#voice-chat-text-0
1 messages · Page 551 of 1
Probably mountain climber. They're super strong.
no?
I think I fixed my stuff
Someone found the bug
It's such a simple bug
Took me 5 days 😭
I love programming man
it's awsome
that's sooo true lol
😭
Nix?
some 64GB microsd cards.
I got lucky 🍀 and wasn't shipped fake ones.
Python install manager
The installer we offer for Windows is being replaced by our new install manager, which can be installed from the Windows Store or from its download page.
@wind raptor
are you sure?
Hit that link
read the first 2 lines
Python is evolving
Change for the sake of change 🙂
It installs somewhere
easy to remember command
I'm a Debian monk, surviving years without updates
in theory yes, life without updates is so much better
in what theory
I think updates hitting eMMC soldered storage should be taken into account.
if an app hits eMMC hard, it should be dropped from the appstore
I mean I'm on the fence about updates. They're needed but at the same time I myself just don't want the latest features.
Smart TV Theory
and reality, a smart TV that never got an update runs better
A linux user home be like🥲
I see @crude nova
unexpected
Fanta was invented in wartime Germany due to import restrictions
@primal shadow
A New York bank linked to Fritz Thyssen, a German industrialist who gave early financial support to Hitler
-- https://en.wikipedia.org/wiki/Union_Banking_Corporation
I could go on. There are many more links. US companies had business interests and many many subsidiaries in Germany.
Though it looks like they might have done business during the war, and the government was unhappy
because as a whole ,teh US government was at war with Germanyt
the actions of free will having Americans does not dictate the policy of the nation
but doing biz at the same time
I was just on the way out.
@sudden palm 👋
!voice
Can’t talk in voice chat? Check out #voice-verification to get access. The criteria for verifying are specified there.
@frosty spoke 👋
@torpid crane 👋
Hey
@raven lance 👋
Hello
look for this -->
@wind raptor give me 5 then let's play
working
it's functioning
(it was reverse case: I joined with a VPN, noticed it's an okay server, unVPNd, then rejoined)
@fading eagle 👋
!voice
Can’t talk in voice chat? Check out #voice-verification to get access. The criteria for verifying are specified there.
With what do you require help?
im new to python and i want to know where to test my skills as a beginner
a beginner
Test, well, you might like to try your hand at some projects.
!kindling
The Kindling projects page contains a list of projects and ideas programmers can tackle to build their skills and knowledge.
thank you
@abstract bobcat 👋
Hii
@wheat cove 👋
Hello
@upper basin do you git tag your code frequently?
you can publish pre-releases if they're usable at all
0.0.0-a.426
I added pyproject.toml support for the thing I use to partially automate that
(for now just as an experiment to test whether the plug-in system works at all, haven't yet tested on real projects)
That's a nice feature.
used to be Cargo.toml-only
now partially supports pyproject.toml and package.json in addition to that
also will need to add [tool.uv.workspace] support
pyproject.toml still doesn't yet have tool-agnostic workspaces iirc
doing the tag automation also really helps with properly doing releases, from my experience
https://github.com/parrrate/opentagit/actions
(it's mostly done for that purpose)
did you really just have to use a calculator for 480/60
2014
@upper basin I see you're consulting the ancient wisdom
target directory for one of my projects is getting somewhat non-small
rustc situation
@upper basin you can try guessing how non-small that is
(Rust's build artifacts)
455
(gigabytes)
I somehow constructed a method chain so complicated that rust-analyzer crashed
it's only, like, 18 methods
(was)
now it's ~22
had to .boxed() a few times to unbreak it
out of 12TB available total, okay-ish
hehe
(I just never ran cargo clean there)
Gotcha
30TB if the other disk is included
mostly just result of this
it accumulates over time
for a single compilation, it's around 10GB disk + 10GB RAM
with 20~40MB result
all the source + intermediate artifacts
Rust caches those for incremental compilation
so that it doesn't take half an hour each time to recompile
almost exclusively because of Laravel or some other specific tech
(I don't learn languages from books)
currently adding an option to do --sign=false on the CLI in case someone actively refuses to sign commits and tags
yes
didn't hear the question
walkdir is the default approach, yes
there's not much point in asking upfront
just use it; if it doesn't fit, write yourself or use something else
(if you're already aware of walkdir existing)
if you want to customise it, just borrow the code and link to the original
Have you been trying the button from time to time?
yeah
The activity blocks, I'm guessing.
yeah i don't have Been consistently active for 1 hour 30 minutes in text chats
for some reason
@upper basin with ✨Agentic AI✨'s help, you could've optimised that 30 minutes to be 3 hours
Well, you're in the home stretch if you keep at it.
alright
@gusty thicket 👋
Copilot
🛬💥
ruff includes isort
import pandas as pd
import torch
import torch.nn as nn
from torch.utils.data import Dataset, DataLoader
from sklearn.preprocessing import StandardScaler
import numpy as np
from numpy.typing import NDArray
import matplotlib.pyplot as plt
from itertools import product
import tqdm
Best to not.
from itertools import product
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import torch
import torch.nn as nn
import tqdm
from numpy.typing import NDArray
from sklearn.preprocessing import StandardScaler
from torch.utils.data import DataLoader, Dataset
yeah it's ethically weird
i wouldn't do it basing it off on my personality
@upper basin
red red red
this almost looks like Red, the theme
Is it true that developers are bad at finance?
real estate developers?
