« first day (738 days earlier)      last day (1037 days later) » 

00:00
Dang
Well at least it isn't fatally erroring
well... infinite replace element in jyxal is no more
using semicolon as the separator messed node aliases up
so i had to find another character to do it
and thought ¢ does the job, as it kinda looks like something that could be a separator and it is the least used monograph in the corpus
sad, but had to be done
maybe ill add a digraph for it
ok, ¢ is now ÞI in jyxal
 
1 hour later…
01:25
!!/prod
@ľýxäł you are not a privileged user; ask someone to grant you permissions if you believe you should have them (user id: 521865)
!!/prod
Weird.
01:46
hmm
lemme check console
local variable repo referenced before assignment
!!/prod
yep that's the error that appears
 r = requests.post(
                f"https://api.github.com/repos/Vyxal/{repo}/pulls",
                headers={
                    "Authorization": "token " + STORAGE["token"],
                    "Accept": "application/vnd.github.v3+json",
                },
the {repo} here isn't defined - should it just be Vyxal/Vyxal? @lyxal
02:09
@hyper-neutrino it should
alright i'll just commit directly then and break my own rules kek
nvm i can't
hyper-neutrino opened PR #65 (Vyxal/VyxalBotSE) (Vyxal:update-prod-patch → Vyxal:master): [fix] repo is not defined for update prod command, should be Vyxal/Vyxal
@hyper-neutrino would you like the process expedited?
are you able to force it as org owner
No
But I have other ways
02:15
are you able to abuse alt accounts
but yeah sure, i don't see how this could be wrong lol
Maybe...
3 reviews is probably more than necessary; is it possible to require me to review it but not have a number limit otherwise?
@VyxalBot damn, who is this person
i definitely know it's not lyxal
Vyxal-Bot approved on PR #65 (Vyxal/VyxalBotSE): "I can't wait to start making PRs!"
Should be mergable now
02:20
hyper-neutrino merged PR #65 (Vyxal/VyxalBotSE) (Vyxal:update-prod-patch → Vyxal:master): [fix] repo is not defined for update prod command, should be Vyxal/Vyxal
hyper-neutrino deleted branch VyxalBotSE/update-prod-patch
@hyper-neutrino pulling new changes; I will restart in a few seconds if any updates are available
!!/prod
ah
have you been using too much JS lately
data=json.dumps({
    title: "Update Production (" + todayDate + ")",
    head: "main",
    base: "production",
    body: "Requested by "
    + str(message["user_name"])
    + " [here]"
    + "(chat.stackexchange.com/transcript/message/";
    + str(message["message_id"]) + ")"
}))
02:22
?
Oj
this won't work cuz a: b is not "a": b
?
Oj
*oh
Dang
Too much yaml
hyper-neutrino opened PR #66 (Vyxal/VyxalBotSE) (Vyxal:update-prod-patch → Vyxal:master): this is python not yaml/JS/whatever :P
Lyxal approved on PR #66 (Vyxal/VyxalBotSE): "Ah, sorry about that"
Not-Lyxal approved on PR #66 (Vyxal/VyxalBotSE): "Yeah, sorry about that. "
imma set the requirement to 2 reviews again
hyper-neutrino merged PR #66 (Vyxal/VyxalBotSE) (Vyxal:update-prod-patch → Vyxal:master): this is python not yaml/JS/whatever :P
hyper-neutrino deleted branch VyxalBotSE/update-prod-patch
02:23
!!/pull
@hyper-neutrino pulling new changes; I will restart in a few seconds if any updates are available
Vyxal-Bot opened PR #902 (Vyxal/Vyxal) (Vyxal:main → Vyxal:production): Update Production (25/04/2022)
Lyxal merged PR #902 (Vyxal/Vyxal) (Vyxal:main → Vyxal:production): Update Production (25/04/2022)
02:26
Also poggers that apart from the JSON and url, it worked first time
 
1 hour later…
03:34
peeks
in general, how do y'all decide what character to assign to issues with the label "element request" (e.g. #901)
03:48
We look at the type of digraph and then choose the most appropriate available letter
 
7 hours later…
10:55
@Seggan hey wait what specs did you use?
Because I don't consider the one in the symvy to be a complete thing yet
I think allxy's idea to add functionality that makes it somewhat Turing complete would be a good idea.
Things like self modification and macros/conditional macros could have a use for making shorter expressions
But it'd also allow for symvy golfing to be a thing which could draw more people into vyxal
Or at the very least improve perception of us as a community
Hi there @Razetime!
I'm a chat room admin now!
That means I can sudo cookies from the bot
!!/sudo make me a cookie
@lyxal [SUDO] Here you go: 🍪
See?
Isn't it great?
@lyxal hey, afternoon
@lyxal o.0
!!/sudo make me a cookie
@mathcat No, you sussy baka.
11:08
lyxal you hecker
not funny
!!/sudo heck
fsk
!!/sudo make me a cookie
@Razetime No, you sussy baka.
!!/ make me a cookie
@mathcat Here you go: 🍪
fghfhgdf
yay!!!
11:10
seems like you need to be a bot owner?
hmm maybe
11:47
You need to be defined as a bot admin for sudo cookies to work
that or you need to be pygamer0
wat
pygamer0 is cookie king
 
