#development
1 messages ยท Page 16 of 1
does this look right? I am trying to display the generated picture as .webp here:
okay, so I implemented the changes and it runs the prompt, but the image doesn't show? I see no errors in the Chrome inspector
Only two things I'm getting. The first one is probably a browser plugin
That second picture is just a warning, wonโt break anything
Here are the docs: https://platform.stability.ai/docs/api-reference#tag/Generate/paths/~1v2beta~1stable-image~1generate~1core/post
Am I doing it wrong?
dm
why is the javascript not in it's own file?
It seems to me like itโs just a school project or just quick introduction, and he confesses that heโs only just started learning so it might take a while to recognise and develop best practices
Anyone like my code?
"Python"
math = int(input("Math equation: "))
if math == "0 + 1":
print("1")
elif math == "0 + 2":
print("2")
elif math == "0 + 3":
print("3")
elif math == "0 + 4":
print("4")
elif math == "0 + 5":
print("5")
elif math == "0 + 6":
print("6")
elif math == "0 + 7":
print("7")
else:
print("idk")
๐
im pretty sure you can make that more simple
you could also just do it in a loop
I literally just started learning it a week ago, I have a full time job so I've been reading an hour or so a day
My local barber asked if something like this can be done so I thought I'd try it in JS
Originally used RShiny but it doesn't handle animations very well
i barley know how to program myself im just pointing out a more efficient way of doing it with less code
Thanks for the info
it's a joke cause they're brute forcing addition lol
Not really, every single line is fundamentally broken, if the syntax, typing and logic were not flawed it'd be useless.
I could explain why every line is wrong but I doubt there's any point because it's either a bad joke or you've never actually tried to execute a Python program you've written.
I'ma just say eval exists...
And it should absolutely not be used
hey i don't know if this is the right place to ask, I'm learning the cryptography library in python, and i ran into an issue using padding:
my server receives a padded message from my client, but when i use the unpadder, it returns an empty string.
i have initialized the unpadder like this:
unpadder = padding.PKCS7(HEADER).unpadder()
and call it like this:
msg_length = unpadder.update(msg_length).decode(FORMAT)
Hello guys !
I want to know if anyone here has already or would know how to bios mod a motherboard, because I have some high-quality HP proprietary motherboards, single-socket, and the problem is that if certain proprietary cables are not connected (USB port and other stuff), it asks you to press enter, which is complicated if you're remote. So, I need to bios mod the boards/or try to find a workaround and P.S. I'm too lazy to buy the cables in question, they cost a fortune, thanks.
Here, if anyone is up for it or has had a similar problem, I'd appreciate it.
it looks like you're supposed to call update() with the data, not with a length
the naming scheme might be a bit off, but msg_length is the padded data
idk haven't used this before but the documentation is showing it called like
msg = unpadder.update(msg)
msg = msg + unpadder.finalize()
to get the full unpadded data out
https://cryptography.io/en/latest/hazmat/primitives/padding/#cryptography.hazmat.primitives.padding.PKCS7
Thanks, i tried but when i add the finalize() it raises this error: cryptography.exceptions.AlreadyFinalized: Context was already finalized.
i could do with this information too, i need to update the bios on my hp socket t system to support dual core cpus
Is the firmware flash on a socketed chip? Depending on the era of the device, there are a variety of complications that may exist to stop you from modifying or analyzing firmware. Modern devices can employ a physically uncloneable function that raises the bar to a fairly extreme level.
what in the christ am i looking at
youtube shorts page
Interesting
this page
Yeah ik which one it is
doesn't unload the one you scrolled past
i expected a more... elegant solution for something as big as youtube
Idk seems pretty elegant
People do go back sometimes, and if you're on a shitty connection then it would take a while to load
talking about the whole thing in general
google does as google does
You need to provide a little more detail than that lol. There's a lot of different types of freelancing. Anyways ads aren't allowed here
is there a easy way for a raspberry pi running openPLC to force a another linux pc to launch a music file, and then force close the file later? They are just 2-3 meters away from eachother
Are you restricted in any way on how that communication is dealt with? I Think it makes more sense just to make a lightweight server software on the pc which plays audio via an api
Python fastapi library is pretty straightforward
ugh fastapi
good starter lib but really sucks for anything beyond starting on a project
Never heard of it. But it sounds like the guy I was responding to just needs a super simple way to get messages between machines
I mean for just a simple basic api, it does the job, flask is a nice one tho
the dead pixels are the exact points where they pointed on the screen.
think you might be in the wrong channel there buddy
Hey, does anyone know how to edit dlls ?
I'm trying to mod gta4 but I have two different dinput8.dll. One is for a general fix of the game (FusionFix) and the other one is to increase the vram available to gta4 (rockstar messed something up and if you have an amd card you may be limited to 512mb of vram)
Diff them in a hex editor then just combine the changes into one dll
alright thanks
attempting some video processing in the PL. 4096x3072 24-bit (max) uses a good chunk of resources
just missing a handful of constraints
Does anyone know anything about html
yes
and now it fails timing closure oops
stream interconnect also wasn't the right IP to use there
maybe something like this?
Updated and marked up. No testbench to validate it's actually correct yet
It at least achieves timing closure now. Still haven't made the testbench to make sure it's even doing what I hope
oh
what programing langureses are the best for each catergory
depends what you mean by category cause realistically you can program anything using any language if you have the will and courage
Even "best" is pretty subjective
they all do what u tell them to do
unless its machine learning related then they do what u teach them to do
Well, not all. Some have their limitations
Hey Guys My Name Is Matt Im 17 Years Old Learning To Code Ive Tried A Couple Times Before But Failed Due To Having No TIme Currently Im Learning The Basics Of HTML And CSS Then Im Going To Work More With Javascript Then Hopefully Using Javascript As A Base For My Learning If Anyone Wants To Give Me Any Tips Feel Free To Inbox Me
tip: for the love of God dont type like that
C is the best language at being C, for example.
If ya wanna code logic for the web, you either start learning WASM or JavaScript. One is more powerful and closer to hardware while the other allows you to interact with HTML and whatnot
Personally, I'd go JavaScript for the web ๐
You could learn a CRM like Wordpress or Django too if you wanna go fullstack and not limited to frontend/backend
I was more focusing on the frontend part. For the backend you can go all sorts of languages. Tho if you only want to learn one language and wanna do web, then JS is kinda your best choice. Love it or hate it (and no TS is not it's own language, it's just JS with sprinkles on top)
better learn typescript over javascript any day
basically its own superset as to how c++ is to c
Technically, it still is.
Although it's not strictly a superset, there's some stuff in C that's invalid in C++, but it does not change the fact that C++ still adds more to C.
Im pretty sure that means it's not a superset at all
I mean there was a dedicated wikipedia page for the entire debate iirc
The C and C++ programming languages are closely related but have many significant differences. C++ began as a fork of an early, pre-standardized C, and was designed to be mostly source-and-link compatible with C compilers of the time. Due to this, development tools for the two languages (such as IDEs and compilers) are often integrated into a s...
If the superset doesn't compile the original language then it cannot be a superset
Regardless of what the keyboard warriors on Wikipedia think
I mean i can use c++ compilers to compile c
depending on what I write ig
and which compiler I use
(looking at you, msvc)
Yes that's exactly the caveat that makes it not a superset lol
It does compile the original language, just an older version.
Which leads to newer versions of the original language capable of having more features than the not really but kinda sorta superset
You've just described "not a superset"
Heyyy
Quick question:
I'm trying to compile a c programm using a makefile, and it works perfectly when the code is in the same folder as the makefile, but when I move it to a different folder, the makefile being in a parent folder, the dependencies break, even though all files are properly linked
Does anyone know why?
is gtk.h a fil in your current project tree ?
wait no its not
no, it's a library I installed
Did you installed the gtk development files
yes
since it works when all files are together it appears so
here is the makefile ```
CC=gcc
CFLAGS=--std=c99 --pedantic -Wall -W -Wmissing-prototypes -ggdb3
LD=gcc
LDFLAGS=
DOCS=doxygen
LATEX=pdflatex
Files
EXEC=masterMind
MODULES= source/mainMasterMind.c source/gameplay.c source/display.c source/dataStructure.c source/humanPlayer.c source/computerPlayer.c source/score.c
OBJECTS= source/mainMasterMind.o source/gameplay.o source/display.o source/dataStructure.o source/humanPlayer.o csource/omputerPlayer.o source/ score.o
GTK=pkg-config --cflags --libs gtk+-2.0 -Wno-deprecated-declarations
Rules
all: $(EXEC)
masterMind: $(OBJECTS)
$(LD) -o $@ $(OBJECTS) $(LDFLAGS) $(GTK)
mainMasterMind.o: mainMasterMind.c
$(CC) -c mainMasterMind.c -o $@ $(CFLAGS) $(GTK)
gameplay.o: gameplay.c
$(CC) -c gameplay.c -o $@ $(CFLAGS) $(GTK)
display.o: display.c
$(CC) -c display.c -o $@ $(CFLAGS) $(GTK)
dataStructure.o: dataStructure.c
$(CC) -c dataStructure.c -o $@ $(CFLAGS)
humanPlayer.o: humanPlayer.c
$(CC) -c humanPlayer.c -o $@ $(CFLAGS)
computerPlayer.o: computerPlayer.c
$(CC) -c computerPlayer.c -o $@ $(CFLAGS)
well first thing I notice is that in your screenshot the gtk part of the command isnt present
Run that pkg-config manually. It appears to be empty here
ikkkk, but it's with this make file, sooo I've no clue why it doesn't show up
Makefile shell expansion syntax is $$(shell stuff) not backticks. Nvm backticks also works, but it does mean it will get repeatedly launched instead of evaluated once with $$ syntax
is this makefile a template you modified, or you wrote it ?
wrote most of it, based on a teacher provided template
the output seems fine
https://www.gnu.org/software/make/manual/html_node/Simple-Assignment.html
You may want to read this page and make the appropriate adjustments. Backticks for command substitution isn't described in the manual at all. The shell function is what the manual describes for calling shell commands
Simple Assignment (GNU make)
Shell Function (GNU make)
ok, thanks, i'll go through it
How do I learn filesystem of a unmounted device in linux / c ?
ok, solved, thank you
it was a mix of shell expansion and a flag not working properly together
of an unmounted device ? i dont think you can unless you mount it and query the device for what it is
What's the current state of ESP32 and does the ADC still suck ass?
Coming from Arduino and looking for something simple, cheap and wireless for IoT stuff. From what I seen the 12-bit ADCs on ESP32 WROOM-32 suffer from non-linearity, deadzones of 100mV in the lower and upper range, calibration, Vref accuracy.. worse than the 10-bit on the Uno and Nano. The ESP-S3 supposedly doesn't have the deadzones and has better hardware ADC calibration but can someone confirm that? Also how trustworthy are these devboards on AliExpress? I usually get everything else from there and have no issues so far.
Espressif ESP32 Official Forum
File systems may be identified by a magic number present at a fixed location (usually close to the start). For example NTFS has an 8-byte magic number NTFS at byte offset 3. Idk if every fs has got a magic number, but they usually do.
while i still think that test driven desingn (tdd) sucks monkeyballs. i found yet another reason to dislike it ,... im writing some tests atm (after the programming has been done, and the project is sufficiently interlocked with its own components that creating a test basicly means writing mocups for every other part of the program (and thus rewriting the whole program , but to facilitate a single test, and this for every part... now , if the project itself is complicated enough to warrant writing tests , and the mocups for those tests are in fact the program again minus the to test part... wouldnt that make the tests complicated enough to warrant writing tests for... ๐ no id like to see some TDD guru try to write tests to the 3 or 4th order of the actual code first before anything else ๐
you could always sidestep the deadzone by making a select that switches between a 1/100 and a 1/1 voltage divider , and if the final result on the adc would be 20mv and thus smaller than 100mv select the 1/100 divider and bump up the 20mv a 100fold resulting in an acurate voltage after the divider of 20mv
lsblk -I 8,259 -o name,path,partlabel,label,fstype,fssize
lsblk -o name,path,partlabel,label,fstype,fssize #for al devices
I hate tests as much as the next dev, but you didn't follow TDD. The complexity of adding tests as an afterthought like you're doing is exactly why TDD is so advised.
i didn indeed because i wrote the tests afterwards to begin with, but i tought when done well id like to see someone come up with this beforehand ... as i would have tought well actuall these tests need tests ๐
speaking of testing, I've finally figured out how to use the MPSoC VIP. Configured the interrupt controller and tested a manual interrupt
its like a watch, you can make a test for every gearwheel inside individually but once you want to test if they interact correctly , and see if the solar eclipse indicator is working okay , your mocking up the rest of the watch again to test that functionallity ๐
You might need to take a course on testing or something. Unit tests and functional tests are just separate efforts. If you need to mockup the entire rest of your system to do every functional test, you've done something really wrong.
