures.
So right now my static analyzer can handle three commands: input _, output `, and addition +
here's some typical output:
___+````__`_+_`````________``+``++``
......EE........EEE................E
The top line is the command, the bottom is whether or not there is an error. (In this case, caused by stack underflow.)
The important part of the static analyzer is that it doesn't actually run the code, just does a single pass of the source.
To demonstrate its real strength, I need to implement one more command... give me just a second.
> just a second
 
... so in this complete sub-graph cover problem, the minimum number should be the same if they are disjoint or not. If you have complete sub-graphs that aren't disjoint you merely remove the intersection from one of them (leaving a smaller disjoint complete graph). Insisting they're disjoint does nothing.
 
3:31 AM
the problem is np-complete :/
 
so?
 
it makes it less interesting
 
were you thinking fastest code challenge or code golf?
how is it less interesting?
 
_2:```
.....E
 
it basically means that the algorithm needs to try each edge possibility
 
3:33 AM
The : command makes N copies on the stack of the previous item. Here, the input is duplicated, allowing it to be output 2 times before an error.
 
that is pretty cool
are the only errors stack overflow and underflow?
so basically each command needs to implement "getStackSize(int currentSize)"?
 
__:```
...WWW
In this case, it is duplicate an unknown number an unknown number of times, giving warnings for the output commands.
Since it doesn't know how big the stack is.
There's actually many ways to trigger errors, it's customized on a per-command basis.
 
right, but for a command to know if its going to trigger an error, it needs to know what its being passed
 
Should I make a language?
(Another one?)
I've got a pretty good idea
 
what's the idea?
 
3:38 AM
___:2:```_``
........W..W
This is a pretty good demonstration, since no matter what the 2nd and 3rd inputs are, and whatever the first duplication gives, it knows that there's at least 1 item on the stack (the 1st input, unchanged). Then the second duplication turns it into at least two things on the stack (with an unknown maximum).
 
ok, language design question: I've got a Container<T> interface, with the method T get(). I would return it in cases when I want to return T, but also allow them to do perform actions on the container itself. For example, Iterator<T>.next() would return a Container<T>, so you call prev() or next() on it if you wanted to.
I'm wanting to make a special syntax for this get() function
aka, you could call container..someFunc(), which would automatically expand to container.get().someFunc()
however, I also thought this would be useful for Collections. Aka, you could call mySet<T>()..someFunc() which would return a new Set mapped with someFunc()
does that sound like a good idea?
 
3:56 AM
I'm making a language based on Excel
So you can put values and formulas in Excel cells
And it's called Excelsior
 
10/10 language name
 
@RohanJhunjhunwala I'd be interested in adding Pip to the IDE, but I don't Java. :(
@NathanMerrill I'm not entirely clear on what you're trying to do. Does container.get().someFunc() call someFunc() on the item, or on the container?
 
...that's the magic I'm trying to figure out
 
Yus
 
technically someFunc is on the object
 
4:08 AM
It works so well... the name
So the syntax is going to be something like is:
open derp
To open an excel file called derp
And
cell A1 = 12
To set the cell A1 to a value of 12
Hello? Anyone?
 
@Qwerp-Derp ?
 
I just have a language idea
Probs going to be lazy though
And because of that I'm probs not going to implement it
 
@Qwerp-Derp 0/10 don't rip off VBA
 
@ASCII-only Is the thing done?
@ASCII-only But it's called Excelsior
 
4:13 AM
what thing
VBA is similar enough though
 
The circ cond one-liner
 
Pushed yesterday, should be working
 
Yay
Doesn't work, tho
It returns a NameError
 
uh
what
testcase?
 
circ asdf(n)->cond n->0/1
out asdf(0)
Returns a NameError: n
 
4:24 AM
@Qwerp-Derp IDK, working for me
 
Really?
 
yeah
 
4:35 AM
Just spent 40 minutes walking some person through their vbs issue from their SO question x.x
If they would have given all the information up front and not changed the question 4 times it would have been much much easier
 
4:50 AM
@Dennis - <= and >= worth implementing a single byte dyads ŀ and ġ?
(-a)
 
5:05 AM
@JonathanAllan Hm, maybe. ŀ is taken though.
 
Oh, it is yes; still in comment at top of jelly.py and not on atoms page.
 
It's a quick, not an atom.
 
oh yeah.
ṿ?
Do you think it's worth the bytes?
 
I'm not sure.
 
yeah I'm not that sure
 
5:13 AM
You can usually get away with < and >. Sometimes : even works as >=.
 
I used
 
Which challenge?
 
decanters
1
A: The Decanting Problem

Jonathan AllanJelly, 9 8 bytes Ṁ<¬a⁸g/ḍ TryItOnline How? Ṁ<¬a⁸g/ḍ - Main link: capacities, goal Ṁ - maximum capacity <¬ - not less than goal a - and ⁸ - left argument (capacities) g/ - gcd reduce over list (gcd of capacities) ḍ - divides

 
Replacing with : should work.
 
oh
yeah that'll work!
nice
 
6:17 AM
hello
@Dennis is there an expiration date of the TIO links?
 
@seshoumara No, the links are just the input, code and arguments encoded
 
No. Currently, the code is encoded in the URL, so I couldn't expire them even if I wanted to.
 
Ok, thanks.
 
6:33 AM
yo!
 
Yo yo yo!
 
I can convert a register value into a unary value, I can convert a unary value, into a decimal value, now I just need to convert decimal, into unary, and turtlèd arithmetic should be good
 
Hello
 
6:42 AM
!oγ
The computer in my classroom
 
!u-[*+.r_]' l[#[ (0'9l)( r[ '#r]u)(1'0u)(2'1u)(3'2u)(4'3u)(5'4u)(6'5u)(7'6u)(8'7u)(9'8u)]dd[*l]{*r},u{ l}[ r]l]
almost works
but does two extra :(
not sure whether to just remove them, or find a better way
!u-[*+.r_]' l[#[ (0'9l)( r[ '#r]u)(1'0u)(2'1u)(3'2u)(4'3u)(5'4u)(6'5u)(7'6u)(8'7u)(9'8u)]dd[*l]{*r},u{ l}[ r]l]{#' l}dr{*r}ll"
removes extra two chars from unary representation
 

« first day (2064 days earlier)      last day (2770 days later) »