« first day (1047 days earlier)      last day (728 days later) » 

01:44
!!/status exciting
BOT STATUS: ONLINE!!!!!
UPTIME: 1 DAY, 9:28:15.350055!!!!
RUNNING SINCE: 2023-02-26T16:15:45.186954!!!!
ERRORS SINCE STARTUP: 0!!!
!!/status cryptic
Sorry lyxal, I'm afraid I can't do that.
feel like that should be a feature
!!/help status
01:45
@Jacob !!/status: I will tell you what I'm doing (maybe). If you don't have a sense of humor and/or are a bot developer, run !!/status boring for actual status information.
hm
!!/status
@lyxal I am doing things like inventing a new type of transportation that can travel underwater and through space
!!/status sleepy
Sorry Jacob, I'm afraid I can't do that.
01:46
zzzzzzz
!!/open issue in VyxalBot2 More status variations Status <sleepy|cryptic|goofy|...>
Sorry lyxal, I'm afraid I can't do that.
!!/issue open in VyxalBot2 More status variations Status <sleepy|cryptic|goofy|...>
vyxal-bot[bot] opened issue #17 (More status variations) in Vyxal/VyxalBot2
good
01:48
we need status tingly which puts those asterisks and italics and adds uwu or smth
bot statsaku; onlonichan
02:01
no
we really don't
yes we do
02:17
@hyper-neutrino now that you're here (but probably gone again) would you be able to see if I've implemented SSS and the compressor correctly?
sure
ngl i don't know what sss stands for lol all I know is the jelly compress function is named that for some reason
Smart String Scompression?
Silly sock Sunday
That's the code
def compressString and def SSS are the relevant functions
time to review LongDictionary.txt and ShortDictionary.txt line-by-line
I think I used the 5 list intersection word list
With words starting with non-letters filtered out
@hyper-neutrino probably from compressString
02:44
honestly that sounds in character
@Ginger we really don't
@Jacob we will be fowced to muwdew you if you do this uwu
Jesse what the f*ck are you talking about
Oh wait you're Jacob not Ginger
@lyxal I can condone a bot that does people without their consent, but I draw the line at a bot that uwufies its messages
oh
in The Nineteenth Byte, Mar 11, 2018 at 16:17, by Dennis
I think it was Short String Shortener.
 
6 hours later…
08:37
in The Nineteenth Byte, 5 mins ago, by mousetail
Stack length is kind of useless in a practlang unless you have a way to isolate separate stacks. Since it can't be used in functions. It's only useful for very short bits of code but can greatly simplify those
Y'all need to stop talking about things like this in TNB
It just keeps giving me somewhat bad ideas
Relevant idea: modifier that applies an element to a higher context based on how many modifiers of that type you have
So ©f applies f on the values of the context one up, ©©f two and so on
It wouldn't pop from upper stacks I don't think just peek
©<number>f applies f to number contexts up
A neat side effect would be that ©n allows retrieval of the context variables of outer contexts
Hey maybe this isn't such a bad idea after all
© obviously isn't the final symbol choice
It's just the most context modifier looking thing I could find on my phone's keyboard
And it's also a much better idea than how x and X operated
Because instead of applying to everything, it only applies to one thing
And it doesn't need to be reset after using it either, meaning it could be somewhat viable for golfing
@lyxal it would obviously push its results to whatever context it was called in
It would just peek/pop on stacks further up
 
