#development
1 messages Β· Page 17 of 1
Have you restarted your computer after installing? Are you sure it installed correctly? I remember installing xammp like 15 years ago on uni
some fivem tutorial
Let's see it then
Let's see the tutorial
Checkout our store https://store.rcore.cz
XAMPP default settings are intended for the local development environment, if you want to use XAMPP on a production server, you must adjust the database settings and permissions.
XAMPP download link: https://www.apachefriends.org/download.html
Tiktok: https://tiktok.com/@rcorestudio
Discord: https://...
send screenshot of your XAMPP control panel
This seems like a windows firewall issue
Did you get any errors when startid apache and mysql?
The server is off
which one
Apache and mysql ?
This should be a really easy and straightforward install. Try to reinstall if its not running as expected. If reinstall does not fix the issue try to check out the windows vhost file for conflicts
so reinstall xampp
Reinstall XAMPP into a directory without spaces
or any special chars
or is it cause im running it as admin?
One message removed from a suspended account.
Try plotting the data in a graph. Perhaps visualizing it makes it more clear what's actually going on.
the sql server is possibly already runnin from a prev start... or mssql server is also running on the port ... normally you can kill the process in taskmanager , you can also do a netstat -a -d to find what is using the mysql ports
also xampp does not like it when you start one service and then click on start all services , as it will try to start the already running tasks , and not to restart them, starting them wil fail because they are already running and using the port, when it fails xampp registers the service as stopped and leaves the origial process orphaned
and wow xamps made a nice panel for windows :p on linux we just get :
does anyone know why this doesnt work?
[%0] prompts a picker which works by querying the jnd."Year" first. which already shows all the years. but somehow doesnt work. i've tried replacing [%0] for year and it worked
wow, you are complicating your life so much π
(if that is Oracle DB)
(now i just need to remember how did we do it, give me a second)
SELECT ADD_MONTHS(SYSDATE, -1*(ROWNUM*12)) THEDATE
FROM ( SELECT 1 just_a_column
FROM dual
CONNECT BY LEVEL <= 366
)
WHERE ROWNUM <= 10
If it is oracle DB. the FROM section is just an abuse of dual and connect by to generate pivot table of 366 rows (you can limit it to less if you want) and then you just generate dates by limiting the number of rows by using the rownum π
and if it is a different SQL variant, would be good to say what version of SQL it is and shwat is "doesn't work"
but most likely if it is SQL, then you should use a different placeholder, typically languages use ? as parameter placeholder
It's a different variant called Hana db for SAP
I tried to find so many ways and it still doesn't work so at the end I just made a table and fill it up with 100 years
:PARAMETER_NAME
like this
but question marks should work too
so jnd."Year" = ?
seems like it is identical to Oracle DB, so:
? - ordered parameter, that is first question mark is the first parameter, second question mark is second parameter etc. So the tool will ask you to provide first, second etc parameter
:name - named parameter, so the tool will ask you to provide parameters named name etc
Oracle is usually special...
What if they just made logging buffered and actually log it to the server in the background 
Oh wait....
i want to know how runtime polymorphism happens at runtime
void eat(){
System.out.println("Animal Eating");
}
}
public class Dog extends Animal {
void eat(){
System.out.println("Dog Eating");
}
public static void main(String[] args) {
Animal animal1 = new Animal();
animal1.eat();
Animal animal2 = new Dog();
animal2.eat();
}
}
i cant able to know how this is happening in runtime
Ignoring optimizations, this is usually implemented with vtables. Plenty of videos and pages describe vtables if you want to learn more
thanks
i am having an interview so i just want to know how that works bcoz the interviewer may ask "HOW THIS IS HAPPENING IN RUNTIME"
can you guy be my stackoverflow for a second?
i have a dilema on good practice on backend with automapper
I have an issue with AutoMapper and best practices. I have two DTOs: one DTO has a string status, and the other has an object with a link to the status API to fetch the status DTO. I solved it in such a way that when mapping from the DTO with the link to the string DTO, a function is called that fetches the status DTO from the API and updates the necessary parameter on the DTO with the string status.
What do you think? My mentor says it's not a good practice to put logic in the mapping profile, but another senior developer said this is how it is solved. I am a bit divided because it seems like clean code to me, but I also value my mentor's opinion. Does anyone have experience with this?
oh forgot to mention its .net stack
please ping me if you answer me
why does this
'Handshake' AS "Link"
FROM OITM T0
WHERE T0."ItmsGrpCod" = '[%2]'```
and this
WITH
tahoon as (
SELECT
A."Name",
'Handshake' as "Link"
FROM "@TAHUN" A
WHERE A."Name" = '[%0]'
),
tahoon2 as (
SELECT
A."Name",
'Handshake' as "Link"
FROM "@TAHUN" A
WHERE A."Name" = '[%1]'
),
tahoon3 as (
SELECT
A."Name",
'Handshake' as "Link"
FROM "@TAHUN" A
WHERE A."Name" = '[%2]'
)
SELECT
THN."Name" as "tahun1",
B."Name" as "tahun2",
C."Name" as "tahun3",
THN."Link"
FROM tahoon2 THN
INNER JOIN
tahoon B
ON
THN."Link" = b."Link"
INNER JOIN
tahoon3 C
ON
b."Link" = C."Link"```
works
but this :
WITH
tahoon as (
SELECT
A."Name",
'Handshake' as "Link"
FROM "@TAHUN" A
WHERE A."Name" = '[%0]'
),
tahoon2 as (
SELECT
A."Name",
'Handshake' as "Link"
FROM "@TAHUN" A
WHERE A."Name" = '[%1]'
),
tahoon3 as (
SELECT
A."ItmsGrpCod",
'Handshake' as "Link"
FROM OITB A
WHERE A."ItmsGrpCod" = '[%2]'
)
SELECT
THN."Name" as "tahun1",
B."Name" as "tahun2",
C."ItmsGrpCod" as "tahun3",
THN."Link"
FROM tahoon2 THN
INNER JOIN
tahoon B
ON
THN."Link" = b."Link"
INNER JOIN
tahoon3 C
ON
b."Link" = C."Link"
doesn't work. logically it should
its SAP B1 Hana Version
please kindly enlighten me when you found the issue
Sort of a hardware dev question and just pushing the boat out to see what sorta experience people have had in here.
Anyone ever used one of these 50 Pin D-sub connectors for projects/tinkering and wanna shed some light on how I can go about adapting it. Any help is appreciated ππ»
well either you use some standard on the pin layouts or you just connect pins to your liking
thus creating your own standard it isnt rocket science tbh
anyway anyone?
I guess you'd just break it out on both sides and wire it up however you want. My electric drumkit uses something similar
yep exactly like i said either you follow some standard for wiring or just make up your standard
or use standard cable but wire connectors like you want
Like most things in programming, it is a taste thing. I prefer startup to be small, so hopefully it changes less frequently. Editing an automapper profile to change some logic doesn't feel amazing. If it was around startup code I'd be asking more questions about it. My taste is to have the logic live fairly close to the DTO, if not inside it.
When I'm reviewing a PR, I can look across all the DTOs their transformations / overloaded constructors. Get a feel for how that DTO is effected by other DTOs.
While Automapper isn't hard to learn, others might not be able to debug it.
Basic DTOs with methods to transform one DTO into another.
thank i appreciate your input
Programming isn't up to taste. Everything should be MY WAY and MY WAY ONLY! /s
Every company has a guy like that ^^ XD
idc about taste i care about which things reduce problems/work
hi
i think someone in Java world have seen Express from npm and decided he wants same in Java π https://javalin.io/
what a shame... but thank you for your knowledge for sure. would be on win11 and the why: just out of curiosity. ^^
is there a dip chip computer still bein made to buy? i want to try out breadboarding and embedded more
Arduino or esp32 or rp2040 all exist and are decent options to get started
The demo page looks a bit like Minimal Api's in .net
and both look like Express from Node ecosystem, the original source π
Never really used any JS for back-end stuff. It just doesn't really click for me for some reason.
I only like made a few scripts in js to automate some stuff on a server, which was way quicker to get working in js than in .net.
Also I didn't want to learn python so js it was haha
const express = require('express')
const app = express()
const port = 3000
app.get('/', (req, res) => {
res.send('Hello World!')
})
app.listen(port, () => {
console.log(`Example app listening on port ${port}`)
})
Damn, that's pretty simple
and then there are plugins for various stuff. like static serve, which will then serve files from specified folder so you don't have to deal with every static file one by one (but you shouldn't really do static content via Node, you should do that by the webserver in front of node).
anyway, still can't decide what language to pick for some REST API for my project, didn't do backend for ages, and at same time i don't see myself doing J2EE for this π
I'm probably biased, but .NET is like java, but better in almost every way. It has many tutorials for REST api's and it's very often used for that kind of stuff.
i mean the issue is not picking the language, but the whole ecosystem, i don't want to overengineer stuff when i might be the only one using said app π . Also while something like Quarkus sounds nice, but i don't want to depend on hard to set up environments. Becase relative to a docker image, forcing users to use Kubernetes might be a deal breaker for some. It would certainly be a deal breaker for me
U don't need to use k8s for docker?
To me not having a docker image is a deal breaker if I want to run an app haha. It's just so convinient
of course you can run docker images without kubernetes π
but it seems quarkus might be a reasonable choice even for docker stuff, so i might look at it first https://htl-leonding-college.github.io/quarkus-docker-gh-actions-demo/
https://x.com/ThePrimeagen/status/1819913196808901015 lol close enough
The greatest sign you are a successful tech company:
There is a Mac mini or Mac trashcan somewhere, usually in a closest, that if unplugged untold chaos would ensue
we got a pair of mac minis for iOS automated builds
I think it was twitter that had a trashcan running in a janitor closet that hosted their entire corporate vpn and the janitor moved it to make space and the entire company was fucked for a day until someone found out where said computer actually was
i haven't seen so far anyone try making a framework GPS module
so I'm gonna try doing that myself
https://www.u-blox.com/en/product/max-f10s-module this seems pretty cool
ideally it'd have inside of it a 2-port USB hub such that you still get a USB port off of the module, as to not waste one slot. would be cool to have a type-c one such that you could charge your framework through it and keep it always connected, but a type-c hub like that would be kinda complicated
the SAM-M10Q is better actually, it has its own antenna and stuff. hopefully it all fits inside a framework module
crap the framework IO modules are 30x32x6.8mm and the SAM-M10Q is 6.8mm tall
i think i have a design I'm happy with
my personal favorite is still the Novell Netware server which got walled in and nobody missed it because it was just working with no problems π
BOFH meets Edgar Alan Poe
Wow novell is a name i havent heard of in a long time
memories of some bullshit stuff when i was in elementary school lmao
Use spring boot? π€ in combination with swagger and you got a great interface for rest apis
still going to try quarkus first
I meant just in general, to me Java just feels a bit more verbose than C#. A bit more work to get something working. Also I just love Visual Studio, so I am a bit biased. On top of that, LINQ is just amazing. Java has the Stream api since recently, which is a very nice addition, fortunately.
C# also has swagger/openapi docs, so its kinda the same
You can even export the comments on your methods and exposed public types, and it will just generate documentation from it as well. Java probably has this too, but it's so nice to have.
unity go brr
It indeed does
opinions on godox?
you mean Godot?
Hell of Unicode
yes yes, sorry that got autocorrected
Probably confusing just like all the other game engines
open source, free, not greedy like unity, and the programing language is a little too similar to PY
those are my opinions
It's possible to use 4 different languages for scripting in Godot. Two are custom to the engine (gdscript and visual scripting), but you can also use C++ and C#
oh dam i didnt know that
for python devs, here is the best user interface library for terminal: https://textualize.io
It's amazing, and 10ΒΉβ°β°β°β°β°β°β°β° times better than curses
proof: chatgpt client in the terminal
thanks, just remembered I still have my outlook login on github in plaintext...
np glad to help out;
going forwards, i would recomend setting up vaultwarden to store passwords, and 2fa all my accounts even if the password is 100 characters long;
i'd also look at using alias emails for dev projects to keep your real email secure in the event the data leaks;
**Recommended System Requirements:**
- 64bit Windows 10.
- GPU capable of running OpenCL 1.2 (Most computers from 2012 or later).
- View the README.txt file as things may change between versions.
Ensure Graphics drivers are up to date, and extract all the contents before running the application.
**Minimum System Requirements:**
If your system cannot run [Application], there is a version that supports 32bit Windows Vista and later. However, some features are omitted to improve compatibility.
```I published a game/application on Itch.io, would you change anything about how I've written the `Download & install instructions` section?
Hey everyone! I'm developing an AI-based mobile app aimed at helping people with disabilities
I'm looking for one or two developers with experience in mobile app development and either NLP or computer vision taking into account that it is a co-founder opportunity so it's not a paid role but a chance to collaborate on something meaningful
If you're interested feel free to send me a message!
Nothing special I just published my first google extension and what it does is allows you to highlight text from websites that doesn't allow you to highlight and copy text. :P
There's E books for my school that we use and they make it that you can only copy little by little of text each week and such and if you reach the limit they block you from highlighting the text so I made this.
