/tech/ - Technology★

>>1062254 (OP)

write better op

>>1062291

What's wrong with it?

I want to learn C because Dennis Ritchie invented programming.

I can’t even imagine that I would get a job if there would be no UNIX, no Linux, no Windows, no graphical application, no sound driver, no GUI based tools, no programming languages and still we would be struggling with binary and machine level language. There would not be computerized electronic gadgets no iPhone, iPad etc.

http://www.aliencoders.org/content/person-who-founded-c-and-unix-dennis-ritchie-no-more/

>>1062294

Everything. If you want to start a discussion you have to start it yourself, not just make a thread and tell people what to discuss. This is not a homework assignment for us.

I've rarely used C++ since college, but I'm starting the EOS.IO developer courses soon and will have to dust it off. Not rly looking forward to it, tbqh

>>1062285

struct Node<T> {
    prev: *mut Node<T>,
    next: *mut Node<T>,
    value: T
}

Wow. So hard.

>>1062447

You use english US layout in software. You do know how to touch type, right anon?

I actually use a keyboard with german layout at work, even though I never use that layout. It was the best keyboard there, but nobody used it, probably because nobody here even speaks german.

>>1062455

>>not knowing that the hard part is writing the allocation functions.

Nigger, linked lists aren't hard. You learn how to do those in your first semester at university. After that you never use them again because they suck.

>The guy's introducing functions (mem::replace) that you don't even hear of in the "nice" tutorials that don't go over linked lists.

What the fuck are you even trying to say?

>What the fuck is your language if you can't even give me a clear set of primitives with which I can accomplish all I need?

?????????????????

Your image is an accurate depiction of yourself.

Is there any inherent performance difference between calbacks and promises in node.js?

const fs = require('fs-extra')

// Async with promises:
fs.copy('/tmp/myfile', '/tmp/mynewfile')
  .then(() => console.log('success!'))
  .catch(err => console.error(err))

// Async with callbacks:
fs.copy('/tmp/myfile', '/tmp/mynewfile', err => {
  if (err) return console.error(err)
  console.log('success!')
})

>>1062478

>node.js

No difference. Both perform like shit.

>>1062467

Read the post nigger. I've written (both simply and doubly)linked lists about 100 times in C and C++. What I couldn't manage to do is write one in Rust without triggering the borrow checker.

>>1062488

You literally posted a link to a tutorial on how to write linked lists in Rust.

>>1062458

t. luddite

>>1062254 (OP)

Why are you having issues learning a programming language? Learning languages is the easiest part of software development.

>>1062512

Rust wants to steal both the memory safety crowd and C++ programmers, but they think pandering to C++ fags lets them get away with a messy language and awful compile times. This disgusts memorysafety fags and C++ fags aren't interested in another fat language with glacial compilation, so Rust is stuck with a small but very loud group with much lower standards: hipster webdevs.

>>1062254 (OP)

>end up in Data Science

>it's a thankless meme job where you get shit on by everyone in the company who are desperate to poke holes in your analysis

>all I want to do is program all day and not worry about company politics

Is such a thing possible bois

>>1062512

I genuinely don't know

Go seems to be everywhere all of a sudden but I much prefer Rust

I think it still has the reputation of "don't use this, it's not mature enough" even though it's probably fine

>>1062532

The majority of users are diversity hires and third world labor, that's who it was designed for, real programmers don't need the language to hold your hand to the extent that Rust does.

>>1062549

Maturity has nothing to do with why you shouldn't use Rust.

>>1062254 (OP)

Every fucking time i tried to learn android i would stop at Fragments. Good thing they removed it in the Android 9 update.

>>1062447

Try different combinations of Shift, Fn and Alt Gr. That's probably the solution, but I don't know for sure.

>>1062447

Shift + >

t. äöå

>>1062510

Besides mastering a language and learning some stuff about networks and databases, what else is there to software development? Problems become solvable when you know what you're doing.

>>1062706

Learn how to sleep at your desk. Learn how to spend 6 months writings a minimal CRUD app in PHP. Collect a fat check and produce little to no value. Outsource 95% of the job to consultants for 5-10x reasonable development costs. Use JQuery UI in 2019 for your SPA.

That's been my experience with software "engineers"

>>1062567

Learn Perl6. Trust me.

>>1062404

heh

>starting to realize that objects are just structs with functions attached and the ability to make retard children with

fug

>>1062478

It's most likely more dependant I the specific callback/promise than is measurable objectively. I imagine in the post you made it'd be negligible, no?