3 hours later…
11:54
you know you're having a good time when you run into limitations with int size
oh huh decimals
gee this really is fricked
there should not be decimals
!!/status exciting
BOT STATUS: ONLINE!!!!
UPTIME: 1 DAY, 19:47:50.834324!!!!
RUNNING SINCE: 2023-02-26T16:15:45.186954!!
ERRORS SINCE STARTUP: 0!!!!!
"Hello World!" #C
> "ᶻ4PẆ←jʀu”
"ᶻ4PẆ←jʀu”
> "6helalmountingsQ%"
I am having fun
so much fun
I love porting Jelly code into scala
I love it so much that I even get the language name wrong
me when 6helalmountingsQ%
@lyxal keep this as its own element, for the funny value
okay, so I found a good uwuification library, and I was just looking through the source:
> ***unzips your pants***
> ***sees bulge***
this will take Vyxal Bot's indecency to a whole new level
12:10
@Ginger well you see it's not the element, it's the underlying implementation of sss
I don't know whether it's a problem with the compressor or the decompressor
@lyxal if they're the same algorithm, try decompressing it with the known-good Jelly decompressor and seeing if it works
well you see one tiny problem
@lyxal yeah, keep the broken implementation as its own element
different dictionaries
we here are vyxal corp (tm) (c) (r) (inc) (tm) may steal algorithms, but word lists are something we pride ourselves on having as unique
@lyxal 🫁😶‍🌫️
12:16
also, jelly's word list is a single .py file. Vyxal's word lists are split into 3 files: two .txts and a .js file
different words and different formats
also also different codepage
@lyxal sounds like you need some workflows :b
that's overkill for something that will never need to be generated again
GingerIndustries created branch more-statuses in Vyxal/VyxalBot2
fair enough
> kyangarborannunciative
> nWASTS_ablative
so turns out it's a problem with both the compressor and the decompressor
12:31
huh
so turns out multiplication doesn't work like I thought it did
apparently 3 * 96 * z + o != 3 * (96 * z + o)
apparently distributivity exists
I think I'll write a letter of complaint to the manager of mathematics about this
12:58
I have his number, it's +61-3-8652-1453
mathcat4 deleted branch add-flowey-quotes in Vyxal/VyxalBot2
13:15
compressor now works
Ah'll be bahk.
Well, here we are again.
alright, let's see if this works
!!/status goofy
Bot status: Online🤓🤓🤓
Uptime: 0:00:08.701337🤓
Running since: 2023-02-28T13:25:56.101962🤓🤓
Errors since startup: 0🤓🤓
nice
!!/status cryptic
Obg fgnghf: Bayvar
Hcgvzr: 0:00:19.387605
Ehaavat fvapr: 2023-02-28G13:25:56.101962
Reebef fvapr fgneghc: 0
13:26
and, of course:
!!/status tingly
@Ginger B-b-bot status: Onwinye
wow
!!/status cryptic
Obg fgnghf: Bayvar
Hcgvzr: 0:00:40.674778
Ehaavat fvapr: 2023-02-28G13:25:56.101962
Reebef fvapr fgneghc: 0
lmao
13:26
!!/status tingly
@lyxal Bot status: Onwinye
one second, I'm not sure why it dropped the rest of the message
oh hell no
@VyxalBot jesse we need random keys
oh, it was a bug in the msgify function
13:28
Ah'll be bahk.
Well, here we are again.
!!/status tingly
Bot s-s-status: Onwinye
Uptime: \*\*\*looks around suspiciously\*\*\* 0-0-0-0:00:02.587917
Wunnying since: 2023-02-28T13:28:10.978663
Ewwows since stawtup: 0 (˘³˘)
much better
oh right, chat doesn't format multi-line messages
so I don't even have to use msgify
Ah'll be bahk.
Well, here we are again.
I present to yall: The best thing I have ever added to Vyxal Bot
!!/status tingly
13:29
Bot status: Onwinye
Uptime: 0:00:10.024438
Wunnying since: 2023-02-28T13:28:58.631511
Ewwows since stawtup: 0
!!/status tingly
this is glorious
Bot status: Onwinye
Uptime: 0:00:22.200848
Wunnying since: 2023-02-28T13:28:58.631511
Ewwows since stawtup: 0
oh, and sometimes it'll add random text/emoticons to the statuses
(there's also this one:)
!!/status sleepy
Bot status: Online\*yawn\*
zzzz
13:30
me when backslash
Ah'll be bahk.
wellllll?
oh, captcha :/
SEs captcha is kinds bad lmao
Well, here we are again.
!!/status sleepy
13:32
Bot status: Online
Uptime: 0:00:02.926414 *yawn*
zzzzzz
very nice
!!/status tingly
B-b-b-bot ***pounces on you*** status: Onwinye
Uptime: 0:00:27.289766
Wunnying since: 2023-02-28T13:32:32.820900
Ewwows since stawtup: 0
glorious
needs more *baka* tho
wonderful
!!/status sus
13:33
Sorry mathcat, I'm afraid I can't do that.
hmm, I wonder what that should do
GingerIndustries opened pull request #18 (More statuses) in Vyxal/VyxalBot2
can we get some reviews on ^
@Ginger one line should be randomly generated
lmao, perfect
13:37
"ᵃ&∨„ᵐ(Ko¦”
That should be hello world, but parsing doesn't like it
> uwuipy
because it thinks it's a goofy number
meaning the string quotes shouldn't be allowed in the compressed string
that's a fix for tomorrow
changes have been pushed to main
#C to compress
oh, nice
!!/status tingly
Bot status: Onwinye
Uptime: 0:06:37.118270
Wunnying since: 2023-02-28T13:32:32.820900
Ewwows since stawtup: ***looks at you*** 0
"ᶢ%hŀᵈEė{” --> "Hi yes hello"
13:39
@VyxalBot I fear I have done something I will live to regret
After this string compression pr is done, I never want to touch sss again
It can just sit quietly in the code base undisturbed
@lyxal ?
The site doesn't update from PRs
Only version 3
Needs to be merged first
oh, right
I'll clone it then
 