SAAS
I reluctantly accepted using Error Lens, because underline things were too hard for me to spot
I have weird eyesight
I can see individual pixels when looking at them directly but can't find an error highlight on screen
You got the eye of the eagle
that case where "better than perfect" eyesight is completely useless
That's quick with whatever happened to pylance a week ago.
I went from 0 errors, to launching the window and suddenly getting 1k errors.
I also can't see moving objects well
@terse bone 👋
I wish I could recognise you!
which I discovered when training for big tennis
But I guess I have gold fish memory
Voice Lord mentioned
Btw, I hope you are doing well? @naive basalt
... this is so slow
also I should probably
- stop using Fedora for CI
- use a pre-built container with all the deps
I wish we could have fella back like we used to have back in those old days!
My English is poor
@somber heath we know who will do it in we know which language again
iirc that's why some things were added to the standard library
so presumably productive
@somber heath you still can solve even if you can't submit
one day I'll probably stop bundling the entirety of cargo as a dependency
and just use cargo-metadata instead
it finally compiled
only took 35 minutes this time
@timber rivet 👋
(high enough to actually cause damage to an animal)
Sometime today I'll finally receive voice perms I think
sup chat
Hello stranger
@patent mural we can hear your keyboard
💀
It's not like there's anything it's disrupting
At least it's something to listen to
tbh if you need something to listen to just listen to music, i don't mind the noises the keyboard makes but just thought i'd let him know
@upper basin more random git questions: do you use monorepos/big multi-project workspaces
hi
though I guess that distinction only makes sense once stuff gets published
@scarlet halo did u hear me breating?
no
@dim hazel wtf is that pfp😭 😂
Another exciting day of work
Sure,
These clacks ain't gonna click themselves, so I'm gonna clicky clack until I go home.
Sorry went to watch Dollmaker (Jane Fonda one). Umm, I don't think so. I guess if both quick and qmprs shared the same repo then that would be it, but I usually try to keep them small and separate for easier management.
That's mainly because all of my projects are just maintained by me. If I had a proper team, and was going to brand it as a platform, then yeah then I'd probably put it all in the same repo.
cirq I guess is the one I've seen do that.
!pypi cirq
Usually cirq-core is what people use, but there're ton of other modules there at the same level which I guess would make it technically a monorepo?
yes so the 6th edition is available
https://www.amazon.com/Learning-Python-Powerful-Object-Oriented-Programming/dp/1098171306
@balmy shuttle what did you do today?
very good. you like money?
lolll!! dont u?
i’m not a prada gucci armani guy. i’m the guy you see getting coffee at the gas station at 4am 🕓
Hi
@upper basin answer closed as duplicate
-- StackOverflow
Am I deafen?
no one is talking seems like (in reply to above)
someone literally closed a very old question of mine in math stackexchange after so many years for being too opinion based..this is why stackexchange is dead now
just write 15 papers on your opinion, now it's research-based
❌ submit an AI-written answer
✅ submit an answer backed by unreadable amount of AI slop
Does a Software Engineer(Frontend Focus) need to handle server Devops part?
the whole point of devops is that there is no "part",
devops is the discipline where development and operations are not partitioned based on who is responsible for them
if you (you being a person or a team with shared responsibilities) write front-end code and you are responsible for deploying your own front-end code, that is devops
otherwise it isn't, as far the strict way of it goes
@dry jasper
5 7
as tall as the handgun? no way
lmao
IQ tests are too long, I'd rather play minesweeper for 10 hours instead
@dry jasper IQ is just a way of converting ranking to a number
just a scale
I only ever took online tests
some random one which was shorter and mensa's online thingy
from what I remember, currently only shapes-based tests are considered somewhat reputable when applied to Europe/NA population
because culture-agnostic within that subculture
(if you want to compare the whole human population, it stops working correctly)
or rather just most of the world, not just two specific areas, but those are just a safer guess
hello there
mensa sounds like scam
typical fake "society"
@crude nova if the IQ test is not language-agnostic, it is absolutely useless
like completely discredited
even mensa is aware of that
"very high" is relative
introducing language is one of the easiest ways to make it about culture alignment instead of intellect
but I scored well enough to land in the gifted program
Internet IQ tests are junk, proctored tests ca n show the capacity to adapt
and learn
@upper basin the ones which don't show results out of a certain range are at least somewhat less cursed
like only 60~140
everything else is shown as <60/>140
Intelligence is the ability to learn though
That's the problem
You're seeing a knowledge test
what you know
your intelligence is not your knowledge
intelligence is capacity
knowledge is information
(this is not what mensa test does iirc)
((at least from what I witnessed))
(((at least not within those ranges)))
What? So the only way to test for intelligence is unsolved problems? That's madness
@upper basin the point with those test is in part in how they're structured in a way where you can't study for them, only train
iq tests are scams lol
to you, if you take it for the first time, it will be a "haven't seen this before" problem
as for usability of it, very dubious
they're about as valid as MBTI
Online ones are for sure
why would you ever pay for it
it's pseudoscience at best
IQ test results are good at predicting your IQ test results next year or something like that iirc
allegedly they seem to be somewhat stable, not pure randomness/feelings stuff
IQ test results can predict IQ test results
without useful connection to the rest of the stuff
and IQ itself is, again, just a scale
yeah height is really good at predicting if you're going to be taller next year...your point? just because it's not random doesn't mean it's meaningful
@upper basin So if I solve the unsolved problem, what's my score?
Is this new IQ test a pass fail?
You either solved the unsolved or you're a moron?