>>1062740

>he didn't read SICP

oh anon I am disappoint

>>1062732

>Trust me

I'll do it!

>>1062512

> Crystal

Now do Nim, you fag.

>>1062802

Rustfags BTFO

>>1062447

Those ISO keyboards/keyboard layouts are an abomination. Just get an ANSI (US) physical layout keyboard, and use a "US-International" logical layout if you need regional characters.

>>1062641

Shit.

Don't know why I said shift instead of alt.

>>1062478

How do you put your code in a fancy window like that?

>>1063164

Press Alt+F4 to open a secret 8chan text editor.

>>1062254 (OP)

BASIC was the first language I learned. I learned that from a tutorial series on YouTube. It's an easy language but there are ton of different version of BASIC all with slightly different syntaxes. So looking up code examples I was often finding it hard to get things to work and not understanding why it didn't work when I followed tutorial to a t.

Visual BASIC 2010. Since I had a handle on QB64 I figured I was jump in to VB. It was easy to pick up and as long as I was looking for VB2010 tutorials or code examples I was doing fine.

Gets interesting. I switched to Linux and back then there wasn't much support for BASIC. So I learned C++. C++ is great for creating fast fast binaries but getting libraries to work always seemed to be a pain in the ass and I wanted to learn network programming.

Started learning Java. Did several series of tutorials on Java. Progammed in Java for a while. Got further into networking and got into security. Started learning to hack. Started working with penetration testing tools. Most of them were written in C. So I started learning C. Fucking hated it. Learned enough to look over a program and see how it worked. Other than that fuck it.

Picked up a Ruby book and learned Ruby. Never used it again after that.

Had a friend who was working on cryptocurrency project. Customized a wallet for him based on the world coin wallet. C++/Qt already knew C++. Crash course in Qt/QML made wallet look fucking cash. Have forgotten everything I learned about Qt/QML since then. Never used it ever again.

Had a friend who suggested I learn C#. Downloaded C# books and started trying to learn C#. Could never find a tutorial where the code actually worked. Didn't want to fuck with shit I didn't understand Fuck C#.

Wanted to learn MySQL. Decided the manual was 5000 pages of bullshit and the syntax changes for nearly every major version you have to look things up but you always find an answer that is different syntax than the version you are using. Fucking gave and used flat files.

Learned some PHP and PHP/MySQL applications development because I wanted to build my own search engine. PHP as a language sucks donkeyballs and I've already mentioned PHP sucks. If it install from CentOS and Cpanel or there's an install script that does everything unattended I'll use it. Otherwise it's static HTML or just enough PHP to include headers, footers, navbars, etc so I don't have to repeat code and I can just template pages with some includes.

Javascript. Wrote some shit in Javascript. It's an easy language to learn. It's C like in many regards but more like C++ and Java rather actual C. So the syntax is easy to pick up if you've learned a C like language in the past.

Kept coding in C++ for a long time. Realized I could just use system calls to do a lot of shit. Realized I don't really need C++ when I have a lot of Linux tools that will do exactly what I'm trying to do with C++. Learn BASH scripting and get good it. Coding in BASH a lot.

Friends start sending my python scripts. Look at this. I'm like dude, I'll get around to learning python later. Once you have time invested in learning something and getting good at it you just want to work in that language because it's just more efficient. But I eventually got a python book and learned python. Started using python more and more. Use it a lot now.

Wanted to learn real hacking and maybe earn cash on bug bounties rather than working for the man. I picked up some books on X86_64 and started learning assembly. Did some crash courses in bug hunting, exploit development, and writing shellcode. Low level programming is alright and I'm sure I could have got good at it if I would have spent the time. It takes time to learn. You're not going to learn assembly overnight. It takes time. Must leverage time effectively.

Since most of the programs I write about 100 lines of code and I don't care about optimizations I try to use scripting languages as much as I can. Python or shell scripting is really where it's at for me.

>>1063175

Regex really comes in handy but it's basically a language of it's own. But grep and python's re are pretty fucking useful if you want to sort a larger data set for just a few statements like a phone number, email address, or IP. I would suggest learning it or at least have some idea what it can do and have a cheat sheet. It will come in handy at some point.

Once you know one language it's easy to learn another and if you know programming in general it's easy to write a program in a language you don't even know just by looking each thing up on a search engine as you go like how do I declare a variable in X, array in X, print statement in X, system call in X, and that makes it easier to understand how other people's programs work.