3 hours later…
17:03
!!/status tingly
B-b-bot status: Onwinye
Uptime: 3:30:55.443360
Wunnying since: ***huggles tightly*** 2023-02-28T13:32:32.820900
Ewwows since stawtup: 0
@Jacob look what you've done
17:13
!!/status tingly
Bot status: Onwinye
Uptime: 3:41:26.228734
Wunnying since: 2023-02-28T13:32:32.820900
Ewwows since stawtup: 0
I love it.
Thanks i hate it
@VyxalBot don't let hyper see this lmao
 
3 hours later…
20:39
@mathcat Why not?
@mathcat @HYPER-NEUTRINO !!!!!
idk if that pings
@hyper-neutrino
!!/status
@user I am doing pxeger's bidding
Sorry mathcat, I'm afraid I can't do that.
!!/status tingly
1 message moved to ­Trash
21:39
@user does scala js support reading from resource files?
idk, we need to try that at some point
I don't think it does
shortDictionary: Seq[String] =
io.Source.fromResource(Globals.ShortDictionaryFile).getLines().toSeq,
longDictionary: Seq[String] =
io.Source.fromResource(Globals.LongDictionaryFile).getLines().toSeq,
That's gonna error then
Yes, it's only the default so the tests don't have to do it themselves
Also, the exceptions in the compressor aren't really for debugging - they're really for exiting a function without returning
@mathcat yes
21:47
@lyxal I meant that if something else threw an exception, we wouldn't know it happened, making debugging harder
Like an ArrayIndexOutOfBoundsException, for example
@lyxal on the contrary, i loved touching sss to implement figs compressor
@user for other explicit throws we should have more specialised error types then
ye
Oh yeah there's still the problem that the word lists aren't optimal and that compressed strings need to not return other string characters
Oh dear
I tried compressing Hello world and got "ᵇ∧ṗ“₉→/ᵂṘ]¤ʀκġ!=†jŀṅ"@λḌ”
Looks like my changes screwed it up, let me undo some of them
Where's the original Jelly compression/decompression code btw?
21:56
I linked to what I ported in comments
Compression is from an SE post, sss is found around line 1500 in jelly iirc.
I can't find it in the comments on the PR
@user not on the pr, in the code
Above the function definitions
@emanresuA I used a different source for compression
@user I can only assume that something with big ints' size limit has gone wrong
bigints have a size limit?
If you don't create them properly
Probably
22:01
that shouldnt be
javas bigints dont have that problemo
@lyxal Ah ok, I'll take a look alter
8 hours ago, by lyxal
"ᵃ&∨„ᵐ(Ko¦”
For reference, that's what it should be
Which is longer than jelly btw because of word lists
How large are our dictionaries
It might be useful to have some compression tools builtin to the site
so you can click a button to compress instead of running code and typing the right element
@emanresuA 20k words in short, 220k in large
Our short has less than jelly, our large has more than jelly I think
@Steffan tabs coming in handy here :p
22:07
I've got a TS version of jelly compression if you want it
Different word lists though
And? That's easy to change
@lyxal Yeah, although it probably shouldn't be open by default
@emanresuA we haven't even finalised the word list yet
It's not hardcoded
Can just import from wherever it's exposed in Vyxal
 
1 hour later…
23:35
@VyxalBot I'm sorry little one
*suspends for 9999 hours*
@hyper-neutrino I got the sss and unsss working
problem is that the word lists aren't optimal
any insights on why that might be
is it word list length? Words contained?
uh
by not optimal do you mean decompression isn't returning the best result or it is compressing to something that is substantially longer than jelly
@lyxal oh nice, do you still need my review then?
Might be word list length
@hyper-neutrino yes
400K is chonky
23:37
lemme recall how sss works
@hyper-neutrino longer than what jelly produces
it's the same length using jelly words, but longer with the custom ones
@user *400k split into 2 lists
also 228k words
Oh
I ended up using the 5-list-intersection list
short length:24820
long length:202934
Note that the short list has words at length 5 maximum
otherwise you get
short length:50781
long length:176973
which is probably way too disproportionate
the order of words might affect sss slightly?
lemme test that hypothesis
yeah it seems like ordering the dictionary in decreasing order of frequency helps?? not actually certain if that's true
seems likely
23:45
Hufflepuff tree
because smaller numbers are smaller :P
wow
I mean the thing is it's essentially mixed base conversion and like
if you store a list of 3 numbers from 1-99 as digits, 10101 is not shorter than 19999
wait, my test might've been screwed up by the existence of one of my words in the short dict
though it should specifically only be affected by the frequency of the last/first (forget which) token
yes
oh yeah I am dumb and tested with a short dict word and a long dict word
23:51
yeah nvm it doesn't make much of a difference

« first day (1047 days earlier)      last day (728 days later) »