Do you have to have a badge that says high IQ?
there is no point, just like there is no point in IQ tests beyond that
Ooh, medium
"oh, look, repeatable results, this surely looks like science"
Best source of facts
you want papers? how about nicholas taleb? a mathematician?
would assume, yeah, something like this
fyi the papers are included in the medium
"look, we have statistics, they're, like, existent, and, like, they have numbers, and numbers, like, exist, this surely means something"
(from the second one)
another thing medium is apparently now famous for: personality attacks
Seriously
that 2nd article is fucking trash
Sounds like a salty writer did bad on an IQ test
last time I spent >minute on reading a medium "article" it was some critique of HTMX that ended in a series of ableist and transphobic comments
The ability to recognize patterns?
Patterns are everywhere
Literally
Pattern recognition is pattern recognition
Being able to connect the disconnected
That's why we administer IQ tests to children
because they haven't thought to study how to game them yet
IQ test for college?
lol did you understand the math
I'd probably call it dangerous/useless science rather than pseudoscience, unless you require ethics and meaningless results for it to qualify for science, which, sure, fair
the second article looks a bit like it tries to overwhelm the reader with mathematics
one random thing that I picked out from that was how "dead people score 0 on the IQ score" or something similar
if it meant something else, it was very unclearly written
but
the point is clear there
dummy inputs bias the results
(just, like, written poorly)
look, science: #voice-chat-text-1 message
I still haven't figured out the logistics of procuring a .249878 of a rat
why is MBTI still respected in any way
do people not get disgusted by such idea of classifying people in this way
Y'all are confusing knowledge with intelligence
@calm ginkgo intelligence != knowledge
IQ is capacity
IQ is not knowing
that misunderstanding has been already addressed
A newborn has an IQ, doesn't mean they know physics
"I have a metric for measuring brain power, but it doesn't actually apply to any tasks that use brain."
so clearly not "y'all"
Hello 👋
True
IQ has no meaning!
(with the implication being that obviously this also means education-based test are disqualified)
it has a very specific meaning:
for a specific test, it represents ranking
so yes
@haughty pier Could you explain these scales ?
No. It's like measuring the capacity of a container
and the container is empty
so the capacity is now irrelevant?
15 points represents a single standard deviation
or something like that
Right
Just because it's not yet used?
My car doesn't have as much HP as it does because I don't floor it?
145 is top .15% roughly iirc
The problem with IQ is that, just like any other test an applicant can score higher if it's 2nd or 3rd attempt
so the average is 100, right ?
@haughty pier iirc I did the maths, and that is around correct, yes
your guesstimate is close enough
or mathstimate
193 is one in 3.5 billion
Depends..
In a decent economy and a country with a near 100% literacy rate, yes.
But for developing countries it reduces to 80 or below exception is China
by definition
by how IQ is constructed
also, by construction/definition, it's normally distributed
smart is an unsmart word
Smart and Intelligence are different words, it is as different as want and need
smart and clever bring certain negative connotations to them
"being smart" and "being clever" are often negative comments on someone/something
Tbh, in my perspective a rich and healthy man is an intelligent man!
^ that sounds more like intelligencia than intelligence
... and that's where we hit the ambiguity on what "intelligent"/"intelligence" means
words are weird and somehow we ended up in a situation where "intelligent" is primarily about education, and "intelligence" is primarily about thinking capabilities
Idk how most psychopaths have good EQ and IQ
I wonder what IQ Delta Force operators have.
Watching too much Dexter and Hannibal.
psychopaths "with high EQ and high IQ" are the media-popularised types thereof
meanwhile there are many just random people who just can't comprehend empathy
That's really sad. Some are not even sociopaths, they just don't understand it and are confused.
psychopath + somehow learns to abuse social interactions + somewhat brainful
^ this combination really helps a person get more influence, and therefore we end up seeing more of them
This is how I judge if a man is smart or not :-
- Humble and Respectful both by action and word
- Meaningful talk beyond material existence and people
- Independent Problem Solving and Information retrieval
- Slow but steady increment in skills over time.
not being able to empathise is pretty much the definition of sociopathy (though if slightly reworded)
"humble by action" sounds dubious contextually
sounds like lack of daring
I don't know why, but when I think of sociopaths, it's always those cold assholes.
Such a horrible mental image.
weren't idiot and moron scientific terms?
that's the whole point there: popular image creates that selection bias
I learned java in 4 days, python in two weeks. I was useless despite understanding the code. Self-studying is invaluable but it's only fleshed out when you practice, and that doesn't need a high IQ. Just commitment.
"learned" being was able to write something you chose rather than doing something a tutorial told you to?
I think it's not just about IQ some people hate programming but they are good at other things like Art, Finance, Mathematics, etc
I wouldn't've considered myself as someone who finished learning Rust before I read through the entirety of reference and standard library reference
The smart designs, good practice patterns, those you learn from experience. It's not IQ.
That's fair.
and that's the only language for which I can say that
I haven't read through the standard library of Python
Nowadays, many are hesitant to learn programming because of the AI, not because they are bad at it
I think I read it, but honestly don't remember it.
I did skip parts because, to be fair~~, they're useless shit not worth reading~~
The Armed Services Vocational Aptitude Battery (ASVAB) is a multiple choice test, administered by the United States Military Entrance Processing Command, used to determine qualification for enlistment in the United States Armed Forces. It is often offered to U.S. high school students when they are in the 10th, 11th and 12th grade, though anyone ...
like stuff that got yanked recently
I'm gonna go sleep guys, cheers.