Is this too chonky of a switch case? It's reminding me of the legendary Undertale dialogue switch
What the fuck
if that is JS and the name is same as the property name, you can make it much simpler without the whole switch case
how would i go about doing this? yes javascript
give me a minute to make a code
this is my first js project, im way over my head rn lmao
var s = {
totalStarts: 2000, dtMin: 10000, dtMax: 1800000
};
function setLivelyPropertyListener(name, val) {
if (s.hasOwnProperty(name)) {
s[name] = val;
}
}
function getLivelyPropertyListener(name) {
if (s.hasOwnProperty(name)) {
return s[name];
}
return null;
}
Relevant docs:
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Object/hasOwnProperty
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Property_accessors
https://github.com/rocksdanister/lively/wiki/Web-Guide-IV-:-Interaction will that fit with this? I think it has to be the function livelyPropertyListener(name, val) to take input
app rn is hosted at https://qwikster.dev if you want to see the use case
those things are pretty basic JS techniques, so i suspect it should work in any JS engine
I mean, will it still be able to take configuration changes from Lively Wallpaper's configuration system? I'm pretty sure it calls livelyPropertyListener(name, val) every time the user changes something in the desktop app's configuration UI
you can name function what you want, the names i did only because i shown the use case of both setting and getting from the object
the point was this part for the equivalent function of your code:
if (s.hasOwnProperty(name)) {
s[name] = val;
}
it checks, if the s object has a property with name from name variable; if it does, then it sets the name property of s via s[name] to val value
if the property with name doesn't exist in your s object, then that value and property it gets ignored
so you just have your s object with your settings, and it sets only those who exist in that object. Those who don't will get ignored and you don't have to write a special code for every property - you just add a new property to the object and the code will automatically accept it, based on the fact that you added it into the s object
ohh, I see how this works! is there an easy way to make it so I am able to perform operations on the input (ex. some numbers might want to be in seconds UI side for usability but i need them in ms to use), or should I make it perform those operations where I use the variables?
well you can have special use cases written before, or have a second object for modifiers. Give me few minutes
I think i see how that would work
also are you setting the s object from anywhere else ?
I don't think so, no
use either several if statements or a switch case before the hasOwnProperty if, that would work, yeah
var s = {
totalStarts: 2000,
_dtMin: 10000,
set dtMin(value) {
this._dtMin = value * 1000;
},
get dtMin() {
return this._dtMin;
}
};
function livelyPropertyListener(name, val) {
if (s.hasOwnProperty(name)) {
s[name] = val;
}
}
console.log(s.dtMin);
livelyPropertyListener('dtMin', 5);
console.log(s.dtMin);
the point is that you can have a getter and setter (and store the value to some other variable, in my case _dtMin)
and then the setter simply takes the input and multiplies it by 1000 in my case, and this is abstracted away inside the object itself with those getters and setters
so you would do this set/get only for properties which need special handling
I feel like putting more if statements before the s.hasOwnProperty() call would be simpler in my use case, but i'll remember your method for the next time I do some kind of js project that has more than 3 or 4 special cases
btw even for your implementation, i have one more trick, give me another minute π
var s = {
totalStarts: 2000,
dtMin: 10000,
dtMax: 180000
};
function livelyPropertyListener(name, val) {
if (['dtMin','dtMax'].includes(name)) {
val = val * 1000;
}
if (s.hasOwnProperty(name)) {
s[name] = val;
}
}
console.log(s.dtMin);
livelyPropertyListener('dtMin', 5);
console.log(s.dtMin);
You can check the property name against an array of special cases (in this case properties which you want to multiply by 1000), modify value and then let the normal process continue
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Array/includes
(these are stupid optimization tricks you learn when you code for decades π )
I can tell xD
This is definitely useful, there are several time variables
You can also use the same trick to validate values, so you could for example modify val to boolean for boolean properties, check if what you expect to be number is a number (by adding check via https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Number/isInteger) and so on
hold up i thought js automatically differentiated between char, int, float, and bool?
i am talking about input. In theory someone could call your setter method with string for numeric setting, or number/string for boolean property, could enter the whole text of Romeo & Juliet for your color properties and so on
π
There is a problem detected with your user. Please replace user and press F11
so you can add input validation into that listener, which checks if values you are getting sent conform to what you expect, and reject invalid values by not setting the invalid value into property
Good point, although I do find that unlikely - and if that happens, it's kind of the user's fault - I doubt i need to bother with data sanitization since the calls are handled by an application developed by someone else (that should already have data sanitization?) (unless there's a possibility of code injection?)
Would I not want to do it like this?
with return 0; after that*
sure, but i meant it as an advice in general
for example for your color properties to check by regular expression if the input matches what hex color should be
oh ok yeye
Have I done this properly then?
var s = { //Default settings
totalStars: 2000,
dtMin: 10000,
dtMax: 180000,
velDecay: 0.985,
rMin: 1,
rMax: 6,
rRepel: 100,
rAttract: 200,
repelStrength: 80,
attractStrength: 15,
elasticity: 0.45,
col: #f7eedf,
bgCol: #05050a,
invMouseOver: false,
spawnVel: false,
showPrompt: true,
collision: true,
death: true
};
function hexToRGB(hex, alpha) {
var r = parseInt(hex.slice(1, 3), 16),
g = parseInt(hex.slice(3, 5), 16),
b = parseInt(hex.slice(5, 7), 16);
if (alpha) {
return "rgba(" + r + ", " + g + ", " + b + ", " + alpha + ")";
} else {
return "rgb(" + r + ", " + g + ", " + b + ")";
}
}
function livelyPropertyListener(name, val) { //Settings manager + listener
if (['elasticity', 'velDecay'].includes(name)) { //for percentage
s.[name] = val * 100;
return 0;
}
if (['dtMin','dtMax'].includes(name)) { //for time in seconds
s.[name] = val * 1000;
return 0;
}
if (['col', 'bgCol'].includes(name)) { //for hex color
s.[name] = hexToRGB(val);
return 0;
}
if (s.hasOwnProperty(name)) { //direct usage
s[name] = val;
}
}```
do you want to do the RGB conversion from hex string to RGBA, or you did it for the checks only ?
I would use a regex match if only matching is needed:
var re = /^#?([a-f\d]{3,4}|[a-f\d]{6}|[a-f\d]{8})$/;
console.log("#001122".toLowerCase().match(re)!==null) // true
console.log("#ff001122".toLowerCase().match(re)!==null);// true
console.log("#GG1122".toLowerCase().match(re)!==null); // false
console.log("#FFGG1122".toLowerCase().match(re)!==null); //false
you can also add A-F instead to regex groups and that will remove need for toLowerCase
but otherwise yeah, your code should be fine
It's for conversion - not sure if the code i need those two for have support for hex, so i might as well just convert to rgba
also btw, i would modify val in those ifs and not return
like this:
var s = { //Default settings
totalStars: 2000,
dtMin: 10000,
dtMax: 180000,
velDecay: 0.985,
rMin: 1,
rMax: 6,
rRepel: 100,
rAttract: 200,
repelStrength: 80,
attractStrength: 15,
elasticity: 0.45,
col: #f7eedf,
bgCol: #05050a,
invMouseOver: false,
spawnVel: false,
showPrompt: true,
collision: true,
death: true
};
function hexToRGB(hex, alpha) {
var r = parseInt(hex.slice(1, 3), 16),
g = parseInt(hex.slice(3, 5), 16),
b = parseInt(hex.slice(5, 7), 16);
if (alpha) {
return "rgba(" + r + ", " + g + ", " + b + ", " + alpha + ")";
} else {
return "rgb(" + r + ", " + g + ", " + b + ")";
}
}
function livelyPropertyListener(name, val) { //Settings manager + listener
if (['elasticity', 'velDecay'].includes(name)) { //for percentage
val = val * 100;
}
if (['dtMin','dtMax'].includes(name)) { //for time in seconds
val = val * 1000;
}
if (['col', 'bgCol'].includes(name)) { //for hex color
val = hexToRGB(val);
}
if (s.hasOwnProperty(name)) { //direct usage
s[name] = val;
}
}
good call
i mean in your code you would have to add if(s.hasOwnProperty(name) around the whole code block or to every if statement
function livelyPropertyListener(name, val) { //Settings manager + listener
if (s.hasOwnProperty(name)) { //direct usage
if (['elasticity', 'velDecay'].includes(name)) { //for percentage
val = val * 100;
}
if (['dtMin','dtMax'].includes(name)) { //for time in seconds
val = val * 1000;
}
if (['col', 'bgCol'].includes(name)) { //for hex color
val = hexToRGB(val);
}
s[name] = val;
}
}
Like this
This is why you don't ever let me code things
alternatively depending on what u need it might be way simpler to just do a if (!s.hasOwnProperty(name)) { return; }
and then rest of the code doesnt have to be in the if statement. but that assumes youre doing nothing unless if s.hasOwnProperty(name) is true
i might be over commenting
Bruh how does anyone do software development on linux with that atrocious middle click to paste forced bind
It is in most software yes
Thankfully I figured out vscode can disable it
Can someone help me figure out what I'm missing exactly in this error log with pip install simpleaudio? It tells me that I have to get Microsoft Visual C++ 14.0 or greater. I was given a link within the log and I download it and installed it just for simepleaudio to not be installed because of it.
C:\Windows\System32>pip install simpleaudio
Collecting simpleaudio
Using cached simpleaudio-1.0.4.tar.gz (2.0 MB)
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing metadata (pyproject.toml) ... done
Building wheels for collected packages: simpleaudio
Building wheel for simpleaudio (pyproject.toml) ... error
error: subprocess-exited-with-error
Γ Building wheel for simpleaudio (pyproject.toml) did not run successfully.
β exit code: 1
β°β> [20 lines of output]
C:\Users\Hunter\AppData\Local\Temp\pip-build-env-x54dsws_\overlay\Lib\site-packages\setuptools\_distutils\dist.py:268: UserWarning: Unknown distribution option: 'test_suite'
warnings.warn(msg)
running bdist_wheel
running build
running build_py
creating build
creating build\lib.win-amd64-cpython-312
creating build\lib.win-amd64-cpython-312\simpleaudio
copying simpleaudio\__init__.py -> build\lib.win-amd64-cpython-312\simpleaudio
copying simpleaudio\shiny.py -> build\lib.win-amd64-cpython-312\simpleaudio
copying simpleaudio\functionchecks.py -> build\lib.win-amd64-cpython-312\simpleaudio
creating build\lib.win-amd64-cpython-312\simpleaudio\test_audio
copying simpleaudio\test_audio\c.wav -> build\lib.win-amd64-cpython-312\simpleaudio\test_audio
copying simpleaudio\test_audio\e.wav -> build\lib.win-amd64-cpython-312\simpleaudio\test_audio
copying simpleaudio\test_audio\g.wav -> build\lib.win-amd64-cpython-312\simpleaudio\test_audio
copying simpleaudio\test_audio\left_right.wav -> build\lib.win-amd64-cpython-312\simpleaudio\test_audio
copying simpleaudio\test_audio\notes_2_16_44.wav -> build\lib.win-amd64-cpython-312\simpleaudio\test_audio
running build_ext
building 'simpleaudio._simpleaudio' extension
error: Microsoft Visual C++ 14.0 or greater is required. Get it with "Microsoft C++ Build Tools": https://visualstudio.microsoft.com/visual-cpp-build-tools/
[end of output]
note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for simpleaudio
Failed to build simpleaudio
ERROR: ERROR: Failed to build installable wheels for some pyproject.toml based projects (simpleaudio)
I did a clean install of windows this week and I need this program to function so my hearing aids' bluetooth connection can stay connected to pc. It has a thing if it doesn't hear an audio after 5 minutes, it disconnects to preserve battery but I don't need that at all. 
You don't have a compiler installed to compile code with
Click the link and download the build tools
Just the first one
This is what I got but I still installed, haven't restarted pc if that matters. 
Desktop dev with c++?
Yep
'ight, ty. I'll try that. 
Works, thanks alot :D
Welcome
Has anyone used p11tool in alpine before? Not sure why p11tool --list-all --provider /usr/lib/libcryptoauth.so give me "core dumped" error.
Does someone know what this is??
Its on my Xiaomi phone....
I just questioned my self and on google i can't find enything
And can i use it in developing a phone app??
find out the bundle id of the package, not name
π
Hey, anyone with experience working with esp32 with tight battery restrictions?
I'm trying to build small device that uses BLE for connectivity that will work at least one day on 150 mah battery.
So far I have about 3h. Reducing bt tx power by 24dbm gave me additional 30 minutes or so π¦
I wonder if its even doable, or i have to just go with bigger battery
ESP32 is a series of devices with a fairly wide range in power consumption. The general idea with any of them though is to move into sleep states as much as you can. Which particular SoC do you have? The datasheet reports typical power consumption in the different states
Esp32c3. Can't really go to sleep with it since i need to send data every 1 second. I use deep sleep when not in use
have you set the cpu frequency to 80MHz? If you can estimate your transmit duty cycle, then you can calculate a predicted power consumption.
Consumption is on page 52-53:
https://www.espressif.com/sites/default/files/documentation/esp32-c3_datasheet_en.pdf
10% duty cycle transmit (802.11n HT20) and 160MHz cpu (all periph on) would be approx 28mA or ~5.35 hours on a 150mAh battery. Assuming the esp32-c3 is the only device and you've got zero power use elsewhere
ig the short answer is no for day-long battery life on 150mAh. Even an idle 80MHz CPU with all peripherals disabled and no transmit at all, will get at most ~11.5 hr out of a 150mAh battery
Thank you for your answer. So I will try to lower the cpu frequency, but at the end of the day I will probably have to increase the battery capacity or connect more batteries then. about 12 hrs is a minimum for my use case.
