« first day (562 days earlier)      last day (917 days later) » 

1:52 AM
@user in regards to
if item < 0:
    temp = type(item)(str(eval(item))[1:][::-1])
else:
    temp = type(item)(str(eval(item))[::-1])
you asked
# TODO (lyxal) What is temp for? Why use eval directly without
# checking that it will truly evaluate to a number?
and the answer is that the eval is there just to turn rationals into floats
at least, that's the intention
return sympy.Rational(eval(rev) * sign)
that should work better
@user about your concat helper function
def concat(vec1: VyList, vec2: VyList, ctx: Context = None) -> VyList:
    """Concatenate two lists/lazylists"""
    if LazyList not in (type(vec1), type(vec2)):
        return vec1 + vec2

    @lazylist
    def gen():
        for item in vec1:
            yield item
        for item in vec2:
            yield item

    return gen()
Now's the moment I reveal that it already existed inside LazyList
def join_with(lhs, rhs):
    for item in lhs:
        yield item

    for item in rhs:
        yield item
elif takes_ctx(function):
    return function(*args, ctx)
return function(*args)
That's why safe_apply exists
which I know you know exists because you modified it lol
 
2:28 AM
Wait what
safe-apply errored for me because of the extra context param, that’s why I made takes_ctx
@lyxal im blind lol
@lyxal oic
That whole func confised me so much because temp is assigned to but just ignored
And at the end you just return a Rational(lhs)
I hope all my other “fixes” weren’t as useless
 
If they were I'd have pointed them out
Most of them were legitimately things I'd missed
 
tbh i kinda feel like this would be really hard to get short in vyxal due to its weakness with array operations
might give it a shot
 
3:52 AM
is there [0, 1, 2, 3] -> [0, 1, 2, 3, 2, 1, 0] in vyxal
 
4:07 AM
...thought so, but m and øṁ duplicate the center
 
mhm :/
 
not going to look up the actual elements for doing that then but i'd assume it comes out to like four bytes on one axis
dup, reverse, behead, concatenate
then to do that in both dimensions you transpose at the end and do it twice
 
4:23 AM
@hyper-neutrino oh wait there's øm
 
ninja'd
!!/run 0 1 2 3Wøm
yes bot thank you very cool
 
bot still didn't kill itself properly
💀💀
 
alright one more try
 
4:28 AM
i forgor that localhost isn't a protocol
so get localhost:5888/kill is not valid
 
@UnrelatedString is this what you meant?
 
oh yeah lol thanks
that makes
much more sense now
 
2(ømÞT) is equivalent btw
 
oh yeah of course you can do that
i thought it would come out the same length to use a lambda and the repeat element lmao
 
well
i ended up with 28
so not exactly the 4-way tie we wanted
 
4:37 AM
show us and we can help
 
1
A: Checkered grid with X mark

hyper-neutrinoVyxal, 28 bytes :±+ʁ:v±†vǔd$ʁ:v+∷+ṘvṘ2(ømÞT) Try It Online! damn i am bad at vyxal

input b,a
 
ez 27
:±+ʁ:v±†vǔd$ʁ:v+∷+ṘvṘ2(ømÞT
 
:±+ʁ:v±†vǔd$ʁ:v+∷+ṘR2(ømÞT for 26
 
o
oh there is vectorized reverse builtin?
couldn't find it sadge
 
4:41 AM
R if not given a function
 
:±+ʁ:±†vǔd$ʁ:v+∷+ṘR2(ømÞT for 25
 
then what do you use to reverse a function
 
@hyper-neutrino that's the same output it gives me as the 28
no wait lol it'sn't
because of the footer
@hyper-neutrino Try it Online!
 
oh
oops
i see
 
4:49 AM
@hyper-neutrino an explanation might provide me a way to get 4 extra bytes off
 
5:03 AM
@lyxal done
 
wait when did i save a byte
i think
yeah
 
5:22 AM
@hyper-neutrino upon careful examination, it turns out I'm not big brain enough to golf it
Also, I need ideas for what to put into the short dictionary - a list of 162 entries that are accessible by single non-ascii characters
As in, when you only have half a SCC
I currently have 32 regexes in it
But I need more ideas
!!/hyperping
 
@AaronMiller @Allxy @Ausername @Milk @Wansen @emanresuA @exedraj @hyper-neutrino @math @pxeger @user ^
 
oh wait i am dumb
all 3 of those bytes were lyxal
 
lol
 
lol
...there goes the tie
@UnrelatedString you didn't have to actually edit me out
 
 
1 hour later…
6:58 AM
Hey guys dictionary compression now works
and `ḋ`Ė is the new shortest fizzbuzz
because I totally didn't put ₁ƛ₍₃₅kF½*∑∴, into the short dictionary
no i totally didn't do that for the lols
why would you even suggest that I would do such a thing? ;p
 
whom are you talking to?
 