Somehow not smart enough to defeat Taliban?
smart enough to leave relatively early
That’s political thing
Russia used to be interested in US presence there, but now Russia and Taliban are allies
to the point of, random fact that's sometimes forgotten, NATO had a military facility within Russia
remember that weird world when Russia and NATO were briefly allies? me neither, but it did happen
from my big book of surreal facts
alongside the "did you know that Eclipse is newer than IntelliJ IDEA?"
despite what it looks
can we rule 10 the VC please
How come the Russian average in IQ is higher compared to the US?
I mean the US has much more per capita and facilitates?
✅ if your IQ is too low, you get shot
unfit for communism
Xerox is US iirc?
Norwalk, Connecticut
the what the where
Xerox are also the early PC people
(PARC)
ah, it was Kodak that massively failed on early tech
these, like, three companies easy to confuse with each other
Xerox had PC
Kodak had digital camera
IQ has nothing to do with getting hired, because humans hire people
Kodak buried it because it would (did) impact their sales
Nepotism and favoritism
aaaaaaaaa
The job market is not a skill based system
It's a who you know, not what you know, in most places.
and Xerox gave up on PC because <???>
Just because someone has an incredibly high IQ doesn't mean life didn't leave them them as an unskilled farmer
Trying to answer the question - can someone with a 97 IQ be a good programmer - ChatGPT says yes: https://chatgpt.com/share/69178947-7f2c-800e-9fda-439da68386d0
does US regulate favouritism in hiring?
And just because soemone is a fucking imbecile doesn't mean their wealthy father didn't give them a company to run
no
lol
FREEDOM
FREEEEE DOM!
Free dumb...
free to be dumb
yeah, maybe "gave up" is a bit too optimistic
The US does allow the pigs to refuse to hire high IQ individuals
Discrimination against the intelligent
becasue they won't be pigs
the what
how is this not Onion
Endless debate 😄
Because america
do they do another Godzilla story now
like they did last time stuff got way too surreal
this can't be real wth
Look at the date but it was
@fast gyro Hi
!paste
If your code is too long to fit in a codeblock in Discord, you can paste your code here:
https://paste.pythondiscord.com/
After pasting your code, save it by clicking the Paste! button in the bottom left, or by pressing CTRL + S. After doing that, you will be navigated to the new paste's page. Copy the URL and post it here so others can see it.
the google suggested further questions are truly something
.
Please react with ✅ to upload your file(s) to our paste bin, which is more accessible for some users.
MLP
when I see MLP there's only one thing I can think of
ye
Understood
see this google's random suggestion
one (CV) is about what was, the other is about what is to be
stylistically they might be similar, even sharing some parts of content,
but
the main point of and most content in them is different
Bye tc
@fast gyro ChatGPT has some suggestions, FWIW: https://chatgpt.com/share/69178d4b-f650-800e-8c03-43420be1229a - my personal opinion is to rewrite it from scratch without referencing that version, then compare them, make conservative, judicious edits, and submit the rewrite. But maybe this one has your genuine voice and if you believe it does, then submit it.
cat
oh thanks, good idea, but the SOP took me like 2 weeks to write it, deadline is in 2 days. so not sure if I WIll have enough time to rewrite another one from scratch.
another cat
so many cats
@somber heath said you after hearing that screech
that one is no longer the highest or loudest recording
I did do, like, an octave higher at some point
it's not torturing the vocal cords
but
it might be slightly torturing the ears
it's really low force applied
yeah, it's surprising just how loud it gets from a slight effort applied
it's excessively loud compared to how little strength both diaphragm and vocal cords are putting in
the only strain, that sometimes happens there, comes from me instinctively trying to quiet myself down, because I get startled by my own voice
it's kind of the same effect when you instinctively try to speak louder when the environment is noisy, but in reverse
the reaction bypasses any sort of thinking and that part is slightly dangerous
=> need to be fully relaxed~~ and prepared for ears to bleed from hearing this~~
another funny sound which, as far as I understand, is done differently, sounds more like whistling
this I was not able to reliably replicate since
Does anyone know how to make a stylized outline of an image using python
okay I think I've learned how to make it not as ear-splitting but still high and noisy:
block the loudness by doing it more nasally
(red underlined stuff being the moments where I forgot to do that)
volume just randomly explodes out of nowhere
just making sure whatever I learned to do a couple of years ago is still possible
the other interning
-5 to 255
literals are usually treated like that
*similarly
not guaranteed
your code shouldn't depend on this
!e
print((1,2,3) is (1,2,3))
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | /home/main.py:1: SyntaxWarning: "is" with 'tuple' literal. Did you mean "=="?
002 | print((1,2,3) is (1,2,3))
003 | True
!e
a, b, c = 1, 2, 3
print((1, 2, 3) is (a, b, c))
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | /home/main.py:2: SyntaxWarning: "is" with 'tuple' literal. Did you mean "=="?
002 | print((1, 2, 3) is (a, b, c))
003 | False
Python does some amount of optimisation/deduplication before it actually turns stuff into bytecode
iirc it sometimes does arithmetic as part of that
like turning 2 + 3 into literal 5
!e
print((1000 * 5) is (100 * 50))
a = 1000
b = 5
c = 100
d = 50
print((a * b) is (c * d))
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | True
002 | False
as far as I know, what happens there is that the first turns into literally print(5000 is 5000)
before evaluation
and it might even just do the is pre-eval too, idk
it's not undefined, it's underspecified
UB is something that a compiler/runtime can assume to never happen
or rather, compiler/runtime is allowed to do anything on UB getting encountered
!e py a = 1, 2, 3 b = 1, 2, 3 print(a is b) print(a == b)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | True
002 | True
these both refer to same literal, so you're not constructing them at runtime
!e
# I wonder if this changes anything
a = x, y, z = 1, 2, 3
b = x, y, z = 1, 2, 3
print(a is b)
print(a == b)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | True
002 | True
it doesn't
!e
# I wonder if this changes anything
x, y, z = 1, 2, 3
a = x, y, z
x, y, z = 1, 2, 3
b = x, y, z
print(a is b)
print(a == b)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | False
002 | True
this does
(in response to what I wrote)
((not VC))
chained assignments are generally weird in Python
I don't actually remember the order in which they happen and whether that order is guaranteed in any way
.