I had thought about learning go or rust at some point just because people are always saying learn X but really who has time for all of that. I don't code for a living and I wouldn't want to work for someone else just because I tend to get really pissed when I can't figure things out and make something behave the way I want so I would probably have a cussing fit at the office and fired my first day.

>>1063379

>captcha get's installed

>place immediately turns into /g/

lmfao

>>1063405

Manual memory management isn't hard, chances are it's even harder to use whatever garbage collector you're using and it comes with a performance penalty as a reward too.

>>1063406

Objects are registered with GC at creation and just freeing them at the appropriate time causes segmentation faults. I guess I will have to make all new allocation and free methods for the objects I am using to avoid this headache

>>1062418

That syntax is worse than vomit.

>>1063508

That's just like you own opinion fam

Hi, I've recently started learning Guile scheme.

Things I know already are basic data types, conditionals, procedures and simple I/O from and to console, creating and reading a file.

Is there something cool I can do on this level to test myself? A certain kind of game, program? My mind is so empty.

>actually did tcp right this whole time

>just clear the string before i sent it to the serber

fug my shit up

>>1063550

Learn how to integrate Guile into a C program.

>>1063553

>Learn how to integrate Guile into a C program.

There is a tiny problem, I can't C yet. Tried a bit C++, but guess it's not the same. I'm looking for a scheme-only task.

>>1063508

>muh syntax

Anti Rust shills always resort to this because they don't actually know enough about Rust to post legit criticism.

pathetic

>>1063821

Skip everything except Generics and then look at popular Java libraries.

>>1063828

>skip everything except generics

Even the security and network stuff?

>>1063828

There's no such thing as an anti-Rust shill. Nobody would give a fuck about that abomination if Steve Klabnik didn't brigade

Every HN thread where Rust is ever mentioned

Check Zeming Yu's paper on Rust on arxiv. Rust's safety is a lie. You're only safe from memory bugs in trivial use cases that could have been discovered with a decent static analysis tool. Or you could use a GC language because 95% of use cases don't actually need that small performance boost of manual memory management (and borrow checker and lifetime annotations literally everywhere is not actually less work than manual memory management).

Ask yourself, why are there NO large scale Rust projects on GitHub? Redox is a failure. Nobody uses Servo. Firecracker VM is riding on Amazon's name. Rocket is full of retarded emojis and is less performant than Node and Go. The stars of the Rust world are rewrites of GNU coreutils. Fucking wow. Grep was really holding me back until now.

>>1063863

https://actix.rs/

https://chromium.googlesource.com/chromiumos/platform/crosvm/

https://github.com/intel/cloud-hypervisor

iNtEl iS a sMaLl CoMpAnY tHeY dOnt cOuNt

>>1063896

Nobody is actually shipping actix. Have you tried to write any non-trivial application? No? Here's a preview: Imagine your helloworld server had 80+ dependencies and takes more than 10 minutes to compile on retail hardware.

>Oh but Lemmy uses actix!

The federated reddit that has yet to implement federation does use actix. So we have one vaporware application where some huehue copied an AngularJS shitshow and used bots to upvote himself.

Cloud Hypervisor is an experiment with its first commit just this month. Give it a couple years. It will be abandoned. You might notice Nemu is written in C.

Remember Iron, Nickel, and Gotham? Remember how they were going to be the hotness in web frameworks with impeccable safety and concurrency? Weird how those projects all seemed to start fast and then fizzle within 18-36 months. Why is that? Could it possibly be that Rust is a fucking trainwreck of a language where literal homosexuals just wrote every feature they thought was "cool" on a white board and then proceeded to haphazardly "stabilize" these features into their language. All the mistakes of design by committee are personified in the Rust language. It is a cautionary tale of letting hipsters be in charge of anything at all.

>>1063863

>There's no such thing as an anti-Rust shill.

LMAO

>Check Zeming Yu's paper on Rust on arxiv.

Link pls. Better yet: uploaf PDF here

>Ask yourself, why are there NO large scale Rust projects on GitHub?

Except for those you mentioned right after that sentence?

LMAO. That absolute state of anti Rust shills.

>>1063899

Did Steve Klabnik rape your mother? Calm down, nigger.

>>1063900

Dropbox and Cuckflare are also using Rust in production.

Kys, LARPer.

Hello. Newfag here. I also dont know shit about programming or technology and i have been interested in programming languages for audio/music production

I have been questioning if i should learn Pure/Purr Data or SuperCollider

btw i use Windows lol

>>1063554

Not learning C (or at least the C subset of C++) and skipping directly to a high level-language can lead to bad programing habits. C beats into you how the machine works, so even when you learn to use high-level abstractions and languages you won't abuse them to write bloated inefficient code. So be careful .

