[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/g/ - Technology

[Advertise on 4chan]


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


File: E0eaWaNXsAUk2Hc.jpg (255 KB, 2048x1152)
255 KB
255 KB JPG
old thread
>>81438508

What are you working on, /g/?
>>
First for |
>>
I want to learn Fortran. How can I do that?
>>
In pedo circles, 'pro-c' stands for 'pro-contact' (with kids, sexually), and 'anti-c' stands for 'anti-contact'. To be pro- vs. anti- C is apparently the chief controversy of this community.

Still want to read K&R?
>>
>>81448043
https://fortran-lang.org/
big button right in the center of the page.
>>
opengl be like 'yeah bro it's 200loc to draw a triangle have fun'
>>
>>>81447150
>>81447150

help im retarded
>>
>>81448075
your question is far too specific for this thread
>>
>>81448068
wait til you see vulkan
>>
>>81448068
the complexity is frontloaded
it's the same loc to draw a million triangles as it is to draw one
>>
>>81448093
okay thnx
>>
>>81448068
isn't opengl deprecated anyway?
>>
>>81448068
glBegin(GL_TRIANGLES);
glVertex3f( 0.0f, 1.0f, 0.0f);
glVertex3f(-1.0f,-1.0f, 0.0f);
glVertex3f( 1.0f,-1.0f, 0.0f);
glEnd();

wow so hard
>>
>>81448124
no, opengl now runs on vulkan in angle.
>>
>>81448124
no
>>
File: 1476391701519.jpg (204 KB, 660x751)
204 KB
204 KB JPG
>>81448101
https://github.com/SaschaWillems/Vulkan/blob/master/examples/triangle/triangle.cpp
>>
>>81448124
Always has been
>>
>>81448160
I want to see this in Agda
>>
>>81448160
My kinda functional vulkan renderer is 3.5k loc, the doom 3 vulkan renderer is like 6k iirc.
>>
>>81447039 #
>The people you're going to be learning from in a masters program are going to be people who have dedicated their energy to learning one specific facet of CS really well like operating systems, computer vision, machine learning, etc. If you're not interested in getting really good at one specific thing, there's no reason to go into a masters program. That said, the people you're going to learn from are going to know a lot more than you, regardless of how much you think you already know.
That’s what I’m asking anons. What am I going to learn For a masters at a prestigious university if I can just download their syllabus for courses and self learn it on my own. The masters is remote and if I self study without it I can avoid debt.

I just want some thought on that. Is there anything they teach that isn’t already on the Internet? Like is there any purpose for a masters if I already eat books and material like it’s lunch? I just don’t want debt and worrying over grades
>>
>>81448220
>Is there anything they teach that isn’t already on the Internet?
no
>>
>>81448220
>I can avoid debt
oi vei, we can't allow that
>>
>>81448220
The point of going to school is to get a qualification to get you a job
If you can demostrate your ability as a programmer without a qualification you can also get a job
Either way, knowledge means nothing
>>
File: SOytinereS.png (87 KB, 656x510)
87 KB
87 KB PNG
Working on SerenityOS. :^)
>>
>>81448252
>onions
>>
File: file.png (290 KB, 474x316)
290 KB
290 KB PNG
>>81448294
Yes.
>>
>>81448044
And which side are you on?
>>
Haskell is literally bad.
>>
>>81448447
>error: literally is not in scope. bad is not in scope
>>
>>81448160
wgpu native
https://dawn.googlesource.com/dawn/+/refs/heads/main/examples/CHelloTriangle.cpp
https://github.com/gfx-rs/wgpu-native/blob/master/examples/triangle/main.c

with rust wrapper
https://github.com/gfx-rs/wgpu-rs/blob/master/examples/hello-triangle/main.rs
>>
>>81448075

Did you look at the example?

https://openlayers.org/en/latest/examples/layer-extent.html

Besides that I cant help
>>
>>81448251
>The point of going to school is to get a qualification to get you a job
>If you can demostrate your ability as a programmer without a qualification you can also get a job
>Either way, knowledge means nothing

Imagine the disappointment believing you need spend 4 years not learning anything just to get a job that the only qualification is a piece of paper. Pathetic.

Knowledge is actually everything. A job means nothing.
>>
Is it cringe to star your own repo?
>>
>>81448220
You're paying for access to the people you're learning from
Masters isn't undergrad
>>
>>81449095
only if it brings the total stars to 1
>>
>>> from operator import getitem
>>> getitem(pair, 0)
10

what's the point?
>>
File: 1611800323377.jpg (782 KB, 1598x1970)
782 KB
782 KB JPG
How do I make a custom icon for my gui, using the netbeans editor.
https://netbeans.apache.org/kb/docs/java/gui-image-display.html
I've been using this guide but I want to make a custom icon for a gui thats already made.
>>
>>81448975
Knowledge is only half the battle
>>
>>81448220
I am currently halfway through a masters in machine learning. There is no part of our program that you could not have learned by using online resources. However, the issue is discipline. Are you disciplined enough to keep going when you've been stuck on one task for weeks? Can you persist when just as you understood the last thing, a new, even worse thing comes and hits you?
Can you do all of this without the support of classmates that can assist you when you're stuck, or at least share your pain?
If yes, then go for it. Otherwise, go to university.
>>
File: firefox_zM3zjzL466.png (44 KB, 353x914)
44 KB
44 KB PNG
how much of this do you know?
>>
vulkan taking 1000s of lines of code to write a triangle is kind of a meme, the actual act of writing a triangle is pretty straightforward and doesn't actually take too many lines, there's just a lot of set up that has to go on when you do it by hand
>>
>>81449566
Reading the Practices section gave me cancer
>>
>>81449566
At least half of that is just pure cancer
>>
>>81449600
Based and checked. Software engineering is a meme created in the 90s to justify the existence of Java and the resulting OO exegesis that justified its horseshit.
>>
>>81449600
this
>>
>>81449614
one hell of a meme if all the relevant parts of the industry adhere to structure and practices laid out by it.
>>
>>81449654
Was this post instantiated by the an AbstractPostFactoryImpl?
>>
>>81449669
yes, which derives from the IhaveAJobAndYouDont super class.
>>
>>81449678
>t. bishop of the cult of cargo
These threats are the only reason new companies follow these practices
>>
>>81448160
lol my software rasterizer isn't even that complicated. I'm never writing raw vulkan if anything i'll just use an abstraction layer.
>>
>>81449566
Almost nothing. Software development as a practice seems boring to me. That's why I only care about the theoretical part of CS.
>>
 
* (defun ok () (defun ok () 2) 1)
WARNING: redefining COMMON-LISP-USER::OK in DEFUN
OK
* (ok)
WARNING: redefining COMMON-LISP-USER::OK in DEFUN
1
* (ok)
2
*


uh, I had to try it
>>
How is the job search in tech?