!e
class Test:
@property
def a(self): ...
@a.setter
def set_a(self, _): print("a")
@property
def b(self): ...
@b.setter
def set_b(self, _): print("b")
Test().a = Test().b = 123
:x: Your 3.14 eval job has completed with return code 1.
001 | Traceback (most recent call last):
002 | File [35m"/home/main.py"[0m, line [35m11[0m, in [35m<module>[0m
003 | [1;31mTest().a[0m = Test().b = 123
004 | [1;31m^^^^^^^^[0m
005 | [1;35mAttributeError[0m: [35mproperty 'a' of 'Test' object has no setter[0m
I don't remember how to do this
!e
class Test:
@property
def a(self): ...
@a.setter
def a(self, _): print("a")
@property
def b(self): ...
@b.setter
def b(self, _): print("b")
Test().a = Test().b = 123
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | a
002 | b
okay so this
I had a suspicion that it's done in this order
which I don't find intuitive
for me a = b = c is b = c then a = b
but that might be C bias
Rust just disallows this altogether
there a = b = c is b = c then a = ()
so the order of initialisation is something like this (ordered by which variable first gets the final value)
a1 = a2 = a3 = a0
I don't like this order, breaks the right-to-left assignment flow
a1 = a0
a2 = a0 # here it's not actually evaluated
a3 = a0 # here it's not actually evaluated
I guess the ordering might've been borrowed from what it's short for
a1 = a2 = a3 = a0
# same as
a1 = (tmp := a0)
a2 = tmp
a3 = tmp
(introducing tmp in case a0 is a more complex expression than just a variable)
()
!e
print(())
print(len(()))
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | ()
002 | 0
!e
() = ()
:warning: Your 3.14 eval job has completed with return code 0.
[No output]
and in Rust it just stands for a default empty type
materialised-ish void, within the type system
unlike in C/C++ where void can't be a value type
which is now considered a design mistake
some people also consider first-class reference support in C++ to be a mistake
(references being aliases not values)
C++ and Rust share a lot of the abstract model and general thinking, but Rust is just a better implementation of it
it's not even a just "opinionly better" type of thing any more
C++ standardisation participants are admitting that Rust is now something to catch up to, and a model for how things largely should be
C++ has same compilation time issues
- whatever is added thanks to excessive TMP
incremental compilation is largely good enough
by default, Rust doesn't recompile everything from scratch
most do nowadays
some don't just because their compilers are extremely fast
Jai recompiles everything from scratch in seconds from what I remember hearing
also Go's compiler is quite good at that
there are books which include quite a bit of entertainment value
https://zguide.zeromq.org/
watching talks is also an option
big long video
https://www.youtube.com/watch?v=F8sZRBdmqc0
Bryan Cantrill on "ARC: A Self-Tuning, Low Overhead Replacement Cache" by Nimrod Megiddo and Dharmendra Modha ( https://www.usenix.org/legacy/event/fast03/tech/full_papers/megiddo/megiddo.pdf )
Bryan Cantrill is the CTO at Joyent, where he oversees worldwide development of the SmartOS and SmartDataCenter platforms, and the Node.js platform. Pri...
@full dagger watch talks by:
Sean Parent
Jonathan Müller
they cover more fundamental parts of how C++ is supposed to be written
This talk explains why (and how) to implement polymorphism without inheritance in C++.
The talk contains many C++ tips and techniques, including many new features from C++11. During the course of that talk a key feature from Photoshop will be demonstrated and implemented.
NDC Conferences
https://ndc-london.com
https://ndcconferences.com
this is 100% must watch
it might take more than one attempt to fully understand
pre-processor is a C construct
it's a very simple string processing utility
C++ uses the same pre-processor
you can run the C pre-processor on a Python program if you dare to
@heavy zenith pre-processor is unrelated to the speculative execution
speculative branch prediction is a different concept
(I might've be misinterpreting what's being asked)
no, C very much does error out
you can't lookahead the function name
it's just a concatenation/string cutting tool
it can't really do arithmetic even
for arithmetic you need constexpr and consteval (C++)
not just before compilation
before tokenisation/parsing
whereas in C++ you can do this
if constexpr something {
/* ... */
}
which erases the if if something is false
(this happens during compilation)
if constexpr something_is_wrong_during_compilation {
/* do something that causes a compilation error here */
}
Zig has something like this too
you would often have that code inside something templated
so you check if something is true about a type that's passed in, and do things based on that
Rust, by design, doesn't allow constructs like this
^ I don't promote it, especially given statements like this
this is a limitation of Rust
specialisation, as a feature, is still not stabilised
and might never get stabilised
also doesn't (won't) have reflection stabilised
which Zig already has
and C++ is slowly getting
should be usable by C++29
current mechanisms Rust uses instead are based on raw token processing
complex + might be extending the language too far + quite a lot of effort already invested in other methods + internal politics of the Rust Foundation
one of the approaches to mimic reflection
https://docs.rs/mirror-mirror
General purpose reflection library for Rust.