Or maybe try to use different chip. I do almost 0 calculations, I just need to send number of impulses from reed switch since start and timing
Would BT 5.2 @ 1/2Mbps be sufficient for your application? nRF52/nRF53 have a fraction of the power consumption
30-40 hours would be pretty achievable with 150mAh
I need to send 6 bytes per second π
But I need a simple cheap board with usb of a size of the coin that a random person can buy from aliexpress and flash using webUSB, I don't see nrf52/3 devkits that small
nrf52 dongle exists
Maybe two coins large?
xiao nRF also exists, though may be a bit less accessible
oh its on ali tho
not a dongle however
In that case I could just use bigger battery without need to learn new platform
That format is perfect
you should consider learning an RTOS that supports a range of hardware. That way platform knowledge is transferrable. nordic recommends zephyr rtos for the nrf52/53 and some esp32 chips (incl c3) are supported by that too.
ay zephyr
It even is supported by platformio, might give it a try with that xiao board
Thank you for the advice
I've learned bits of videojs yesterday and found out it can't autoplay video with audio on due to browser policy. What is the workaround/proper way of doing autoplay with audio on? Or is this not achievable using videojs? What is the recommended plugin i can use to do autoplay video with audio on?
iirc it's not achievable at all. Any way around this browser policy would be considered a "bypass" and prolly be quickly patched by the browsers. Not sure what video plattforms like YouTube do, whether they are in some sort of whitelist or whatever. YouTube embeds don't autoplay tho for the same reason that stuff like VideoJS can't autoplay either...
https://www.w3schools.com/tags/tryit.asp?filename=tryhtml5_video_autoplay yup page there says the same basically can auto play on some browsers but typically muted is more likely to work
Think there was a short time where it did work there and advertisers ruined that... This is why we can't have nice things
YT autoplay works for chrome though, with audio on. Not sure if there is a way to add my own site to the browser whitelist or something.
it comes down to having your own browser so you can make your own exceptions
https://developer.chrome.com/blog/autoplay/#new_behaviors
I'm sure they are but YouTube doesn't actually need to be whitelisted. They pass the MEI by being a website dedicated to video playback. Without passing the MEI, an interaction on your page just has to happen to start playing a video with audio.
Learn best practices for good user experiences with the new autoplay policies in Chrome.
github is down where is the panic button
Literally
what do I do now
https://tenor.com/view/sad-gif-7523306793289960933
switch back to real version management

Hey, at least it wasn't GitLab. They lost some customer data a while back because one of their devs "accidentally" deleted the Prod DB and they didn't have any proper backups 
Iirc it was like 5 minutes worth of data
If you're tasked with deleting a database, make sure you delete the right one.
Sources:
https://about.gitlab.com/blog/2017/02/10/postmortem-of-database-outage-of-january-31/
https://about.gitlab.com/blog/2017/02/01/gitlab-dot-com-database-incident/
Notes:
1:05 - The middle bullet point about the account that had 47,000 IPs was never mentioned ...
Still 
Gotta love how in an incident with GitHub that Kevin covered, one of the GitHub devs explicitly mentioned "We're not GitLab" xD
(in this one: https://youtu.be/dsHyUgGMht0)
A brief maintenance accident turns for the worse as GitHub's database automatically fails over and breaks the website.
Sources:
https://github.blog/2018-10-30-oct21-post-incident-analysis/
https://github.blog/2016-12-08-orchestrator-github/
https://github.blog/2018-06-20-mysql-high-availability-at-github/
https://news.ycombinator.com/item?id=18...
I mean dropping the prod db is always "a slight oopsies"
Hey, I do see the benefit with every single one of these oopsies: We (hopefully) get another Kevin Fang video out of them 
I'm at a loss right now, I had a friend that helped me figure out why my program wasn't looping as it crashes on play_obj.wait_done() as shown with faulthandler in cmd with no luck.
C:\Users\Hunter\Desktop\TvStreamerLoop>python -X faulthandler audio2.pyw
Windows fatal exception: access violation
Thread 0x00003a58 (most recent call first):
File "C:\Program Files\Python312\Lib\site-packages\simpleaudio\shiny.py", line 49 in wait_done
File "C:\Users\Hunter\Desktop\TvStreamerLoop\audio2.pyw", line 9 in <module>
import simpleaudio as sa
import time
filename = 'Roblox.wav'
wave_obj = sa.WaveObject.from_wave_file(filename)
print('can we play?')
play_obj = wave_obj.play()
print('here 1')
play_obj.wait_done() # Wait until sound has finished playing
print('yes we can. sleeping')
time.sleep(250)
print('here 2')
Have you actually read the error? @finite dagger
Does it actually play any sound before crashing?
while play_obj.is_playing():
time.sleep(0.1)
Perhaps try this instead of play_obj.wait_done() # Wait until sound has finished playing
It does, It just crashes. I'll try that cmd and see. 
Crashes on time.sleep still 
All I want is for this program to loop so I can keep my bluetooth connection for my hearing aids to pc. 
Uh oh
Seems like I made a minor oopsie in a for loop!
Can you try a different file to play back? I asked chatgpt yesterday and it mentioned something like that.
Question. I downloaded a GitHub repo via tar.gz file but now I need to do git stuff with it... How do I change it so I can do git stuff without disturbing what's in the folder locally right now? (Context it's mastodon so I don't wanna disturb anything that would result in having to scrap my entire mastodon server)
Usually, ya...
you clone a repo to another directory, then you copy over your changes from your work directory. If you don't know what you changed, well, good luck. You can copy over all files, but you might need to figure out what differences are changes in repo and which were your changes
it's mastadon, i think i shd be ok copying it over directly? supposedly the only things that i have changed are the nginx.conf and data folders
git checkout will make identical folders as ZIP/TGZ download, just adds the .git folder with GIT metadata to it. So if you checkout a project, copy over changes, then GIT will think you edited those files and show those files as changes
ooh just copy over the .git folder aand it should work? i'll try that actually ty
either way, but again, remember that your local folder might not match with what the repository says is the latest, so it could show the whole repository as changed, and you would have to figure out what is your changes and what are just old files which were changed in meantime by repository owner
also just to be sure, keep a copy of your original around, just in case you need to try again π
#Code 1
import heapq
from collections import defaultdict
N = int(input().strip())
values = list(map(int, input().strip().split()))
mp = defaultdict(list)
v = []
for A in values:
heapq.heappush(mp[A // 4], A)
v.append(A // 4)
for i in v:
print(heapq.heappop(mp[i]), end=' ')
print()
#Code 2
def bubble_sort_custom(arr):
n = len(arr)
while True:
swapped = False
for i in range(n - 1):
for j in range(i + 1, n):
if arr[i] > arr[j] and (arr[i] ^ arr[j]) < 4:
arr[i], arr[j] = arr[j], arr[i]
swapped = True
if not swapped:
break
return arr
n = int(input())
arr = list(map(int, input().strip().split()))
sorted_arr = bubble_sort_custom(arr)
print(*sorted_arr)
Both of these code do the same thing. The first one is faster. But i don't quite understand how code1 does it.
The problem is sorting an unordered array but you can only swap two elements where X (xor) Y < 4
i know code1 is faster because it's not suffering from the O(N^2) of the bubble-sort nested loop but i don't know how it sorts it
Case 1
input:
5
2 7 1 5 6
output:
1 5 2 6 7
walk through the steps for a single swap
it'll help you grok how the sorts in each work
also an important lesson this code doesn't really teach in either case is "make sure your mega weird function that does dark magic is commented"
i understand the bubble sort one, i'm not getting the heap one
as for the comments, i didn't write the code
oh I assumed as much! you were using them as examples, but whoever made them needed to throw in a few lines starting with # π
when people write stuff like that its ok to not understand it immediately, stuff like that is really why comments exist so people reading your code dont have to go line by line and figure out what it's doing but basically the dark magic fuckery is in https://docs.python.org/3/library/heapq.html
anyone wanna rate my website https://stableinc.one
- Dark/Light theme doesn't toggle properly. (It works but no animation.
- The domain name flash thingy is too fast.
- hyper links have bad contrast. On lower end monitors it's probably not even visible (or you know, "lower end" eye sight)
- The FAQ section will look better if you allign it in the middle. (Make it half the width, and allign the whole block in the middle, text can stay left-alligned)
- The profile pictures will also look better if you allign middle.
- When zoomed in, the dark/light theme toggle clips behind the "about us" cards, sometimes on top, sometimes behind.
- I would recommend obfuscating the email address in a way that makes it properly readable, but not easy to read for machines (to prevent bot spam, however it's personal prefrence)
- FAQ "What is you discord". spelling error. Also make the link clickable/hyperlink.
Other than that, I like the design. It's clean and has potential.
hi thanks im not the best dev. do you think you could help?
ayy elijah get of mc
I don't have time cuz I am working atm, and I'll be very busy with personal life this week.
But you got very far! I am sure you'll be able to make adjustments
k
this image is to demonstrate how the FAQ section could be changed btw, didn't change anything else
also joost you know how to get a free vps π¦
No idea, I host on my own hardware.
lucky
Oracle Free Tier is perhaps something you can use
does not let me signup
I'm off for work again. Good luck with your website. If you need help, I might be able to provide some this weekend or next week
k
one more thing, fix the border-radius on the bottom-left and bottom-right corners.
serbian... hmm
That would be me Larsy
You get an alert everytime some new IP visits the site ?
Heh, very nice
Lol why would you want that
thats bizzare
||i hate my life just seen visual studio for mac is being discontinued||
Visual studio, not visual studio code, from what I can tell
Good, it was bad anyways
VS on mac was very different than VS on windows (not to be confused with VSC)
what was even the use for VS on Mac ? Can't really think of any
Xamarian
i suspect they will push it to VSCode
like hell, now there is even partial functionality of SQL Developer in VS Code https://marketplace.visualstudio.com/items?itemName=Oracle.sql-developer
The ip π
can anyone recommend me cheap s3? i need 1tb budget 5$ month
Save your money then buy hdd
you're not gonna get that much sotrage for that price, however, one of the cheapest options I have found is Cloudflare R2. They dont charge egress fees either which makes them cheaper compared to most alternatives
Be careful with bandwidth, a lot of vps providers limit his much you can transmit, (or charge you for overages)
this is why you stay protected
Might? It does infringe on GDPR
An IP adress is considered personal identifying information, and not just is agains Discord ToS, it can follow up a whole lawsuit against ya...
At least as far as I know GDPR considers IP addresses personal identifying information, since it can be traced back to a rough location (in some cases as accurate as the city they live in)
Womp womp u agree by clicking into the website find my t&s also ips get deleted after 1 day
That isn't allowed by GDPR. User needs to give explicit consent, and neds to be able to see Privacy Terms before agreeing
idc -> deletes the code proving they in fact do care kek
You can always log it locally on your server, as long as it isn't shared with third parties it should be fine. I doubt you will get in trouble for it though, it's usually the bigger fish that they're after.
always sanitize your inputs guys https://fxtwitter.com/_andychung/status/1826021061957398799
Our team found a great bug on the apple support forum where they don't sanitize html in the discussion title - in this case causing the whole page to .
I just started college today going for a cybersecurity and my professor for my beginners coding class said something that is making me consider going for a computer science major. Does anyone have a recommendation or insight on what would be a better major?
what did they say?
CS is super saturated right now, hard to find jobs
Mainly that with a Computer science degree you can still get a job in cybersecurity but not vice versa.
Seems like more job opportunities with a computer science degree
perhaps, both markets are pretty saturated right now tho
You can look at job postings in your area or in a big city to see what tech stack is in demand. But stuff will change in those few years so I guess it is better to go for more general degree and specialize later. Right now it is a pretty tough time for junior positions worldwide.
yes they are saturated but they are also always growing.
would the general degree be the computer science?
After a quick perusal of job listings for Security Architect / Cyber Security Support etc they all mention Bachelorβs degree in Computer Science, Information Systems, Software Engineering or a related field or even just a University Masterβs degree or equivalent experience . From what I can see it is more about experience and certifications. So I would say computer science is a good general choice.
Alright I'll keep that in mind
The FE is wrong, even if the BE is corrupt the FE should be rendering it as text not code.
It's way way more nuanced than that as there are couple more articles regarding the logging of IP addresses. while it is a PII, they're definitely not the same level as something like a home address.
and on other side BE should remove all 'dangerous' code π
wait what?
not fair
im on windows but i was thinking to switch for team mac
hey guys
i wanna learn how to code
any suggestions on where to start?
anyone here good with css files and firefox?
doubt
You can call me Sensei. What would you like to know, young one?
i have installed tree tabs as an extension
and i am trying to mash together a css file so that it looks nice
but i cannot get it to work, i'd also like to try and frost the top portion of it so that it shows my background, but i have no clue how im going to do that
@marsh star Are you trying to customize firefox with css?
yes
this is as far as i've gotten, the minimize, downsize/maximize, and exit buttons are in the wrong spot
i have succeeded
i am programmer now π€ (i copied code from github and only adjusted pixel values)
now i just want to get the top toolbar to be frosted somehow
I thought you were referring to web dev on firefox. Customizing firefox with css is... not a good idea. Per Firefox's own documentation this feature is deprecated and may no longer be supported in the future.
https://support.mozilla.org/en-US/kb/contributors-guide-firefox-advanced-customization#w_what-do-users-need-to-know-before-they-use-userchrome-css-customization
What you need to know about Firefox customization with the userChrome.css file and how to help users get more help or revert their changes.
i can always just delete the css file if needed
You wont have to -- if Firefox stops supporting it, then the browser just wont look for the css file period. It wont load at all.
If you want to get your feet wet on the web-dev side of things it's pretty easy to get started with python and flask.
i doubt they'd close that path off though, considering they needn't do anything on their end for css files to work
i've no clue about web-dev or anything, is that a more permanent way to customize firefox?
