« first day (2336 days earlier)      last day (2802 days later) » 

04:02
;_; parsers are hard
hey lemon what do i do when life throws crying lemons at me
obviously you have to burn lifes house down to make an example out of him
normally you would do this with the lemons but the lemons are crying too much for that to be effective
brb
ok it turned out I was being stupid again was the reason why it was borked
yes that's usually the reason for code bork
developers have a higher tendency of being stupid while doing their job
ACTUALLY IT IS STILL BORKED
Wait false alarm
04:14
I need to advance the parser some more so that it can recognize when a thing is just negative stuffs and ()
hmm
so far it parses all the ()s in a list into an int to save time when it runs
but I should make it so it recognizes negatives as well
I could also make it so that it recognizes stuff like ()()[()[()()]]
but that seems like the user could optimise that stuff themselves
actually if I do it right that might be a tagalong effect. but anyway I will stop writing about it in here until I either: finish it, or mega bork it so bad I can't help crying in chat
ok it looks as if my parser is done (except for the execution and newline expansion, better do those now)!
04:37
I called the parsing function lispify because it makes it lisp like
my interpreter uses recursion rather than the third stack because recursion is good and yes
04:58
@ATaco Even though half the letters are gone I can still read it you know
@ATaco 65248 not 975...
@PhiNotPi Can we help :P
@ASCII-only I might need some help from a python handler.
(python handler = someone who knows python)
@PhiNotPi ok i know python
i mean my main language is JS but i wrote a language in python so i suppose that counts as python handling experience? :P
> wrote a language in python
TBH that sounds pretty qualified
I need advice
it's a golflang though so not complex python
05:03
if program_list[0] == self.command_index["<"]:
@PhiNotPi ok then i might be able to help
I was doing this because it parses it to numbers, but is this a loss if I just use a dictionary anyway?
the answer is probably yes
@DestructibleLemon what
I'll keep people updated if I run into any actual issues. Right now I'm just planning how I'm gonna solve the problem itself.
but unfortunately python does not have #define
@ASCII-only I mean I was trying to make it so it parses all the commands into numbers in lists, but then using those numbers might obfuscate
but then it ends up defeating the purpose if I use a dictionary anyway
I guess comments
05:05
@DestructibleLemon yes
it's faster to lookup the command in a dictionary
hmmm, I just realised, how does ≤≥ work when it is inside another pair of parens?
@DJMcMayhem can you help me with this language design question?
@DestructibleLemon what kind of question is that (also what do they do)
≤≥ is the halt command
initially I thought this would be annoying, but I think I can fix it if I tune up the parser
done!
05:17
CMC: given two positive integers a and b, where b>1, find the next integer from a which shares a divisor with b other than 1 and -1.
So I realised there was a huge flaw in my session recovery logic, which essentially kept people logged in all devices as long as one device stayed logged in. So I had to re-write basically my entire session handling script.
@LeakyNun example pls
10,6 -> 12; 9;6 -> 10
how do you make a programming language consider flags?
@DestructibleLemon consider?
05:26
like how brainflak uses -a and -A
@DestructibleLemon well handle the flags separately?
in fact, is the - needed?
@ASCII-only do I use these as arguments?
so how do I take arguments to my program like that?
@LeakyNun JavaScript (ES6), 50 bytes: f=(a,b)=>(h=(a,b)=>b?h(b,a%b):a)(++a,b)>1?a:f(a,b)
had to implement euclid hcf <_<
@ASCII-only btw quetion: can we have multiple applied candidates for 1 root?
e.g.: (U extends T) param take T candidates are U & T
i don't think possible
but idk
not sure how to resolve
idk if I should 1D filter or filter the applied plane too
05:38
ono wat confuse. Like function is: func f(a: T)
f(a: myVar) and myVar's candidates for the type it could be are U and T
Worse approach, but how I thought of solving it. a=>b=>{for(;;){a++;for(c=2;c<b;c++){if(a%c+b%c==0)return a}}} 61 bytes.
@ASCII-only I don't know how to handle the flags!
@ATaco ಠ_ಠ you use block lambda ;_;