The entire point of TDD is to ensure the very design of your product takes testing into account
since every small piece depends on one or more outcomes of other packages , and not linearly the easiest way to create the mocup would be to rewrite the original code that would have created the object in the first place... closest thing i can compare it to is to a sysfs subtree parser, you can write tests for each individual small parser, but as soon is you want to test if a node is parsed correctly , there is nothing to compare it against that is less cumbersome and less error prone than to compare it against another parsed version of the same original sysfs subtree
Please just take a course or study someone else's tests ideally relevant to what you're doing. I'm not experienced enough to try to teach you how to properly build tests or correct your misunderstandings. Test methodology is a very wide topic anyways so even if I was more experienced I may still not know the most ideal test method for whatever you're testing.
the point of TDD is that you can write tests for your interfaces/functions/whatever else BEFORE implementing the thing itself, so you already know the outcome your implementation should result in before writing any code
You dont always have to follow it, i dont even myself since I like to write code while I have an idea but I usually end up wasting time writing and testing the code if i write the tests after everything
Plus thers instances where writing tests after writing code can falsify your results because now the tests are most likely more tailored towards your implementation rather than something more generalised and less biasd towards a specific implementation
finally got it timed perfectly. Now to fuck it up with pblocks so I can use DFX
can someone help me with neovim on macos
it renders my theme properly when within tmux but breaks when not in tmux
It could be related to how Terminal handles things
try a different terminal application
sure that is one way to deal with the issue, but i was really hoping to find a fix without switching emulators
given i cant uninstall this terminal emulator anyways; can only install a new one in addition to this
Tabby is nice.
MacOS terminal has this weird thing where it doesnt like to provide the utilities that things often rely on
Electron terminal emulator is a little bit cursed ngl
There's also Hypr which is suposedly good
though my goto is Alacritty on Linux and windows terminal
If you're on Mac, I've heard good things about warp but I'm not sure of its compatibility with neovim
Iโve always used iTerm
Yeah I like iterm
On Windows I just roll with windows terminal. It gets the job done I guess
Ya, better than command prompt
Yeah, itโs awful. And actually more complicated than that if you want it to be more accurate
yeah the answer is do notttt use regex lmfao
or more specifically, use someone else's code, if they use regex then its on them
Cleaned top up a bit and split the static and dynamic portions of the design. Nearly ready to export the platform and move over to Vitis. Just need to create v_dfx's pblock now
wtf is this ๐ตโ๐ซ
Hey folks, has anyone tried to build up or use a combo of open source libs/tools for ID/Passport data scanning off images? Does not require to be rock solid as for KYC/legal reasons, i'm looking into it for auto-completion reasons for a form as a UI/UX aid for our users so they don't have to fill it in all by hand.
cursory search has not been very encouraging and for now we are using a 3rd party API. but things are getting pretty costly and we are thinking of building out something that we host ourselves.
There is at least 1 open source tool that can do it, however its macos only and hasnt been updated in 9 years. My guess is there are legal problems to making tools like that in the open
btw, if theres any x86-64 nerds in here: i tried to reimplement strcmp in assembly, kinda looks like this
section .text
global _ft_strcmp
_ft_strcmp:
mov rax, 0 ; make sure thats reset
mov rcx, 0 ; make sure thats reset
loop:
mov al, BYTE [rdi] ; take char at current position into 8 bit register
mov cl, BYTE [rsi] ; take char at current position into 8 bit register
cmp al, cl ; compare em
jne end_loop ; end if they dont compare
cmp al, 0 ; first char is null ?
je end_loop
cmp cl, 0 ; second char is null ?
je end_loop
inc rdi ; increment pointers
inc rsi
jmp loop
; useful info about movzx, and alignment issues ; https://stackoverflow.com/questions/22621340/why-cant-i-move-directly-a-byte-to-a-64-bit-register
; https://www.felixcloutier.com/x86/movzx
end_loop:
movzx rax, al ; this moves al (8 bit register) to rax (64 bit register) by setting the rest of the register to 0
movzx r10, cl ; same thing here cause we cant sub 64 bit reg with 8 bit reg (alignment)
sub rax, r10
ret
Is there any way to optimise this and kinda make it look better ? It does behave like the original function so far
oh there are like a bunch of libraries and tools that are easy to google. The problem is that they are more general OCR stuff and I have not found any more ID/Passport specific stuff. I don't need a ready solution, just somethign that's geared into that direction and might require training and stuff, but that's something we can do.
Is the possibility to develop internally possible ?
Yes, but not from scratch
We are a startup, we have limited resources, but a need for this amd pretty big one
So it's a cost/benefit/resource thing for us
right i see
Hey yall can you guys help me with this problem I'm having https://stackoverflow.com/questions/78462150/dialogueevent-not-registering-dialogueeventlistener-to-list-of-listeners
FPGA floorplanning
Sick
What would be the best way to e.g. check if something would be due in a PHP program live? Crontab?
Vivado is the best software in the world and doesnโt need any more bug fixes or optimizations, especially with their installer ๐๐๐๐๐
I love the way text turns into zalgo and/or vanishes if you have it running long enough. I've had no issues with the installer besides being slow as shit though
Installer pins any cpu to 100% while doing nothing
Itโs quite fun to watch lmao
But yeah I love vivado bugs ๐ฅฐ๐ฅฐ๐ฅฐ
$3k software and you have to use โunofficialโ official tools to properly use most features ๐ญ๐ญ๐ญ
Has vitis gotten any better with synthesis?
3kb error message lol
It's free for (most) Zynq-based devices (except RFSoC)
Thatโs lucky, I wish it was free for vup ๐ญ
if i saw this i would just give up
i would call this a "phantom error" because of how complicated it is
you should recode the bits it mentions or take the hard path and debug
i like to call them phantom errors because its either a ghost causing the error (making you recode it) or just really complicated
i made little snippets of code to plug into my browser and it had an error I couldnt debug, I recoded it and it worked!
had no i dea what it was, the code is the same as the error-ridden one.
ghosts man, they come when you are stressed
I mean it's kinda understandable lol. People working on $300-$2000 zynq boards would be seriously put off by $3000 software to program it. If someone is working on a $9000-18,000 vup board, $3000 software could be considered just a part of the cost.
Oh itโs certainly understandable but definitely annoying at times
I wish it was free for 7 series the most
I can understand newer parts but decade+ is still locked
Though I guess anything is better than ISE
Looks like most of the 7 series is supported, besides the entirety of the Virtex-7 lineup
Oh kintex-7 is in the free tier?
I totally thought that 325+ or so was paid, that would be super cool if they changed that
Limited kintex-7 support
Not sure why anyone would be considering such legacy devices when Alveo exists and is fully supported
Wow
is making games in python cursed
making anything in python is cursed, so technically yes, but don't let that stop you
i made 2 games
rock paper scissors and black jack, both with betting
message = input("what message do you want to print? ")
print(message)

generic OCR is your best bet if you want it for free, or google mlkit text recognition https://developers.google.com/ml-kit/vision/text-recognition/v2 if you talk about Android/iOS.
We explored this at work, and for actual ID documents, we ended up going commercial ๐คทโโ๏ธ
i mean mlkit can give you a pretty good description of the actual photographed item, and then you just need to figure out what is what
yes. that is how most of the software does it. Or have a CLI application run 24/7, with sleep() used to wait. But cron is imo a better choice if the cost of the setup is low enough - i mean setting up database/LDAP/messaging connections.
and for generic solution, you have to go for the base project used by mlkit - tensorflow
Got it, thank you
what's the word size of flash memory on the ATmega328P in bits?
It's 16-bits, but there's no time like the present to learn to read a datasheet.
Section 8
https://www.microchip.com/en-us/product/atmega328p#document-table
I did read the datasheet and found some information about page size. There is 256 pages in flash memory and 1 page equals to 64 words but I couldn't find word size. Section 27.5 on datasheet.
riddle me this batman: why does this code run without any issues
print("Encrypting your username...")
enc_username = rsa.encrypt(_secure_username_c.encode(), _secure_enc_key_publ)
print("Encrypting your password...")
enc_password = rsa.encrypt(_secure_password_c.encode(), _secure_enc_key_publ)
print("Done!")
print("Saving for comms...")
with open("fb.sgres", "w+") as f:
f.write(str(rsa.encrypt("True".encode(), _secure_enc_key_publ)) + "\n" + str(enc_password) + "\n" + str(enc_username))
_secure_usr_username = rsa.decrypt(enc_username, _secure_enc_key_priv)
_secure_usr_password = rsa.decrypt(enc_password, _secure_enc_key_priv)
but
this code:
```python
try:
with open("fb.sgres", "rb") as f:
lines = [line.strip() for line in f.readlines()] # Remove newline characters and skip empty lines
print("Lines read from file:", lines)
# Get the content of the file from the lines read
file_content = b'\n'.join(lines)
print("File Content:", file_content)
if len(lines) >= 3: # Ensure there are at least 3 non-empty lines in the file
enc_password = lines[1]
enc_username = lines[2]
else:
print("Error: Insufficient lines in the file.")
except FileNotFoundError:
print("Error: The file does not exist.")
except PermissionError:
print("Error: Permission denied to access the file.")
except Exception as e:
print("An unexpected error occurred:", e)
print(type(enc_password))
print(type(enc_username))
print("Encrypted Username:", enc_username)
print("Encrypted Password:", enc_password)
try:
print("Decrypting...")
_secure_usr_username = rsa.decrypt(enc_username, _secure_enc_key_priv)
_secure_usr_password = rsa.decrypt(enc_password, _secure_enc_key_priv)
print("Decryption successful!")
except DecryptionError as e:
print("Decryption failed:", e)
raise DecryptionError
except Exception as e:
print("An unexpected error occurred during decryption:", e)
raise Exception
_secure_usr_username = _secure_usr_username.decode("utf-8")
_secure_usr_password = _secure_usr_password.decode("utf-8")
does this:
Loading encrypted data
Lines read from file: [b'G\x8c\x1f\xe0y,Qq\xd2^\xe1\x9dD!I\xef7\x0b]6\x97J<\xf9\xf2i{\x8b\xd4\xa1\x050\x84\xec\x90\xd8\xf7\x1b\xbd\x92\x8d\x0eHg\x86\xd9\xb4\xb2\x13\x1f\x0e\xbdE\xf4+\xaa+\xd4\xeb\xcf\xfbM6\xbe\xbe\x11H-\x1fA\xbbp\xecNo\x8d\x84\xf3\\rM0|x\x13q!\xb3r\xea\xc7u\xa2N\xe2\xfb\xd9.|]\xf4\x01', b'\x8c\xb6\xca}\xa2}\xd4;\xcb\xe0\xbdlu!Fm?\x80"\xbd\xc1\t\xb8\x1bd', b'\xae\xbd{\x02$\x13\xf5\x01\x86\xca9\xf3=\x97\xd2\x08m\x86\xc4\xe8U\xc1\xe8\x10\x19\x12\xdb"ow@c\t3\xc65\x07\xb1\xc4\xcft\xfc\x90\xdc\xb5cy\xd06z\xfd\xfc\xbc\xb8\xeb\xe9\xeea6q&\xe7h\x85!\xdfH)\xdc\x16\xb0\xf9\xf1\x18?\xc1`\xeb\xa7L\x9bL\x12\xd3\xc6\xea\xf2\xe4\x10\x8f\x08rE/p\x89\xb0\x06\t\x14\xcevM\xf76)\x024\xb4\x05\xe1S\x07\xad\xa7\x05=I-26g\x84c\xe4Z\x9a\x84', b'\xb1R\x8e\x1d\xb5n\xed\x9c\x98\xe5\x1af\xca\xc6\x02\x9a\xd0\x0e\xb0AZ\xe7xs\x12\x8c\x05\xc5\xd0pf\xc8(\\`\x84\x06\x85u\xb9XE\xef\xc0\xa8KBr\xfc\x13\xe7T\x93\xee\x07\xbe\xde\xb9N\x102\r\x90\x8d\x0f\xd4\x80\xb5\x8f\x82\x17\xc2\x05\x9e\x1a\xe0\x0c\x9dlg\x14`\x99', b'\x10b\x92\xb5hr\xa5HF\x0f\xfa\xc6M\x80\xe7u\x9ea\xb5&\xce\xe5\x7f\xff\xc7\x8d\xd9\xffR`\xa3\xb4 \xa1\x91D6}\xef\xf0\x83\xb91q']
File Content: b'G\x8c\x1f\xe0y,Qq\xd2^\xe1\x9dD!I\xef7\.....
<class 'bytes'>
<class 'bytes'>
Encrypted Username: b'\xae\xbd{\x02$\x13\xf5\x01\x86\xca9\xf3=\x97\xd2\x08m\x86\xc4\xe8U\xc1\xe8\x10\x19\x12\xdb"ow@c\t3\xc65\x07\xb1\xc4\xcft\xfc\x90\xdc\xb5cy\xd06z\xfd\xfc\xbc\xb8\xeb\xe9\xeea6q&\xe7h\x85!\xdfH)\xdc\x16\xb0\xf9\xf1\x18?\xc1`\xeb\xa7L\x9bL\x12\xd3\xc6\xea\xf2\xe4\x10\x8f\x08rE/p\x89\xb0\x06\t\x14\xcevM\xf76)\x024\xb4\x05\xe1S\x07\xad\xa7\x05=I-26g\x84c\xe4Z\x9a\x84'
Encrypted Password: b'\x8c\xb6\xca}\xa2}\xd4;\xcb\xe0\xbdlu!Fm?\x80"\xbd\xc1\t\xb8\x1bd'
Decrypting...
Decryption failed: Decryption failed``` (i shortened some of the output, so it can fit into 1 message)
27.5 is boot loader lock bits. That's not the place to be looking for the word size. Section 8.2 is in-system reprogrammable flash memory.