the problem is this one is happening at runtime
currently, you cannot conditionally construct extra code based on type information, only based on code text (after tokenisation)
both C++ and Rust have some level of Lisp-like "just make it somewhat work" with regards to these compile-time metaprogramming functionalities
meanwhile Zig implements all this through first-class primitives
generics in Zig are just functions which take types as input and return types as output
...
to quote from this,
even though this wasn't specifically directed at C++
in Python, don't forget to from e when re-raising within except
so that it explicitly chains errors
in Rust there are .context methods on certain result types which allow to build a trace
also if the error gets tracing::error!d, that includes the information about all the nested spans it happened inside
in C++, idk, you just hope you'll be able to debug it from a core dump
in C, iirc Kernel guidelines say this: return 0 when ok, unless it's a check function, in which case return 1 if and only if the condition is true and there were no errors, otherwise (error or false) return 0
Zig uses error sets, similar in part to Rust, but there you can't attach any additional information to the error
want to do Result like in Rust -- invent your own
depends on how often you hit them
performance-wise, exceptions optimise for happy path, errors-as-values optimise for error path
ACCU Membership: https://tinyurl.com/ydnfkcyn
C++ Exceptions Reduce Firmware Code Size - Khalil Estell - ACCU 2024
Universally, exceptions are avoided in embedded systems C++ code, with a preference for error codes as out parameters, return values, or sum types like std::expected<T,E>. This avoidance stems from two main issues: the...
this explains, specifically for ARM, the whole path from throw to catch
@full dagger stored errors are a potentially terrible idea, unless you always force them to surface eventually
for debugging, look into tracing and stuff alike
this is not really error handling in this case
you need tracing/logging, if you want to later inspect it
this is in parallel to actually surfacing errors to users of your API wrapper
you need to rephrase the API in terms which more explicitly allow partial parsing
class MaybeParsed[Field]:
def get(self) -> Field: ...
class ParseError[Field](MaybeParsed[Field]):
def __init__(self, e: Exception):
self.__e = e
def get(self) -> Field:
raise self.__e
class Parsed[Field](MaybeParsed[Field]):
def __init__(self, field: Field):
self.__field = field
def get(self) -> Field:
raise self.__field
class Response:
field1: MaybeParsed[Field1]
field2: MaybeParsed[Field2]
also sounds somewhat dubious
is the API documented?
mildly recklessly maintained
presumably
whenever possible, don't make matters worse, abort early
unless the situation is totally hopeless and no one is going to fix it
you can just record everything you send and receive
if you build some structure (e.g. database) based on what your received, you can then re-compute the state from this log
which, for example, allows fixing a bug in how you process that data, or introducing new features
@frozen path 👋
@somber heath can you hear me okay?
How To Play Multiplayer In Minecraft Without Sign In! - Android, IOS, Windows, PS5, Switch
Looking to play multiplayer in Minecraft without signing in? Watch this video to learn how to play with friends on Android, IOS, Windows, PS5, and Switch!
My Client: https://mcpecentral.com/frybry-200k-client-for-minecraft-bedrock-edition
My shader: http...
@unkempt herald 👋
Mm
sorry i can't talk right now
hello
@kindred narwhal 👋
You'll need to visit the #voice-verification channel if you haven't, already.
@whole bear @nova flax 👋
@grave oracle👋
👋
i got an ad named conquer country
the ad content was about the best way to nuke someone
it looks like a pompom
How do you do @somber heath ?
I changed my name
I would like to learn python
Would you like to help me learn?
I have done some basic like variable, loop, control statement
Dictionary and set
Both are form of list?
{a: b, c: d, e: f} dict
{a, b, c, d, e, f} set
Yes
!e py print({1, 2, 2, 3, 3})
:white_check_mark: Your 3.14 eval job has completed with return code 0.
{1, 2, 3}
Hmm
!e py my_set = {1, 2, 3} result = 2 in my_set print(result)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
True
Set is form of Data Structure in which every element is distinct.
!e py my_set = set() my_set.add(1) print(my_set) my_set.add(2) print(my_set) my_set.add(1) print(my_set)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | {1}
002 | {1, 2}
003 | {1, 2}
!e py a = {1, 2, 3, 4} b = {3, 4, 5, 6} c = a.intersection(b) print(c)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
{3, 4}
!e py a = {1, 2, 3, 4} b = {3, 4, 5, 6} c = a.difference(b) print(c)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
{1, 2}
!e py a = {1, 2, 3, 4} b = {3, 4, 5, 6} c = b.difference(a) print(c)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
{5, 6}
!d set
class set(iterable=(), /)``````py
class frozenset(iterable=(), /)```
Return a new set or frozenset object whose elements are taken from *iterable*. The elements of a set must be [hashable](https://docs.python.org/3/glossary.html#term-hashable). To represent sets of sets, the inner sets must be [`frozenset`](https://docs.python.org/3/library/stdtypes.html#frozenset) objects. If *iterable* is not specified, a new empty set is returned.
Sets can be created by several means...