Everyone here and everyone who will read the above
It's an announcement aimed at a general audience
 
ok
 
7:42 AM
!!/run `testing the api`
 
@lyxal testing the api
 
Hmm
!!/run λ0|k≈℅;£0{:₁<|⟨⟩4(¥†J)B},
 
[@lyxal: 59490557]
STDERR:
[Errno 11] write could not complete without blocking
 
🤔
Idk why that doesn't work
 
8:34 AM
!!/run1
 
@PyGamer0 1
 
hey look i made a quine
 
!!/run `:qp`:qp
 
@lyxal `:qp`:qp
 
Actual quine lol
 
 
1 hour later…
9:56 AM
@lyxal ... Just use constants. I suggested either [nonascii][ascii] or leaving them alone. Having a short dictionary leads to parsing nightmares.
@UnrelatedString ... I had a 26 that didn't even work
I also had a 22 that also didn't work
 
10:12 AM
@emanresuA it doesn't
It's not a matter of parsing
It's a matter of decompression during transpilation
I've written a working, relatively simple, decompressor
Groups of 2 non-ascii take priority over single nonascii
So something like `¡¡¡` sees it as ¡¡ and ¡
Spaces after a single nonascii are removed to make including them easy
But only the first space
So like ¡ a is the same as ¡a
That's so you can have ¡ ¡ and have it treated as if you just concatenated two individual copies of ¡
(¡ is just an arbitrary nonascii here - I'm on mobile, and it's the first character on my phone's keyboard that is on the code page)
 
 
3 hours later…
12:51 PM
@PyGamer0 That’s not payload-capable
 
@lyxal why not put the actual text output of fizzbuzz in the dictionary in that case lol
 
@pxeger nah that'd be too overly optimised lol
@pxeger isdo is the command set of deadfish
 
oh ok
 
yeah I figured that we should stop overspecialising for just fizzbuzz, and start overspecialising for more challenges :p (that's a joke please don't quote this out of context lol)
3
 
@lyxal how about some common Unicode characters that aren't in the codepage? (which would otherwise take lots of bytes to include using 1234C or another method)
alternatively (and this would require quite some additional complexity), make some of them refer to n the most recent decompression strings to reduce repetition
e.g. if ȧȧ -> Friday then ȧȧ¡ -> FridayFriday
where ¡ has n=1
or maybe something like hello¡ -> hellohello?
although choosing the right boundary would be difficult
 
 
6 hours later…
7:01 PM
is there a shorter way to concatenate a and b on a space? `egg`ð+`bruh`+ assuming you can't hard code the strings
 
`egg``bruh`"Ṅ
does that work
 
yeah also how do you do the compressing string thing?
 
!!/run `Hello, World!` øD
 
@spooky-neutrino `ƈṡ, ƛ€!`
 
alright cool
is there a way to combine the top 2 items in the stack like this (a|b) -> ((a|b))
 
7:15 PM
W (wrap) should do it
!!/run 1 2 W
 
@ruse ⟨1|2⟩
 
For the first time in forever, I actually remember a Vyxal element lol
 
lol
 
doesn't that combine all of them
!!/run 1 2 3 W
 
@spooky-neutrino ⟨1|2|3⟩
 
7:18 PM
!!/run 1 2 3"
 
@spooky-neutrino ⟨2|3⟩
 
:( I don't actually remember a Vyxal element
obtw I think W is broken in the rewrite
Can't check anymore because I accidentally generated a couple hundred failing tests, but the one for that was failing, I think
I love it when you fix over a dozen builtins and think you only have half a dozen to go, and then you get 187 F's
 
kekw
also
@spooky-neutrino i literally used " here
 
Then you should've explained to Hydrazer what you were doing better smh
 
not criticizing hydrazer
i'm criticizing you :P
 
7:22 PM
bruh you expect me to read what was said before?
in The Nineteenth Byte, 13 secs ago, by ruse
...I'm not having a good day
The defense pleads tiredness
 
the jury finds the defendant guilty and sentences him to go to sleep
yes ik the jury doesn't do sentencing
 
o/
@spooky-neutrino Looks like the jury is also guilty :P
 
sage moment
 
7:53 PM
what now
 
8:31 PM
@Hydrazer smh we use Python, not sage
Sage is cool but we don't really need that extra stuff, sympy's good enough
 
9:23 PM
lol i mean to say sadge
 
10:07 PM
sadge
 

« first day (562 days earlier)      last day (917 days later) »