Since all AVR instructions are 16 or 32 bits wide, the Flash is organized as 2/4/8/16K x 16
Section 8.2 Titled Clock Sources
Please use the up to date datasheet from the manufacturer if you're going to discuss it
think the issue is the conversion to string before writing to file, you should either write it in binary to the file or encode it in some way you can decode it again after reading it out of the file.
like if you want the file to remain ascii, you should encode it with something like base64. unless python has some way to decode the string representation back idk if it does
ah i guess there is ast.literal_eval(bytes_string) if you want to convert back from what you've written to the file into a bytes object
depends on the type of game , and what you are willing to ofload to faster langs... like imagine a (whats the genre collecting things in the 3d world that is persistent with a million items and combinations) where every item is having to be programmed in c , it would bump the price to new hights (all them c programmers arent cheap) while with python doing alle the items would be really fast to develop comparitively , the modding community would love it , and you can still do the core engine and network code in c ..
but if its as good as it promisces to be maybe mojo is fast enough to also do that part ?
mojo is the name for python++ except that it makes things faster not slower :P\
Lol mojo
you're writing in UTF8 and reading as bytes, should probably sync those up
I'm not sure how literal eval factors into this though?
just because my eye fel on it , but did you know f.readlines() is the default so py lines = [line.strip() for line in f.readlines()] is identical to : py with open(file,'rb') as f: lines=[l.strip() for l in f] and that this wont remove empty lines as far as i know. but this will: ```py
with open('main.py','rb') as f:
lines=[l.strip() for l in f if l.strip() != b'']
because str(bytes) destructively converts bytes to a string repr of it
It's the same as str.decode / bytes.encode
but it would be more proper to do it like e.g.
with open("fb.sgres", "w+b") as f:
f.write(base64.b64encode(rsa.encrypt("True".encode(), _secure_enc_key_publ)) + b'\n' + base64.b64encode(enc_password) + b'\n' + base64.b64encode(enc_username))
with open("fb.sgres", "rb") as f:
lines = [line.strip() for line in f.readlines()]
dec_password = base64.b64decode(lines[1])
dec_username = base64.b64decode(lines[2])
I'm on mobile so reading code is an awful experience but it seems to me that you're writing an encrypted blob and then trying to read it as though individual sections are encrypted rather than the whole thing
you can also benefit of my remark so it seems ,. f.readlines is the default so you can drop the .readlines()( ๐
not that using it is wrong, but les to type less to read is easiier to plow trough code
>>> pw = "mypw"
>>> bytes(pw, 'utf-8')
b'mypw'
>>> str(bytes(pw, 'utf-8'))
"b'mypw'"
>>> savedpw = str(bytes(pw, 'utf-8'))
>>> import ast
>>> ast.literal_eval(savedpw)
b'mypw'
>>> savedpw == bytes(pw, 'utf-8')
False
>>> ast.literal_eval(savedpw) == bytes(pw, 'utf-8')
True
he's saving string literal bytes to the file I think
instead of just the bytes
literal eval has no place here
is there another way to convert the string "b'mypw'" back to the bytestring b'mypw' ?
yeah python3 and bytes ,... some things were easier to grasp with python2
but yes using eval is janky which is why i suggested b64 encoding it before saving it to the file instead
so it remains ascii but doesn't need eval to decode
you should not have this string in the first place, which is one of the first things wrong here
yes that's what i said, "the issue is the conversion to string before writing to file"
right, but like i said thats one of many issues
the biggest one is that he's trying to read an encrypted blob as if it were unencrypted
actually i just took a look on my laptop and i might be wrong on that
curses on mobile formatting
(honestly id just do the one blob instead of multiple, but whatever floats his goat i guess)
with open("fb.sgres", "wb") as f:
f.write(rsa.encrypt(password + '\n' + username, _secure_enc_key_publ)
with open("fb.sgres", "rb") as f:
password, username = rsa.decrypt(f.read(), _secure_enc_key_priv).split("\n")
this works too
i just like ascii files so you can see what you did without hexdump lol
not that its that important but that way if you know the username you halfway there of decrtypting it ,.. ?
i wouldnt encrypt a username myself but that seems to be what theyre going for
looks like they're hacking together some kind of credential manager
better than just keeping them plaintext like i do lol
marginally, if the key is right there
fwiw those tend to leave things like usernames/websites unencrypted too, to provide them to the OS for tooltips lol (not always usernames, but certainly other metadata)
huh didn't know about this encoding this is kinda handy
>>> b'\x35\xe5\xFF'.decode('iso-8859-1')
'5รฅรฟ'
>>> b'\x35\xe5\xFF'.decode('iso-8859-1').encode('iso-8859-1')
b'5\xe5\xff'
oh that's just latin1. but yeah apparently it can encode all of 0-255 as characters that can be decoded back
ty guys
this worked
I just needed to modify the code a bit
with open("fb.sgres", "wb") as f:
f.write(rsa.encrypt(_secure_password_c.encode("utf-8") + b'\n' + _secure_username_c.encode("utf-8"), _secure_enc_key_publ))
with open("fb.sgres", "rb") as f:
_secure_password, _secure_username = rsa.decrypt(f.read(), _secure_enc_key_priv).split(b"\n")
Would this be an appropriate channel to chat about REGEX?
Probably, what you wanna know? I have very limited knowledge about it though
Why is it that every html template that uses bootstrap that i saw so far end up looking like a clusterfuck
The code itself that is
The website looks like well... A bootstrap website
Did they not just days ago deprioritize flutter's multiplatform goals?
Maybe mobile and web is close enough to some. RIP desktop apps
it exists for desktop too. and Kotlin multiplatform and flutter coexisted for some time, it's different use cases
Flutter is more rapid, Kotlin multiplatform is more native (if you want to go that way)
I'm aware of that, I've used it. It appears you haven't heard the news of the flutter team being restructured and desktop support improvements being removed from the roadmap a couple days ago
Or rather two weeks ago
well restructuring is more or less "we get rid of expensive employees in US and push the work to lower paid EU counterparts"
if anything, the EU counterparts are usually just as expensive, think further east or south
not really, we talk about $140k for US vs 70k in EU for same position
Python team was asked to move to Germany, Flutter is going to Mexico team etc.
And in the end, for same position in US vs EU, you can usually expect salary drop of 30-50% for literally same job, before taxes even come into play
I think if its the same people on the team they'll also need to pay US taxes
I forget exactly how that works
since when can websites block inspect element and right clicking?
they technically moved devops positions to another country. Income taxation of person who moved to that other country is sadly a personal issue of that employee, the employer doesn't have to pay anything in US.
right click since forever (oncontextmenu). Inspect element can't be blocked, but they can make your life miserable for a while by calling the debugger call which stops execution. But you can disable debugger globally for the tab
I mean the employee would have to pay US taxes if they're a US citizen even if they live and work in another country I think (unless they renounce their US citizenship/plan on never going back to the US)
i found a way around, control shift i didn't work but control shift j did
wait you meant some keyboard shortcut for it ? i never did inspect element via keyboard, it was either mouse context menu or by clicking on the icon on debugger panel ๐
oh ok. also if anyone is curious this is the website that blocks right clicking and control shift i shortcut https://www.newsunzip.com/articles/what-its-like-to-work-inside-nycs-windowless-long-lines-skyscraper/
yeah, they literally just override oncontextmenu for right click
and this for onmkeydown:
document.onkeydown = function(event) {
event = (event || window.event);
//alert(event.keyCode); return false;
if (event.keyCode === 123 || event.ctrlKey && event.shiftKey && event.keyCode === 73 || event.ctrlKey && event.shiftKey && event.keyCode === 75) {
if (show_msg !== '0') {
show_toast('inspect_elem');
}
return false;
}
if (event.ctrlKey && event.keyCode === 85) {
if (show_msg !== '0') {
show_toast('view_src');
}
return false;
}
}
a completely idiotic naive approach, as if one can't call those functions via the browser menu through mouse
Havving used quite a bit of langs and of course they are subdividable in many ways , but the one im curiouse about is this split where one is way more common and popular then the other, yet i dont see the benefit for it, so : { } or no {} there are only 4 languages that i know of thta dont use the {} as a code block designator : python, VHDL,COBOL,BASIC. and 3 of that are not really applicable to computer programming today.
Weird thing is : Other than opening up allot of discussions for where they should go , i dont see the benefit of using them. Okay the other option is forced indenting like python does so you cant do: ```
def function(){
verry='verry'
flat='structure'
}
so since every other lang uses them (while they could be used for something else (thers never enough symbols to use) they must geve a huge benefit some how or enable you to do stuff theat not using them prohibits. problem is that i cant seem to find any. nowim defenetly not the lifetime hardcore programmer that some of us are , so what are the benefits for using {} ?
i will never accept languages which use indentation for code block demarkation. It is just way too prone to errors
using indentation for such task is a bad idea in a world where we can't even agree if we use tabs, 2 spaces, 4 spaces or 8 spaces ๐
plus using {} allows you to do crazy oneliners, of course at expense of readability
i more often forget to open the { then i forget to indent ๐ (closing the } is fine (as i automaticly // blockname behind it to indicate what im closing
you can't arbitrary create a scope in python can you? everything is just function scope?
although i rarely use that even in C++, usually scope my variables to the function anyway
some ideas just catch on and nothing else provides enough of a benefit to justify switching to it i suppose
javascript has to have them though since it's sent over the wire as text
i think i've have the final sane answer on that but i wouldnt force anyone to change his or hers habits on that since there is a way currently in use that all of them are vallid as long as you dont mix them.. (ps: the only sane way i found is ๐ it makes sence to use a diffferent char for indent and for allignment , and the only guarantee allignment keeps alligned is by using spaces for that , so tabs left for indentation) (there are more reasons but...)
also how is indentation more prone to errors ?
with {} you have clearly demarkated start and end. with indentation, let's say you have something like
python code line 1
python code line 2
python code line 3
python code line 4
you go stupid and you delete the indentation on line 3
python code line 1
python code line 2
python code line 3
python code line 4
poof, completely different code
yeah you mean bytes or bits over the wire i guess wich in turn makes it not really a matter of have to since you can use anything to indicate end of block at that point. but i get your point ,since javascr has to be typed ... but even then it would only be a problem if a space/tab were a bunch of zeroes indistiguishable from datafail
with {} you do:
code line 1 {
code line 2
code line 3
code line 4
}
you delete the indentation, code executes still same:
code line 1 {
code line 2
code line 3
code line 4
}
and i mean unintentional change in code which changes the execution, because you went from one valid code to other valid, but functionally different code, just because you deleted some indentation by mistake
hu ? im missing that because in the one you decide do delete the significant part and in the other the insignificant part , you should delete the { or the } there ```
code line 1 {
code line 2
}
code line 3 {
code line 4
}
code line 1 {
code line 2
code line 3 {
code line 4
}
```
that and it's at least 3 extra bytes per line (or does it accept tabs?) to transmit, idk with javascript they always remove all whitespace before sending it
seems equally prone to errors there
tab is just a single character just like (any/most) others
whitespace would probably compress out pretty well if you like gzipped it though maybe that's not a big deal
that is why you can havve your editor pick how many space to use for a tab
no, because your case is invalid code. In case of Python, that is still a valid code
in the end, it is down to personal preference, you can always just use python if you like indentation block languages ๐คทโโ๏ธ
okay : c will fail to compile, javascript wil just deal with the invalid code and run with it
yeah my point was also more like if we'd have {} availeble for other syntax stuff we would not have to doe the weird constructions we now sometimes do
been a minute since i worked with javascript but i thought it crashes if a document is invalid
cause it has to parse the whole document first even if it doesn't run
with the amount of javascript used today,.. that would have made the web unusable
why would C fail ?