Actually they have already made the decision. They will stop supporting it. The bottom lone is that it is a time-cost to maintain that code.
no... If you want to learn more about html, css, and javascript you can create a simple python-flask application very easily.
i'm just not sure i'd be able to recreate this effect on my own, i used tutorials and copied tons of code from github since i know next to nothing about it myself
ah okay, so at some point it'll just revert to the default state
Correct.
i don't really want to learn much about code, i just wanted to make my firefox look good while also making the vertical tabs fit nicely
when I was researching vertical tabs I saw that Mozilla was considering adding native support, so maybe by the time they shut down css support they'll have implemented vertical tab support
Hi,
i have been developing a small project like youtube where you can upload videos and stream it on different quality
i had finish the upload encoding part using ffmpeg running on ProcessBuilder convert into m3u8 then i will write a master m3u8 file which has various video quality file path depends upon the bandwidth
#EXTM3U
#EXT-X-STREAM-INF:BANDWIDTH=800000,RESOLUTION=640x360
360p/Ghost of Tsushima Directors Cut 2024.08.15 - 23.57.37.01.mp4_360p.m3u8
#EXT-X-STREAM-INF:BANDWIDTH=2800000,RESOLUTION=1280x720
720p/Ghost of Tsushima Directors Cut 2024.08.15 - 23.57.37.01.mp4_720p.m3u8
#EXT-X-STREAM-INF:BANDWIDTH=5000000,RESOLUTION=1920x1080
1080p/Ghost of Tsushima Directors Cut 2024.08.15 - 23.57.37.01.mp4_1080p.m3u8
now i want to stream back seamless in the frontend
is there any best ways to do coz i have seen a solution to use Video.js (VHS plugin)
Is there a tool like GitHub Codespaces or Amazon CodeCatalyst but for local environments? Like something where I could take a blank VSCode/JetBrains/other IDE and have a profile to install all the plugins/tools I need?
Yeah its called a package list. You can export an import-able list of your vscode plugins and then just copy the config for settings...
Also dont use "codespaces"...
HANDS!
I spent my week learning how thermal receipt printers work, and controlling one with JavaScript.
Hooked it up to the @getsentry webhook API and now I get print outs of my issues as they happen
While... impractical? I absolutely love it. I want to implement this on the receipt printers we have at work.
Learning is hard π₯² setting up kubernetes for the first time to work with my homelab servers in a cluster. Ultimately to run clusterplex to test transcoding across worker nodes to spread workloads. Came into this from using proxmox and docker thinking I had a leg up. Nope back to school.
I think the most intimidating part is not knowing all of the yaml variables for kubernetes.
Iβve watched some hour long YouTube videos on Kubernetes, and for learning purposes they use linode VPS which does the setup for you or minikube which runs a local kubernetes cluster which seems like the wrong way to go if Iβm trying to make replicated pods be on other machines I host.
Anyone have any good tutorials Iβm willing to drink the kool-aid just need some good videos to get my feet wet.
Seems to me docker swarm might be the easier solution, but kubernetes is the better way to go but significantly more complex.
What flavor of kubernetes are you using
Some do some of the work for you so it's easier
Does anyone happen to know how to get Idea to see value classes?
Honestly havenβt chosen it yet. I figured Iβd research the hell out of kubernetes before jumping in and possibly frustrating myself from lack of knowledge lol.
yeah I had a little cluster, but kinda gave up after not heading into the IT/dev space. Tried using k3s
I reccomend you just start trying
I wanted to go k8 simply because thatβs what production servers typically use. Figure if I can immerse myself in it enough their is certifications I could get for Kubernetes. But itβs going to be outside my wheelhouse. Still nothing not worth learning.
i can't find myself getting into kubernetes over Docker for my development use cases. It is just overly complicated for non-business, single-node uses cases
it makes sense when you want to move servers around, when you want all the load balancing features, when you want the high availability etc
but when you got a single node for few apps, the whole configuration angle is just a nightmare
one needs to learn all the stuff like ingress, where is konfiguration to what...
I think in the short term for what I want to try with clusterplex Iβll probably just use docker swarm as I think I can get that up and running in 15-20 mins. And tinker with kubernetes on other proxmox vmβs I can spin up in my cluster. Iβve been using LXC in proxmox as my containers, but would prefer docker since itβs more versatile.
for home use, i never needed my containers have HA, so i just run them on one node via docker compose, and if i needed them elsewhere, i just redo them elsewhere (haven't thought about data backups yet tho)
In my case I just use OS mount points for container data, and let backblaze sort the rest
just watched this, my brain still hurts https://www.youtube.com/watch?v=2yplBzPCghA . I mean i get it, it's reproducible, HA etc... But in comparison just starting docker containers this is way, WAY too complicated for a normal home use
Hosting my own services using a homelab has been an absolute dream. However, my first homelab setup had some mistakes, and so I decided to rebuild it from scratch, and share with you how I did so.
Links to my hardware:
Beelink EQ12: https://amzn.to/3zK4Pbu
32GB Memory: https://amzn.to/3Lkotgy
2TB SSD: https://amzn.to/4eV117n
Ubiquiti Switch: ht...
I went with docker swarm for my project. Actually works really well, just need to push hardware encoding to it since docker canβt see the hosts devices yet. As much as I want to still learn K8, for what I wanted to try it would be overkill.
my home use needs are all done by 2 docker instances - one which does majority of stuff, and one on my 'server/NAS', which does media related work. If i ever learned/needed to learn k8s, it will be for actual software development, so the whole setup would be ready for k8s from design perspective. So no monolith, but that's about all of it
I run single-node docker swarms in production on a bunch of servers and it's been great so far. Chances that you need the scalability of k8 are pretty slim, so all it really does is make everything more difficult to set up and debug.
i meant more in context of "i am developing something new from scratch, with chance of doing microservices in Quarkus, adding authentication through Keycloak or authentik, with some relational DB and needs file strage too" - now that is prime candidate for something which could be few docker images, but also a well behaved kubernetes setup with better scaling than the docker variant
Hi, I'm asking in lots of servers what design they prefer out of the two in the images, just want some people's opinions
im trying to make a site, not sure which one i prefer
The first one, but with the colours fixed so that they're legible
hm ok
Just a tip, if you can't see it at 10% brightness, the contrast isn't high enough
huh, i wonder what happened for Apple to drop this requirement
ah nevermind, they added it and then instantly removed it π
Damn one downside to docker swarm that k8 can take advantage of is amd/intel gpu hardware acceleration isnβt supported in docker swarm. Their is scripted workarounds, but clearly something that should work out if the box doesnβt. Guess Iβll pop in a few older nvidia gpus.
That wasn't on my bingo card for 2024
Does anyone know of a better way to deal with a lot of vscode tabs? I need to jump around like 25 files repeatedly to do some task. And opening them all, scrolling between the top tab bar and finding each one is really slow. Not sure if theres like a vertical tab bar or something that can fit them all at once
click on the top search box ?
it has the recently opened list
otherwise i suppose just reopen them by name
How would i make my own version of android, like idk debloated and install it on an android device? Like a custom os of some sort xd
do you know hwo to make your own linux
do that first. easier to find help and documentation.
And equips you with the skill to go mobile
I was planning on doing something and run my own custom OS on android devices, might just go with linux fr
how do you think android is made
real xD
yup, sums it up
have fun with hardware compatibility
I would just install something like LineageOS or GrapheneOS.
However custom roms don't work on any phone, some manufacturers don't allow other versions by restricting access to the bootloader
this code requires organising in C# but ive never learnt C# before, just wondering if anyone can shed some light on how the syntax in this language works
id rather not go for trial and error and try to actually understand how this works
i dont want to just give you the answer
but essentially, the problem itself can be divided into three segments
the main function which calls FtoC and CtoF
and FtoC and CtoF
you can tell what the bodies of FtoC and CtoF are based on the two return statements here
the former of the two code blocks takes C and converts it to F, the latter takes F and converts it to C
from there it's just piecing the other bits together which is relatively simpkle
i still dont understand what all the bracketed sections are for, when piecing parts of the code together i always run into errors because i cant figure out how to properly arrange it with the different signs like semi-colons, etc
every opening bracket { needs a closing bracket }
semicolons go at the end of each line, basically tells the compiler "hey this block of code is finished, what comes after it is separate"
so basically in between the brackets i have to have (for example)
{
return (C * 1.8) + 32***;***
}
so the brackets indicate the start and end of a subroutine or other sub-section of a program like for example calling FtoC and the semi-colons indicate the ends of the blocks of code within those subroutines?
im very sorry if im being an idiot here lol, i just suck at logical thinking in a coding perspective
yes
it takes practice
brackets at their lowest level determine a section in code
i see
how the individual programming language chooses to use them is up to it, c# is pretty verbose on that fromt
@nocturne kernel i can check ur answer if youd like
its fine, still working on some stuff atm, im gonna get my teacher to review it so he is relatively familiarised with my skill level
i appreciate the help :)

no problem
idk rust but its prob just auto generated
solid "its complicated"
think im going insane
does it work if you ctrl c the not-working version and then run the same not-working version again
'cause its not even getting to the clearing of the terminal when you run the not-working version. but my thinking is every time it didnt work it was a fresh terminal and it was hung on some weird terminal fuckery that for some reason fixes whenever you try it a 2nd time regardless of the unrelated print line
If it's auto generated u should be able to just do the generator and push the new thing and ignore conflicts
Unfortunately it's not that simple in this scenario
Oh that stinks
Seems like i'm falling in love with OOP more and more. Can't wrap my head around it at first but now it just makes sense
Its ok we all had a time of such foolishness π /s
its ok to write bad or slow code if you just write a comment saying TODO: fix this
KEK
nope i i tried both from a used and from a new shell(new terminal window)
no it really is somehow the unrelated print line that does it weirdly enough , im suspected something with timeing is going wrong (hence why i included the tamestamp everywhere but so far no actuall differentce between the order of things in the two
i use select in python to check if sys.stdin is changed , so i know the key input was for me when asking pynput to store every keystroke in a 1 key buffer so i can fetch whatever key was pressed wich i also then use to figure out how many chars are printed and need to be read from stdin (if you read one to few , stdin wont clear and select for the next keystroke wont work, but if you read 1 char to many the read command hangs untill there is one more character to be read, wich is kind of a nogo if you trying to make a nonblocking keydetect π (arrow keys send 3 characters , space and enter send two (esc+n for enter)
damn do i regret booting to windows alreadi you would think that the start menu would at the verry minimum find stuff thats in the startmenu or on the system path... but damn is windows getting worser worser each time
, found the issue (i hope + i think) apparrently the python printstatement does not flush by default, adding flush=True , seems to fix it
(without the end='' the the terminal would flush on '\n' but in noncanonical mode the terminal should flush on every character , emediatly with ICANON
or atleast thats how i understand the termios debacle, had to use the C manpages for termios to find out as there is no mention of it in the python termios docs
When I have to write a GitHub tutorial because the developer instructions are lacking. π₯² and still.. canβt 100% make it work.
Follow this guide for Intel Quicksync HW Transcoding for docker swarm. Follow steps on each swarm node. https://www.linkedin.com/pulse/docker-swarm-reducing-plex-cpu-utilisation-60-reis-holmes Edit...
Does anyone here have experience in Ionic (with Vue) and can help me with figuring out how to use ion-modal?
https://stackoverflow.com/questions/78957478/how-can-i-add-and-differentiate-multiple-ion-modal-elements-ionic-framework-v
next fun one lol :
https://docs.python.org/3/library/termios.html this one doesnt mention anything about this : (from here https://man7.org/linux/man-pages/man3/termios.3.html)
might be saying <instance of List> != <other instance of List>
thats is not == i believe
so i think if they compare equal this should be okay :S
a dumb use case for the keystroke thing i made , its not blocking what not really shows in the clip but hey, im better in ansi stuff then the primeagen π as this doesnt rewrite lines only the stuff that actually changes changes (hence no flickering) π
gotcha my mistake. and also thats sick ^
still has buggs in it and can only do folders atm, gonna make it do dicts and lists aswell so youcan feed it anything to be a menu even a directory with config files , and it should brows the actual configs , (got that nearly but some hickups since apparenly my simple configs arent equal :P)
also to make it extra hard on myself i decided that it should be part of the terminal stream and not wipe or lose any scrollback history :
still has some bugs as you can see
of and it only reads the next menu on actually selecting the previous one, so it should take new files into account during the thing
Alright, I'm not a js wiz, but I have this working well until I fell into this case. So I want the parent page to be made bold, in the first navigation bar, but I don't want that to be the case for the "Baseboard Home" (since both are /baseboard), so how would make an exception when the second navigation bar is in use. Here's some excerpts of the code:
JS:
const navItems = document.querySelectorAll('.nav-item, .nav-item2');
navItems.forEach(item => {
if ((item.href === window.location.href) || (item.href !== window.location.origin + '/' && window.location.href.startsWith(item.href))
){
item.classList.add('active');
};
});
});
HTML:
<nav>
<div class="menu-icon" onclick="toggleMenu()" style="">β° Main Menu</div>
<ul id="nav-list">
<li><a href="/" class="nav-item">Home</a></li>
<li><a href="/baseboard" class="nav-item">Baseboard</a></li>
<li><a href="/downloads" class="nav-item">Downloads & Tools</a></li>
<li><a href="http://blog.duplexsw.com" class="nav-item">Blog</a></li>
<li><a href="/help" class="nav-item">Support</a></li>
<li><a href="/about" class="nav-item">About</a></li>
</ul>
</nav>
<nav id="navbar2">
<div class="menu-icon" id="menu-icon2" onclick="toggleMenu2()">β° Baseboard Menu</div>
<ul id="nav-list2">
<li><a class="nav-item2" href="/baseboard">Baseboard Home</a></li>
<li><a class="nav-item2" href="/baseboard/intro">What is Baseboard?</a></li>
<li><a class="nav-item2" href="/baseboard/beta">Download</a></li>
<li><a class="nav-item2" href="/baseboard/docs">Documentation</a></li>
<li><a class="nav-item2" href="/about/contribute#bb">Contribute</a></li>
</ul>
</nav>
Any help would be greatly appreciated
yes, I am coding this mostly from scratch
compare the pathname https://developer.mozilla.org/en-US/docs/Web/API/Location/pathname
Alright, I will
This is a stupid question but is there any way to tell a compiler to place some data in the CPU cache? I have something I want to run as fast as possible, for fun
Or is it up to the CPU
You can't directly tell the CPU caches to do anything. Instead you have to design code that is inherently cache friendly. It's a fairly complicated topic. Many have made conference talks about this that are on YouTube.