> I'm looking for a scheme-only task.

Here's a fun one:

Write a symbolic derivatior for dx:

Eg: you give it a string "x^3+y*x" it should give you something along the lines of "3*x^2 + y" .

If you know the language well it shouldn't take more than a day.

Here's an outline of how to do it:

First you transform the string into a flat symbol-list, (x ^ 3 + y * x) then into a scheme-like expression-list-tree (+ (^ x 3) (* y x)) (trough parsing), then apply the derivation rules, then simplify maybe, then transform it back into a string

https://en.wikipedia.org/wiki/Differentiation_rules#Elementary_rules_of_differentiation

https://arxiv.org/abs/1902.01906

I work in Java right now. Tired of NullPointerExceptions and the pain of checked exceptions, I suggested we move a bunch of our code to Kotlin, which was approved by the project lead. Learned a bunch of Kotlin, worked through the Koans, and built some sample stuff, and then actually moved a bunch of our code to Kotlin. Things were cleaner and more succinct, I was able to eliminate tons of the null checks. We were using lambdas everywhere, which with checked exceptions looks a lot like this:

try {
    try {
        optional.ifPresent((Type item) -> {
            try {
                // This throws a checked exception
                item.doSomething();
            }
            catch (CheckedException e) {
                throw new RuntimeException(e);
            }
        }
    }
    catch (RuntimeException e) {
        if (e.getCause() instanceof CheckedException) {
            throw e.getCause()
        }
    }
}
catch (CheckedException e) {
}

Thanks to no exception checking, Kotlin makes it look something like this:

try {
    optional.ifPresent { item ->
        item.doSomething()
    }
}
catch (e: CheckedException) {
    throw e.getCause()
}

Our code was way nicer, way safer, and easier to maintain.

The Project Manager (over the Project Lead's head) denied the merge because nobody in the company knows Kotlin except me. I insisted that our Java developers could easily learn Kotlin in a day or two, but he mumbled something about maintainability and went back to his office.

Fucking shit working on a team sucks sometimes.

>>1063961

Doublepost. Just to add, Java's checked exceptions are fucking garbage. On their own, they are an okay feature that force more safety, but Oracle fucked up the design by making them completely not fucking work with lambdas, and by not properly extending types like Runnable to throw Exception (or at least fucking giving a CheckedRunnable type or something for backwards compatibility), making you have to wrap all your fucking checked exceptions in tons of goddamn cases and unwrap them on the outside. All types being nullable by default is also shit.

Working with Java for the past year. It's better than Python or Ruby, but I'll take any fucking language with non-nullable types if I can get it. Even Rust. I actually quite like Rust, and a lot of the nicer bits of Kotlin resemble Rust quite a lot.

Java is the only goddamn language I've seen that makes Lambdas more painful to use than just creating a goddamn callable class.

>>1063958

>Not learning C (or at least the C subset of C++) and skipping directly to a high level-language can lead to bad programing habits. C beats into you how the machine works, so even when you learn to use high-level abstractions and languages you won't abuse them to write bloated inefficient code. So be careful .

Don't worry I've learned C++ enough and I'm not all green about computers. I'm learning scheme, because Guix System uses it and because of my autism - I was tired of C-like syntax.

>Write a symbolic derivatior for dx

Sounds cool, will check it out, thanks. Maybe I'll post the code.

>>1063967

>Sounds cool, will check it out, thanks. Maybe I'll post the code.

Parsing infix notation is really the hardest part. If you need a tip, here is an algorithm that can do it: shunting-yard algorithm

>>1063965

>>1063961

Been writing Java for over 10 years and whole-heartedly agree. I was so excited when Java 8 came out, only to be disappointed when I ran into what you're describing (as well as the insanely convoluted stack trace). I simply don't use checked exceptions for my custom exceptions these days. It's a shame your company won't move to Kotlin; it's such a better language. Almost all Kotlin jobs are Android development (bleh).

Since you're stuck with Java, you could create some utility interfaces to abstract away the re-throw, as described here: https://dzone.com/articles/how-to-handle-checked-exception-in-lambda-expressi

>>1064051

>muh exceptions

Just don't write code that fails. What's so hard about that?

>>1064052

This, if you write code that has errors or bugs then you're literally no better than niggercattle.