In places like San Francisco, cops execute political dissidents to ensure kids of rich daddies have no competition. We note that this is capitalism and not communism because kids of rich daddies have actual skills in manufacturing LSD.

You could never do such a thing because all you know how to do is program, which is what makes the kids of rich daddies special. They have the power of a god, because this is a democratic society and spoiled kids get the best jobs.

Are you willing to betray your country over a couple of bucks?
>>
never go full schizo
>>
>>81450061
in J this is just
   ok =: 3 : ('ok =: 2&['; '1')
ok''
1
ok''
2
>>
>>81450061
>>81450101
in Lisp this is just

(defmacro dd (name x y)
`(defun ,name () (defun ,name () ,y) ,x))

(dd ok 1 2)
(ok) ;; 1
(ok) ;; 2
>>
if i have some flag variable in memory set by one thread without any synchronization effort, how long can i expect for it to take to propagate through cache and shit and be seen by another thread on a potentially different physical core, in say C/C++ for instance?
i want to put up a flag raised asynchronously by one core for whatever reason but i dont care how long it takes, the thread will go to sleep by a condition variable or semaphore or smth, and i dont want the thread reading the flag to go through the performance hit of trying some synchronization to read one flag like that, so it can take a couple of hyperperiods of whatever cyclic schedule its running
>>
>>81450343
Completely unpredictable. Use proper synchronization.
>>
>>81450404
boo
>>
Writing code for Arduino and MATLAB serial communication. The Arduino is set up to take readings of an LED and display them, but now I'm trying to get these values to go to MATLAB. I was trying to put them in an array in Arduino as they are measured, but that doesn't seem to work. I'm also having issues getting the serial monitor output I'm seeing from the Arduino to be useable in MATLAB. Any help would be appreciated.
>>
File: 1620095653196.png (64 KB, 353x914)
64 KB
64 KB PNG
>>81449566
>>
once I figure out how to read data from a continuously changing file/buffer/interface, I'll really be in business
>>
Is there an X automation utility that will let me move windows across different monitors?

My display resolution in xdpyinfo gives the full resolution of all my monitors together but xdotool and xdo don't want to move a window past the limits of the monitor it's currently on.
>>
What are some good books to read if I am new to programming?
>>
>>81449654
>everyone else is injecting jewish source code into their veins I'll do it too
all management fads are roughly the same bullshit (some common sense + religious doctrine attached) that cycles every few years.

80% of workers are useless in this meme economy if not a detriment to the code base. the most useless are most attracted to fad roles (black belt in 6 sigma). most everything in that image is engineered to give the 80% a sense of purpose.
the 20% do not need to be babysat in standups, they're the reason the company was able to quintuple hiring in the first place
>>
>>81450760
streams
>>
>>81451023
https://wiki.installgentoo.com/wiki/Programming_resources
>>
>>81451023
The AWK Programming Language
>>
>>81450955
I think I remember crunchbang (ages ago) doing this with wmctrl. If your WM supports the stuff it needs to do the job, that is.
>>
>>81451058
thanks
>>
import numpy as np
rng = np.random.default_rng(100)

for _ in range(10):
xs = rng.normal(0, 1, size=10000000)
print(f"Mean: {xs.mean()}, std: {xs.std()}")

outputs:
Mean: 0.0004317489262153253, std: 1.0002369215804099
Mean: 8.086035251684091e-05, std: 1.000178641253615
Mean: -0.00014205432016323283, std: 1.000546603543541
Mean: -0.0001558874056962429, std: 0.9998324474887033
Mean: -0.00019211435369352394, std: 0.9999673731511028
Mean: -0.0008230974186607831, std: 0.9996894317928174
Mean: -4.247869469658985e-05, std: 0.999957514894894
Mean: -0.0001003962315409098, std: 0.9997719339321975
Mean: 0.00012897240150309505, std: 0.9995039757281178
Mean: 8.884798338503783e-06, std: 0.9998447477838814


That's 10 tries to get a distribution with a mean very close to 0. Isn't it surprisingly far off in some cases? I am generating 10 000 000 numbers after all.
>>
>>81451364
Never mind, just have to lower the deviation to get what I want
>>
File: 1615333918388.png (39 KB, 654x445)
39 KB
39 KB PNG
>reddit in my referring sites
Does GPL let me forbid redditors from using my software?
>>
>>81451485
No
>>
>>81451494
free software isn't all its cracked up to be
>>
Does anyone know how to "hard code" a circuit in LogicWorks 5? I can't package binary switches, but I need this one red line to always be true. Pls halp
>>
>>81451485
You can use the poison pill approach. Dedicate your software to something reddit hates but wouldn't get banned by shithub in the README, like Trump and Christianity. Redditors won't be able to use it without being attacked by their "friends". Same technique works with datamining, you can niggermark images to keep info from being reposted there.
>>
File: 1610356967142.png (10 KB, 517x91)
10 KB
10 KB PNG
>>81451559
You reminded me of some of my favorites
https://sqlite.org/codeofethics.html
>>
>>81451485
Why are you so insecure?
>>
https://www.learn-js.org/en/Arrow_Functions
Why do you have to use const to declare arrow functions?
>>
>>81451829
You don't have to. You can use let too. However, usually you don't want to change it afterwards, so it makes sense to use const.
>>
where i can find cmake.h for windows?
>>
>>81451912
Is it some project that writes a cmake.h as part of the build? In the CMakeLists.txt is there a configure_file() line with cmake.h?
>>
anybody in /dpt/ does competitive programming? or does problem solving exercises for fun?
I want to find an alternative place to project euler, any recommendation?
also, python gets way too slow in difficult solutions (and my shit inefficient code skills doesnt help..). is c++ the best language for this type problems?
>>
>>81452035
You could probably use a language with a GC and still get good enough performance. I doubt it's worth dealing with manual memory management for most of the problems.
>>
File: solve this.jpg (11 KB, 285x285)
11 KB
11 KB JPG
>>81452035
I used C++ when I was still doing competitive programming in school (the other options were Java, Pascal and C).
Advent of Code is language-agnostic like Euler.
You can also try doing exercise from IOI's test bank.
>>
>>81452035
i don't do competitive programming, but i do have a lot of fun optimizing my programs and coming up with weird efficient ways to represent algorithms. i'm pretty inspired by the demoscene, i'd love to participate in it someday
>>
Any ideias on good rust projects?
>>
>>81452739
Build an app!
>>
>>81452739
dilator
>>
>>81448975
If you want to get a job, which is what he was asking, then you need qualifications or ability
>>
>>81452739
rewrite some random student projects in rust and do a pull request just to see their reaction
>>
working on my tile renderer, my script is hilariously messy but sometimes you just have to get your hands dirty
>>
>>81448043
Why would you do that?
>>
>>81447883
all computer operations are boolean because they operate on bits.
>>
>>81451485
Just dedicate the work to God and say it is intended to fight the synagogue of Satan.
>>
>>81451485
Also start off your introduction with a Terry Davis quote.
>>
>>81451585
>is-odd
the sheer amount of self shilling on this github readme
>>
foo(1, 2, 3); //OK
...
foo(1, 2, 3); //ErrorInvalidValue
//This indicates that one or more of the parameters passed to the API call is not within an acceptable range of values.