#include <stdio.h>
int main() {
double a, b, product;
printf("Enter two numbers: ");
scanf("%lf %lf", &a, &b);
// Calculating product
product = a * b;
// %.2lf displays number up to 2 decimal point
printf("Product = %.2lf", product);
return 0;
}
and
#include <stdio.h>
int main() {
double a, b, product;
printf("Enter two numbers: ");
scanf("%lf %lf", &a, &b);
// Calculating product
product = a * b;
// %.2lf displays number up to 2 decimal point
printf("Product = %.2lf", product);
return 0;
}
are functionally identical programs
def multiply_numbers():
num_1 = float(input("Enter the first number: "))
num_2 = float(input("Enter the second number: "))
product = num_1 * num_2
print(f"Product of {num_1} and {num_2} is {product}")
and
def multiply_numbers():
num_1 = float(input("Enter the first number: "))
num_2 = float(input("Enter the second number: "))
product = num_1 * num_2
print(f"Product of {num_1} and {num_2} is {product}")
are not (not sure if the print statement would crash or not, i did very limited python coding in my life)
document parsing is browsers job. Javascript will not crash because of bad HTML structure, as it is not going to interpret it in the first place. It will work with whatever DOM structure browser created from the HTML document
code could fail if it expects some elements to be present and they aren't, but that is another story ๐
i mean in his example where he had unbalanced braces
hi there, so ive been working on php lately and right now, im encountering this error
Fatal error: Uncaught ValueError: Path cannot be empty in E:\5thSemesterWebDevProject\imgUpload.php:26 Stack trace: #0 E:\5thSemesterWebDevProject\imgUpload.php(26): getimagesize('') #1 E:\5thSemesterWebDevProject\myassignments.php(2): include('E:\\UniversitySt...') #2 {main} thrown in E:\5thSemesterWebDevProject\imgUpload.php on line 26
where i change the file name in query selector in this
let input = document.querySelector('.imageFile');
the code on that line 26 in pfp is this
$check = getimagesize($_FILES["image"]["tmp_name"]);
does anyone know what the issue could be here?
unbalanced braces would fail in JS too
idk about in the browser i would think you'd get a console error about the script tag being invalid, i was just thinking about in the nodejs interpreter
browsers are obscenely fault tolerant with bad documents though yeah
you should never manipulate the file input, you probably set it to invalid value and nothing got uploaded. print your incoming data, but most likely you uploaded nothing
yeah thats what seems to be happening. without the file name change, it works fine. but changing the name generates that error.
the reason i need to change that name is because of the input form i already generated
that "browse" and dragover function relies on that name alteration.
that must be some weird library
oh btw, the original name of that element is ".uploadForm input" and im changing it to a specified name mentioned above
is it just a transient state while user is editing things? don't call the function until interaction stops and catch and handle the error so it doesn't bubble up
https://developer.mozilla.org/en-US/docs/Web/API/HTML_Drag_and_Drop_API/File_drag_and_drop this is how you do it, you process the obtained file and submit it yourself, for example by converting it to base64 or uploading the blob ?
you prolly already know this but its actually html that is fault tolerant, well not really it has defined rules , what it does in such a case.. liek with : html <p>some text<br /> <h1> my title</h1> some more text </p> if you for instance forget to close the p tag it wil do it for you like : ```html
<p>some text<br />
</p><h1> my title</h1>
some more text
</p><-- will bi ignored then
it will make the second part a floating piece of text not in tags , because a <h1> tag may not be inside a <p> tag so it will close it for you ๐
srry was wrong it will also reopen the ptag right be fore the other one so : ```html
<p>some text<br>
</p><h1> my title</h1>
some more text
<p></p>
ig i will have to redo it
and people wonder why their css doesnt work for the second part of the text ... ๐
(and most teacher arent aware either
(i even remember the primeagen being baffled lol )
The ultimate example of trouble from digital promiscuity. HTML is more tolerant than C because it has to be. Professor Brailsford explains.
Deep Learning: https://www.youtube.com/watch?v=l42lr8AlrHk
Secure Web Browsing: https://www.youtube.com/watch?v=E_wX40fQwEA
AI Game Playing Challenge: https://www.youtube.com/watch?v=5oXyibEgJr0
Convolut...
after more iterations than I care to admit, got the video top level architecture done and achieved timing closure
dont understand a single thing here but it hurts my brain
FPGA video processing, with a static and dynamic parts. Though until I can figure out how to setup the Vitis platform, the dynamic part is just a loopback
Static is fixed at platform creation and consequently at boot in the device. It has scaling, chroma resampling, color conversion, and change detection. Along with the video read and write DMA to the system's DDR4.
Dynamic can be reconfigured live after boot.
the static:
dynamic:
Hey everyone, anyone good at typescript, nodeJS and SSE? I got an issue with that and i just can't figure out the issue.. going crazy :D
On big codes that demarcation is very much no longer clear
Where statements become nested to oblivion you very much rely on indentation anyway
I've seen some really crappy java codes with no attention to indentation be absolutely unreadable
while possible, most will at least try to move big nestings into functions instead at least
Does attiny85 use SPI for programming or it's a custom communication?
Yes, the ISP is SPI. page 5
But it uses 2 input pins?
SII(Serial Instruction Input)
SDI(Serial Data Input)
I was looking at different programming named High Voltage Serial Programming. You are right it's SPI.
i finally deployed my first dynamic website.
But fuck i focused on the backend so much while learning i neglected the mobile front-end
at this point i'll need a complete CSS re-write
Also still need to set up a CDN for the large file size data
Can't be storing vids in an SQL database
Not trying to be rude, just trying to shift how you ask questions because of the article.
Yes you're right.
After a lot more debugging i found out it was an issue with my reverse proxy...
Yeah. Itโs a pet peeve of mine. People do it at work or on chats. Iโve been guilty of it too. Hopefully you enjoy the article though!
nah your totally right to do so, I do the same our time is limited, and some people just scroll through answer once and leave again, which is their good right.
I also do the same. want another one? https://nohello.net/en/ I use this one too ๐ since it takes too much time.
NoHello.net, DontAskToAsk.com, and https://xyproblem.info/ are all some of my favorites.
My all-time favorite has got to be this article from 20+ years ago.
http://www.catb.org/esr/faqs/smart-questions.html
Before You Ask
Before asking a technical question, do the following:
- Try to find an answer by searching the archives of the forum or mailing list you plan to post to.
- Try to find an answer by searching the Web.
- Try to find an answer by reading the manual.
- Try to find an answer by reading a FAQ.
- Try to find an answer by inspection or experimentation.
- Try to find an answer by asking a skilled friend.
- If you're a programmer, try to find an answer by reading the source code.
When you ask your question, display the fact that you have done these things first; this will help establish that you're not being a lazy sponge and wasting people's time.
Use meaningful, specific subject headers
On mailing lists, newsgroups or Web forums, the subject header is your golden opportunity to attract qualified experts' attention in around 50 characters or fewer.
- Don't waste it on babble like โPlease help meโ (let alone โPLEASE HELP ME!!!!โ; messages with subjects like that get discarded by reflex).
- Don't try to impress us with the depth of your anguish; use the space for a super-concise problem description instead.
One good convention for subject headers, used by many tech support organizations, is โobject - deviationโ. The โobjectโ part specifies what thing or group of things is having a problem, and the โdeviationโ part describes the deviation from expected behavior.
oh cool , I know about all 3 above those ๐ฎ but not the catb one thats a new one ๐ Thanks!
it sounds a bit harsh these sites, but its for the good of learning and quick answering.
should update on 8. use an AI to explain snippets of code if coding ๐ its actually a thing I do to explain library or gain an example when sourcecode is not giving one
Yeah. Iโve read that one and have shared it to people at work. Itโs a little different working asynchronously.
Itโs also in the Other similar problems section of Donโt ask to ask, Just ask.
Really enjoy this. I say, if you canโt provide your work / steps, research / links, or where you looked, then you havenโt done enough to understand the actual problem you are facing.
A good question provides the right scope to help you obtain the answers you seek.
petition to pin this so it can be linked when someone inevitably ignores it
Are vim keybinds worth learning even without using vim itself?
in my personal opinion, not really
I've been using Vim keybindings in other editors for years. Use 'em if you like 'em.
there is also http://www.total-knowledge.com/~ilya/mips/ugt.html
it is always morning when person comes into a channel, and it is always late night when person leaves. Local time of any member of channel is irrelevant. Resistance is futile.
Anyone know where I can buy a compact ultra sonic fingerprint sensor for hardware design for consumers
Any time I cite DAtAJA or NoHello or XYP or any of the others people get passive aggressive ๐ญ
"Well, I'm sorry. I'll be sure to never express tentativeness in this chat going forward!"
(This was an actual message sent in response to me posting the don't-ask-to-ask link)
sucks to suck for them tbh
just tell them its for future references ๐ . it helps you and me! and most of the request / questions is to help you , its to help me , help you better.
Man... if only there was an easier way. Always having to call the compiler and faxing it code is quite tedious ๐
https://youtu.be/pJ-25-pRhpY
Hi, I'm Lex (they/them)
I made a very silly fax service.
In the video it is running on York Hackspace's internal phone system.
It will hopefully also be running at EMFCamp very soon, on the copper telephone network.
Source code is here: https://github.com/lexbailey/compilerfax
Enjoy :)
I just saw that exact thing on a different server haha
Does someone know how to host a website on a own server and make the website public?
What in particular are you having trouble with?
How I should host it so have a Proxmox Server, I could run a virtual machine or a container. And how I can connect my website to my domain
is it a static site? web app?
as for domain, you just create a DNS record pointing to your IP
you can use cloudflare to hide your IP and use their proxy
Itโs just a porfolio of my projects
ok? but still what type of site is it
does it neeed databases? a bunch of html, css (static), etc?
so it's a static site?
Yeah
fyi you can host it for free on cloudflare pages without having to mess with hosting it on your own server
That's exactly what I did say ๐ญ
Well let them be sour ๐๐ป and ignore them until they have anything useful to say
It also depends on where they host it too. If itโs in an ISP the IP address will roll at some point.
Theres a cron job script that will update the IP address if it changes.
Also, the database really shouldnโt matter as you arenโt exposing that to the internet. Iโd only make it accessible locally and just do a vpn / ssh tunnel back to my network and then connect to the database that way if I needed to do remote work.
They got the mods of this particular server in on it, that's the thing.
Some people are just so aggressively inefficient
yeah, or use a dynamic dns service
I was asking about database since that still would need to get setup on the server
instead of just spin up nginx/apache
If youโre hosting yourself, they most likely are looking to save some money. Otherwise, host on a cloud provider.
There's a bunch of free ones. I used a script for cloudflare dns
just found out one of the weirdest workflows for creating projects, ever ๐
reading about Kotlin multiplatform, i started searching for the wizard inside Android Studio ๐
except the "wizard" is at https://kmp.jetbrains.com/ LOL
Create your first multiplatform project using the Kotlin Multiplatform wizard for Android, iOS, and Desktop, or use one of the pre-made templates.
"fill out a web form and download a ZIP with your project"... what a way to do this
Uhhhh guys git isn't working anymore for me and I have no idea why :( I'm a programming noob so I've been trying to learn about git, and I changed some of the default settings, then I tried changing the default directory and I think that might've fucked something up. Github desktop doesn't recognize any of my repos anymore and I can't use git commands in git bash. Pls I need help :(
I even reinstalled git
I get "fatal : unable to access 'filepath' : invalid argument" as the error. The filepath seems to be the git config but idk it doesn't seem like the right one actually
Ok nvm I got it working ๐ I had to change my environment variable because it had a really scuffed path because I definitely misunderstood a git command
Bro wtf.. Epic has some nerve to tell us how to code C++.
Section in picture.
those are coding standaards if you want to contribute to the engine. It is their project, they set the rules
It clearly says here at Epic Games they follow these guidelines. Meaning anyone wanting a job there won't unless u follow these guidelines and doubt any plugins or scripts you submit for the asset store will be accepted if it's not adhere to their coding standards.
To ignores these as simple "eh it's just epics thing" isn't great for anyone wanting to be part of the industry or anyone who is in the industry that might eventually have to adhere to the guidelines if the company decides to use their engine.
It's saying u either write pseudo code that won't offend anyone or just don't have a future in the industry
that is literally not what it is saying, it is not that deep, those are their coding standards. in fact, the first two sentences in the document are literally
"At Epic Games, we have a few simple coding standards and conventions. This document reflects the state of Epic Games' current coding standards. Following the coding standards is mandatory."
Now, the way I interpret this, is very much "following the standards (at Epic Games) is mandatory, and this documentation clearly lays those standards out for you to prevent confusion.
you do not have to follow them, if you do not work for Epic Games, though it will help you get better support if you do because they follow the internal guidelines everyone is beholden to at the company, meaning all the people who actually follow them are more likely to be understood....
I am just before writing this right now refactoring old code of mine that used slave, master, blacklist and whitelist to use more modern terminology, I agree that some of the things they wrote in there are slightly on the wilder side of "why though", like abort, execute, or native, but for the most part it seems to be entirely fine
Your fbi agent when you search "how to execute a slave" on google: 
Hey i just made one of my first real websites that wasent just a test, can yall check it out and give me some tips?
http://ashare.mooo.com/
2 things
- The connection isn't encrypted
- There doesn't seem to be a filesize limit
(after a while it throws this error, but it still seems to continue uploading)
and when I try to upload a different file it just seems to throw an error
c#
could someone help me with fixing this?
Commit -> GitLab CI/CD docker image build -> GitLab container registry -> Openshift
the problem is that locally the image from gitlab registry is running with no problems
but deployment fails with this error
yeah people were uploading bad crap so i purposely broke the website
also did a very, very bad command and had to reset the whole thing, was able to save the project folder from that burning building of a OS
*made everything be owned by a user that wasent root, had no ssh access, and the only connetion i did have was a SFTP connection i forgot to close. Used that to download my project and let the os burn down
If you can, check your permissions on your deploy server. I had troubles locally with permission mismatch with some being of the actual user, and node_modules and other files/folders being owned by root
if i am reading it right, your issue could be caused by Openshift, as it is running images as non-root random user. You could have those folders existing in the image, with set permissions for root, which obviously means the docker image commands under non-root user can't create files you want.
And locally it runs because i guess you run it under Docker and not Openshift ?
I have a feeling I might have broken something ๐คฆ
you are obviously sending a JSON object into a ParseInt method
Yes, however said JSON object was instead supposed to be a simple 0-0, which does actually get parsed by the utility method properly
Hi, I'm looking for some insight into program permissions, if anyone is available of course to chat about it for five minutes.
does anyone know gnuplot here? id want to load a column from sql, evenly space all the points (or the last 125) from that column along the x axis, and then have te columns values be the Y axis.. i have made a little bash script, but its very broken, and im really struggling to understand gnuplot and its requirements.
this is my script tha ti have glued together:
#!/bin/bash
# Set up variables
graph_title="Luftfeuchtigkeit"
output_file="LuftFeucht.png"
# Query the database and format the output as space-separated values
data=$(mysql -u root -ppasswordhereiguess -e 'Select LuftFeucht from Sensoren' Mars | tr '\n' ' ')
# Write the data to a file
echo "$data" > temp.dat
num_rows=$(wc -l < temp.dat)
# Generate the Gnuplot command file
cat << EOF > plot.cmd
set title "$graph_title"
set xrange [-0.5:{NUMROWS}-0.5]
set yrange [*:*]
unset key
plot "temp.dat" using 1:2 with linespoints title "$column_name"
EOF
sed -i "s/{NUMROWS}/${num_rows}/g" plot.cmd
# Plot the data using Gnuplot
gnuplot -persist plot.cmd
# Convert the graph to PNG format (optional)
convert -density 300 plot.ps $output_file
# Clean up temporary files
rm temp.dat plot.cmd plot.ps
this complains about undefined Y values.