>>1064053

>no-coding this hard

You are unworthy to quote St. Terry

>>1064053

based

>>1064078

unbased

>>1063968

The only language I like and I'm comfortable with is Lua. The problem is, it doesn't have the kind of community and support other languages have. I sometimes actually have to read documentation and figure out my own solutions. Of course there's probably no employer looking for a Lua programmer either.

>>1064161

>Lua

Just curious, how did it come to be that you only use Lua? I see it used as a supplementary language paired with something else, so I find it odd.

>>1064215

I wanted to make games, so I started with pygame since everywhere you look people recommend python to inexperienced programmers, but I didn't like it. I didn't care for the space-sensitive syntax, and there seemed to be this tug-o-war going on between python2 and python3. I ended up using Love2D after looking for an alternative and I just found Lua easy to work with.

>>1063405

Read M Timm Jones's books on Linux development. Shit will make more sense.

>>1063828

Is there a fun programming book to learn Rust? Like hacking with Rust or game programming with Rust or just sick and tired or boring dry programming books. I've read hundreds of thousands of programming books. You know what they need? Fucking dank Pepe memes and shit that makes you laugh. And autism jokes.

/tech/ we should start a b& and do some open source collaborations. It could be fun.

>>1064227

Sorry meant to say hundreds of thousands of pages.

>>1064227

>/tech/ we should start a b& and do some open source collaborations. It could be fun.

There are very few actual programmers on this board.

>>1064227

Rust is not a productive use of your time.

>>1064227

There are a few git-* repos that we can be found in. Nntpchan is a good start. Pleroma if you're OK with Lains.

>>1064161

> I sometimes actually have to read documentation and figure out my own solutions.

Oh no, the humanity. Anon, you're not going to learn anything by copy-pasting Stack Overflow, reading the documentation is the normal thing to do.

>>1064295

Pretty ironic since Ada was created by a Turkish Jew; John Ichabod.

>>1064326

A graph seems like overkill.

Here's one way to do it that calculates the result dynamically (because I was too lazy to store the groups in a list and "clean" the list when 2 groups are "merged". That's probably how a normal coder would do it.)

#include <iostream>
#include <fstream>
using namespace std;
int main()
{
    ifstream  myfile;
    int n,p;
    int **st; //st[s] states which groups the student s is in;
             //the value (*st[s]) contains how many students are in st[s].
    myfile.open("input.txt");
    myfile>>n>>p;
    st = new int*[n];
    //here is where a nice coder would use a list of the group pointers
    int result = 0; // we calculate it dinamically;
    int tStudents = 0; // total number of students so far
    for(int i=0;i<n;i++){st[i] = NULL;}
    int a,b;

    for(int i=0;i<p;i++){
    myfile>>a>>b;
    if((st[a]==NULL)&(st[b]==NULL)){
        st[a] = new int;
        st[b] = st[a];
        *st[a] = 2;
        result += tStudents*2;
        tStudents+=2;
        continue;}
    if((st[a]==NULL)){swap(a,b);}
    if((st[b]==NULL)){
        result += tStudents - (*st[a]);
        st[b] = st[a];
        *st[a]+=1;
        tStudents+=1;
        continue;}
    if(st[a] != st[b]){
        result -= (*st[a])* (*st[b]);
        (*st[a])+= (*st[b]);
        delete st[b];
        st[b] = st[a];}
    }

    myfile.close();
    cout<<result;
    return 0;
}

>>1064334

Ignore that. This code is wrong when multiple people belong to the st[b] group.

>>1064301

O(n) Don't think you need any special data structures.

>>1064301

was bored so I solved it in O(n):

from itertools import accumulate
from operator import *

def c(it):
    F = {}
    R = {}
    for a, b in it:
        F[R.get(b,b)] = F.get(a,a)
        R[F.get(a,a)] = R.get(b,b)
        F[a] = b
        R[b] = a
    n = []
    while F:
        A = next(iter(F))
        i = 0
        while A in F:
            i += 1
            A = F.pop(A)
        n.append(i)
    return sum(map(mul, n[1:], accumulate(n, add)))

Yo I love writing Rust! Even though I am shit at it right now

>>1064394

based

>>1063958

>>1064046

One more question about this parsing thing.. It should be done by using operations on strings?

I have finished reading how to design programs working on cat in racket rn

>>1064357

Who is this semen demon?

>>1064539

You normally parse the string to a logical structure (a tree), and operate on that.