Sweet
Didn't know any of this. Ty
What's a good course for computer architecture
I wanna learn how to make processors
This series from Ben Eater on YouTube is a great series explaining how to build your own computer, he goes deep into how it all works and explains things quite clearly, easily one of the most welcoming pieces of content to view if you want to get into a deep dive
Thanks
Hey guys, any experienced developers/engineers here?
I was doing research on whether it's too late to get into development in 2024 because of AI and it's impossible to find the answer because everyone has a different opinion
So I was wondering did anyone else here do some research on this and what's the conclusion?
It is never too late to get into development, but, you absolutely have to temper your expectations, a lot of people go into the industry thinking theyre going to work at MS/Google or some other FAANG company, but the reality is that a lot of them wont even get a chance at that. So long as you know why you want it, and can accept that you're not going to be the next silicon valley CEO, youll get on fine
I would be more than happy with even a low paying job in tech since that is already twice as much as the jobs I did before (construction work and similar)
I know a little bit of Python and C (wrote a few little games in PyGame, finished Harvard's CS50)
Although not very related, I have the CompTIA A+ so I'm not a complete novice in IT.
Do you think I can land a job within a year if I study and practice for 8 hours a day?
I mean, so long as you are building things which can be put on a portfolio, and those things are fairly solid, theres a chance that someone might hire you, yes. A lot of companies in my experience are no longer looking for degrees and are instead moving to a more "prove you can" approach by going with portfolios over certificates. Knowing your stuff will never hurt though
Yeah a very good portfolio is my plan, what would you say are the chances of getting hired within 12 months though?
that would depend on where you are, what companies are located nearby, and what sort of skill you can pick up in a year, and so that is something I likely cannot answer for you
I have a 12 month contract for a part-time job where I only work 2 hours a day, meaning I literally have a full year that I can use to become a developer
I'm located in Germany and I really enjoy studying and improving
I dont have any software buddies located over that way, so I cannot really comment, however I do hear that the job market there is on the up, though that might only be a state level thing, the only thing I can say is that it might be worth a shot
No honest professional developer thinks AI is gonna replace coders. Only the C-suite of AI companies and those they've paid handsomely parrot that rhetoric. Be suspicious of any dev saying AI coding is of adequate quality, as even juniors are mostly capable of recognizing how mediocre its output continues to be.
And don't get me wrong, the fact that AI tech can do what it does is pretty amazing. It's just not actually that useful in anything more than the most rudimentary of problems that have already been solved.
How is a supercomputer hooked up to a database going to replace anyone?
There is no intelligence in GPT
Why are you asking me? My stance is clear in the first sentence
You write something on Discord, someone responds to you and you wonder why?
You replied to me with a question that ignores everything I said. Can you not read?
People on Discord are so dull, itβs a bit absurd
My man you asked the person who just explained why ai isn't going to replace anyone, how ai is going to replace people
That's like saying that you're annoyed that the chocolate tea pot melted when you filled it with boiling water. Or lighting a firework then being angry when it explodes
Why
Yeah youre right, in the end, we aren't gonna see carpenters put snippets of code together
I've experimented with it, and it honeslty cannot provide the same quality that writing the code yourself does. It struggles to follow a lot of basic principles when it comes to trying to develop a full solution with it
I asked on a different server if Web Development is still worth it in 2024, and a guy told me he was a Web Dev for 7 years and got replaced by a drag and drop builder 
sounds like buddy had a boss who didnt realise how drag and drop works lol
Must be a troll lol
π makes sense if he's only making wordpress websites
Yeah I kept replying to him for a few more minutes and hes definitely a troll
Is NameJet a legit site for backordering a domain?
I was wondering. How do yall use github or git. like in university you just push to the repo without much resistance but when you are contributing to a public repo, you have to fork, make a branch, make your changes, commit that branch, commit to the forked origin, and then create a pull request.
am I missing something here or is that how most workflows go?
ok
Is there an easy way to push hugo generated files from a private gitea to a github repo in actions?
No
i don't believe you
yes, welcome to workflows
usually you make a series of changes on your dev branch before the PR so it's not a big deal
Anyone here familiar with dnscrypt_proxy? How do I verify the outgoing request from dnscrypt_proxy to public dns server is indeed encrypted?
wireshark
Damn so you do have to fork to work on something. Ok thanks for the info. Is there any other way that people go about this? Like lets say its a big code base. I cant download 200 gb?
ok
Well yeah you have to download the project to make a change
Git shouldn't be used for big files, so having 200gb's of text is so big you probably won't even write that much code in your lifetime
Big files, like database exports are usually shared via another route
backtracking the disaster that was earlier this year xD
Good on them
Couldnt care less about the professional tier pricing, but that fee was ridiculous
Yeah. The whole thing was a mess
they also did it at a time where open source alternatives like godot were at an all-time-high usability standpoint
had they done it 2 years prior, and i don't think it'd have had the same effect
How hard is it to create and use stuff like Lottie or Rive on your website? Or does it require two professionals, one for Lottie and one for Web Dev?
For rive at least, looks like they have packags to run it on web. shouldnt be hard
Fill me in on Lottie or Rive? I am totally out of the loop. I can make all sorts of animations in CSS and JS, write my own SVGs and stuff, but am lost on what these editors do and deliver as a final product. Do they bundle SVGs with JS or something? At some point I am going to need to actually just try it all out...
Ok
bro my website is requesting a file that dosent exist and returning with status 200?
i thought it was just cached but now im just confused
yeah
hi
What do people think about the various (potentially) upcoming Java features? Things like:
valueclasses- Condensers
lazyfields- Classfile API
- Vector API
- Primitive/better generics
- Flexible constructors
- Nullability constraints
Ok
well, check the server. Maybe you have it set to return that code in the web server config, or maybe you are serving static files via your own server which returns 200 for everything
me happy to be finally on Java 11 feature set when we do stuff π
At least you have some of the QoL stuff past 8.
and in java world, i am a simple man, i return to Java world only here and there. My current 'plan' is to make some backends in Quarkus
its a free website it aint got that stuff
eat bean and egg burrito
it's not a language, it is a library/server for coding backends https://quarkus.io/
very simply put Quarkus is when you take best parts of J2EE, combine it with rapid development cycle build system, very fast startup engine and modular architecture (you add only modules you need), optimized for building fast starting servers (so it is good for Kubernetes).
Does it do any shenanigans with pre warmed JVMs?
Hi guys, I'm currently using InteliJ version 2024.2.1 on MacOS Sonoma 14.7, currently I'm experiencing a bug where I press "New Project" or "Open", a Finder pop up appeared on my screen and when I clicked the folder named "CS-Java", it froze the entire program and I have to hard kill it (this happen multiple times and pretty random too). So how can I solve this issue ? Here is the picture of the error
Btw, there is another program which froze too alongside with InteliJ, which is something called "Open and Save service (InteliJ)" (idk if that is the actual name of that program)
(ABSENT)
Interesting... Have you tried updating to the latest IntelliJ version and newest MacOS?
yea i have done that tho, but still get the same error π
I have not used IntelliJ on Mac, but I suppose you might have to reinstall it. Maybe somebody else has a solution, but that's the only solution I was able to come to think of
https://intellij-support.jetbrains.com/hc/en-us/articles/207241085-Locating-IDE-log-files
If you know the rough time when it last froze, then find that in the logs. Can't promise it'll clue you to a direct solution but there's usually at least some indication of the problem.
oh yea i have found the issue, it was just some of the buggy extension from the marketplace tho
But somehow another issue arise, my project root directory was constantly missing and it said something like "indexing...", how can i fix this too π (i can only saw this 2 directories)
let it index
no. but it moves bunch of stuff into the compilation step, plus you can even use GraalVM for even faster startup
That sounds like Leyden.
yea but it constantly disappearing and idk why
although the indexes thing has completed
have to downgrade to intelij 1.6 tho
Fr
Any good open source database clients?
People like dbeaver, dunno if it's open source
DBeaver is solid and it's at least on GitHub, so it might be open source? https://github.com/dbeaver/dbeaver
I use DBeaver at work but I prefer DataGrip by JetBrains for personal stuff
I think HeidiSQL is also open source
I need to learn the basics of python in 2 weeks
Python is very simple to get into. I recommend working through some coding website like projecteuler or similar and working through the problems, as well as using websites like geeks4geeks to understand in layman's terms how to use stuff
The hardest part about Python is gonna be if you're not used to how programming logic works, but it's relatively simple to get into
The biggest gotchas are that it doesn't like different whitespace in the same file, whitespace matters (unlike most languages) and it has a generally funky syntax.
The nice thing is that most modern IDEs, even VSCode with Prettier or something like that will autoformat it for the most part. It might not fix it if you put something outside a function that was meant to be inside, but it should keep it relatively clean for you
If you have a .edu email I generally recommend getting into Jetbrains tools, but VSCode is great and you can't go wrong with it
I do pay for Idea and friends because they are so good.
oh same, I got it for free in college and when I graduated I used their "former student" discount and I'll never get rid of it
it's one of the few yearly subscriptions I have that I think is 100% worth it
I cannot learn something by myself someone has to be there to keep me focused
It gets cheaper for the first few years too.
Remote Java, C, C++, and other bad languages too.
Well I guess Go is decent.
I recommend getting the GitHub student starter pack instead, you get JetBrains' tools and a bunch of other stuff for working with data, hosts, copilot, and a lot of other cool things.
I am desperate and can't find the issue here?
has anyone tried uploading a file to an API with Multipart/form-data content type? I can not figure out where I am going wrong. I need to send an json file as a descriptor and send the file.
upload.html
<!DOCTYPE html>
<html>
<body>
<form action="upload.php" method="post" enctype="multipart/form-data">
Select image to upload:
<input type="file" name="file" id="file">
<input type="submit" value="Upload file" name="submit">
</form>
</body>
</html>
upload.php
<?php
$target_url = 'https://10.12.6.10:7777/api/v1';
$api_key = "api key";
$ch = curl_init();
if (function_exists('curl_file_create')) { // php 5.5+
$post['saveGameFile'] = curl_file_create($_FILES['file']['tmp_name'], "application/octet-stream", $_FILES['file']['name']);
} else { //
$cFile = '@' . realpath($_FILES['file']['tmp_name']);
}
$jsonarray = [
'function' => 'UploadFile',
'data' => [
"FileName" => "test",
"LoadFile" => "false",
"EnableAdvancedFileSettings" => "false",
]
];
$data = json_encode($jsonarray);
$data_size = strlen($data)*8;
$post['data'] = curl_file_create("data", 'application/json', 'data.json');
curl_setopt($ch, CURLOPT_URL,$target_url);
curl_setopt($ch, CURLOPT_HTTPHEADER, [
"Authorization: $api_key",
"Content-Type: multipart/form-data",
'Content-Length: ' . $_FILES['file']['size'] + $data_size
]);
//curl_setopt($ch, CURLOPT_HEADER , true);
//curl_setopt($ch, CURLOPT_NOBODY , true);
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false);
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, false);
curl_setopt($ch, CURLOPT_POST,1);
curl_setopt($ch, CURLOPT_POSTFIELDS, $post);
$result=curl_exec ($ch);
$result_code = curl_getinfo($ch, CURLINFO_HTTP_CODE);
$error = curl_error($ch);
echo "<pre>";
echo "Result = " .$result;
echo "\nResult code = " .$result_code;
echo "\nError = " .$error;
curl_close ($ch);
What's the error?
I have tried sooo many things. right now with that i get Error = operation aborted by callback
Receive the form input and debug it first without the file
That seems to be fine. I don't have access to the api side, I tried contacting the developer but have not gotten a response.
<?php
// Replace with the URL of the target API endpoint
$target_api_url = "https://api.example.com/upload";
// Check if a file was uploaded
if (isset($_FILES['file'])) {
$file_tmp_name = $_FILES['file']['tmp_name'];
$file_name = $_FILES['file']['name'];
// Prepare the data for the POST request
$data = [
'file' => new \CURLFile($file_tmp_name, 'application/octet-stream', $file_name)
];
// Set up cURL options
$ch = curl_init($target_api_url);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, $data);
// Execute the request
$response = curl_exec($ch);
// Check for errors
if (curl_errno($ch)) {
echo 'Error: ' . curl_error($ch);
} else {
// Handle the response from the API
echo $response;
}
// Close the cURL handle
curl_close($ch);
} else {
echo "No file uploaded.";
}```
You're mixing, JSON, with multipart and octect... what's the API specification?
to be fair it is this, I had changed some of the json stuff for another sever I was in. I have my reasons.