Yes, this script is partially AI Generated, as the examples that i have found online seem too complex. id appreciate if anyone could bother fixing this, or atleast tell me why and what doesnt work-
does anyone have tips for setting up mediawiki? i fucking suck ass at php since ive never learned it (always thought it was insecure or something)
i do have it set up since softaculous saved me but i dont rly know how to do anything else
https://cdn.discordapp.com/attachments/1228693549567049838/1246563065088970792/20240601-2031-23.7901818.mp4?ex=665cd7d2&is=665b8652&hm=1699ffee636e53019cd1a77f547af703492d59878489ff35f84f6835c7b982b2&
anyone know why when im in intellij, i delete something, i save it, and i build, it just reappears?
can anyone help?
Hopefully i found the correct project online (Chaikew/Raven-b3), downloaded, resolved all the needed libraries and it compiled just fine
My guess is that you didn't download all the necessary libs/deps
No, I modified the code and I edit it and it just reappears :/
Maybe because you're actually editing build output
Don't edit stuff under this directory
Ah yeah, I forgot lol, thanks for the help, went outta my mind ๐
It happens, glad we could sort it out
absolutely fumbled my java exam today and i'm pissed
because i misunderstood the question
Now that i understood the question 5 mins after, i am able to make the intended program in like 20 minutes with proper OOP arch.
We're using Azure DevOps Server 2020 Update 1.1, with 2 projects, our project, and the org shared project with package feeds, and one of the pipelines in our project is using the DownloadPackage@1 task to download a package from our maven feed in the shared project, and it successfully downloads it, or so it thinks, but it is a 1kb file. Does anyone have any idea what could be wrong? I can successfully get that package from the feed when depending on it in a maven project.
YAML is trash change my mind
On a related note anybody know how to stack multiple x-field-extra-annotation in OpenAPI generator without having to do \n (which works but is ugly):
type: integer
format: int64
example: 10
x-field-extra-annotation: "@jakarta.persistence.Id\n@jakarta.persistence.GeneratedValue(strategy = jakarta.persistence.GenerationType.IDENTITY)\n@jakarta.persistence.Column(name = \"id\")"```
- making
- a
- list
``` results in buggy crap so that's not the answer
The prior gives me this at least
@jakarta.persistence.GeneratedValue(strategy = jakarta.persistence.GenerationType.IDENTITY)
@jakarta.persistence.Column(name = "id")
private Long id;```
but that's expected
Note the missing tabulation though
I can also make them on one line but thats uggo. Can't manually format the java file since it gets auto-generated -.-
this is pretty much a YAML question, no ? https://yaml-multiline.info/
Find the right syntax for your YAML multiline strings.
Yeah and a bit of an openAPI gen question too. Because technically should be separate additional annotations.
I just space delimited them for now. I mean it's not the end of the world. It's an autogenerated schema class that looks messy on its own
I was told I shouldn't directly use schemas it as Spring JPA entity classes anyways. This is temporary measure.
unfortunately what happens with x- items is literally up to the tool which uses the OpenAPI spec, as anything x- is a non-standard extension
have you tried what these guys did ?
Same effect as \n but this is honestly what I'll do for now, thanks. It gets rid of outer quotes so I don't have to escape ones in the annotations. Also makes the REST API yaml look way cleaner so that's something
Student:
type: object
x-class-extra-annotation: |-
@jakarta.persistence.Entity
@jakarta.persistence.Table(name = "student")
properties:
id:
type: integer
format: int64
example: 10
x-field-extra-annotation: |-
@jakarta.persistence.Id
@jakarta.persistence.GeneratedValue(strategy = jakarta.persistence.GenerationType.IDENTITY)
@jakarta.persistence.Column(name = "id")
username:
type: string
example: abominableCider69
x-field-extra-annotation: |-
@jakarta.persistence.Column(name = "username")
firstName:
type: string
example: John
x-field-extra-annotation: |-
@jakarta.persistence.Column(name = "first_name")
https://bunchy7ss.github.io/
im proud of my baby, and no one can tell me otherwise.
it may have no css, and no javascript, and was made in 2m, but im proud
yes.
This is the greatest thing I've ever seen, thank you for sharing this โค๏ธ
Hi yall, I need to do a system migration for which the database is the main issue rn. The current db has about 350 tables, which needs to come down to 30. Quite some data can be truncated but what would be the best strategy for converting/transferring the data? Both dbs are mysql. Do you have some recommended tools or strategies for this?
are there any plugins that automatically input an ';' at the end of every c# line?
for vs code
Could try to find a linter?
ok random "challenge", i'm looking for amusing examples of spaghetti code
preferably something not too high level
i could look it up myself, but im curious what you guys found over the years
ioccc
I'm making a Mastodon bot/framework. Any suggestions for the name?
for whut
looks like they are doing math
or something that involves calculations
it's a program that converts an image to a spectrogram
genorating millions of increasingly longer sin waves is pretty computationally expensive
the length of the sin wave increases so it can fit more sin waves in the spectrogram
awsom
my websites honestly follow this
if it works in windows 98/95 its good
(that is actually not a joke. i have a 98 VM and test out the website and see if it runs. ill take a screenshot when i can)
just Mastodon Bot ๐ค
i am still laughing at the level of joke this repository is going for https://github.com/auchenberg/volkswagen
lmao, this is golden
so i have an idea of a project to do (export Plex library metadata into Kodi compatible NFO files), but can't decide on development itself.
Should i use it an excuse to learn a new language (Rust, Go or Python?) or do it in something i am fluent in.
Should i do it as CLI, desktop or web based UI ?
Hate making design decisions LOL
What languages are you fluent in?
lately i did mostly frontend, thus JS/TS & other web technologies. Also did a bunch of Java/J2EE, PHP. I wouldn't count my use of Pascal from 2 decades ago as a relevant skill these days. Also did some of the Android & iOS native technologies, so light Android Java/Kotlin & iOS Objective C/Swift skills (but only very light, to get me by to create Cordova plugin interfaces between the native code and JS)
also dipped my toes a little bit into Dart/Flutter
also probably forgot random small technologies we used/tested at work along the way
I'm Stefano, I'm the Founder of Kaizen Media. I'm offering my custom web design & web development services for free. Experience in this space 4+ yrs Shoot me a DM and Let's Build
-Fully Coded
-No Templates used
-Mobile Friendly
Can you think of any solid reasons you would need a new language?
reason is why not. One can never know when it will come handy
Well, learing a new language is a time investment, is spending that time to learn a new language worth it? Or would you rather spend that time mastering a language you already know?
It's always handy to learn a new language, never know when it might come in useful
i guess i could still do Kotlin multiplatform, even if i don't want to make mobile platforms, thus allows me to learn the Kotlin language & Compose multiplatform functionality before i might need to use it at work. Not sure it is ideal toolkit for this use case, but oh well, not my problem /s
Well if you did a bunch of java and just want to get a working program out the door you could just use that
If they already have to learn Kotlin MP for work at some point anyway, why not learn it now, and be more productive at work later when they are more experienced? Seems like a reasonable time investment
I agree
fun times starting https://www.theregister.com/2024/06/10/fortune_200_oracle_java_audit/
Ah yes, Oracle - the Legal Company that also happens to own (and sometimes write) software.
https://bonkersworld.net/organizational-charts
i wish
well, it didn't start well ๐ . Tried to run https://github.com/LukeHagar/plex-api-spec/blob/main/plex-media-server-spec-dereferenced.yaml through https://github.com/cjbooms/fabrikt/ and it immediately failed with a weird, hard to debug error inside the generator. I guess i will have to try https://github.com/OpenAPITools/openapi-generator instead
Rip
i mean it was throwing me an error inside the squareup codeblock generator, but with no context, so couldn't really figure out what was wrong
Looks like it is time for hand crafting
i mean sure i can do requests by hand and parsing XML, but i would prefer not to
i want to learn programming how ever the only issue is that i won't focus long enough to learn it
dear god, Kotlin has some serious idiotic parts ๐
had to make the package name to start with in_. instead of in. ๐
https://github.com/OpenAPITools/openapi-generator processed the OpenAPI file just fine btw
Damn, that's kinda lel
Hey guys,
This is Arsalan from CAMB AI -- we've spent the last month building and training the 5th iteration of MARS, which we've now open sourced in English on GitHub https://github.com/camb-ai/mars5-tts
We've have also been featured on VentureBeat: Check it out here.
We'd really love if you guys could check it out and let us know your feedback. Thank you!
MARS5 speech model (TTS) from CAMB.AI. Contribute to Camb-ai/MARS5-TTS development by creating an account on GitHub.
Need the collective brain on a problem. We are recording an mp3 voice message in the webapp via the https://github.com/samhirtarif/react-audio-recorder , that works just fine. We get the file to the server, mime type is correct ( audio/mpeg ), file name has .mp3 extension. I upload it to whatsapp media endpoint, it accepts it. Then I send audio type message ```Array
(
[messaging_product] => whatsapp
[to] => 1234546566
[type] => audio
[audio] => Array
(
[id] => 306114195907326
)
)
is this PHP ? have you checked the exact message you are sending as JSON ? Because https://developers.facebook.com/docs/whatsapp/api/messages/media/ specifies it as:
"audio": {
"id": "your-media-id"
}
So it does matter if your PHP to JSON conversion makes:
"audio": {"id":"xxxx"}
or
"audio": [{"id":"xxxx"}]
from it
The official documentation indicates that in order to send such messages, the conversation must be initiated by the user.
could it be that your media file is the first message in the conversation ?
no that part is handled
wall other messages work just fine, only audio has this stupid issue and Guzzle client makes it hard to debug...
you could try temporarily to upload it to a server and send link instead of id, at least to test if there is some issue with the uploaded media:
{"messaging_product":"whatsapp","to":"+xxxxxxx","type":"audio","audio":{"id":"1507724556813303"}}
@cloud knot it does send out correct JSON format
do you have webhook set up ? Some errors are aparently only communicated via webhook
we have, but i've not been checking what arrives for it
okay, we gonna need to update some code for our gateway to process some system messages, maybe it's been sending it, but we never knew that some errors are webhook only
on positive note, the annoying part (Gradle configuration of openapi, automatic building, fetching of the OpenAPI schema etc) is done, so now i can actually start coding in Kotlin and figure out how the hell does Kotlin & Kotlin/Compose Multiplatform even works. Even if i only do a desktop app for now
ive been thinking about building my own pokedex with the pokeApi in java. And as a beginning to my java script journey of course. ive bern coding backend for almost 10 months now with java. would you say building a pokedex is a good intermediate project to learn some things in frontend dev and improve my backend dev skills even more?
backend is mostly whatever you want to use. You want to have a well defined "contract" for frontend, that is the REST API and the structures in it; then you can play around with backend technologies as you progress through the skills
if you do your REST API definition well, you can replace a Java J2EE or Spring backend with Quarkus microserves running on Kubernetes, then replace them with whatever next ๐
so, I have a very specific problem with HTML
I have several audio tags in a page
I click to open the options to one of them and it always opens the last one, no matter which one I clicked
Any ideas to fix it?
what do you mean by click to open the options ? Native browser menu or some JS functionality ?
if JS functionality, how do you initiate those opening of "options"?
Native browser menu
I made this codepen to demonstrate
ok, so I just checked it. It's a visual bug, the menu opens only on the last one, but the "context" is for the one I clicked
Context menu opens fine on my end, did you try different browsers ?
I only tested in chromium-based browsers
works fine for me in floorp/firefox
Hello @everyone , i am Chablis a full stack developer. I am available for any project
i am new in this server
Pro tip, you won't make friends by pinging everyone (even if you could do that)
this is a duplicate of crbug.com/341741271 which has been fixed starting with Chrome version 127.
we are at Chrome 126 now i think, so next version ๐
the microsoft chart works very well for intel lol
@echo off
echo starting
REM Add ngrok authtoken
ngrok config add-authtoken YOUR_NGROK_AUTHTOKEN
REM Start ngrok
start /b ngrok tcp 192.168.1.195:3100 --log=stdout > NUL
echo ngrok complete
REM Wait for ngrok to initialize
timeout /t 5 /nobreak > NUL
REM Get public URL from ngrok
for /f "tokens=2 delims=:" %%a in ('curl -sS http://localhost:4040/api/tunnels ^| jq -r ".tunnels[0].public_url"') do (
set PUBLIC_URL=%%a
)
echo ngrok curl complete
REM Split URL into components
for /f "tokens=1,2 delims=:" %%a in ("%PUBLIC_URL%") do (
set PORT=%%b
set URL=%%a
)
REM Remove leading "//" from URL
set URL=%URL:~2%
echo %URL%
echo cloudflare
REM Update Cloudflare DNS record
set JSON_DATA={\"data\":{\"port\":%PORT%,\"target\":\"%URL%\"}}
curl -X PATCH "https://api.cloudflare.com/client/v4/zones/YOUR_ZONE_ID/dns_records/YOUR_DNS_RECORD_ID" ^
-H "Authorization: Bearer YOUR_API_TOKEN" ^
-H "Content-Type: application/json" ^
-d "%JSON_DATA%"
REM Wait for the DNS update to complete
timeout /t 1 /nobreak > NUL
REM Change directory to Minecraft server location
cd /d C:\Users\DELL\Desktop\paper-1.20-17
REM Start the Minecraft server
java -Xmx8G -jar paper-1.20-17.jar -nogui
can anyone help me witht this
the cloudflare section is not working
{"result":null,"success":false,"errors":[{"code":9207,"message":"Request body is invalid."}],"messages":[]}
it automatically starts an ngrok tunnel
then tries to update a cloudflare dns record
before finally starting the minecraft server
i think the backslash in set is not required, thus you are sending those slashes to cloudflare as well, and they are like what are these backslashes doing in the JSON
What programming lang should I learn today?