ok...
>>
>>81453364
...
changeFoosAcceptableRangeOfValues(1, 2)
...
>>
What is the most efficient way to determine whether x or y is double of the other number. We don't know which one is the bigger one.

At the moment I am using

if (list[x] == list[y] * 2 or list[y] == list[x] * 2):
return True
>>
>>81453873
this is almost certainly a silly thing to be worrying about.
>>
>>81453873
who is we
>>
>>81454010
>I am caring about being efficient and always seeking if there are more efficient ways of doing things instead of using the first brute force path I first thought of when I was 18 for the rest of my life
>This guy: This is a bad personality to have as a software developer. Dont question things. Dont learn things.
>>
>>81454085
this is almost certainly a silly thing to be worrying about.
>>
File: 1478739648720.jpg (47 KB, 645x968)
47 KB
47 KB JPG
can you guys spot why the fuck I'm adding
input[0] 2 times in this vector sum?
>input = {1 , 2, 3, 2, 1}
>output --> {1 + 1 + 2 + 3 + 2 + 1}
should be
>output --> {1 + 2 + 3 + 2 + 1}
__global__
void vectorSum(const int* const input, int* const output) {
__shared__ int temp[BLOCK_SIZE];
const int globalIdx = threadIdx.x + blockIdx.x * blockDim.x;
if (globalIdx >= INPUT_SIZE) {
temp[threadIdx.x] = 0;
return;
}
temp[threadIdx.x] = input[globalIdx];
for (int i = 1; threadIdx.x + i < BLOCK_SIZE; i*=2) {
__syncthreads();
if (threadIdx.x % (i * 2) == 0) {
temp[threadIdx.x] += temp[threadIdx.x + i];
}
}
__syncthreads();
if (threadIdx.x == 0) {
atomicAdd(output, temp[0]);
}
}
>>
>>81454112
No, it's very important, in fact more important than the question asked
>>
File: objection.png (20 KB, 448x313)
20 KB
20 KB PNG
>>81454010
possible floating point == following multiplication by 2
>>
>>81454137
Multiply which one with 2? x or y?
>>
>>81452293
>>81452119
thanks anons

>>81452358
I do it for fun too, dont care for results. its a good way to find new methods of doing things and solving stuff feels great
>>
>>81454120
nvm, I'm a double-retard.
I was calculating wrongly the single threaded sum. the cuda one is ok.
>>
>>81454091
Caring about being efficient isn't the same as being efficient.
>>
>>81454091
The odds are, your "optimized" solution works slower than normal solution after compiler optimization, and is human-unreadable on top of it. Even if it is faster, it's likely to be in a part of program that isn't a bottleneck, so you wasted time and effort optimizing the code that doesn't require optimization.
Maniac microoptimization is usually a bad habit of new coders.
>>
File: 2021-05-04_13-56-25.jpg (11 KB, 706x125)
11 KB
11 KB JPG
Why would I even need to cast uint as int?
>>
I want this thing
I'd make an interface for exporting gameboy camera images in png, the way god intended

I actually have a pico already, but I can't solder for shit and don't want to slice up my only link cable
>>
>>81454174
kys
>>
>>81454352
C# isn't C, you are not going to get buggy automatic coercion for this.
>>
>>81454322
You must be shitting me, do you have a CS degree or are one of those bootcamp idiots?
Algorithm complexity is different than Pajeet OOP complexity. You should always aim to use <n^2 algorithms, which by default means using the proper data structures(ie. if you need to search in a collection repeatedly, use a hashset over an array or linked list).
>>
>>81454290
>>81454322
retards, i know not all optimization is good. that is why I am sitting down and trying to have a discussion. if this is the most efficient way of doing it, just fucking say it.

sometimes we think we know the most efficient way of doing something but turns out there is another algorithm with some math involved that is much much more efficient and I didn't even know I didn't know it. that is why I am asking.

it just felt like there should be a more efficient way of doing that operation (including the assembly and machine code that we get after compiler optimization and compiling)
>>
>>81454357
Just run some wires to an arduino
>>
>>81454528
In this particular case, just write down the possibilities. It's either:
1. y is 2x => x/2x = 1/2
2. x is 2y = 2y/y = 2
In truth, it doesn't matter for this particular algorithm since it's a constant complexity, but you can have one less extra math operation if you have to. Keep in mind floating point numbers will be trickier to handle.
>>
>>81454357
better yet, a program to make the GB printer print out 4chan threads
>>
>>81454528
It depends on a massive amount of environment we don't have. This "Algorithm" is trivial. The question is unanswerable.
>>
>>81454562
Division is much more expensive than adding a number to itself (which is what *2 becomes).
>>
>>81454522
First, algorithm complexity has nothing to do with premature optimization.
Second, algorithm complexity often doesn't correlate with measured performance.
People like you should stay away from optimization because you have absolutely no idea what are you doing.
>>
a >> 1 == b