Steps being tokenization (splitting the string into base entities; eg (25 * x) + (2*y) becomes tokens of (, 25, x, ), +, (, 2, *, y, and )

Followed by parsing, which takes it to a tree: (+, (*, (25, x)), (*, 2,y)))

And then if you want to eg calculate the expression, you recursively calculate its children: x and y evaluate to whatever number, 25 and 2 evaluate to themselves, multiply applies multiplication to its two children, getting back a single result; the two multiply outputs are fed as input to the add function, and theres your final result

The main reason to convert it to a tree is to make eg rearrangement trivial: if you want to make one of the multiplies the top-level operation, you just have to move the nodes around

Which is how you’d do things like optimizations with logical equivalences, eg if you were doing a sequence of matrix operations, you can swap the nodes to make the smaller matrixes come first, to reduce size of future of operations and reduce overall work

And thats the basic structure of a compiler

>>1064372

Thanks but I'm not used to python can't really read this could you explain the logic what's going on here

>>1064802

There are three steps. Lines 5-11 create the groups. 12-19 count their sizes. Line 20 computes the number of pairs. The groups are created using a graph. Each student is a node. F stores the forward edges, R stores the reverse. Every student in a group will be in the same connected component. Given a pairing a,b, we splice the relevant groups together. If a student isn't in the graph yet, it's as if they are in a group all by themselves, such that both forward and reverse edges point to themselves. This is what R.get(b,b) and F.get(a,a) mean. For the second stage we continuously pick an arbitrary student and use F to find all their group members, counting and removing them as we go. For the final stage we compute the sum of the products of every pair of groups. Suppose the group counts are a, b, c, d (for 4 groups). We have S = b*a + c*a + c*b + d*a + d*b + d*c. If we factor we have S = b*a + c*(a+b) + d*(a+b+c). This is the expression the final line computes.

>>1064227

Thanks, I got the Jones linux application development book on library genesis and it seems really good.

>>1064677

That's a dude

>>1065129

Question still applies.

I remember seeing a picture of programmer challenges, ranging for simple stuff like printing all the even integers in arrays to much tougher things. Does anyone have it saved?

>>1065273

https://i.warosu.org/data/g/img/0651/18/1521215724300.jpg

I've been translating the Wolfenstein 3D and Doom source code into modern C++ and NASM, to re-learn the former, and to get a hang on x86 and x64 Assembly. So far, I've been doing alright. I've been considering reworking both games' graphics code with SDL or Irrlicht, but for now, I'm going one step at a time.

>>1065307

>modern C++ and NASM, to re-learn the former

What was your previous experience with the language?

>>1064229

Haven't you seen the do your own web downloader thread? There are a hundred programmers that are regulars here, at least.

>>1062706

>problems become solvable once you know what you're doing

That's the hard part. If you write some big (+2000 loc) program for yourself, you still have to figure out what the specifics are. If some retard business man gives you something to program, they usually have no fucking clue what they actually want. You have to figure out what they want to accomplish and where this or that feature is implemented logically.

Bottom line is writing up some kind of design which will be the base of what you are going to program. The design will also keep you on point as to what should be in the program to be deemed usable as a starting point. Its easy to get sidetracked without it and waste time on some interesting shit not progressing development.

Programming is just modifying input to output, every retard can do that.

>>1066013

Yes there are at least 9 people on this board who can write one-liner shell scripts. I stand corrected.

>>1065307

How are you accounting for interrupt usage? Or are you writing it to still run off of DOS?

>>1064301

Learning Haskell so I thought I'd give this a shot.

import Data.Graph
import Data.Tree

pairwise :: [a] -> [[a]]
pairwise xs = map (\(x, y) -> [x, y]) (zip xs (tail xs))

parseLn :: String -> (Int, Int)
parseLn s = let [x, y] = words s in (read x, read y)

go :: String -> String
go s = let l:ls = lines s
           (n,_) = parseLn l
           g = buildG (0, n-1) (map parseLn ls)
           ps = concat $ map sequenceA (pairwise (map flatten (components g)))
       in show (length ps, ps)

main :: IO ()
main = interact go

Output for the example data is: (6,[[0,2],[0,3],[1,2],[1,3],[4,2],[4,3]])

Do any of you have any experience handling Java project dependencies using maven? I'm trying to figure this out but boy I'm not doing too well

Strings are terrible. Fuck them and fuck parsing everything in unix land.

>but it's easy to parse

No it's not.

oh shit nigger strtok "extract tokens from strings" damn that sounds like a good way to parse structured data. lmfao read the man page.

>"aaa;;bbb" would return aaa and bbb

Yeah because obviously the ;; is not an empty field or anything. Thank you helpful string util.

>>1067020

That's not strings being terrible, that's strtok being terrible. Use a real lexer.