COW Lang
What's that?
It's a programming language :)
What can I do with it?
Random stuff, your imagination is the limit
hi guyss!!
i have learned MERN stack, want to learn nextjs can anyone suggest some good free resources
powerpc assembly :p
both annoying and useless at the same time !
ADA
And if you want something useful learn go if you havent already
Both ends up at the same spot
If it works it works
Though then there's that third cat who jumped
Hello! Do any1 here know about llm and python integration ?
There's lots of tools for it nowadays. If you're using OpenAI they have their own python library, there's also other stuff like LangChain for other types of ai providers.
Yeah, LangChain is quite popular in the LLM space
Thanks! i'm trying to use an ai model offline with python, but everything that i came accross like lm studio use openai api but don't support function calling, which is important
Hello guys, I built a native app in Unity that runs on a meta quest3 and a react app that will be hosted somewhere else.
Now I want the native app and the react app (the instance that will be opened on the browser) to talk to send messages to each other.
I do not want an external server to be involved for every transaction.
I tried WS but the HTTPS protocol (on which the react will run) will not talk to WS, and I am getting errors with the certificates hashings.
I tried WEBRTC but it seems too complex and there are not many resources out there for Unity WEBRTC with React for text.
I tried running a local server within Unity app and this approach works but the messages can only go from the native app to the quest but not the other way around.
Can someone please help me out with this problem?
You have to have some sort of process listening to a port in order to receive unsolicited messages
webrtc needs a STUN server or something im pretty sure
yeah webrtc should work, but that is too complex and I do not want to spend that kind of time on it...
can you pelase suggest something better
Yeah it does need a STUN server, I did use that but it didn't workout
realistically turn, stun only really works over lan
for a start if you want websocket communication you almost definitely need to be using it in an insecure context, as much as that sucks
If ur on lan you can just expose a port and be done with it
hey guys, im playing around with js rn (relatively new to programming, only did backend stuff like java and c#). Im trying to build a pokedex and have tried the following code. the first fetch returns an url to the Kanto region endpoint. is it possible to fetch the returned url like i did? what is my mistake here? :PepeHmm:
I dont fully understand what you are trying to do/what the api does, but fetch() and then() are both parts of async promises, so keep in mind that the program will continue while the code in those are processed in the background.
The final .then() doesn't return anything, so it's a promise<void>. If you change it to (y => {log...; return results}) you'd get closer to what I think you want. However, that is still just a promise to eventually resolve a result, if you want to use that url in the let kantoDex part, you'd need to make that piece of code asynchronous so that the code can actually pause and wait for pokedex to resolve to a result
you got two options:
- you move your code in the callback. that is the "then" function. See the bottom of the example https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Promise#basic_example
- you use async/await, if your Javascript engine supports it https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Statements/async_function
Things have gone horribly wrong when you start vectorizing your Golang for performance gains
Like making use of SIMD functions in cpu? How do u even do that in go
Not necessarily. If your workload is math-heavy then it's a godsend
The issue is Go not having to use SIMD. I try to avoid GPU and SIMD when doing Go because it usually means youโre in such a hot loop you should be using C(++)/zig/rust to get proper hardware and memory controls
Fair
@here i recently changed my personal site's color scheme, criticism?
At first glance, not enough color contrast. I don't personally follow WCAG myself and I have 20/15 vision, but even I have a hard time reading the text because there's not enough contrast. https://accessibe.com/accessscan?website=deusvult.neocities.org
Thank you; i have poor vision as well, however it is acceptably corrected with just my glasses. (i am severely near sighted.)
Coming in here with 20/20 vision, on a best case scenario dark room with bright screen.
There is no way anyone is reading that without straining their eyes, needs much higher contrast on the text. If you want light text, you can try a darker outline on the text with maybe 1px thickness and keep the existing text colour, but for anyone with any level of reading issue, the site as it stands is as good as incomprehensible
i want to learn python, c++ and c# but i can't do it by myself unless someone is there
It's funny cuz I actually did end up making a website for a biology assignment
I couldn't find an app for android that can log the value of battery temperature and current consumption in a file. Is it possible to use qpython to write a logger instead?
if you want it directly in android, then you need to create an app, which has a backgrond service, which periodically does this.
Or if you don't want to do it, you can just install a home assistant companion app on your phone and let it report information to HA instance and use the data from there (not sure if it reports tempeature)
Thanks for the suggestion. Will check out the home assistant part.
as i said, if you want it yourself, an android app with background service which reads that data and reports it is an option
To add to the HA discussion, these are all the stats that HA can pull about the battery:
Max update frequency is 1 minute for HA App
College: here's the instruction set for a extremely specific education-specific subset of ARM assembly that has to be run in a simulated environment
Also college: we won't tell you how to use it
your quad word is misaligned bro!
I get it. I think the HA approach is probably simpler to achieve.
well it is pretty easy to read, but, node cannot find index.js, so, where is index.js located? because it isnt in the current directory
yeah, it requires no coding for mobile and if you got HA server already, all it takes for you to set up an action when value changes to let's say publish values to MQTT server or whatever other target
I have a native unity app running in the background of my quest and this app will launch the browser onto this device and run in the background, and now this app is not listening to the user speech in background, and is not playing audio, how can I solve this
I got this really weird C++ error
Idk if this is right I donโt know c++ or any language but try return(main)(int); no if that will work
add )
it asked me then to add a semicolon
That's just not valid C++. There's no way to fix it. What is your intention?
Ngl from my small I should have known you do return 0 I think
That's assuming their intention is to return 0
Itโs strange that I know that even though I have not learned c++
That style of return syntax is widely used across languages, not even just those that are C-like
unrelated but i know that pfp, that card is so annoying
I may or may not play the deck irl
it was a pretty cheap deck to assemble ill admit
Most of the cards when I got them were at the most expensive price besides the Dino cards
oof
But I regret nothing
What is your intention?
to end the program if the variables inputted are below the minimums in a conditional
convention is to return a non-zero number for errors and zero for success. You can define your error numbers with an enum or just directly return an integer literal.
enum Error {
NONE = 0,
INVALID_INPUT = 1
};
int main(int argc, char** argv) {
if (error_condition)
{
return Error::INVALID_INPUT;
}
return Error::NONE;
}
or
int main(int argc, char** argv) {
if (error_condition)
{
return 1;
}
return 0;
}
if you're trying to end the program from outside of main, you can't really. Not without it being a crash anyways.
So we have this 11k loc go file. Will break that up into a smaller file later. But anyway -- my coworker uses a mac and his vscode autocomplete and stuff is almost instant, mine takes over a second on windows. Anyone know if this is a configuration difference or mac is just magically instant at it?
What kind of storage device is your project and user's home folder on? vsc stores intellicode data in appdata
my C drive is a p31 gold 1tb. Almost completely full, might be harming speed.
Project itself is stored on a second drive, crucial p3 4tb
maybe the two drives?
Performance degrades a lot beyond 80-85% capacity. You can benchmark your drives and ask your coworker to do the same if you want to see how substantial the difference is. However, if you have sufficient RAM the impact of storage should lessen as things get cached. If they have apple silicon, their RAM is also potentially higher performance than yours.
They have a high end mac, m2 ultra 192gb and some huge ssd. And of course their SSDs and ram are crazy fast -- but my pc isnt a slouch either, 12600k 64gb ram. Not crazy fast, but im surprised that a file that big just turns intellisence into mush on my machine
What speed is your RAM?
just ddr4 3600
To put that M2's RAM into some perspective, it has 27.7x your RAM's bandwidth. It's CPU is also substantially faster. Even if your machine has no potential issues, it's still gonna be noticeably slower
Yeah im aware of the speed difference. There doesn't really seem to be a middleground though. Either its instant or its ~3+ seconds. So I figure it might just be a system config thing
Either way. Looks like it's just system speed and obviously the solution is to break that huge file into multiple
unfortunately pc will never get fast ram ๐ time to buy an epyc cpu
Intellisense just happens to be a very RAM performance sensitive workload where M2's unified memory really gets to shine
Its just fact of the matter m2 ultra gets 12 channels of ddr5 (or is it 4 idk) where you can only get that speed on epyc
EPYC doesn't officially support 6400MT/s DDR5 and the M2 ultra has 16 channels of that. Apple silicon is pretty crazy
i tried installing pip 2 times it still isint recognized as a command pls help
close the terminal and open again?
tried 5 times
What's the OS, (Windows i presume), if so, don't forget to add it to path if you didn't already.
Yes I forgot to add it to path that was the problem
here is the full code:
using namespace std;
int Milk = 0; // ml
int Flour = 0; // g
float Egg = 0; // whole
int Butter = 0; // ml
int Salt = 0; // mg
const int minMilk = 24;
const int minFlour = 13;
const float minEgg = 1;
const float EggPerBlin = 0.4;
const int minButter = 15;
const int minSalt = 284;
int main() {
cout << "Hello!" << "\n\t";
cout << "blinmaker starting up! \n";
cout << "How much milk do you have?";
cin >> Milk;
cout << "How much flour do you have?";
cin >> Flour;
cout << "How many eggs do you have?";
cin >> Egg;
cout << "How much butter do you have?";
cin >> Butter;
cout << "How much salt do you have?";
cin >> Salt;
return 0;
if (Milk < minMilk || Flour < minFlour || Egg < minEgg || Butter < minButter || Salt < minSalt){
cout ;"no blin today";
return(main;);int());
}
}```
not sure about others, but that whole return statement makes no sense. you are returning main (what is main here? the entry point function) from main and whatever is behind it ?
you return a number; and everything else starting with ( and ending with ) after that return is syntactical nonsense (imo)
you literally just return 1; (or any other valid int) there, not return (main;);int());
also the whoile last section is unreachable code
the first return 0; literally stops the execution, so the if and anything after is not executed
This was already explained the other day, not sure why they didn't follow the advice
How do you change the font that every part of VSCode uses? I can only change the font of the editor portion, not other parts like the explorer
#include <iostream>
using namespace std;
int Milk = 0; // ml
int Flour = 0; // g
float Egg = 0; // whole
int Butter = 0; // ml
int Salt = 0; // mg
const int minMilk = 24;
const int minFlour = 13;
const float minEgg = 1;
const float EggPerBlin = 0.4;
const int minButter = 15;
const int minSalt = 284;
int main() {
cout << "Hello!" << "\n\t";
cout << "blinmaker starting up! \n";
cout << "How much milk do you have? ";
cin >> Milk;
cout << "How much flour do you have? ";
cin >> Flour;
cout << "How many eggs do you have? ";
cin >> Egg;
cout << "How much butter do you have? ";
cin >> Butter;
cout << "How much salt do you have? ";
cin >> Salt;
if (Milk < minMilk || Flour < minFlour || Egg < minEgg || Butter < minButter || Salt < minSalt) {
cout << "No blin today" << endl;
} else {
cout << "You have enough ingredients to make blins!" << endl;
}
return 0;
}
@jovial garden try this
Async Javascript, AJAX, APIs and Classes OOPs in Javascript neccessary to start ReactJS ?
OOP not so much, the rest yea.