There you go. Gay and useless.
>>
>>81453873
This is the best you can do. You can always try things like this on Godbolt and see what the compiler does to optimise them.
>>
>>81454642
oh and a can't be odd, add a check for that too.
>>
>>81454606
>First, algorithm complexity has nothing to do with premature optimization.
Then stop being a vague imbecile. By promoting this you're just making other idiots parrot it and shit programming up even more.
>Second, algorithm complexity often doesn't correlate with measured performance.
It obviously depends on the situation and the constants, especially if you're working in some shitty webshop where you operate on 10 element arrays, but it literally costs you nothing to use a decent algorithm.
>>
>>81454642
You can write it as a/2, compiler will turn it into bitshift for you anyway.
>>
>>81454642
>>81454666
This won't even work, every value of a has some other value (either a+1 or a-1) that give the same value after shifting.
>>
>>81454664
So you're admitting that you have no idea what you're talking about, you fucking mouthbreathing son of pajeet. Now backpedal all the way out of the thread. Retard.
>>
>>81454606
>Algorithm complexity often doesn't correlate with measured performance.
>People like you should stay away from optimization because you have absolutely no idea what are you doing.
Not Anon but holy shit, you're a special kind of retard if you believe that
>>
>>81454010
you're trying to optimize an operation that will most likely be reduced to a couple instructions regardless, it's pointless (as long as those things are int/int64, not floats or unsigned)
what you SHOULD be worried about is the access pattern on those lists to make sure you're using your cache lines properly.
>>
>>81453873
>>81454646
Whilst you can't do this in Python, as this check will fail most of the time its probably better to avoid short circuiting on the or.
>>
>>81454841
It's true, many algorithms that have lower time complexities only end up out performing on larger inputs. It's why efficient quick sort falls back to insertion sort for small partitions, or why naive matrix multiplication is best for sizes below 100.
>>
>>81454642
what if it isnt an int
>>
>>81454935
Quicksort isn't better than insertion sort in terms of complexity, though.
>>
>micro optimizing multiplication by 2
>in Python
>>
>>81455200
It has better average performance
>>
>>81455214
Given the following discussion, I believe it's a medium-effort bait.
>>
>>81448252
whatcha working on
>>
How far are we from scraping data from pixels? I know it's been done but I mean is it worth it or have any perceived advantages?
>>
>>81455539
>How far are we from scraping data from pixels? I know it's been done but I mean is it worth it or have any perceived advantages?
What are you talking about?
>>
>>81455539
>How far are we from scraping data from pixels? I know it's been done but I mean is it worth it or have any perceived advantages?
huh?
>>
>>81455539
I don't know but we've got a long way to go in AI text generation.
>>
>>81455200
>average o(n logn) is not better than average o(n^2)
ok
>>
>>81448160
And thats not even all of the code. Its another 3300 lines (see the vulkanexamplebase.h include)
>>
>>81448068
Use opengl 2, then its less than 15 lines to setup everything + render a rectangle
>>
>>81451485
>>81451520
Use the license+NIGGER license https://plusnigger.autism.exposed/
>>
>>81448160
>le vulkan is too verbose
The vast majority of that is initialization-time configuration structs. After you write this boilerplate and put it into functions, it's not really that much code. The difference is, state is no longer implied, like it is with openGL, so the driver isn't going to make bad guesses now.
>>
For what reason should I learn Julia?
>>
gonna learn gRPC and more async Rust and be based
>>
Sitting in Russian city Vladivostok, time is 11PM learning how to implement linked lists in C. Cyкa блять.
I have no degree, but I making my way towards earning some money through web development, also learning math with physics, I don't know what the fuck am I doing.
Ask questions if you will.
>>
>>81455979
>Sitting in Russian city Vladivostok
How long does it take to travel to anime land from there?
>>
>>81455979
пoчeмy нe нa двaчe
>>
>>81455997
It's really fast, 2 hours in general, and really cheap. Russian currency sucks balls. It's like white Nigeria.
>>
>>81456017
вeздe ycпeвaю
>>
>>81456040
>It's like white Nigeria.
what do you mean by that
>>
File: FREE CS.jpg (32 KB, 747x525)
32 KB
32 KB JPG
>>81455979
Haхyя тeбe linked lists? It's one of data structures that got somewhat obsoleted by CPU development, as it's not cache-friendly.
And its implementation is trivial as long as you understand pointers.
>>
>>81456067
It means that to buy something in another country - Japan for example, is to spend your half-year earnings unless you work there.
>>
>Khronos released provisional Vulkan video encode/decode extensions
Holy based.
>>
0 and 1
on and off
high and low
don't need more than that
>>
>>81456087
It's 17 chapter in Prata's book, linked list, binary trees, queue etc. For some reason I need to understand that.
If you experienced what would you recommend to do next? C++?
>>
What are some languages/editors where you can single-step and modify the code while it's running?
>>
>>81456193
GDB?
>>
>>81456193
C# with VS is the only one I've worked with. I'm sure there's others.
>>
>>81456139
c++ has lists and red-black trees implemented for you as a part of standard library, all you need is to familiarize yourself with their interface. It's nice to know how they're implemented, but I think you will get it on common sense level once you're more familiar with language, so there's no benefit in pushing it now.
What's next depends on what you want to do. C and C++ generally have different applications, even if you "can" use C++ in embedded hardware. If you want to learn C++ go with Lippmann's book, C++ primer 5th edition. It's translated to russian (dunno about translation quality, I read it in english) and covers everything starting with basics.
>>
>>81456100
Do you still have an opportunity to get into a uni for free? If yes, try to take it when you have some money and get into Moscow's top unis or at least SpB's. My friends who didn't just fuck around in school all got in and so did I, so far I'm glad I didn't stay in my shithole. Environment actually decides a lot. Being around smart people who know shit and are willing to help each other has been extremely beneficial, plus most people in year 2+ already work and can get you past the initial screening straight to the interviews.
>>
>>81456193
That feature is called edit and continue.
>>
>>81455200
>>81455665
Quicksort is O(n.log n) average, O(n^2) worst-case. Getting worst-case behaviour requires consistently choosing bad pivots, which is statistically unlikely. Specifically, if the size of the smallest partition remains constant as n increases (the probability of which is inversely proportional to n), you get O(n^2) behaviour. If the size of the smallest partition is proportional to n, you get O(n.log n) behaviour. The worst-case behaviour matters if you need hard real-time guarantees or need to defend against complexity attacks.
>>
>>81456234
No worries I read Prata's book in English. Thank you.
I want to do game programming, that's what math and physics for. But I have no degree, so I got a hopes for a web dev as income, maybe I'm wrong, but web-dev is easier than looking for job in C field with no degree.
>>81456316
>Do you still have an opportunity to get into a uni for free?
Yep, I just got really unlucky. I have an opportunity to get into uni for free.
>If yes, try to take it when you have some money and get into Moscow's top unis or at least SpB's
That was my plan! I wanted to travel to SpB and stay there, will try to get into math-focused uni. I study required fields right now to get into it. Got really big passion towards learning now that I understand things better. But troubles with income etc. This summer I plan to go for it.
>>
>>81456411
>Getting worst-case behaviour requires consistently choosing bad pivots, which is statistically unlikely.
not if the input is sorted already
>>
File: 1619788249905.jpg (45 KB, 480x960)
45 KB
45 KB JPG
>>81447984
>What are you working on, /g/?
Hey /dpt/, what do you think about this?