JS golf tip: never use block lambda
reccomend use recurse
I use block lambda when it demands I do.
but if100% sure impossible (which is not possible that it could be impossible) you should do:
05:40
Well yeah, I'm not very good at JS.
@DestructibleLemon what. use argparse
a=>b=>eval("for(;;){a++;for(c=2;c<b;c++){if(a%c+b%c==0)break;}a")
@ASCII-only oh btw can i have ur input on VSL app
@ASCII-only how use
@Downgoat ???
should it have config so you can change between modes (e.g. type check mode, ast printer mode)
05:41
@DestructibleLemon :| look at bottom of charcoal.py
@Downgoat yes
@ASCII-only thanx
@ASCII-only where should button be ^^
or should I add button
@Downgoat ninja'd!
wat ;_; how y u do dis, we never post about same thing?
05:42
because your caret now points at my message
muhahahah
idk what you're taling about
vsl>d
TypeResolutionError: Use of undeclared identifier d. If you wanted to reference a function, make sure you specify a function type somewhere.
TypeResolutionError@staging.vihan.org/VSL/app/main.bundle.js:18738:65
emit@staging.vihan.org/VSL/app/main.bundle.js:4812:19
resolve@staging.vihan.org/VSL/app/main.bundle.js:18430:17
resolve@staging.vihan.org/VSL/app/main.bundle.js:18683:13
modify@staging.vihan.org/VSL/app/main.bundle.js:10338:14
transform_once@staging.vihan.org/VSL/app/main.bundle.js:10875:13
The following commands are available:
  help - displays this dialog
ok maybe that is too much error message @Downgoat
@DestructibleLemon not sure wat confuse you never declare d
05:42
@Downgoat :D work very well
no I mean why is there so much error
@Downgoat except bad tree
like in python:
>>> d
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
NameError: name 'd' is not defined
@DestructibleLemon transformation/deduction passes are incredibly complex so don't expect sane stack traces
it is short and sweet in python
05:43
wtf python only displays one line of stack trace O___o
python is sane
no wonder python program look like carpet
carpet = carp = crap
it only has a small stack trace for simple things
05:44
@ASCII-only :D made in react so we can also really easy expand and add new thing
if it is a bunch of recursion layers deep it will be longer
How do I do anything in VSL?
@ATaco there is no compiler yet
@ATaco well STL is not included in REPL by default so make sure you run that:
@Downgoat carpet python?
05:45
2 days ago, by Downgoat
@primitive(Integer) class Int {}
@primitive(FloatingPoint) class Double {}
@primitive(String) class String {}
once you do that you should have those three root classes
you can then declare variable & function and watch magic type deduct to magic
@Downgoat :( why tree look so bad pls
;_; halp not expert tree maker like u
i try very hard (like 7 minutes tops) :(
@ASCII-only this is monstrous
@DestructibleLemon what ono what i do wrong
anyway thanx I guess
05:47
@Downgoat but is in node.js
@ASCII-only it is just very big
and confusing from bigness
Hey, there
@ASCII-only oh this is scope printer
@DestructibleLemon just look at the ArgumentParser part
05:47
i am using ATM for debugging type deduct
quetion: should i call type deduct or type infer
@Downgoat i mean steal from node.js
porblem is i can't do recursive like in node.js
This is what is ATM:
@Downgoat but can still do pretty tree
pls
all u have to do is special case last child
wait does node.js AST tree work
there was lot of change
05:50
there was lot of change in node spec.
@Downgoat idk have not check probably not i just port so we have in js
ono
btw we're not rewriting in C++ like crazyguy say right
like C++ is very hard
@Downgoat no because manual memory manage is vrey hard annoying
oh yea tru
and also very hard to do cross-platform
like valgrind go ono u have 8 byte leak
and we have to spend a month finding leak
05:52
:O valgrind = sheep confirm???
CMC: do you have a doggo/puppo
@Downgoat no
@Downgoat yes
@Downgoat *pupper
05:53
@ASCII-only you failed the CMC
ono u can have more than one of same function
k then seeya
\o
@Downgoat pls see my second most recent message
@DestructibleLemon :O pix pls
06:00
I have 2x Doggo.
@ASCII-only wat do u mean
"Read this message, then do what it says"
@Downgoat i can have like million function with same signature = ono
@ASCII-only that is becaus repl
repl alow redeclararion
@ATaco +1 this sounds really accurate
@Downgoat but wouldn't it override
06:02
Yea
@Downgoat so why are both in scope at same time D:
"Read this message. If the result of doing what it says is the same as the result of doing what this message says, finish following this message."
Forkbomb!
Can a problem with 3 possible outcomes, like X wins, Y wins, tie be considered a
?
Also @MartinEnder and @Dennis, do you mind if I create part 2 of this challenge? This time, I will add the note regarding targeted voting from the beginning, and I will sandbox the post if you agree.
@Mr.Xcoder No, But iirc there is a tag for that.
@ATaco what tag?
@ATaco Oh, I wasn't aware of that tag. Thanks!
Let me just double check that, to be sure.
However, I don't think it's the case here
No, I doesn't really fit the tag description... Anyways, I will sandbox the post.
Yep, classification is the go-to tag for "I have a few classes of things, determine which of these classes the input falls under"
@ASCII-only because i basically create new scope ever REPL execution which inheits old one
06:15
Yeah, doesn't really fit
@LeakyNun hi
and hello all
@Lembik hi
Hola.
aloha
Yá'át'ééh!
06:25
@Lembik hi
@ATaco como estas?
@LeakyNun I can't stop wanting to edit your math.se question... I don't understand " Clearly the answer is also at least $n$, since one can pick a fixed $a$ and $b$, and argue that $\{ab+c \mid c \in C\}$ has exactly $n$ elements."
In fact I feel it is wrong
do you mind if I change it?
@Lembik it's thy question; change it howsoever thou willest
thanks
@Lembik You are about to hit 10kin total! Congrats!
@Mr.Xcoder too many accents, lol
06:28
@LeakyNun so.. why is it at least n?
is that obvious?
otherwise I will just delete that part
@Lembik I gave a justification.
That's Navajo (the language).
@LeakyNun it's the line I pasted above?
that doesn't make sense to me
yes
why not?
why should that have exactly n elements?
06:29
why not?
oh I see what you meant!
sorry :(
@LeakyNun No habla el Español.
@ATaco but you just spoke it
@Mr.Xcoder Are you a native speaker thereof?
@LeakyNun at least the question is getting a few upvotes now
I can say various forms of hello across various languages, I don't speak most of them.
06:31
@LeakyNun Of course not
That's a Southern Athabaskan Language
I'm from Romania
oh
Romania is the only Romance language that has managed to preserve three cases iirc
In my language: "Salut!" or "Buna ziua!"
@LeakyNun What do you mean by "three cases"?
In my language: "G'day!"
@Mr.Xcoder /s the lowercase, the uppercase, and ...
no, the nominative, the genitive/dative, and the accusative
and the vocative
there are 5
06:34
are they different?
For example?
Genitive and Dativ are related
It's hard for me to give examples in English @LeakyNun
no, I mean the word
What word
??
06:35
I thought the genitive and the dative are the same word although they have different functions
like the inflection is the same
like you add the same ending
@LeakyNun It's not related to the word itself. They both point to structures that express posesion
But in different contexts
I'm only referring to the word itself
Russian has cases
I think there's 6
The form of "Dative" and "Genitive", the similitude between them is irrelevant @LeakyNun
I am only talking about the forms
of course I know that genitive and dative are different functions
06:38
Ok, let me give you some examples.
I mean, other Romance languages have merged the forms of the cases
Sadly no-one counts Australian as a Romance language.
@LeakyNun Genitive: "My grandmother's glasses are old." - Here "glasses" is in Genitive (in Romanian), Dative: "I told him that ... " - Here "him" is in Dative. The common ending of the words "Dativ" and "Genitiv" only denotes that they both express possession.
why would it be and why is it sad?
What's more romantic than "Oi sheela! Fancy a shag?"
06:41
@Mr.Xcoder I said I know they mean different things
I just said the forms in Romanian are the same
@LeakyNun I know you knew.
True
and English actually has two cases, but the second case has been highly regularized
cookies for figuring out what the second case is
@LeakyNun I think you will end up eating those cookies by yourself :))
You just used that case :p
Oh really?
Where specifically, if you are willing to share with us @LeakyNun
06:47
well, "yourself"
I give up
the genitive case...
king -> king's
book -> book's
Really :\\
highly regularized?
regularize to become 's for all nouns
Oh, ok.
06:50
0
A: Sandbox for Proposed Challenges