I'm working an a react native app with EXPO and I can't get notifications to work. They do arrive but they don't play any sound or vibration even though I set these things. They also don't appear as banners, they just land in the notification center
const notification = {
content: {
title: 'Low Charge!',
body: `Hey there! You only have ${batteryLevel}% battery remaining! Please charge your phone!`,
sound: 'notification.wav'
},
trigger: {
seconds: 2,
channelId: 'lowBattery'
}
};
if(batteryLevel < threshold) {
Notifications.scheduleNotificationAsync(notification);
}
async function registerForPushNotificationsAsync() {
let token;
await Notifications.setNotificationChannelAsync('lowBattery', {
name: 'Low Battery notification',
importance: Notifications.AndroidImportance.MAX,
vibrationPattern: [250, 0, 250, 0],
sound: 'notification.wav'
});
//rest of stuff that's required (permissions, token &stuff)
I'm making the app for Android
Did, I added maths to say how many blins you can make: here's the full code, now there's an error about a missing semicolon:
using namespace std;
int Milk = 0; // ml
int Flour = 0; // g
float Egg = 0; // whole
int Butter = 0; // g
float Salt = 0; // g
const int minMilk = 24;
const int minFlour = 13;
const float minEgg = 1;
const float EggPerBlin = 0.4;
const int minButter = 15;
const float minSalt = 0.284;
int main() {
cout << "Hello!" << "\n\t";
cout << "blinmaker starting up! \n";
cout << "How much milk do you have? ";
cin >> Milk;
cout << "How much flour do you have? ";
cin >> Flour;
cout << "How many eggs do you have? ";
cin >> Egg;
cout << "How much butter do you have? ";
cin >> Butter;
cout << "How much salt do you have? ";
cin >> Salt;
if (Milk < minMilk || Flour < minFlour || Egg < minEgg || Butter < minButter || Salt < minSalt) {
cout << "No blin today" << endl;
} else {
float FlourBlins = Flour / minFlour;
float MilkBlins = Milk / minMilk;
float EggBlins = Egg / EggPerBlin;
float ButterBlins = Butter / minButter;
float SaltBlins = Salt / minSalt;
cout << "You have enough ingredients to make" min(FlourBlins, MilkBlins, EggBlins, ButterBlins, SaltBlins) " blins!"<< endl;
}
return 0;
}
in my experience with push notifcations with firebase - make sure the notification file in available as resource, you seem to send "notification.wav" but they seem to send path starting with ./ in their examples - maybe that is where the sound is broken ? For vibration, have yous et vibrationPattern ?
also do you do that build type and have that plugin ?
yeah, I did set a vibration pattern (it's in the original message) and have these lines in app.json
Also, I'm trying to make it on-device only, it doesn't need to connect to the internet, it shouldn't actually since it's just a small tool
this is correct, right? I did build it with EAS and tested the APK on my phone but it didn't work
i haven't done expo, so i can't help you with their specific functionality, i was talking in general from Android notifications experience
check if the file exists in res/raw in Android app structure. Also i found this ? https://github.com/expo/expo/issues/16447#issuecomment-1935664116
i made a game engine
You forgot to add << on line 39
whats your guys thoughts on this question for my assignment
Write a method checkCharacter() which has 2 parameters: A String, and a specified index (an int). Method checkCharacter() checks the character at the specified index of the String parameter, and returns a String based on the type of character at that location indicating if the character is a letter, digit, whitespace, or unknown character.
public class FunWithCharacters {
public String checkCharacter(String word, int index) {
public static void main(String[] args) {
FunWithCharacters labObject = new FunWithCharacters();
String word = "happy birthday"; // Set string
int index = 2; // Set specified index to 2
System.out.println(labObject.checkCharacter(word, index));
index = 5; // Set specified index to 5
System.out.println(labObject.checkCharacter(word, index));
word = "happy birthday 2 you";
index = 15; // Set specified index to 15
System.out.println(labObject.checkCharacter(word, index));
word = "happy birthday!";
index = 14; // Set specified index to 14
System.out.println(labObject.checkCharacter(word, index));
}
}
is what I have so far, the only thing written by me is the public String
the entirety of my thoughts are: weird wording on what its asking u to return (unless there was some markdown where the question was presented), my first instinct is that it wants you to return a string like "letter" or "digit" but that's not immediately obvious just vaguely implied
but also you can do it we all believe in you 
Is it bad that I use f strings in python?
I think they are fine
Anyone have experience creating datasets for machine learning? Pls DM me i have a project im trying to do but i dont know how to do it
If someone has knoweldge in coding in c# pleasd dm me. Working kn a gta project and could use help:)
I'm not usually much about the DMs but what are you trying to make a dataset for training (image recognition, image generation, text generation)? What kind of model LLM, stable diffusion, YOLO? What's the question?.... Why the hell is slow mode enabled in the dev room
Ffs if people have questions just ask them why does everyone want a personal tutor to DM them. Public discussion leads to more people learning from the convos and less time wasted. Yes I know how to program in languages...
current codebase:
using namespace std;
int Milk = 0; // ml
int Flour = 0; // g
float Egg = 0; // whole
int Butter = 0; // g
float Salt = 0; // g
const int minMilk = 24;
const int minFlour = 13;
const float minEgg = 1;
const float EggPerBlin = 0.4;
const int minButter = 15;
const float minSalt = 0.284;
int main() {
cout << "Hello!" << "\n\t";
cout << "blinmaker starting up! \n";
cout << "How much milk do you have? ";
cin >> Milk;
cout << "How much flour do you have? ";
cin >> Flour;
cout << "How many eggs do you have? ";
cin >> Egg;
cout << "How much butter do you have? ";
cin >> Butter;
cout << "How much salt do you have? ";
cin >> Salt;
if (Milk < minMilk || Flour < minFlour || Egg < minEgg || Butter < minButter || Salt < minSalt) {
cout << "No blin today" << endl;
} else {
float FlourBlins = Flour / minFlour;
float MilkBlins = Milk / minMilk;
float EggBlins = Egg / EggPerBlin;
float ButterBlins = Butter / minButter;
float SaltBlins = Salt / minSalt;
cout << "You have enough ingredients to make" << min(FlourBlins, MilkBlins, EggBlins, ButterBlins, SaltBlins) << " blins!"<< endl;
}
return 0;
}```
And I'm getting this error: C:\Users\SubSpruce\Code\blinmaker.cpp|39|error: no matching function for call to 'min(float&, float&, float&, float&, float&)'|
So basically I want to train an AI to predict stocks so I need a training dataset to do it. Now I do not want to write out 12500 datapoints by hand, but I have no idea how to do web scrapinf
Where do you expect to get this min function from? The built in static function on the math lib only takes two params and returns the minimum between the two https://www.geeksforgeeks.org/c-sharp-math-min-method/
A Computer Science portal for geeks. It contains well written, well thought and well explained computer science and programming articles, quizzes and practice/competitive programming/company interview Questions.
current codebase, refactor with nested mins:
using namespace std;
int Milk = 0; // ml
int Flour = 0; // g
float Egg = 0; // whole
int Butter = 0; // g
float Salt = 0; // g
const int minMilk = 24;
const int minFlour = 13;
const float minEgg = 1;
const float EggPerBlin = 0.4;
const int minButter = 15;
const float minSalt = 0.284;
int main() {
cout << "Hello!" << "\n\t";
cout << "blinmaker starting up! \n";
cout << "How much milk do you have? ";
cin >> Milk;
cout << "How much flour do you have? ";
cin >> Flour;
cout << "How many eggs do you have? ";
cin >> Egg;
cout << "How much butter do you have? ";
cin >> Butter;
cout << "How much salt do you have? ";
cin >> Salt;
if (Milk < minMilk || Flour < minFlour || Egg < minEgg || Butter < minButter || Salt < minSalt) {
cout << "No blin today" << endl;
} else {
float FlourBlins = Flour / minFlour;
float MilkBlins = Milk / minMilk;
float EggBlins = Egg / EggPerBlin;
float ButterBlins = Butter / minButter;
float SaltBlins = Salt / minSalt;
cout << "You have enough ingredients to make" << min(min(min(min(FlourBlins, MilkBlins), EggBlins), ButterBlins), SaltBlins) << " blins!"<< endl;
}
return 0;
}```
https://www.scrapehero.com/open-source-web-scraping-frameworks-and-tools/ I would lean towards apache nutch if you don't need the GUI for anything and can get it all done through http requests and parsing pages/responses. If need a GUI for some reason would look more towards puppeteer or selenium based stuff like cypress (used for integration and e2e testing but is a general browser automation tool) @modest crystal
Using web scraping frameworks and tools are great ways to extract data from web pages. In this post, we will share with you the most popular open source frameworks for web scraping and tools to extract data for your web scraping projects in different programming languages like Python, JavaScript, browser-based, etc.
With regard to stock prediction it is a thing that high speed trading bots programmed by physicists have basically "won" but there may be some "space" for employing llms or other AI systems to analyze trends and whatnot and help to make longer term choices (as is most llms pretty bad at the long term planning and "true understanding" of dynamics in the world and spatial reasoning and whatnot but lots of room for training specialized ais for now still)
Also somewhat funny all the crypto hype over like the last decade and people probably would have been better off just dumping the money into Nvidia stock instead of buying up cards and/or crypto for the most part
Generally speaking the "problem" with scraping the public web (legal details aside) is typically lots of garbage info copy pasted between lots of servers and higher quality data is usually gonna be behind paywall of some sort and usually with throttling and policy that stops you from taking the data and running (or at least let's them chase you)
good luck getting your hands on a hft
Yeah wasn't really sayin you should do that just sayin it's one of those places computing and statistical analysis and whatnot has been applied a bunch already and hard space to find a niche I think (not financial advice but personally just keep basically majority of stuff in "safeish" slow growth mutual funds and mess around with mess around money in stonks and other things)
not really planning to use this bot to trade more than like 1k just kinda like a project
All good, you see Michael Reeves goldfish investment thing? Pretty hilarious
yea its somehow better than some investment bankers lol
most of these i have tried (the python ones) but they never seem to fetch the correct data
im tryna get the revenue and profit stats from yahoo finance
Get the detailed quarterly/annual income statement for Microsoft Corporation (MSFT). Find out the revenue, expenses and profit or loss over the last fiscal year.
but it is not fetching anything
Yeah a lot of those sites have like paid API access too so do what they can to block out the scrapers, if you put in "enough effort" can usually work around things but some sites put in more or less effort to try and "protect" their data ultimately anything sent to your client and decrypted is readable though
Network tab in chrome and save as har and can parse through whatever you've been browsing while the network panel open.
i tried selenium too but it just doesnt grab anything it just comes as blank
Lots of little make a http request and process the response are written in python since the string processing/parsing is pretty simple and you don't usually need like multi threading or other stuff to just batch process text
Yeah selenium is made as a testing tool so it isn't like geared towards scraping but can be used to generically interact with a browser
Yah that's one option for sure
You might want to just see what pricing is like for the different apis if they have a dev console (Google basically will give you anything for a price and a lot of free usage for sake of development)
i dont want to pay tho thats why im just tryna scrape
Right but I'm saying you should see what you can get "free tier" and/or what the cost is since time is money
Google it ๐ hehe but really I dunno off hand
try making the same request without running headlessly and see where its getting caught up
alr
how do i do that
my current code
I know like Google maps gives you a good amount of "free tier" stuff but they make you get an API key at least to track requests coming from your services and to throttle things eventually or make you pay but takes a while to hit those limits
well thats requests not selenium
i have two versions one with requests and one with selenuim wait
shit i deleted the selenuim one
chances are just using selenium will magically fix any parsing you do because the data is loaded dynamically and selenium will actually run the javascript on pages required to load that data.
ok let me just ask gpt to write me a selenuim script lol
@pep\
@silk eagle it just throws a gaint wall of errors at me
puppeteer is always a choice, as it technically runs a real browser behind it, but you might have hard time getting the data out of let's say charts
for example live updates clearly come through a binary data from
wss://streamer.finance.yahoo.com/ ๐
most of the stock website use charts tho
what i meant that if you can't get the data from network requests and you need to interpret already processed data from charts, good luck
if you can reverse engineer how the data received via web socket is processed, it is much more usable
good luck to me i guess
what does it do
if i understand it correctly, the basic idea is that you connect to wss://streamer.finance.yahoo.com/, send a {"subscribe":["ticker1","ticker2"]} message to server and you are then receiving binary data.
this.connection.addEventListener("message", this.handleWebSocketUpdate))
And then message received from backend is sent in to that block of code above. What exactly that code does is a good question. Atob does base64 decoding, so the first few lines creates an instance of quotefeeder object from that binary data, and then the data is processed. now what is the quotefeeder ? good question
https://stackoverflow.com/questions/58330057/python-web-socket-interpretingreading-data -> https://github.com/yahoofinancelive/yliveticker
someone made a python implementation of it, use convert it to whatever platform you want to use it in
ok thanks
yliveticker is good but yfinance is for historical data, which is what i need
ok, so you want it for the table shown on the page ?
recommend rate limiting yourself tho, otherwise they might put your IP on block list
"rate limiting" - not requesting everything in rapid fire succession, but give some pauses between requests
i mean if you are interested in 1000 tickers - you don't send 1000 requests right one after another, but add let's say a 1 second pause between each call
alright
yes, it will take 15 minutes to download stuff like that, but server might give you a pass because you do reasonable amount of requests, not all at once.
ok i set it to 2 seconds
for example when Yahoo had YQL, they had a 1000 call/day limit for unathorized requests. Or there are for example this https://developer.yahooinc.com/dsp/api/docs/traffic/info/rate-limits.html, where that part of yahoo allows 60 GET requests per minute - so exactly 1 request per second, but only 8000 requests per day
so it depends on the service, but it is good to not overwhelm the target when you do scraping, so you don't put a target on yourself
ok
there is also a library called yahooquery which does exactly that if anyone wants

Hey guys I need advice.
I want to build a simple application that has a ui, a database and can plot the data.
I know a bit of sql and also python and matplotlib.
I already did a small and easy project where I had to get data from an excel table and plot it, and I used python with matplotlib and tkinter, and I did not really like tkinter, but i don't know how i would create the ui without it.
I know I am really picky, in fact i also hate VB (at school we used MAUI and i hated it).
Also i'd prefer to not have to use JS as I'd like to wait to start learning at school in september.