>>1062567

what's a good non diversity hire language? I've been surrounded with indians for 5+ years in java.

>>1063164

>>1063171

don't be like that the screenshots hurt everyone's eyes it's a bbcode code tag

>>1063821

if it's your first language do the top part, skip gui's java gui's are trash, do JNDI. JAXP, JAXB and not on the list but JAX-RS and JAX-WS; you'll get yourself a jerb.

>>1063961

learn about optional and stop instantiating class variables to null without setting them in the constructor

>>1067020

for maximum linux autism learn awk.

echo -ne "aaa;;bbb\nccc;;;;ddd" | awk 'BEGIN{FS=";";OFS="-";}{print NF,$1,$2,$3,$4}'

what really sucks is when people invent there own non standardized formats, this looks like a ghetto version of csv. also kind of pointless to think about parsing without knowing what you're throwing it into.

>>1067095

csv is already a ghetto format.

>"This is a "quote"","\"","oops",lol

Let me know what your csv parser gives back.

>>1067144

xsv flatten --no-headers test.csv
0   This is a quote""
1   \",oops"
2   lol

>>1065307

>structure and interpretation of computer programs

>1996

I've seen people talk about this book. Is it just a meme, or is it acutally worthwhile to read it in the current year?

>>1067616

>scheme

It's a meme.

>>1067631

>weeb image

Actually, I was memeing. Read SICP. It's not a meme.

>>1067625

>It's a meme.

You're wrong, it is actually a language designed better than for example C or C++. I had used C++ and it's nothing special, e.g. its syntax is inconsistent and sometimes ugly. Scheme strings are implemented better than C strings, null-terminated shit, it causes more errors, than have advantages.

>>1067646

>designed better than for example C or C++

Not especially hard tbh.

>>1067819

No thanks.

I like being a heterosexual male.

>>1067819

>Rust isn't in this degenerate image

based

>>1067819

Ruby's mine.

>>1067646

Lisp is the gayest shit in the universe.

>>1067646

No useful software has every been written in lisp. Virtually every noteworthy piece of software for which performance is critical and runs directly on computer hardware is written in c/c++.

>>1068093

>Lisp is the gayest shit in the universe.

t. JavaScript web developer

>>1068094

>No useful software has every been written in lisp. Virtually every noteworthy piece of software for which performance is critical and runs directly on computer hardware is written in c/c++.

Emacs is partially written in emacs lisp.

GNU Guix distribution uses scheme (lisp dialect) as it's main programming language, the package manager is written in scheme and GNU Shepherd - an replacement for sysv init is written entirely in scheme. And Guile - GNU's scheme implementation runs now 4 times faster thanks to JIT compiler, a VM and new low-level bytecode. If a language is fast actually depends on a interpreter/compiler you use. And LISP has different advantages from being fast, for example being hackable - defining syntax rules yourself allows you to create a domain-specific language.

And please stop being such a C weenie. Better learn Rust or something, because it is fasterrrr and saferrr

>>1068093

based

>>1068094

based, except Emacs exists

>>1068109

based

>>1068112

based

>>1067819

>PHP

>good intention ruined by error prone tendencies

>Perl

>an awkward lesbian shut-in

This is actually a pretty fair portrayal. If JavaScript had some annotation stating that she is a schizophrenic with crippling dissociative disorder, then I'd say that's dead on.

Could you draw Golang as a obsessively organized but Downs Syndrome affected kid with rich parents? I'm also surprised no Lisps made it.

>>1068159

Core Emacs is written in C, with its Lisp interpreter also written in C

>>1068201

>Lisp interpreter also written in C

In what Lisp interpreter is written in doesn't matter. There are LISP interpreters and compilers written in LISP, for example chez scheme is written in chez scheme, and parts of GNU Guile written in C are going to be rewritten in scheme. And if emacs used C instead of LISP, that would be a nightmare, you would have to recompile everything and fuck with unnecessary low-level stuff (yes, you don't need low-level to extend a text editor).

>>1068203

>are going to be rewritten in scheme

Says every academic for the last 30 years. Go back to writing equations on the whiteboard.

Programming on *nix is garbage and everything is a fucking joke.

>oh shit how come stderr doesn't print anything

>fuck it probably tty faggot ancient alien magic

>let me set up some log file

>oh look suddenly the log file isn't being written to

Literally nothing to do with the asshole problem I'm trying to solve anyway. Thank you to the worst tooling system ever invented.