I have managed to implement all the rest of the API with no issues, but this one is not working.
the full reference is here https://satisfactory.wiki.gg/wiki/Dedicated_servers/HTTPS_API
This page is a copy of the HTTPS API section of the DedicatedServerAPIDocs.md document distributed to every player, with some community notes and edits.
is "data" supposed to be there or is it supposed to be $data
want to learn python what book is the beast i live in australia
kangaroos are the beast if you live in australia
honestly books probably aint it with python
for the languages like C/C++/assembly/etc those languages were built for the books world but python was invented in a post-books society so the best way to learn would be finding something more interactive to mess around with
the formerly mentioned languages you need to know things in advance. python you need to learn the things by breaking the things because otherwise you're basically reading pseudo-code that is already 80% python
and ton of tutorials/guides... like let's say https://github.com/Asabeneh/30-Days-Of-Python (no idea if it is good or bad, the point is that one doesn't need books. they are nice but not required)
sometimes i marvell at pythons simplicity , dunno how long but i had no id that you could pull a division on Path objects to concatenate them into a new Path π
I haven't done webdev since the Coldfusion days. What's the most popular way to host an unlimited scale database? Spanner?
Not very into this topic, but mssql allows sharding, which basically means host a database spread across multiple database servers.
hi guys, tried everything but can't get an answer to this question. I have a java maven project with a property and a dependency inside DependencyManagement. Is it possible to set that property at runtime? I tried using System.setProperty("property", "value") but with zero success.
That depends on the data you're storing and what kind of scale you need
There's no one answer to what database should I use
<vector, data> pairs and being able to handle bird app scale without a sweat (with extra funding) would be a plus. I'm not looking to innovate in the cloud data space, just trying to serve data to a Unity app.
There is no handling bird app scale without sweat lol, that's like an entire teams full time job.
But it sounds like you're just doing a ton of reads without many writes?
Really? Dang. I would have thought there was an off the shelf solution that handles scaling for you like GCP or something. I've never tried so I have no clue.
Ya. Tons of reads, less writes. But still a lot of writes. Probably similar post vs view ratio as bird app.
If you've got the cash for it, GCP or azure do offer some self scaling offerings
They start at free tiers right? I would have thought if a prototype showed promise, getting funds to scale would be the easier part of the project challenge.
Though maybe if the costs are too high to get more quota, that might not be true.
I think they'll give you some credit but it's not a truly free tier
Booo megacorps. Bring back free shit for small time people.
I usually just host my own stuff on rented servers, it's cheaper but comes at the cost of maintaining it yourself
For your POC phase at least you could probably run something like Postgres on a little vps
GCP is probably fine for my CRUD app.
Most do have free tiers, but 5GB in blob storage and 1GB for a nosql db ain't much
https://cloud.google.com/free/docs/free-cloud-features#storage
Yeah but it's for actual db stuff
I don't know know if Spanner can be treated like a vector db.
"Vertex AI Vector Search" hmmmm
Yeah I don't know much about storing vectors
i don't undestand what are you trying to achieve. maven dependencyManagement is a maven internal thing, and dependencies are just a build time thing which can translate into the manifest file inside the result JAR file. So not sure how you intend to get that into runtime ? If you mean just get a property from maven build into an application build by that maven project, then just generate a properties file at build time you can then load in your app as it will be put in the classpath by being in the generated JAR file ?
nobody responds to "i need help" because nobody knows if they could actually help
just say what your question or problem is and if someone can help they will
What is the worst part of C#?
Howdy all, I'm looking for someone to help debug some wierd apache issues (specifically mTLS securing some but not all of a site). I'm wondering if anyone here is interested?
The fact that its just microsoft java. It fills a need nobody had. If you want wierd runs on redistributables but is also fairly machine agnostic- java. If you want something that requires specifics in the target use c/c++ or rust or golang. c# is like- specific target but also runs on redistributables- it's just like- theres something that does the job faster / with less hastle / just- better in all circumstances.
owh okay
I've written a game or two before.
uhmm
I do game dev.
Has anyone tried out Cursor yet?
Hello, network! Iβm a skilled full-stack developer specializing in the JavaScript ecosystem, with extensive experience building e-commerce platforms, online stores, and real estate websites. Iβve delivered these projects using React + Node and Vue + Laravel for clients.
Additionally, I have hands-on experience with React Native for mobile app development in crafting user-friendly, responsive websites that deliver seamless experiences across devices.
I have completed my recent tasks, and now I want to work with new reliable clients on long-term projects. If youβre looking for a dedicated developer to bring your vision to life, Iβd love to connect!
pls contact me via dm.
Dang. I'm not reliable. Sorry!
The file format is pretty bad, but the fake cross platform stuff is reallly bad.
You typically can't have dependencies at runtime. It's how the code is compiled. You can pull it if you do something called lazy loading but that's more for scripting languages from what I know. If you are talking about getting a property, you are better off setting an environment variable. But that it is at build time too. If you want something that is dynamic, then having it set in a external source like a database or API may be the way. Just depends on your use case.
DLLs:
What stuff should I install for python development?
Python
and an ide like pycharm if you are completely new
anyone else's school forced them to use asp (absolute dogshit) for a CRUD app and then you accidentally overengineered it or just me?
well, at least its useful for job prospects π€·ββοΈ
like i need to hack it all the time to make it respond with json
protected void Page_Load(object sender, EventArgs e)
{
Response.Clear();
Response.ContentType = "application/json; charset=utf-8";
Response.Write(GenerateJson());
Response.End();
}
like look at this garbage
have fun with it
i mangled a lot of my class projects bcs i thought it was stupid to implement it the way they want, just go for it
forreal they had a requirement to have some stupid mechanisms so instead I built in pure JS a client side router to make it a single page application
OSGi says hello (but that is yet another can of worms)
wait till you find out about existence of Objective C (for iOS and Mac development)
bro reads the ECMA specs. I suspect ECMAScript would be just as bad if not worse in specs
This is an ECMA spec.
I can't tell how wide the FieldList and MethodList values are supposed to be from this.
are you (re)implementing internals of it ?
I'm just trying to parse the CIL file for a project.
Does anyone know how I can add a DataGrid with resizable headers that displays data from an SQL database in my MainPage XAML?
noone told me cloudflare caches by default and that led to like 45m of "whys my CSS not working even with ctrl F5" π
I mean that's the whole point of cloudflare, a CDN
Just flush the cache when you make a change
i just wanted a registrar but omg am i loving all these features
i used to use fasthosts as my registrar but its not the best
I don't like having everyone at one place, can lead to a bad situation
So I use porkbun as registrar, CF for DNS and CDN
110% this. I keep my domains registration, DNS records, hosting, and CDN completely separate. More annoying to manage, but it means I can easily replace the DNS records, hosting, or CDN if there's a problem.
im hosting ts on a 2018 mid range laptopπ fine for like me and my freind to learn and play with stuff tho
Does anyone happen to know how I can configure the Java compilation tasks on Android for Jabel?
Severe skill issue
If CF disappears I have bigger issues than not having access to my domains
If I had a separate registrar that disappeared then I'd have cloudflare but no domain
So I see zero point in paying more to have them seperate
That's the entire point.
I know but I thought you had to turn it on, I'm so happy I have cloudfkare now that the websites more complicated tho
One trick would be to put the IP of the server in your hosts file.
i think cloudflare also has an experimental mode or dev mode to disable caching to a part of your site
I need help getting this running... Very simple but I'm unable to get it up as I don't know anything about coding. https://github.com/NelsonDane/auto-rsa?tab=readme-ov-file#cli-tool-installation-
If you can help please dm me
You just put hashes on the ends of all imports, like all major frameworks do.
Or..... Don't develop in production
unless they didnt realize it was enabled by default which was the OP
When it's a simple personal site there is no dev environment
No need to
Skill issue
Wrong. There is a dev environment, it just happens to also be the production environment.
True
well i hear twitter uses prod environment for development wasnt there a whole 'pulling a twitter' or something
meanwhile i am watching Wordpress implode not because of software quality (or lack of), but because their CEO is killing their own product it seems
Dont have any public personal projects and u wont have the problem!
Good.
Good
Is it a good idea to work on 1 big project rather than several mini ones for learning
Generally I would say smaller projects are better, but it realllly depends what you're trying to learn
so im making a backend for my app but it seems to run fine but vs seems to have a fit
Like old school editing php files over ftp π ?
Have you used a Virtual Environment? If so, that would make sense, if VSCode isn't launched from within the venv... Otherwise, restarting VSCode usually fixes missing dependency errors
damn found a bug in a project , rapported it , they cant see the issue,... i figured maybe my explanation is bad but this should do it... if anyone here can spot the issue say so (meaning im not completely nuts for thinking this is a bug)
actually i do think the ceo is kindof not wrong here,... wpengine could either pay to use the wp trademark name or rename their product if they dont plan to contribute, ... , opensource by definition does not require any contribution but as far as i know using a trademarked name has verry little to do with opensource and is a seperate cn of worms , so if youwant to ride on the name , and the price for that is contribute or pay up... i think thats pretty fair
tbh never used wp, as back when i dabbled with some webdev (when jquery was still the hot new thing on the block) wp was know as an awesome remote shell with neat blogging functionallity :), and for my personal stuff i used and still do modx
wich is a kindof really hands off approach to cms and is more like a php templating sytem that includes a database storage framework for your stuff or so π
- WP is not trademarked. Wordpress is.
- asking for a "contribution" at 8% of gross reventue (not profit, revenue) is absurd even for likes of Oracle.
- the whole fake job inteview saga
- it is not the Wordpress organization asking for money, but Automattic,
In short, while yes, it would be nice if WPE contributed financially to Wordpress (they contribute with code/plugins/events), the whole thing is pretty much the Automattic's CEO ego trip and has nothing to do with open source financing
September 24 2024
The abbreviation βWPβ is not covered by the WordPress trademarks and you are free to use it in any way you see fit.
September 25 2024:
The abbreviation βWPβ is not covered by the WordPress trademarks, but please donβt use it in a way that confuses people. For example, many people think WP Engine is βWordPress Engineβ and officially associated with WordPress, which itβs not. They have never once even donated to the WordPress Foundation, despite making billions of revenue on top of WordPress.
until 2024 it was not confusing and it is still not covered by trademarks, so as said, it is the Automattic's CEO's own invented claim and Automattic could lose very badly with this
Anybody building or thinking about agents that spend money?
sounds like a bad idea
does anyone here have experience with runnign ngrok on truenas?
hello i have a question
When I set up the site with nginx, I want to proxy via caddy server via cloudflare.
via /etc/caddy/Caddyfile
that is not how trademarks work , try setting up a fastfood chain Micky D's and see how fast your being sued , and lose the lawsuit, because , mc donalds is also known as,... and your in the same bussiness so confusion is possible , hece WP is covered under the same deal ,... MS Microsoft same deal
also they cant really deny that the WP in WP Engine stands for wordpress , as there mission statement is :WP Engine empowers companies and agencies of all sizes to create, manage, and optimize their WordPress websites with confidence.
also note that WPE is a publicly traded company , they inherently are the bad guys imho, since there sole mission is to extract as much value from their users and spend as little as possible to achieve that maximum extraction . since if they arent doing that they are technically breaking the law. if it were a private company i might have looked at it differently but since they really kind of have to screw everyone over i cant see how i can take their side...
oh no someone took the automattic koolaid
They explicitly said wp is not trademarked
The abbreviation βWPβ is not covered by the WordPress trademarks
In compose multiplatform, is it possible to prevent scrolling a TextField from scrolling its label? For example with this code
TextField(
username,
{ username = it },
label = { Text("e.g., @id:example.com") },
modifier = Modifier.requiredWidth(300.dp).requiredHeight(50.dp).horizontalScroll(
rememberScrollState(),
reverseScrolling = true
)
)
I would prefer for the label to always stay as it's positioned in screenshot 1, never screenshot 2 (or anything else).
i know but MS is also not trademarked by Microsoft , but guess what will happen if you start selling MS Lindows , and have it be an operating system, people might think its actually form microsoft because of the MS in front and Microsoft wont like that and if its a bad product they have more than enough reason to shut it down asap... WP might not be covered by their Wordpress Trademark, but when a company uses WP and Sells Wordpress , and as an extra , when their product breaks or for other stuff points people to Wordpress or people go knocking there for support instead of on the WP engine door, i think that that is enough proof of trademark violation in terms of WP being synonymous with Wordpress. i also can see that the 'you do with it as you want' remark was made in good faith and not meant as you can use it to run my stuff into the ground , and maybe aswell to not get any WPWW members overagitated , or being obligated to also actively get them to stop using WP WW , since www and wp , also confusing. anyway. i also cant see any reason why he should be obligated to continue to pay for bandwith and be forced to let wpengine access wordpress.org ... if nothing else there would not have been a problem if wpengine had wordpress.org mirrored and cached for their users, but that would have moved the costs to them so ofc they werent gonna do that
at the same time if you don't enforce a trademark, that doesn't matter legally
so WP Engine did nothing wrong
i have stuff i made and released under the unlicense, so in theory you could take that bag of bugs and go sell it online for big $$ , and if people findout you pointing them to me , and i get a name for being a scammer because of it... i would not like that either , but the license is not stopping you from doing exactly that
the lawsuit was verry wrong in my eyes
he wont give us his stuff anymore π¦
and we need his stuff to make more $$
they could have just forked the project and made it better and theirs to control ...with their support and their community , if they really cant deal with the guy that made it... yet they chose to go the lawsuit route so they dont have to do much efffort themselfs.