If you have any ideas please share.
tldr: need a way to create an application with database, ui and data visualization, preferably without js or .net
p.s.: I do not speak english as a first language, and excuse me for any technical error
Looks good to me
Since WHEN did typing HTML in vscode got so easy??????
pretty sure since vs-code existed, it's always really been geared towards web dev, some things still require extensions to get completion or compilation for but has always been pretty web friendly/focused out of the box
maybe you're confusing it with visual studio or something,visual studio is more geared towards C# or desktop application development out of the box, sometimes used in combination with Unreal engine or other things
tkinter is definitely a popular option for UI development in python another alternative for desktop UIs you might want to look at is PyQt. UI libraries in general are not that great to work with hence a lot of things just using a web based UI (wide compatibility and lots of folks who know HTML/JS well enough to hack things together relatively quickly, CSS is a nightmare but it gets the job done). If you wanted to make a web API in python and have a web UI you could use FastAPI, SQLAlchemy for ORM so you don't need to write much SQL by hand just dealing with objects in python and mapping them to tables, Alembic is good for migrations if you want to go the extra mile and make it more "real world". It is also typical to architect things in a way that you have a REST API with JSON requests/response bodies and then build your client UI as a separate application that consumes the API but for classwork with limited time (or POCs) is typical to skip on things and take short cuts to get things done.
Also, if you wanted to have the python layer just like output preprocessed web pages might be worth taking a look at Django or Flask, both allow for making web APIs, Django includes ORM as well so would not need SQLAlchemy if using that, Flask is a bit more "lightweight" than Django but still lets you handle setting up "route handlers" easily and has some html templating I believe.
Fucking end me
I already have one collegue who is writing pyramids of doom, but appearantly I also have another collegue who can't code at all. wtf is this crap
I don't know much about coding but this does not look too good
At this point just start watching some begginer C++ tutorials
The code is literally doing unnecessary api calls which it does not need, or even use lol. On top of that it doesn't even await the result, which is why the green squiggle line is there.
I want to get news data for a certain company for the last year. NewsAPI only allows you to go back a month. Any help?
Or perhaps just grab market sentiment
Their C# example code shows that you can set a "From" property in your request. Have you tried that?
@modest crystal
I guess you need to pay if you want articles older than 1 month
Perhaps if that specific company publishes it's own news, it could be scraped and then be kept updated via an RSS feed.
I should have specified this before, but i need it for the top 500 companies
accidentally made the messiest program know to man https://paste.pythondiscord.com/FDYQ
theres exactly 1 comment. good luck
why ar your varible in your loop?
didnt rly care when coding, its an abomination now
please move your varible outside of your loop
use chatgpt to make your code easy to read
good now add comments
What in the if statements
May I ask an absurd question of curiosity,
What if we made a system/program to use the graphics card of a thunderbolt connected device
In theory something like that would be doable on a phone with adb and BlueStacks
But on the scale of 2 windows machines, i should expect that there would be more room for possibility given the connection strength and so forth
like, an eGPU?
which versions of python supports cryptography.fernet?
I have 3.12 but I imported cryptography.fernet and
its saying it doesnt exist
wait nvm forgot to install it
How can I make it so that Docker doesn't keep my GPU awake on my laptop? I've tried setting the desktop dashboard to only use integrated graphics but that didn't help. Going to the file path of WSL through Task Manager doesn't even open anything
If I recall correctly there is a setting in docker desktop that will basically put containers to sleep if they're not in use. You can try that. If it doesn't help, you may actually need to stop the container that is using that GPU device. If it's an Nvidia card, I believe there is also an option in the Nvidia setting somewhere to disable staying awake when the GPUs in use. Don't quote me on that but I swear I saw one in there.
Does anyone here understand in meta all-in-one SDK in Unity?
thermally modeled mosfets in ltspice using many cores. 35min to get 306us, avg 146ns/s. I wonder if only more cores will help or if 3d v-cache affects performance too
If the workload involves moving a lot of data between ram and cpu yes
Though I'm unaware of anything that benefits from cache outside of games
That's not quite what extra cache is for. It's specifically for repeatedly accessed data, typically code. Being able to keep more code in caches is great for games since they're usually big pieces of software. Moving lots of data is what bandwidth and raw processing performance is for. I have no knowledge of ltspice's implementation, so I don't know whether there's a lot of code or repetitive data accesses in the algorithms it uses.
If you know the size of the plane and the amount of (square) tiles it's divided into, how do you calculate the size of one individual tile?
nevermind you just divide it I'm stupid
I compiled a version of my application for 64bit Windows 10, 32bit Windows Vista, and 64bit Linux.
How would you distribute each version? Should I have all 3 in the same .zip file, have different downloads for Windows and Linux versions, or have a separate download for each of the 3 versions
Some bundle, some separate by OS, some separate all of it. It really just boils to what's more convenient for you.
Ya
on github etc they usually just give out source as one zip, then exe/whatever for each platform separately
Hey guys! I'm new to coding and just made a Discord bot called Records. Would anyone be open to testing it?
It's supposed to stores memories in Discord by letting people save messages others have sent, as well as view anyone's saved messages. people can also delete their own saved messages.
I can send the invite link for the bot if anyone's interested!
Yeah on releases page typically different files for different platforms targets so can just download what you need or get the src and build from readme if a package for your target OS doesn't exist or need to make changes and rebuild/bundle, they could just grab src from git though assuming GitHub or gitlab or the like being used to host the packages
Could do the "Uber zip" with all platform versions included but just making people download extra junk and copies of things they'll never use too
Literally everything benefits from caching, caching just means keeping copies of data close to where it will be used, prefetch loads nearby memory into caches to basically have the data at the ready so there is no need to request it from ram or request it from disk if it's already in a CPU cache, if the disk has a cache and need to load something that happens to be a cache hit (something in there) then will use that over accessing the disk, cache is just a spot to dump copies of things along the way so you don't need to go back to the source
Redis is an "in memory"/ram cache system to basically keep disc contents loaded and ready to reduce access time, the l2arc in zfs is a similar disk to ram type system, caches exist at multiple levels for DNS hence the "it's always DNS" problems, caches in browser, it's caches all the way down ๐ข
fun facted do you know one of the main reason why cod an games like it are so storage heavy its becuse insted of not downloading all the weapon an skin shit thats all animated in bits an pieces they downloa it all in one go even if you dont have it unlocked in game its there so when you do warzone rememeber this when there is liek 100 of yall there is so much crap on the back end that its not even funny why you have to run a high end pc for high fps there is some other technical aspects to this but the mojority of games run like this
side note we dont need to have loading sreans any mroe the loading screan is jsut a error check and ids check now besides slower inter net checks
a few gaems i do play run a thing that is used when you unlock the skin the loading animation or transfer animation is the game downloading the skin that you need/ unlocked
there's definitely certain kinds of workloads that benefit more from extra l3 cache than others though
productivity apps generally handle too much data for it to make much of a difference while games have certain bits of code, data and assets that are very frequently accessed, which is why 3d v-cache benefits them more than productivity apps
(usually)
because of course with that said there are productivity apps that handle the extra cache well
Sure it varies how often you're going to get cache hits for any given task but it's not "just good for games" it is good for processes that can make use of the caches and end up with hits more than misses
https://www.amd.com/en/products/processors/server/epyc/4th-generation-9004-and-8004-series.html
3D v-cache is definitely not just for games lol. Nobody is gaming on these.
Unless they're used for game servers
Does anyone if it's possible to auto-archive files in specific folders to my NAS when they aren't accessed for x days, while keeping them accessable in a similar way Onedrive does? (Onedrive downloads it when you try to open it, and then it opens it. All files not physically on the sysem will still be shown in the file explorer)
Yes it is possible
My code isn't working. ๐คฌ
I'd rather just know how to make good comments than put my entire code into ChatGPT, it doesn't think like me, so I probably won't even properly understand "my own" comments. It's not hard to make understandable comments. I don't have nothing against using AI for coding, but just for minor corrections or questions and stuff. I wouldn't completely overhaul my code by an AI.
You would be correct
You can ask for help if you don't know how to fix it btw
We all suffer from a lack of skill
thanks google, now persistent notifications arent persistent anymore
it's everything i ever wished for
for everything to break for no reason whatsoever
what factors should i take into account when deciding what third party libraries (not standard/golang.org) to use in go?
- Why am i using go
license, authors, mantained status, complexity, code quality
etc
isnt the point to make your own portfolio
anyone here competing in the Google Gemini Developer Competition
Hey there. Is there a way to readout every media stream running with codec etc os-wide?
blame developers/companies who abuse everything the platforms provide. Unfortunately all these things have to go, because people will abuse it for fingerprinting or spamming/annoying users
yes
What OS? But probably not, detecting if a program happens to load up some codec at runtime and uses it to decode or encode some stream of data isn't really like an OS level responsibility. You could probably monitor what processes are using the hardware decoding capabilities but not much else about what they are doing with the data inside that process... What's the actual goal and target system though?
This is why we can't have nice things
anyone wanna source code a cheat for me

can someone help me test out a web app i made for a class the website is https://elitemedia.site
their is no liking comment or profile function yet
its going through cloudflare tunnels so idk if the upload speed is good
In this video I discuss a GitHub attack vector that can let anyone access data from deleted or private Github Repos.
Read the blog post about this Cross Fork Object Reference bug on truffle security
https://trufflesecurity.com/blog/anyone-can-access-deleted-and-private-repo-data-github
My merch is available at
https://based.win/
Subscribe to ...
psa: DO NOT store personal data or api keys on github, private repo or public;
it coming along nice, In one of the imgs upload you expose your OpenAI key please hide it for your safety
upload speed is not bad, What are youtr target file types
ok
uhh let me check
i forgot to add a file restriction i was working on file size limit
see idk what to do i create a docker file of it and host it on my server and to expose it i use cloudflare tunnels its slow tho idk why
{'png', 'jpg', 'svg', 'gif', 'mp4', 'webp', 'ogg', 'webm', 'bmp'} is this good?
that is all html allows to show in videos and image format
@blissful geyser
you dont use DevOps
i pushed the fix for the file extention thats allowed
no.
how should i
i need to fix it to not make multiple of the same files be uploaded
when you pushed first did it work, where did you push it
i made my own python system that auto makes the docker image and redeploys with the same data from before like if user uploads images the images are not deleted only the code does
i use a ubuntu server in my house running docker and portainer and to expose i use cloudflare tunnels
and my python code connects to docker to make it happen
im still learning lol
i know there is a better way to do that
well you have to figure out that on own, debug
me too
im gonna have to delete all user media that was uploaded people are uploading some not so nice things
i am not a python expert , but in java and Golang we Threads to handle concurrent user request's
this is what my folder for the code looks like
its simple but i want to make it in another lang
i made a big mistake before if you want to know about it i can tell you
and my teacher looked at it when i did not know people did that with it
its was a V1
we got 2 more months for the project
what lang
do you have on github
idk mostly python i uploaded what i made when im learning
i think i had some js on their
i had to add hcatcha to my contact form bc i would get stuff from seo company bots
i did not pick recatcha bc its only 100,000 request a month on their so i pick hcatcha bc they give 1 million a month
i was dumb couple months ago when i started too i have my contact form data gettign saved to a TXT file lol now i got it getting emailed to me
we live we learn
so all the logic is in main.py
do you have it on gitHub
i can upload the prod one now
ok can you send me a link
ok give me a sec
@blissful geyser
i also got a web chat i made but its broken
@blissful geyser https://chat.rapter.pro the manage profile is broken it does not update the db when users change something on their profile
where do you store data
good practice is important, you need to separate the logic from the handlers/routes
i store media in a folder it is created once the code runs
its named uploads
app.config['UPLOAD_FOLDER'] = os.path.join(os.path.dirname(__file__), 'uploads')
should i seperate them using diff python files or smt diff
this looks fine for one file
use SQlite, storing it on a folder is complex(Profile Details), if you want to store images use a BLOB
i store files thats uploaded in the uploads folder which connects to a docker volume so it stores their
and ip address and other cookie data are gonna go in a mysql db that i run once i get it set up
wait the app runs on docker
thats what the docker file is for
i have a stack on docker with a db volumes and every thing thats needed
im gonna have to push it to docker hub in a image for my teacher
ohhh , i was confused
- if it is a fork/can be forked (always private repo can't be forked)
- if you do by mistake, delete is not a solution, revoking the old compromised key and issuing a new key is the solution
one of the few acceptable uses for a pre-commit hook imho is scanning for secrets
guess the db use
My RISC Architecture:
https://docs.google.com/spreadsheets/d/14R31eohE9jVamZtl5pJgyjpkHrFt1Kq9b2unfl_MZSI/edit?gid=0#gid=0
Instructions
OPCODE ,Instruction,Mnemonic,Definition,Parameters,Format,Note,Pseudoinstructions,Equivalents,Extras,Code
0,1,0,1,1,Interrupt,INT,Trigger an interrupt,-,1,MVI [DEST], [IMM],MVU [DEST], [31:10],Recommended interrupt subroutine,DNT
0,1,0,1,0,Move upper immediate,MVU,DEST(31...
anyone got experience with Firebase App Distribution (https://firebase.google.com/docs/app-distribution) ? All i need is if it distributes apps separately or through Google Play/App Store test channels. If separately, then it is fine for my use case.
Something to do with DnD perhaps ?
@lavish pollen
What programme did you use for this?
What are you using, just base Php?
xampp
What's the issue again?
when i try to access localhost/phpmyadmin on google
what port is the server running on
idk
Find out lmao
how-
same thing
Windows?
yeah
Read the docs lmao
Yep! A combination tool for DND lore and story plotting
idfk anything abt coding
Read the docs