ArthurBubble the brackets! There are a few questions on this site about balancing brackets, and checking whether brackets are balanced. I propose it's now time to use those balanced brackets for something! In mathematics and programming, brackets are like bubbles, isolating everything inside form eve...

Then, is "x of the object" still used to express possession?
Old English cyning -> English king
Old English cyninges -> English king's
@Mr.Xcoder sure
Why do you consider that a case and not the difference between subject/object?
I/me, She/her etc.
because that distinction is only there for the pronouns
whereas the distinction between nominative and genitive is there in all nouns
@LeakyNun Precisely the same situation as for the apostrophe-s things.
06:55
@feersum no, "king's" itself is the genitive
OK, but what's wrong with two cases sometimes having the same spelling?
well then they have merged
@feersum Mainly, merging helps avoiding confusion.
@feersum you can call it the nominative/accusative/dative case
hi I'm back
07:02
Why are some users in stack exchange chat greyed out a little?
@Mr.Xcoder they have been inactive, and over time the avatars grey out little by little while inactive
@DestructibleLemon But are they present in the room?
@DestructibleLemon Thanks!
@Dennis and @MartinEnder
57 mins ago, by Mr. Xcoder
Also @MartinEnder and @Dennis, do you mind if I create part 2 of this challenge? This time, I will add the note regarding targeted voting from the beginning, and I will sandbox the post if you agree.
^
(they do tend to be sleeping sometimes though)
07:08
@DestructibleLemon Good point....
they'll get to the ping when they do, still
My sandbox challenge is ready, and I think I will post it. If you disagree with the idea, I will not upload it on Main
Just let me know.
16
Q: Generalized Quine Generator

ZgarbThe Challenge In this challenge, you specify a source language S and a target language T. Your task is to write the following program P in the language S. If a valid program Q in the language T is given as input to P, it will output a valid program R in the language T which takes no input and ou...

+ 600 OMG
Ignore the above comment ^
07:23
0
A: Sandbox for Proposed Challenges

Mr. XcoderMartin vs Dennis - Round 2: Who has more upvotes? stack-exchange-api code-golf This is the sandbox for the second round in this series of Challenges. Here is Round 1, but this one is a completely different challenge. Your task is to write a program or function, that, given the ID of a questi...

@Mr.Xcoder You can always ping someone who has spoken in the room, even if they are not present.
@Adám Yes, I know that.
BYE!
Hi
@Adám Isn't there some sort of rule that they must have spoken in the room in the past fortnight?
@Arjun Have not heard of such. We could maybe test that by finding a room one of us have not spoken in for over two weeks?
I've pinged @Arjun there.
@Arjun It would seems so, as your id wasn't suggested.
@Adám Nothing happened.
@Arjun Oh no, now I cannot delete my msg. :-(
Jim
Jim
Hullo
@Adám do you have CMC?
07:48
@LeakyNun You want a challenge?
sure
I find your CMCs to be of much quality
@LeakyNun Thank you.
So do you have one?
CMC: Given a strictly ascending list and a list of any numbers, return the index of the gap in which each would fit in the ascending list.
E.g. 1 2 3 4 and 7 3 6 24 3 4 2 (one indexed)
You see, my compliment stands
07:53
@LeakyNun :-)
I don't get the challenge
why is it not 4 3 5 2?
@LeakyNun Because each one is judged anew.
Then it is an unnecessary layer of complication
@LeakyNun I.e. you're not supposed to actually put them in there, only return which gap they would go in.
07:55
It would be something like +/@:<\ in J, but I don't have a computer
@Mr.Xcoder Note or not, anything based on votes isn't going to encourage good behavior...
@LeakyNun No, it is a useful operation, e.g. for labeling observations.
@LeakyNun I can do it in a third that length in J.
And I think the ranks in J are messed up
@LeakyNun Ranks?
@Adám do you mean something like this: (≢⊣)⌊⍳?
07:57
@Adám good for you
@KritixiLithos No, that gives the right result in this case, but not otherwise.
could you give another example?
1 2 3 4 and 7 3 2 1.54 3 2 1
ooh
I get it now

« first day (2336 days earlier)      last day (2802 days later) »