Wordpress is shit anyway, so they nobody should expect anything but for it to break
What's better
Writing custom websites from scratch...
I know kinda elitist, but it really is the best.. Sitepad is not bad I heard, and there are some other ones that aren't bad either, but yeah wordpress is very insecure
What if you need to edit? I don't want to write a CMS
WordPress is fine as long as you use the right plugins
Well, yeah, but damn is it ever buggy and insecure
Also, there is one starting with C that is actually really nice... can't remember its name tho
Contao it's called
Doesn't have block editing
Well
Clients want that. WordPress really is the easiest solution and works for most sites
I don't know, I don't use WordPress, as I only build my own websites
Over 100 clients and it's been fine
There you go, if you use it it's not bad
LOL the bot just deleted my message
I was like, it had me raging like hell
It was buggy, looked AWFUl and was a mess overall
I mean have to customize it, default theme won't get you far
NASA.gov runs on WordPress, and most people want basic info sites which WordPress is good at
yeah, but the rabbit hole goes so far down, I prefer to just build stuff myself, as I feel I embarrass myself if I use WP
WP is good if the client doesnβt need to get help from a web developer to update the information on the website.
Well then what's the point, right? Then the website can simply be created from scratch using HTML, CSS & JS
Strapi as REST backend & CMS, whatever you desire as frontend. (and yes, Strapi has it's own limitations, mostly down to business features like customization of content ACL and stuff)
and apparently even that is not true anymore since Strapi 4.8
I just mean incase they have no technical knowledge WP is easier for a client to update themselves depending on how you build it. If they have to rely on a web developer sure that keeps you paid, but might swamp you with work depending on if your doing this solo.
i want to point out that this was one of my first statements on the topic π , it might sound positive but it really isnt :p
srry i misremembered the quote a bit : here is the original :https://bash-org-archive.com/?0949214
<azonenberg> wordpress is an unauthenticated remote shell that, as a useful side feature, also contains a blog
btw .RIP .io domains
With the United Kingdom giving up sovereignty of the British Indian Ocean Territory to Mauritius[17] (but maintaining the military base on Diego Garcia via an initial 99 year lease), it is possible under IANA rules, the .io domain will eventually have to be phased out within the following several years,[18][19] although historically, some exceptions been granted, as was the case for .su.
lol seems to me they want ot make the web more complicated and more dependable on searching things vs knowing things , adding TLD's like .vlaanderen and scrapping .io
nah, it's just country stopping existing
like there are no .cs domains
well ya but , i was merely saying shame that they deleted a 2 letter tld, as they are easy to remember (and will get more traffic because of that) then a .vlaanderen tld as nobody remembers a tld using that domain painted on the side of a car π
visit now: dev.io or dev.vlaanderen (or any othe obscure long tld)
i have a dot net and a .be , the dot net was allot cheaper than the .be :p
(and i got the .site and .xyz and .tech were almost free :p)
(just registered the names without hosting so thats why the .tech cost 50ct for a year or so)
and couldnt they move the tld from coutnry to generic ?
meaning input/output, or interactive organisation, or international organisation :p
π
havent updated the online repo for it but π
wasnt originally meant to be used as its main function is a python dict , but that allows access to its keys trough .dot.notation π but ..
they could, and they might. And the .vlaanderen is just a paid TLD. If you got money, you can have what you want
you missed my point 100%
.vlaanderen was just an example for a tld that you wont remember when seeing it , .io is , so imho things would be better if they kept io, and moved it from country tld to generic tld instead.
i didn't miss your point, i told they might. But also this is ICANN we talk about, their decisions are often very stupid
This week has been an intresting week, forced to learn either javascript or html whatever it may be (its suposed to be a .js file but has HTML doc listed above) to do an implementation a sister company did, i origionaly became involved due to the BI component, and they asumed it be a 1:1 translation....
Hahaha nice. Well JS is weird, but after having used it for a good 5000 hours, you kinda get the hang of it. But yeah TypeScript is so much better
And I actually really like the way JS does certain things.
Whenever I have to use Java (which I have for university), I donβt like it, but at least we donβt have to use Python, as that would be literal torture
tried to learn Java when I was 12 and the guy making the videos was a professor at hong kong university and his mic was horrible so I moved to different languages
none of them are that bad
Welp, i got most of it done, just ran into an issue with a bit that used an api link, seems to have had a limiter or something in it, but past that i got it running with our information, so im pretty proud of myself for getting this far with it in just 2 days
as always allot of the times i think fireship captures my sentiment on the WP the best , https://www.youtube.com/watch?v=mc5P_082bvY
Find all the best dev content at https://daily.dev/fireship
Automattic CEO Matt Mullenweg recently started a crusade against WP Engine, a profit-driven company based on WordPressβs open source platform, which has evolved into a lawsuit that could blow a hole in the internet. Find out why WordPress is locked in a civil war and who is responsible...
The professor isn't the problem, Java is, a big one.
JavaScript >> Typescript
as typescript adds unnecessary complexity to the codebase.
I really disagree with this, there are so many bugs prevented using TS over JS when working in a team π
Exactly. For simple applications, I would agree that JS can be better, but from even small scale apps, TS is so much better
sure when working alone
but with big projects with multiple developers/teams typing solves problems
also typescript prevents dumb errors like are tenums this method expect likethis or likeThis, or maybe LikeThis ?
there used to be a saying about java , you can look it up if you want it goes like:
saying java is good because it works on al platforms , is like saying a*** s** is good because it works on all ...
my personal feeling about langs is that you can lean allot form the setup , like what do i need to install, configure , buy? , and type before i can do a hello world , save it and then run that hello world on a different system
and a big part of that is the what do i need to type, (important in this is is autocompleted words count as typed since even if you only had to type a+ctlr+space or so , you have to read it , and when looking at someone elses code in the future having to read 5mile long lines for pages on pages is verry tiresome. java while running fast still needs a jvm to run in , but thats the least of its problems , the number of files and lines that have to be there in order do print a hello world is insanity , if a c compiler can be cancelled out against the jvm, C needs :c #include <stdio> int main(){ printf("hello world") return 0; }to work , python needs the interpreter to be installed , or a compiler like nuitka to compile a binary to be portable, but all you have to type is```py
print('hello world')
```python2
print "hello world"
10 PRINT "Hello World"
java on the other hand requires you to type: java class HelloWorld { public static void main(String[] args) { System.out.println("Hello, World!"); } } and here is what at the verrry least is wrong with that,
- to much ,way to much
- class .. if a class is a factory for objects , if i i only need 1 why do i need to make a factory for it ? ,
- class .. we build a factory where do i tell it to make the object
- why isnt
system.outnot the default target of println ? , of yes OOP ... you can only call methods and methods only exist on objects so actually calling a funciton directly and passing in the object allongside woudl be herresyprintln(System.out,"hello world")so no defaults - if i dont require a method to take any arguments why do i still have to include the m in the definition , now if i give the program arguments when i run it it also has to reserve memory for storing those arguments... possibly and probably on the stack
the printing thing is pretty silly yeah
but at the same time there is system.error and others that all implement those print functions so its easy to explain why it's that way
can i printt more than one line with println? if so why not just print? :0
print(system.error, "error") ?
hence my remark for the default, if system. out was the most used , you could set it as a default , in the jvm if not specified , reducing the number of text required , if you did want to print to error, you could overide the default by passing it in
in pyton , also oop allot print is actually a builtin but a wrapper for : ```py
import sys
def print(*a,**k){
end=k.get('end','\n')
flush=k.get('flush',False if k.get('end') is not None else True )
sys.stdout.write(' '.join(a))
sys.stdout.write(end)
if flush:
system.stdout.flush()
return None
its one of the advantages of running in an compiler or somehting like a jvm, and even a compiler , since what printf does in c is allot more than what you type... it also prints to stdout by defautl , and you dont have to specify , in java you do
*not that on linux printing involves opening a file /dev/stdout , and writing to it , but if you had to do that each time you wanted to print smth to stdout and wirte that code each time.... its why std libs exist imho
only java does not allow for that simplification due to how the language works
and yes im one of them bad programmers that on various occasions use 1 letter variable names , rare but i do , mostly my variable names are 3 to 4 letters, holding just enough information to know what they are inside the scope. okay the next is a bit for making fun but its how java (and oddly enough i see it more and more in other places these days) but the first is code i actually wrote ,the second is using meaning full full variable names as they say , even had to change my wrapping rules to make the screnshots as i go by a softwrap at 80, hardwrap at 12, and i never let it go to that , as if you need lines that long your doing something wrong
and i forgot to expand the first p in ppid to parent
Hey, anyone participating in this year's hackCBS?
How about kotlin then π ?
tbh i never seen kotlin code , so i cant judge i have heard of it and it sounded a better language than java , but for the fact that it also uses the jvm ofc something you cant optimize across or what is an extra step in the communication between devs and hardware , if that is good or bad not sure but i tend towards badd
havent used it myself either, know a guy that uses it and really loves working with it but he hate(d) java for quite some reasons you'd specified.
i'm just a nodejs dev =]
if there was a way of running c/other native compiled code (or whatever) using a java ui , as front , id be all for it since that would mean direct control and a portable ui across platforms but
Qt can do something like that
i hate js with a passion ever since jquery highdays and vowed to never touch it again π
Oof, i do dislike jqeury, but really love working with node (ts)
almost all of my tools don't need that high performance stuff but when i do, i'll just write a native c module for it
its just that i can move quickly using node (ts) as the spec of clients will change a LOT sometimes ;p
the fact that ts exists at all should tell you enough about js π
i even avoid js on my homepage , so far could keep it at zero but still working on it , has been a while since i did any web stuff, so my hompage is more like a 2010 throwback in terms of technoloy(i run apache2 because i know it , with php and mariadb/sqlite that serve html5 and css3 :p
I do use js on all my web stuff (nextjs), really loving it so far!
REally wanted to invest some time in golang, seems quite nice to work with π
JVM is just one target for Kotlin, not the actual output
you mean JNI ?
didnt know about that, if calling a c function is as transparant as calling one from zig or python, an kotlin is what they claim it is, i could see some use for it , my favo combo for graphical interfaces is pyqt+c backend (havent tried with zig backend yet but) , just becaus pyqt lets me construct gui's procedurally , and doing it like that , uour basicly just designing it on the fly in your head as it comes , and it mostly works and the day when python is to slow for a gui , wich i cant see ever happening , i can still tarnspile it to c and compile that to a native binary
ps my homepage i mentioned hehe it will throw you back 15 years just by looking at it (under construction still as its for personal use and it runs locally on a rpi4b)
ps truth be told , if what i remember from doing java still holds up , and you want a visual basic like experience for quitckly throwing together a gui , the java thing will be a much better experience than using qtdesigner , wich in my eyes is even harder to use that just programmign the gui
not sure how it is in other languages, JNI example you can see here https://www.baeldung.com/jni - in short you create Java class with "native methods", then running a tool creates the header file for it, you then create a cpp file next to it with same method and it's implementation.
so in Java you will have:
private native void sayHello();
Header file for it will be:
JNIEXPORT void JNICALL Java_com_baeldung_jni_HelloWorldJNI_sayHello (JNIEnv *, jobject);
and then CPP for implementation is for example:
JNIEXPORT void JNICALL Java_com_baeldung_jni_HelloWorldJNI_sayHello
(JNIEnv* env, jobject thisObject) {
std::cout << "Hello from C++ !!" << std::endl;
}
and this is the replacement for JNI https://docs.oracle.com/en/java/javase/21/core/foreign-function-and-memory-api.html
The Foreign Function and Memory (FFM) API enables Java programs to interoperate with code and data outside the Java runtime. This API enables Java programs to call native libraries and process native data without the brittleness and danger of JNI. The API invokes foreign functions, code outside the JVM, and safely accesses foreign memory, memory...
which will be probably less steps, as you will just call a method from dynamic library ? https://docs.oracle.com/en/java/javase/21/core/calling-c-library-function-foreign-function-and-memory-api.html#GUID-E7255CE9-5A95-437C-B37A-276B6C9B5F4D
The following example calls strlen with the Foreign Function and Memory API:
doing c stuff from python looks allot simper tough: in essens this is all you need to get goin: ```py
import ctypes
incc= ctypes.CDLL('./inc.so')
Cint = ctypes.c_int
Cint.value=22
incc.faster(Cint)
meanwhile in MySQL
anyone know any barebones OSes that have the minimum amount of programs installed? i would like to develop a dedicated os but i don't have time to build everything from scratch.(Sorry if this is the wrong channel)
Alpine Linux ? That is what is usually used for Docker containers
thanks i will check it out, it is meant for a raspberry pi console
Mainly for use in the tty
jeah next thing : lets store json tables in sql and tank performance even futher , then blame intel and amd to not make any progress anymore π lol
json as text inside sqlite3 db's as binary blob in mariadb or so
reminds me of customers2 π sec
Recorded live on twitch, GET IN
Article
https://jimmyhmiller.github.io/ugliest-beautiful-codebase
By: Jimmy Miller | https://twitter.com/jimmyhmiller
My Stream
https://twitch.tv/ThePrimeagen
Best Way To Support Me
Become a backend engineer. Its my favorite site
https://boot.dev/?promo=PRIMEYT
This is also the best way to suppor...