>Recently graduated EE, looking for jobs related to signal processing
>Most if not all ask for matlab and/or C/C++
>Already a complete freak at matlab
>Have actual work experience with C++
>Learned to program in C but haven't touched it in years, never did anything "advanced" with it
>Im now reading the C book (k&r) every day after work
>Plan is to eventually start working on writing a library (or most likely a set of libraries) that implements a bunch of numerical methods I'm interested in and that I've previously studied at uni (gauss-seidel, newton rhapson, various interpolation methods, runge kuta, etc)
>If that goes well I'll also write a signal processing library for the computation of some important things such as convolution, correlation, autocorrelation, fft, Fourier spectrae, etc.

What do you think? Should I add anything else?
>>
File: Screenshot_1.png (6 KB, 715x102)
6 KB
6 KB PNG
how do i turn a list of dictionaries to a dictionary in python?
>>
>>81448044
>oddly specific insider knowledge
>'apparently'
>'this community'
just fuckling neck yourself already, pedo scum
>>
>>81456505
It's not a dictionary, it's a set.
set( dictionary.values() ) doesn't work?
>>
racHElbythebay
>>
Looks almost like a proper functional language
type TrimRight<T extends string> = T extends `${infer R} ` ? TrimRight<R> : T;

TrimRight<"lol "> // "lol"
>>
>>81456810
disgusting
>>
>>81456590
i can't do

set = dict.values()

because there input is not a dictionary. it is a list. look at sample data. it has a square bracket at the beginning and end.
>>
>>81448044
wtf who the fucks knows this. get help nigger
>>
>>81456810
is that java? wtf did you do to it? :D
>>
>>81456954
YOU know it, right now. So it must be some other form of cnility that prevents you from sympathizing with how I came to know it.
>>
if i have some for loop that calls malloc and free for an auxiliary array within the loop itself, how can i properly move that to the gpu in openmp?
>>
> create sentiment analysis AI model
> 32% accuracy
> rates almost every sentence a 3
i no longer fear skynet or tesla self driving becoming good anytime soon
>>
>>81455838
Scientific Computing. For engineering, Julia simply isn't mature yet.
>>
File: loads shotgun.jpg (553 KB, 2540x1440)
553 KB
553 KB JPG
>>81448133
wow i posted 10% of the code
>its not that long So easy
>>
>>81456988
Looks like TypeScript
>>
>>81456590
even if the input is a dictionary.

for some reason set(dict.values) gives me

dict_values(['S009', 'S005', 'S005', 'S007'])

where did the S001 etc go?
>>
>>81448160
pretty simple logic, it's basically a long list of statemets
>>
>>81456505
set((value for _dict in sampleData for value in _dict.values()))
>>
>attend course about UX
>case study: reddit
Hard pass.
>>
>>81457464
Not programming
>>
>>81455626
>>81455634
Sorr was taking a shit after smoking some weed when I wrote that. Essentially rather than using webrequest and regex to scrape data automatically, there's a method to scrape text from pictures instead
>>
File: Screenshot_1.png (50 KB, 600x264)
50 KB
50 KB PNG
"More than one entry per key not allowed. Which means no duplicate key is allowed. When duplicate keys encountered during assignment, the last assignment wins."

Yet pic-related following works.

I guess it is only during the first initialization that duplicates inside the same brackets get removed. Just like sets.

>>81457324
I ended up using the code below because i am a java fag.

set = set({})

for dict in list:
for value in dict.values():
set.add(value)
>>
>>81457662
DUDE WEED LMAO
>>
>>81457707
Yeah man coding high can be a blast especially when it helps you come up with solutions after banging your head on a wall for hours
>>
>>81457464
>getting triggered simply by seeing the word "Reddit"
autism
>>
>>81457666
>set = set({})
Don't do this. Now you can't create another set.
>>
File: 1481237839630.jpg (362 KB, 1106x1012)
362 KB
362 KB JPG
>>81457783
let me guess... you need more?
>>
>>81457725
I don't know how some people can get anything done on weed. I'm slow as fuck when programming high
>>
python-logo-master-v3-TM
>>
>>81457765
>t. plebbitor
>>
>>81457783

I mean you still can
>>
>>81457666
set_ = set(v for d in l for v in d.values())

Beautiful and, most importantly, Pythonic.
>>
from typing import List
class Solution:
def isIncreasing(self, nums: List[int]) -> bool:
"""Check if list is in increasing order."""
for i in range(len(nums) - 2):
for j in range(len(nums[i::]) - 1):
if nums[j] <= nums[j + 1]:
pass
else:
return False
return True
def checkPossibility(self, nums: List[int]) -> bool:
"""Check if list *could* be in increasing order, if the position of one value was changed."""
pass

I'm trying to a leetcode problem. I'm trying to think of a way to implement
checkPossibility
, I'm kinda lost at what to do. Is there something smarter than at every position of nums making a new list of lists with all the possible combinations of moving one value and passing that through
isIncreasing
?
>>
>>81457980