!e py a = {1, 2, 3, 4} b = {3, 4, 5, 6} c = a.symmetric_difference(b) print(c)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
{1, 2, 5, 6}
!e py a = {1, 2, 3, 4} b = {3, 4, 5, 6} c = a | b print(c)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
{1, 2, 3, 4, 5, 6}
!e
my_lists = [1, 2, 3]
in_lists = 2 in my_lists
print(in_lists)
my_tuple = (1, 2, 3)
in_tuple = 2 in my_tuple
print(in_tuple)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | True
002 | True
!e py a = {1, 2, 3, 4} b = {3, 4, 5, 6} c = a & b print(c)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
{3, 4}
!e py a = {1, 2, 3, 4} b = {3, 4, 5, 6} c = a ^ b print(c)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
{1, 2, 5, 6}
!e
"""Set is distinct and is always unordered mathematically because abc is same as cba in set"""
sample_set = set()
sample_set.add(2)
sample_set.add(4)
sample_set.add(6)
sample_set.add(8)
sample_set.add(9)
sample_set.add(1)
print("Sample set is : ", sample_set)
sample_set.pop(6)
print("Now the sample set is like : ", sample_set)
sample_set.clear()
print("Set is empty ro clear? : " sameple_set)
:x: Your 3.14 eval job has completed with return code 1.
001 | File [35m"/home/main.py"[0m, line [35m19[0m
002 | print([1;31m"Set is empty ro clear? : " sameple_set[0m)
003 | [1;31m^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^[0m
004 | [1;35mSyntaxError[0m: [35minvalid syntax. Perhaps you forgot a comma?[0m
!e
a = [1,2,3,4]
b = [3,4,5,6]
c = a ^ b
print(c)
:x: Your 3.14 eval job has completed with return code 1.
001 | Traceback (most recent call last):
002 | File [35m"/home/main.py"[0m, line [35m3[0m, in [35m<module>[0m
003 | c = [31ma [0m[1;31m^[0m[31m b[0m
004 | [31m~~[0m[1;31m^[0m[31m~~[0m
005 | [1;35mTypeError[0m: [35munsupported operand type(s) for ^: 'list' and 'list'[0m
!e
"""Set is distinct and is always unordered mathematically because abc is same as cba in set"""
sample_set = set()
sample_set.add(2)
sample_set.add(4)
sample_set.add(6)
sample_set.add(8)
sample_set.add(9)
sample_set.add(1)
print("Sample set is : ", sample_set)
sample_set.pop(6)
print("Now the sample set is like : ", sample_set)
sample_set.clear()
print("Set is empty ro clear? : ", sameple_set)
:x: Your 3.14 eval job has completed with return code 1.
001 | Sample set is : {1, 2, 4, 6, 8, 9}
002 | Traceback (most recent call last):
003 | File [35m"/home/main.py"[0m, line [35m13[0m, in [35m<module>[0m
004 | [31msample_set.pop[0m[1;31m(6)[0m
005 | [31m~~~~~~~~~~~~~~[0m[1;31m^^^[0m
006 | [1;35mTypeError[0m: [35mset.pop() takes no arguments (1 given)[0m
Oh
guys in fact, i use python to quickly solve our quiz
we have quiz about set theory
i use python to lazily solve bunch of questions at ones
primitive data types are simple
!e py my_dict = {"key 1": "value 1", "key 2": "value 2", "key 3": "value 3"} print(my_dict.keys()) print(my_dict.values()) print(my_dict.items())
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | dict_keys(['key 1', 'key 2', 'key 3'])
002 | dict_values(['value 1', 'value 2', 'value 3'])
003 | dict_items([('key 1', 'value 1'), ('key 2', 'value 2'), ('key 3', 'value 3')])
!e
unicode_values = {"a": ord("a"), "b": ord("b")}
print(unicode_values)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
{'a': 97, 'b': 98}
I'm confused
ord = ordinal
!e py prices = {"banana": 7, "apple": 2} print(prices["banana"]) prices["guava"] = 9 print(prices)
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | 7
002 | {'banana': 7, 'apple': 2, 'guava': 9}
dictionary is actually self explaining... its a dictionary, it can give you the decryption of a value on my example you have a character and its unicode value
!d dict
class dict(**kwargs)``````py
class dict(mapping, /, **kwargs)``````py
class dict(iterable, /, **kwargs)```
Return a new dictionary initialized from an optional positional argument and a possibly empty set of keyword arguments.
Dictionaries can be created by several means...
Gotcha
i use hash map for stuff like event calls where i use the key to tell what event type it is and the vaalue of it holds the function what the program should do on that event
Hmm
wdym by hashable?
hash map is just a universal name for dictionary
if you can convet it into intiger
I think they are different?
hash(object, /)```
Return the hash value of the object (if it has one). Hash values are integers. They are used to quickly compare dictionary keys during a dictionary lookup. Numeric values that compare equal have the same hash value (even if they are of different types, as is the case for 1 and 1.0).
Note
For objects with custom [`__hash__()`](https://docs.python.org/3/reference/datamodel.html#object.__hash__) methods, note that [`hash()`](https://docs.python.org/3/library/functions.html#hash) truncates the return value based on the bit width of the host machine.
!e
print(hash("Hello"))
print(hash("Hallo"))
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | 2838597395510877043
002 | -8191158806854847247
you see, a single change will change its output on extreme and unpredictable way
hash is a good way to store a value without needing to know its content
lets say as a key on hashmap (dictionary)
dictionary are just like a lists but you index the values using non-intiger key
but you are welcome to use intigers as key
!e
""" Dictionary are data structure that stores key value pairs"""
sample_dict = {"name" : "OpalMist", "Quality" : "Smart", "Aura" : 800}
print(sample_dict["name"])
print(sample_dict.get( "Quality"))
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | OpalMist
002 | Smart
!e
a = {1:"1", 2:"2"}
print(a.get(3, "0"))
:white_check_mark: Your 3.14 eval job has completed with return code 0.
2
Uh!
!e
a = {1:"1", 2:"2"}
print(a.get(3, "0"))
:white_check_mark: Your 3.14 eval job has completed with return code 0.