My work project has few json objects in the postgres db π
storing json in an sql db is fine*
*Assuming you are not ever attempting to query anything specific in the blob, filter by anything in the blob, or index anything in the blog
also for the love of god link to the OG source https://jimmyhmiller.github.io/ugliest-beautiful-codebase
I am trying to setup a Jenkins server and I cannot for the life of me, figure out why it fails
I can always access the login page for the webserver and login, then I login and try to go to any subsite and I just get a "server isn't responding" error. The server has no error messages
If I am still logged in and directly navigate to the subsite, the entire thing falls apart after I click the next sub link
There are no console errors
to be fair, JSON in SQL can be other than blob
like postgresql has json type (exact copy) and jsonb (decomposed binary); and you can have indexes on jsonb for example
Just released v2 of my project written in react (nextjs) and typescript (bun) if anyone is interested in playing https://wordrama.io
Play Wordrama, Wordle, Puzzles, and more!
Anyoine goign to the OWASP benelux days 28th of november here?
apparently its fine for postgress as it has specialized code to deal with it ? (i have no clue but thats what i made of it so :D)
wlll since i was referring to something he said that was not in the og file , thejson in the db , id dint not see the need for that , as i covered my bases :p
this comment somewhat scares me , (i might be wrong here but) as it seems you just jumped over the actual problem with storing json in a sql database ...
it didn't. what is the general issue with storing JSON in DB ? Performance and searching. If you can index JSON for perforemance reasons and you can search it too, what is the issue ? I mean i have yet to use it, but then i also haven't used the other 50 crazy features PostgreSQL has
i would personally never use JSON in DB, but i can see special use cases where it would make sense. For example as "document storage" similar how you put documents into SOLR.
what would be wrong with storing text like
\x31\x31\x30\x31\x30\x31\x30\x31\x31\x31\x30\x30\x31\x31\x31\x31\x30\x31\x31\x31\x31\x31\x31\x30\x31\x31\x31\x30\x31\x31\x30\x31\x30\x30\x31\x31\x31\x30\x31\x31\x31\x30\x31\x31\x31\x30\x30\x31\x31\x31\x31\x31\x30\x30\x30\x31\x31\x31\x30\x31\x31\x30\x30
π
besides obfuscation ofc
your now doing a lookup to do a lookup to get a value where you could have , just gotten away with doing a lookup , ofc there is exceptions , and not only that , each time you need a single key you get the whole json so depending on the size of the json ... if your changing a value it gets even more fun , especially when more then one client is using that json ... π
worm
that is if you insert JSON as is. But that is not the only way to insert JSON in databases, there are optimized data types which allows indexing, and are stored like what would be a series of tables connected via joins. just because there are wrong ways to do things, that doesn't invalidate the usable ones.
And obviously one shouldn't use it for everything, but there are use cases where it is the best option.
Just because something can be abused it doesn't invalidate the whole technology/idea.
i explicitly excluded postgress from that list , since it apparently has code to deal with that, so in that case the json is part of the database not perse a record in it ... wich is a whole different story.. in 100k records go find a record that has a match for key is some value where the key value is stored in json , , the data baser has to check every record for a wildcard string match wichis defenetly awesome in terms of cputime and memory use , but once found it has to return the whole json any way , now you can search that json for what data you want , and everything else can be discarted , if the json is not that big data wasted is not that big ... now what if you would have done a little extra work and just stored the json not as jeson but as records in a table , so youcan do an exact mach on a key wich is o so much less memory hungry and allot faster, now it can return only the value of the asociated keys , and thats it...
one valid usecase and not for json but for storing databases inside databases as records is when you store your virtual hosts(like an appache virtual host not a full vm ofc) files in a database instead of a filesystem (as bin blob with key as filename )so you can spin of a virtual environment and load files form data base but keep them in memory on an as accessed basis and leave no clutter when no longer needed...
by the way if you like puzzles: this:
\x31\x31\x30\x31\x30\x31\x30\x31\x31\x31\x30\x30\x31\x31\x31\x31\x30\x31\x31\x31\x31\x31\x31\x30\x31\x31\x31\x30\x31\x31\x30\x31\x30\x30\x31\x31\x31\x30\x31\x31\x31\x30\x31\x31\x31\x30\x30\x31\x31\x31\x31\x31\x30\x30\x30\x31\x31\x31\x30\x31\x31\x30\x30 is an acutal text π
hint: ||9 letters long ||
another hint: ||ascii is 8bits but theire is also 7bit ascii identical to 8 bit for the lower 128 chars when since they all have a leading zero anyway) ||
solution: || jsoninsql ||
how to get there: || ```
\x31 = ascii "1" and \x30 = ascii "0"
so its a binartystring stored as ascii 0 's and 1's
group them by 7
31 31 30 31 30 31 30
1101010 = j
... :)
(its how valve stores Booleans key = "1" ) π
(thats 24 bits to store a single bit value okay for it to be human readable you need it to be 8 but damn the " " around the 1 :P)
Hello, I have been trying to send 32 bytes packets over a serial connection on the raspberry pi pico. I am trying to send these about 1kHz. That would leave the UART bus being about 1/3 full at a bitrate of 900,000bps. What would be a better way of offloading that information? Is sending that information over USB better? Can the UART go faster like 3,000,000bps? I am attempting to do some data logging. Capture the data to look at latter and the packets are binary formatted (MAVLink) not ascii.
yes, UART can be clocked higher on the RP2040. Whether your receiver can clock that high is another question. USB FS would be slightly higher bandwidth at 12Mbps
Supports a maximum baud rate of UARTCLK / 16 in UART mode (7.8 Mbaud at 125MHz)
page 416
https://datasheets.raspberrypi.com/rp2040/rp2040-datasheet.pdf
isnt serial speeds mesured in BAUD . like the default is usually 9600BAUD ?
(for a serial console)
there is a linux tty tool that lets you simulate such a connection , aswel as set up a real one by connecting a virtual console to the serial over whatever carrier (like you can do it over usb,eg you dont need a pci-e serial expansion card with a 9pin dbus connector or so)
Technically yes. However for signals with a binary symbol, baud and bits per second are equal so the distinction isn't important
okay true, but if nothing else using baud to google search for might yield more results, ps i think if the proc of the pico can handle the troughput , usb should be faster 3mbps, is 400KiB/s even usb1.1 did ~1MiB/s on a good day
What? Why? Are you sure they do that? Normal rdbms's just stores it as a single bit. Perhaps it's converted to something like that after it's fetched from the DB?
Byou might have to pauze and look for the "1" and "0" cases where its clearly a bool https://youtube.com/clip/UgkxkmAZjBAIX_TvN6JSVrG-KQunngps1YB3?si=BedkXyVhlqxS0uKp
18 seconds Β· Clipped by Hoefkens J. Β· Original video "the txt file that (almost) ruined valve's trading economy (several times)" by f4mi
But valve has β¬β¬ later in the vid. The start gorcibg redouwloads of 8mb file each connection for every player, so 8 mb is not allot but x a couple million a day ...
sometimes fixing these things after many years of using them is just too likely to break something
plus its already a json file innit so whats the difference in size between having a "1" and rewording your keys to use camelCase instead of snake_case
like yeah technically wasteful but the waste isn't a problem as much as explaining to every new person on the project (employees at Valve assign themselves to whatever project they want, for the most part) that your booleans are strings
So, using USB serial, the speed problem is resolved, not I have a speed problem around i2c reads. My solution is going to be switching to SPI at around 10x the speed of i2c.
Been experimenting with DMA today, attempting to make memory dumps of ram sticks to extract the encryption keys for the disks.
could someone please explain to me how this benchmarks web-gpu interface is able to access so much of it?
I thought a 100% gaming load or a stable diffusion generation was all it could offer and somehow, someway, this is even heavier
Disclaimer, do not run this link on a system without a DGPU
Cuz WebGL can just use the GPU to render graphics, directly using the GPU.
There isn't really a reason why it shouldn't be able to use 100%
that app puts an insane load on the gpu
more than ive ever experienced from an uncapped framerate or stable diffusion gen etc which brings util to 100%
thats more what I mean in this question
No idea why. When I was developing a 3d renderer for my previous job in Three.js it also was super heavy on the GPU without optimalisations.
Having the amount of polygons greatly reduced the load to something reasonable.
The model you're seeing looks like it has a lot of polygons, and it's spinning as well. Probably the reason why it's quite heavy.
so the simplicity allows a greater quantity, which translates to a heavier load is what you're saying
No, I'm saying I don't know why it's heavier, and I've also experienced it when I was developing a 3d thingy for the browser in the past when the poly count was high.
heres the source
from what I can tell, its running the math - and constantly rerunning it to follow the parameters made
I wonder what about games is different than this.
and maybe if the alogorithm that displays utilization is flawed somehow
games offload many aspects to a cpu
got a take on this at all? @drowsy elbow
I don't know @simple temple . Been a long time I worked on that project, and I used a library to do the heavy lifting.
what small project in rust should i work on that would take longer than 3 days?
any project that you estimate you can do in about an hour Β΅
i actually know a really good one, and where it may seem more like a "rewrite it in rust" i can assure you its not , i did an attempt at fixing it but im not a c++ dev , C i can do , at first i just cleaned up the terminal interface of it , did a pull request on that but still hasnt been merged , but when i tried expanding it so it would list the device names , also noticed the way the program works internally makes not much sense anymore and all its good for is detecting wether or not the device is active as the troughput is calculated using libpcap and it assumes usb1.1 framesizes so good luck getting accureate speeds with usb3.2 or so... anyway: have a look at https://github.com/aguinet/usbtop, if done right you might end up in almost every major linux distro π
if you want or need help in doing the tui interface using ansi , i cant do rust but i can do ansi pretty well by now (= only update the specific values on screen or doing stuff in line as oposed to full screen redraws)(and still )
Ok
I don't think my knowledge and skill is good enough for that so instead I'm just gonna make a tui tool for something
Idk what yet
average github commit
POV you are my coworkers submitting PR
π luckily its just a private project... most of it is just adding c++ dependencies to the repo
I probably did something wrong when adding the boost library... don't know if it is supposed to be like 180 megabytes in total
update: most of the boost library is useless for me so I will swap it out for a more lightweight edition of it
i usually have huge commits like that... when i am doing the initial project commit, mostly when transferring from a different git/svn repo π
or doing a reproduction case for a bug. Still a lot π
yeh
ChatGPT: makes the same mistake 3 times in a row.
me: Get's mad, don't provide any more details
Also ChatGPT: Appreciated, here's the exact code you're looking for
(First it came up with this crap)
meanwhile business is like "if you are not integratng AI, you will go bankrupt"
(at same time OpenAI has 3.5B revenue, while generating 5B loss /s )
yesssss dude
dont worry AGI is right around the corner and the record losses will soon turn green TRUST ME BRO
the most depressing thing i seen last week was when a young coworker had copilot open in their editor while screensharing to me. I can only hope he uses it as a "shortcut" for stackoverflow content
wishful thinking
My coworkers openly say they use cursor and spam the living fuck out of it all day long to make 10 thousand line long files of cobbled together garbage
meanwhile i jump from topic to topic in my work
one day I will just say I write organic code and charge 3x more for my work
locally source organic etsy code for sale
i mean in last weeks i did:
- Swift iOS code
- Java Android code
- JSF web & Java content
- Angular/Typescript app
- pure Javascript stuff
- set up some Gitlab CI builds for Android & iOS apps to push them to Firebase App Distribution
And probably some other stuff i can't even remember π
is this what they call 10x developer ? /s
no thats when you add rust into the mix
What Javascript Framework I need to learn if I want to get money in my pocket in the future
Must be fun when A.I. gaslights a whole company in being profitable, while in reality it's making record losses.
Java Spring
Nah idk. Java spring isnt javascript xD
If you learn a js framework it's probably outdated when you're ready to apply for jobs
"Making a one file contained inference code is easy! Won't be long too"
The 'one file inference' code:
Iβm ngl I donβt understand a single line
Anyone has experience compiling PyTorch from source for ARM CPU and CUDA GPU?
FACT: every day, 12,000 javascript frameworks are released.
FACT: every day, 12,000 javascript framework bootcamps are released.
FACT: every day, 12,000 job applications are ignored.
i still laugh at that instance where AI meant "away indian" (for real)
Java story - today i wasted whole work day on figuring out a coworkers bug... what was the bug ?
he used the JSP c:forEach tag in a JSF page instead of ui:repeat, so the page broke only if certain elements were duplicated via c:forEach.
Yeah, i had 'fun' today.
I love PHP, lotta people keep creating technologies to replace it, but it still goes strong. As long as you sanitize your inputs, there is very little to worry about.
Modern PHP is not an issue, all the legacy code is. I mean what feature does PHP actually lack these days for an interpreted language (and even that is solvable if you set up a bytecode cache)
Laravel is great
Raw php can be great for creating basic web scripts
For example the one used in the recent network boot video
I've been doing this thing lately where I've been using shared memory to communicate between PHP and a C/C++ or even Java program. It allows PHP to do it's this without calling exec
Great for when you have a Active backend that's interfaced though ajax requests
Didn't know that was a thing possible in php and java, how does that work?
but in the end, the point of PHP is - PHP3 was horrible. PHP4 introduced some things, PHP5 was big jump and by PHP8 while the language could run many PHP3 scripts, there are also tons of new features
i would comare it to Java 1.4 vs Java 21 π
Java has libs to connect to shared memory, I usually structure a process queue and have Java simply flag an item once completed and have PHP handle shifting the Queue to prevent concurrency issues
I haven't used php in ages, but since I got addicted to C# I don't want anything else running on my server
Like aspx or something?