Why would you use set constructor here?
>>
File: gui.png (103 KB, 1208x642)
103 KB
103 KB PNG
How do I set up an image icon in java netbeans for my gui, I'm following this guide

https://netbeans.apache.org/kb/docs/java/gui-image-display.html

But it doesnt tell me how do I use the image on an already made gui.
>>
>>81449128
You can create a partial function and pass it to another function.
>>
>>81452739
How can people have no project ideas despite using their computers for 8+ hours a day?
>>
>>81458929
hours of no productivity leaves the brain in a mushed state, no critical thinking, no drive to do anything.
>>
what does this command actually do in c++?

vector <int> dimension = getList()
vector <int> idxV = dimension

I assume the getList function returns a list which is assigned to dimension list and then dimension list is copied to idxV
>>
>>81458929
they have ideas but immediately discount them as unachievable. If they identified a knowledge gap as responsible, they could at least come up with project ideas in that direction, but 'failures' like "I don't have enough time" or "I don't have enough people" or "I don't have enough money" or "I'm not google" give them nothing to go on.
ex.
"I want to make an MMO"
good outcome: "but I don't know shit about client/server networking. I should read beej's guide and write a series of increasingly complex servers."
bad outcome: no, that takes a huge team of people and I'm just one guy. Man I have no ideas!
>>
>>81459089
copy constructor
>>
>>81459089
yeah
>>
>>81459089
Vector<int> dimentions gets copy initialized from vector returned from getList. Then idxV gets copy initialized from dimensions.
>>81459150
initialization calls regular constructor, even if you use =.
>>
>>81459261
huh, so when is copy constructor called?
>>
>>81458929
I pretty much stay the whole day working and studying for uni exams, My only free time is at weekends where I give up on studying for uni to study things I really like. Until now I used this spare time to study rust and work on a flutter projects I was into, thanks for nothing
>>
>>81459304
When you copy already initialized vectors.
If you add as third line idxV = dimension or vice versa, it will call copy constructor. You can see it for yourself with minimal example
class A
{
public:
A(const std::vector<int>& v) : m_data(v) {std::cout << "constructor";}
private:
std::vector<int> m_data;
};

std::vector<int> func()
{
return {1, 2, 3};
}

int main()
{
A inst = func();
}
>>
File: 1614027986618.webm (2.26 MB, 560x1000)
2.26 MB
2.26 MB WEBM
frens, can somebody help me understand time complexity of this?

https://www.cnblogs.com/apanda009/p/7964875.html

dimensions can be any number : 4D, 5D etc.. I assume the generic complexity is b ^ d because of recursion where b is the branches per depth and d is the max depth of the tree. so branches per fanout will be average of dimensions and max depth will be max of the dimension. i am so confused. please respond
>>
>python in terminal finds numpy
>pycharm doesn’t
Why do redditors and fags on this board like this stupid language. I can’t wait to be done with this course so I can return to comfy R and Rstudio
>inb4 filtered
Yes, I don’t want to learn a bunch of nerd shit about computers to just write some code and get on with my life. Why is a language without native arrays and matrices so popular fuck me
>>
>>81459523
O(n^d), where d is array dimension.
>>
>>81459922
is n the len of the dimension array? in the case of [8, 1, 3] it is 3^3? shoudlnt the time complexity be dependant on max recur depth ( in this case 8 )
>>
I've done literally nothing at my programming job the last 2 years. Neither has anyone else. They don't give us anything to do, it doesn't even matter if we have our laptops on our not. How do I explain this to a future employer? If I make impressive enough portfolio projects (working on it right now) then will that ease the nerves of my next boss?
>>
Are return addresses in Assembly stored on the stack in order to restore them? (RISC-V specifically)
>>
>>81460050
would it not look good for an employee to request work to do in such a case?
sounds like everyone is just leeching of leniency.
>>
>>81460188
RISC-V doesn't have stack instructions, so returning from a subroutine is just jumping to the address stored in a register. By convention, register x1 is used by the caller to store the return address for the callee. If the callee subroutine wants to call another subroutine, it will need to save x1 somewhere so that it knows where to return to.
>>
File: 1610715001792.png (189 KB, 1673x1342)
189 KB
189 KB PNG
>A tree is well-formed only if it has a root label and all branches are also trees.
Is this because if a tree had a branch that didn't itself have branches, that branch of the tree would in fact be a leaf?
>>
File: swing_fucking_sucks.png (48 KB, 1000x756)
48 KB
48 KB PNG
Any java sirs know if I can align the components in Swing's Flow layout manager? I would use the grid layout but the displayed rows and columns must adapt to the window's size
>>
>>81460442
>it will need to save x1 somewhere so that it knows where to return to
This is what I mean, I can just store it on the stack, right?
>>
>>81461222
You can store it on a stack, but there isn't necessarily a stack. If the subroutine is non-reentrant you could store it in static memory or even use self-modifying code to encode it into a jump to immediate instruction.
>>
>>81461222
STORE MY ANUS IN THE STACK
>>
File: 1594522115563.jpg (31 KB, 346x510)
31 KB
31 KB JPG
>>81461257
Thank you anon
>>
>>81461161
A tree is still a tree even if it has 0 branches. Leaf is just a more specific name for that case.
>>
>>81461161
the definition is important because having a complete/balanced tree allows you to assert a bunch of properties regarding search times and whatnot.
the fact that one branch would be a leaf is not the important thing.
>>
>>81459523
please sirs help me analyze the complexity here. is there a recur tree analyzer online? i googled but those ones dont work for complicated functions
>>
>>81461220
>desu desu desu
based desu
>>
would you work at google or amazon if offered? I dont know if i am self sabotaging but i dont want to participate in an evil company
>>
>>81461725
i'd be afraid google has a secret copy of all my search history and use it against me
>>
>>81461725
google interview is hard to clear. amazon is too toxic. id work for google but too low iq to prep for their interview
>>
If I have a class with a lot of data and I create a vector with that class, if I vector.clear() it, will all the objects inside get destroyed, or will they "stay" in memory but become inaccessible?(sorry, I'm still learning about programming)
>>
>>81461855
Yes, vector will destroy all those objects using their destructors.
>>
I'm so tired of pressing Shift when writing code, it's drives me insane.
>>
>>81461916
I only press tab
>>
>>81461425
My understand reading
is_tree()
is that it iteratres through every branch, and every branches branch... Just to check that it has the type list and is not empty. Sorry what do you mean by balanced and complete?
>>
Holy fuck guys i just got the results back and i do indeed have HIV. I don't know what to do. This is so surreal.
>>
>>81461960
_(<>{*&
>>
>>81461751
>secret
>>
>>81461975
Complete trees and balanced trees are particular kinds of trees with additional restrictions. That anon assumed the question was about those, but it doesn't seem to be.
>>
>>81462022
Disregard, wrong thread. Sorry.
>>
>>81462025
#
>>
>>81462064
+
>>
>>81456476
That assumes you're choosing the first or last entry of the array as the pivot. Which is dumb, because it results in O(n^2) behaviour for sorted or reverse-sorted data. If you're using a single, deterministically-chosen element as the pivot, you normally use the middle element.