0
like this
!e py print({}.get("Nonexistent key.", "Default fallback."))
:white_check_mark: Your 3.14 eval job has completed with return code 0.
Default fallback.
I like how python is easy to write compared to Java
good thing about Java is quite good for DSA!
!e
public class swt{
public int findMax(int[] nums, int k){
int n = nums.length;
int sum = 0;
int l = 0;
int max = Integer.MIN_VALUE;
for(int r = 0; r < n; n++){
if((r-l+1)>=k){
max = Math.max(max, sum);
sum -= nums[l];
l++;
}
}
return max;
}
public static void main(String args[]){
int[] nums = {1,2,3,4,5,6};
int k = 3;
swt pack = new swt();
System.out.println("Output : " + pack.findMax(nums, k));
}
}
:x: Your 3.14 eval job has completed with return code 1.
001 | File [35m"/home/main.py"[0m, line [35m3[0m
002 | [1;31m [0m public int findMax(int[] nums, int k){
003 | [1;31m^[0m
004 | [1;35mSyntaxError[0m: [35minvalid non-printable character U+00A0[0m
I can't run java?
my bad
this was sliding window in java
numpy is super fast at its work tho
I decided to make a custom code instead
as I was solving DSA problem on leet code
medium level and easy level
Btw I guess I have to use dictionary in some way to understand it better
so far your explanation was good
and I learnt so many new stuff
I was confused about it
try to code more complex project and youll learn even more on just a short period of time
I think code is 5 minute work and real work is understanding question and creating a solution in mind
@whole bear Hello!
Take yout time opal
!e py text = "abcdefg" for i in range(len(text) - 2): print(text[i: i + 3])
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | abc
002 | bcd
003 | cde
004 | def
005 | efg
🤮
Hmm
O(n) is time complexity?
!e
text = "abcdefg"
for i in range(len(text) - 2):
print(text[i] + text[i+1] + text[i+2])
:white_check_mark: Your 3.14 eval job has completed with return code 0.
001 | abc
002 | bcd
003 | cde
004 | def
005 | efg
I guess it can't be lowered below O(n)
sliding window is used to reeduce the O(n^2) to O(n)
O(1) is quite rare
!e
my_set = {10, 20, 30, 40}
print(30 in my_set) # O(1) average
!e
my_set = {10, 20, 30, 40}
print(30 in my_set) # O(1) average
:white_check_mark: Your 3.14 eval job has completed with return code 0.
True
for i in range(1000): # O(1000) → O(1), constant
print(i)
How come this is O(1)?
As per GPT it is O(1)
Hm
Now I know how it O(1)
For n size it has to do just 1 operation
for it would be O(n) it has to do n operation for n size
I'm confused
damn
as per it is O(1)
I guess it is confusing
moment you replace it with n
it says O(n)
!e
for i in range(1000): # O(1000) → O(1), constant
print(i)
Yep
I guess AI is tripping
Time and space complexity is weird realem
@terse kayak Hey sulfur how are you?
Good
@whole bear 👋
Would you like to talk more about problem solving?
Halting problem?
You know so much about it Opal!
was criminal?
I think it is shame that human are judged based on the orientation, colour, ethnicity.
Human should not be judged based on those irrelvant factor for role!
Human are quire materialistic in nature, no other animal is as materialistic as human!
chimps and monkey are close relative to human
I heard a monkey can adopt the orphan monkey and they treat is their own!
Evil opal
they would be curious
and fascinated
but would like to repeat the behaviour with rock or pebble
Hover board is marketing
just like Gbps in internet speed
And if you do
10 Gbps is 1.25 GBps
pure marketing
Btw
Did you had lunch?
I just had lunch..
claiming refund is hard
they made it intentionaly harder to claim
how to be knowledgeable like u?
Opal?
I gotta do something , brb...
Have nice day everybody!
It was fun time here!
i mean, i do study and i forgot things so easily
yeah like putting some emotions, so it can be stored in unconscious
ahhh yes

how old are u?
im here listing
@dire nacelle 👋
dinner time see ya
im making a terminal based calculator that use prat-parsing to evaluate a expression/math equation
i need dont know what name i should give the file that holds the evaluator and parser
i have this structure
src/
|-app/
| |-main.cpp
|-ui/
|-???/
| |-evaluator/
| |-parsing/
:D
lol
numbers are in superposition
I clicked "Remove AI Chatbot"
it went away
for now
Fixing that right now actually.
a single person can't even read all the specifications required to implement a web browser
@wise loom android is linux
it's like saying "difference between ubuntu and linux"
inside yes
no, there is no extra qualification as "inside"
it is linux
@grave viper embezzlement 🚀
I discovered jujuliagava today.
tf was that
Made my day.
not a fan of gothic...
It's a sophisticated taste.
but hey it sure interesting
She reminds me of a young Jane Fonda.
!rule 4 @last ingot
4. Use English to the best of your ability. Be polite if someone speaks English imperfectly.
hello
its quiet here
im good
wby
lol
really quiet
what are u doing
@paper wolf @wise loom
click my pfp itll show u
k
oh i see
you're writing a...
calculator?
using a pratt parser
or a full interpreter
or even a compiler
a simple one
@lavish gazelle can you go dms i have question
@grave viper use starlette if you want truly barebones
or whoever was asking
I might've mispinged
FastAPI and Django are about different approaches to software
FastAPI is more library-oriented
there you build stuff out of small stuff rather than just having one big thing provide everything with fixed structure
you can use starlette along side fastapi, which is what ive done before