2 hours later…
13:30
@sporeball so for your example, "ends with" is a string operation, so it's a string digraph. That means it starts with ø. Then, we look for an appropriate letter. øE would make sense because it's Ends with. øe is already taken so that's a no-go. øė and øĖ are available, but ascii should take precedence over unicode. øP is unavailable, but would be preferred because "starts with" is already øp. So from the possible choices presented, øE would be the digraph of choice
that's the general idea of what character to assign to element request issues
anyone know what happened Wednesday?
14:34
@lyxal ones in the symvy repo + some of my own edits to it
i.e. i added arcsin/cos/tan and functions are prefixed
What library are you using for symbolic math for jyxal?
none
im directly compiling it to jvm bytecode
if you didnt see my earlier announcement about natively supporting symjy in the code instead of as strings then converted to functions via a symbolic math lib
1 message moved to ­Trash
there. pointless convo resolved
👏👏👏👏
15:59
what
fsk
> NOBODY CAN TRASH ME - mar 12 at 14:16 by mathcat
is a pinned message in trash
I am the original Trash king
16:24
oh no
we don't need yucky java symjy
ah, but it is SymJyxal, not SymJava
there actually is a symja lib
lol
so will symjy compile to symja?
symjy will compile to JVM bytecode
like jyxal
whatever that is
.class/.jar files
16:27
oh, jar files
the stuff that is eventually jit-ed to native code
i saw that
so then you'll have to reimplement the sympy logic
no
im directly compiling the short form
but symvy compiles to sympy code
symjy compiles to java bytecode
so you have to implement the logic
of what sympy does
unless you use symja
but im not doing symjy -> symja -> jvm, its directly symjy -> jvm
16:29
which means you have to implement the logic, right?
so i dont have to implement a bunch of logic, just the mappings for operator -> bytecode
Right
Which means logic
You have to implement the code
the logic is already done for me in the math lib
i am confused
i am confused
sympy does complicated math stuff doesn't it
yes
16:30
so symjy does too
unless you use smth like symja
so you mean java has a so great math library that it can do all your wonderful powers?
symjy is more of a short algebraic notation than a short symja, unlike symvy
@Steffan big math lib, yes
so it can solve comp equations and stuff, whatever sympy is supposed to do
only operation i had to implement myself so far is logb
solve integrals
ok
sympy.integrate I mean
well...
i havent taken calculus yet so idk what integrals or this dx stuff is
16:32
yeah idk really either
some weird mathematical weird stuff
but anyway idk if java can do all that by its own self
? java is 100% turing complete, ofc it can do anything by itself
you mean the stuff is prewritten for me?
yes
that's what I've been asking
because otherwise you have to write the code urself
why is my username so similar to urs
steggan
4
the only code i have to write is to parse the AST tree to the relevant operations. almost all operations/parsing is done for me
cool
gtg
o/
(idk if its \o or o/)
16:35
its o/
i think
 
1 hour later…
17:39
you could combine it into ø
17:53
Steffan153 closed issue #898 (Two-byte strings malfunction with backslashes at the end, Vyxal/Vyxal)
Steffan153 merged PR #904 (Vyxal/Vyxal) (Vyxal:fix-backslashes-at-the-end-of-a-two-char-string → Vyxal:main): Make backslashes at the end of a 2 character string include the backslash
Steffan153 deleted branch Vyxal/fix-backslashes-at-the-end-of-a-two-char-string
 
2 hours later…
20:16
can smbdy review github.com/Vyxal/Vylight/pull/6 ? should i just merge?
its cool how vyxal scala can run in the browser
we wont need pythonanywhere and stuff anymore
you can just use github pages, netlify, surge, etc
anyway the parser needs redone a little for the new transformer system
 
1 hour later…
21:25
To be fair, Python can also compile to JS
But yeah, Scala Vyxal in the browser and Jyxal locally will be great
21:41
question: would there be any interested in a relay between discord and here
@user just wait till jyxal get multi-target compilation lol
was toying with the idea but decided to shelve it for until after 1.0
btw for everyone who came here "more" recently, I am currently offering a 50 - 150 bounty for implementing at least 7 elements in Jyxal
3
and here is the list of yet unimplemented elements: github.com/Vyxal/Jyxal/issues/4
22:01
@Seggan Oh yeah Kotlin should let you target JS
Scala isn't needed anymore :(
the compiler yes, but i meant the compiled code to be multitarget
@hyper-neutrino +1 to that
@Seggan Whoa, how would that work?
Is there a JVM written in JS?
@user abstraction of the low-level constructs or even an IR
perhaps even compilation to native code
22:31
LLVM perhaps?
Does its own optimization too
for IR or native? IR will probably be a bunch of objects in memory
LLVM for native sounds interesting tho
but as I said, this isnt high on the priority list
For IR
And of course, LLVM would turn that into native
@Seggan Of course, it'd be extremely hard
The sort of thing I'd like to do as some undergrad research project
now i know who to hit up :)
tbh jyxal itself is just me exploring how far i can go with my skills and also learning about compilers and stuff
that doesnt look right
seems proguard has a weird idea of what is redundant and what is not
oooooooh
proguard actually creates a new method
its smarter than ii am
23:29
Steffan153 merged PR #6 (Vyxal/Vylight) (Vyxal:multiline-comments → Vyxal:gh-pages): add multiline comments
Steffan153 deleted branch Vylight/multiline-comments
23:51
Seggan created branch Jyxal/symjy_work
@Steffan this is basically how I read y'all's usernames sometimes
Seggan opened PR #7 (Vyxal/Jyxal) (Vyxal:symjy_work → Vyxal:master): SYMJY
wait, is the bot hardcoded to use Vyxal as the repo name?
> Vyxal:symjy_work → Vyxal:master
I'd have to check what's being substituted
Because if it's repo owner name then that's understandable
@lyxal steggan and seffan?
23:59
@Steffan pretty much
we should switch
btw im seggan

« first day (738 days earlier)      last day (1037 days later) »