You can defend against complexity attacks by using a strong PRNG to select the pivot element. That still doesn't help for a hard real-time requirement, though.
>>
>>81462025
so change your keyboard layout to suit your own needs instead of using some typewriter setup.
>>
>>81462278
You don't have enough characters on the keyboard to type everything comfortably. Now if someone comes up with 8 row programmable keyboard that would b enice. I could learn that.
>>
>>81462022
We told you about Rust bro...
>>
>>81454528
What you're doing isn't optimization, it's bikeshedding. At best you might save a couple cycles handcoding it in assembly vs compiler magic.
Worry about orders of magnitude, and then, after you're positive you've got a good overall design, and you've done proper profiling, you can start sweating a couple math operations where you're positive it matters.
>>
>>81462022
They are treating these I believe, your doctor should be able to tell.
>>
File: a.png (75 KB, 1328x736)
75 KB
75 KB PNG
How can I use CTRL-SHIFT-Y without getting carpal tunnel as a man with small hands?
>>
>>81461725
No, trading your time for money isn't worth it if you don't like what you're doing
>>
I've got some documents and I had to recreate them in digital form but holy shit, doing it in latex was a mistake. I'm wasting literally hours trying to figure out how to position tables, how to set specific size etc. Guess I'll try doing it in html or libre office draw
>>
>>81462637
I'd suggest remapping Caps Lock to Ctrl or using your right index finger to press Y, maybe do both.
>>
>>81462637
rebind capslock.
>>
>>81458069
Yes, you can group the values into sublists such that the first element of the sublist is bigger than the next values of the list, then reduce to length, and if the list is all 1s it means it is perfectly sorted, if it is one 2 and the rest is 1s it means that there is only one out of order, and anything else is more than one out of order. I'll show it to you in Haskell:

Prelude Data.List> groupBy (\x y-> x > y) [1,6,7,2,3]
[[1],[6],[7,2,3]]
Prelude Data.List> map length $ groupBy (\x y-> x > y) [1,6,7,2,3]
[1,1,3]

Prelude Data.List> foldl (\(a,b,c) x -> if x == 1 then (a+1,b,c) else if x == 2 then (a,b+1,c) else (a,b,c+1)) (0,0,0) $ map length $ groupBy (\x y-> x > y) [1,6,7,2,3]
(2,0,1)


I know it is a bit ugly as a one-liner , but that 3-tuple basically means "if there is (x,1,0) there is one out of order, (x,0,0) all in order and anything else is more than one out of order)
>>
>>81462637
easily
>>
>>81462637
>>81462929
>small hands
wait nvm
>>
>>81459323
It's not true tho, you would've had a thousand different ideas on how to improve your productivity and study more effeciently then.
>>
>>81462902
Oops, this doesn't work all the time, let me rework it.
>>
>>81463051
>oops this doesn't work
typical hasklel
>>
>>81462637
You can use two hands.
>>
>put off coding for years
>made a checkers game in javascript in my free time

feels so good.
>>
>>81463091
I can't believe you chuds still believe in the rshift/rctrl conspiracy theory.
>>
>>81453873
> not(abs(x/y-1.25)-0.75)
>>
can I achieve this, except in C++?
def f(a):
def g():
print("yo"+a)
return g
>>
>>81463179
now who is we
>>
File: 1615054282436.png (378 KB, 895x672)
378 KB
378 KB PNG
>>81463051
>>81462902
desu I'm a pytoddler and cant read Haskell. i just learned how to use a lambda today.
>>
>>81463182
auto f(std::string a)
{
auto g = [=]{ return "yo" + a; };
return g;
}/code]
>>
>>81453873
x+y in [1.5*x, 1,5*y]
>>
need to learn kotlin to do an inventory app
>>
>>81461916
the_absolute_state
>>
>>81463077
:9
>>81463198
you can implement a bubble sort, but the first time you swap two values, raise a flag. If the flag is up , you won't do any other passes, and if in the comparation you find you don't want to swap, then terminate the sort.

Then check if the list is ordered. if it is not, then there are more than one elements out of order.
>>
>>81463450
comparison*
>>
>>81463077
There you go, cunt

>>81463198
Alternatively (actually a better way):

Find the first element out of order, remove it from the list, then check if the list is in order.

import Data.List

remo [] = []
remo (x:[]) = [x]
remo (x1:x2:xs) = if x1 < x2 then x1 : remo (x2:xs) else remo (x2:xs)

fun x = and $ zipWith (<) y (tail y)
where y = remo x
>>
>>81463591
should be

remo (x1:x2:xs) = if x1 < x2 then x1 : remo (x2:xs) else (x2:xs)
>>
>>81463618
>oops this doesn't work
>again
>>
>>81461220
>if I can align the components in Swing's Flow layout manager?
You can't. You need a different layout manager that thinks in terms of giving each contained component the same amount of space.
>>
in c++ what would be a way to avoid any overhead with the decorator pattern?

let's say i wanted A, B, and C to be supported extensions of a base class, i could use the decorator pattern and use some overhead to "pretend" like it's the class ABC
but is there a way to use templates somehow to make it so that it's as fast as if i created dedicated classes for each combination? basically a way of writing the equivalent of explicitly writing the A, B, C, AB, AC, BC, ABC classes by hand? is there a term for this?
>>
>>81463880
god this is stupid
just write what you want
>>
>>81463880
>is there a term for this?
a waste of time

what performance impact does using the Decorator pattern have that you think you need to optimize this and not something else?
>>
File: kuyashii.png (85 KB, 1137x793)
85 KB
85 KB PNG
>>81463849
Then it can't be done with the default ones? They're not that many and I've read their descriptions at least three times by now, I don't believe that I'm missing something.
I figured out a hack to align them: I've set every label to have a monospace font and a char limit.
My current problem is figuring out how to do a scrollbar, if I just add the current panel to a scroll pane, I also have to supply bounds for the complete back-end view, which I can't possibly know before-hand since each directory has different amount of files.
>>
>>81463744
nah, in the end it achieves the same thing, this just makes it faster.
>>
have you thanked your compiler today?
>>
>>81463450
Okay I wrote a bubble sort for the first time. I tried implementing raising a flag if I encounter a swap (I always will). But I'll need to swap likekly multiple times, even if only one number is out of place while doing bubblesort. I'm a bit confused what you meant by compare, I don't think I understand your second sentence.

class Solution:
def bubbleSort(self, nums: List[int]) -> List[int]:
for i in range(len(nums) - 1):
for j in range(len(nums) - i - 1):
current = nums[j]
next_ = nums[j + 1]
if next_ < current:
nums[j] = next_
nums[j + 1] = current
#print(nums)
return nums
>>
Why can't static analyzers prove a program is free of UB at run-time?
>>
>>81464634
that's what the U is for
>>
>>81464037
>what performance impact does using the Decorator pattern have that you think you need to optimize this and not something else?
it's being hooked up to an interpreter so i'm worried that every single instruction could hit the stack of decorator functions
>>
>>81464606
It was a convoluted idea, it was trying to put the first wrong in the correct position to then see if it was in order, but it is much more practical to just remove the first element that is out of order, then check if the list is in order.
>>
File: 1607018736312.jpg (197 KB, 1000x1000)
197 KB
197 KB JPG
>>81464775
G E N I U S
E
N
I
U
S
>>
>>81452810
Holy kek
>>
dumb question but is a jump table inherently more insecure than not having one? like if i have code like this, would the fact that i have an array of function pointers that i build at runtime make my code more insecure than not?

typedef void (*Handler)(int);   

void func3 (int a) { ... }
void func2 (int a) { ... }
void func1 (int a) { ... }
void func0 (int a) { ... }

Handler jump_table[3] = {func3, func2, nullptr};

if (someCondition) {
jump_table[2] = func1;
} else {
jump_table[2] = func0;
}
>>
>>81464955
code would be protected by W^X and the jump table might not be. It's a discussion that begins with failure though: you're assuming that you've already lost memory security and that a hostile actor can poke around in your memory space.
>>
>>81464862
in J this is just
   (([: }. [: , ' ' ,. ]) , ,.@}.) 'GENIUS'
G E N I U S
E
N
I
U
S
>>
>>81465005
>(([: }. [: , ' ' ,. ]) , ,.@}.)

Miserere mei.
>>
>>81465028
>have mercy
ok. how's this?
   rest    =: }.
onside =: ,.
andthen =: @
flatten =: ,
append =: ,
spaced =: {{ rest flatten ' ' onside y }}
genius =: {{ (spaced y) append (rest onside y) }}
genius 'genius'
g e n i u s
e
n
i
u
s
>>
>>81464992
it's something i'm thinking about for a microoptimization but i haven't tried it so i don't even know if i'll get a performance boost

>you're assuming that you've already lost memory security and that a hostile actor can poke around in your memory space.
could a bug that should be fixed if caught but isn't extremely destructive into an extremely dangerous one?
>>
>>81465005
from typing import List
class Solution:
def isIncreasing(self, nums: List[int]) -> bool:
"""Check if list is in increasing order."""
for i in range(len(nums) - 1):
for j in range(len(nums[i::]) - 1):
if nums[j] <= nums[j + 1]:
pass
else:
return False
return True
def checkPossibility(self, nums: List[int]) -> bool:
"""Check if list *could* be in increasing order, if the position of one value was changed."""
popStore = None
for i in range(len(nums) - 1):
for j in range(len(nums) - i - 1):
current = nums[j]
next_ = nums[j + 1]
if next_ < current:
popStore = j
break
if popStore is not None:
nums.pop(popStore)
return self.isIncreasing(nums)
#print(Solution().isIncreasing([1,2,3])) # True
#print(Solution().isIncreasing([2,1,3])) # False
#print(Solution().checkPossibility([4,2,3])) # True
#print(Solution().checkPossibility([4,2,1])) # False
#print(Solution().checkPossibility([3,4,2,3])) # False
#print(Solution().checkPossibility([-1,4,2,3])) # True
print(Solution().checkPossibility([5,7,1,8])) # True

fuckkk this strategy doesn't actually work. in this last one i remove the number 7 but it's still out of order. but i could of moved the 1 forward............ AAAAAAAAAAAAAAAAAAAAAAAAAAA
>>
>>81465005
here's my take
(, , }.) ' ' ,.~ 'GENIUS'
>>
>>81465005
>>81465222
I really need to learn J
>>
>>81448044
why do you kow so much about nonce vernacular anon?
>>
>>81465005
has Jposting gone too far?
>>
shit I started a new project today and there's so much to do I completely froze
>>
>>81465498
i spent 3 days designing my project instead of writing code
i started setting up the cmake, gave up, considered writing it in rust, realized i don't know rust, then went back to c++ and finished the cmake
>>
>>81465094
Now it doesn't look like something out of the necronomicon.
>>
File: 1608011655993.png (6 KB, 235x214)
6 KB
6 KB PNG
def genius(word: str) -> None:
print(' '.join(word))
for i in range(len(word) - 1):
print(word[i + 1], '\n')
genius('GENIUS')
>>
File: hmm.png (42 KB, 860x1005)
42 KB
42 KB PNG
It would be cool if J/K/APL had become embed in other languages for array processing kinda like what regex is for string pattern matching.
>>
>>81458354
First of all, "Since we specify only a filename, Java assumes that the image is in the same directory as the class. ImageIcon supports formats such as GIF, JPEG and PNG. "

Second of all, using netbeans palette tool, you can set the icon

http://prntscr.com/12i4d35

http://prntscr.com/12i4clq

Here is the code generated

jLabel2.setIcon(new javax.swing.ImageIcon(getClass().getResource("/bug1.png"))); // NOI18N

http://prntscr.com/12i4flb

http://prntscr.com/12i4fz6

Sauce: Java How to Program 10th edition
>>
new thread

>>81465890
>>81465890
>>81465890
>>
>>81465747
april is an embedding of APL in Common Lisp
https://github.com/phantomics/april
>>
>>81465894

Somebody make a real thread please
>>
>>81465919
seethe



Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.