>>1068218

not an argument

>>1068258

based

>>1068259

>loves LISP

>literally a ni/g/ger

You can't spell LARP without LISP

>>1068261

>sucks cock for a living

wrong board, faggot

>>1068262

LISP has never been posted in any of these threads in 5 years. What a totally pointless LARP.

>>1068218

>Says every academic for the last 30 years. Go back to writing equations on the whiteboard.

Lol, but it actually is going to be rewritten in scheme.

http://wingolog.org/archives/2019/05/24/lightening-run-time-code-generation

<I needed simple because I would need to debug any problems that came up, and I didn't want to add more complexity to the C side of Guile -- eventually I should be migrating this code over to Scheme anyway.

>>1069359

> examples with shitty variable names, old non-var variable declaration, annoying italics and colors of members, bizarre and haphazard indentation without any discernible style or logic turning everything into a confusing fucking mess, shitty operator-based string interpolation, and shitty dated GUI libraries

Java is over-verbose, messy, bogged down by historical mistakes and misdesign, but the person who wrote the code in your images would make shitty code in any programming language. Most of the mistakes making that code look shitty are not the fault of Java.

>>1069390

It wants you to use dynamic programming. This means we will need to break down the original problem into a simpler problem. The way we do so is instead of just solving the problem, we a remove the first actual word from the string and solve the problem again. Since you doesn't know the length of the world in question you'll need to branch into two cases. The first is that the word you detected is a word from the actual sentence. The second is that the word you detected was actually just the beginning of a different world. For example if you were given "age" you couldn't just infer "a" is a word since ge isn't a word. You would need to detect "age" as he word.

The problem stated that our alg must have a worst case of no worse than O(n^2). Let's check the worst case of the alg above. The worst cause would be branching after every single letter of the input string of length n. This means the total iterations would be n + (n - 1) + (n - 2) + ... 3 + 2 + 1. This is the same thing as taking the sum of the first n natural numbers. As you'll recall Guass' formula for this is (n^2 + n)/2. Since we are only worried about what happens at the asymptotes (since this is big O notation), we can simplify this to n^2. This means we have satisfied the constraint for the worst performance of the algorithm.

I don't quite understand the second part as you can get multiple solutions. "Again" can be "again" or "a gain." I assume you just output the first solution you find though.

>>1069397

>shitty operator-based string interpolation

Guy that's learning java and wrote the code here. What's bad and "shitty" about it?

>and shitty dated GUI libraries

Are you saying that java devs need to know javafx?

>>1069402

I guess it means if all the substrings are vaild dictionary words, just print the newly created string. Suggests one must give both an analytical answer and a pratical example.

>>1069482

The main shitty thing is that it's basically an inefficient hack. Behind the scenes, the Java compiler translates the operators into a StringBuilder with a bunch of appends. If you're doing a bunch of them, it's going to be far slower than true interpolation, like String.format or something of the sort.

The main point, though, is just that it's ugly and less readable, with a bunch of syntactic noise:

"My name is " + name + " and I am " + age + " years old.  My job title is \"" + job "\"";
// versus
String.format("My name is %s and I am %d years old.  My job title is \"%s\"", name, age, job)

"My name is $name and I am $age years old.  My job title is \"$job\""

Javafx or whatever you like. Swing is just hideous and the API is disgusting. Also, use a formatter or IDE. The majority of the readability issues there are just your fucked up indentation, redundant information, and Swing cruft rather than Java itself.

So I've used Python for years and am very comfortable with it. I've been using Golang over the past couple months and appreciate its simplicity and very useful standard library. I can see it replacing my use of Python for most cases.

I would like to work with something that has good performance and can run on minimal hardware. I can read (not write) C and don't want to deal with having to define really basic operations. I really don't want to invest the years it takes to get good enough with C++ to write working software. Give it to me straight, /tech/. Is Rust really the best alternative to C++?

>>1069509

lua/luajit unironically. Simple and elegant, will remind you a lot of golang, just with less bloat. Also don't let yourself be fooled by the fact that it's an interpreted language, it's positively tiny and in some scenarios, a well-written luajit program can easily keep up with C (yes, that C) speedwise. It also runs on very minimal hardware and in restricted environments. I've had lua scripts run in DOS on 100 Mhz Pentiums with 8 MB of RAM. It doesn't get much more minimal than that. Try that with rust or go.

I love this language since it's so minimal and doesn't tell you how to do things, at all. Want OOP, functional or procedural? Implement it, nothing is stopping you. It's pure elegance. Way underrated imho. The freedom can be overwhelming for some though since you can do whatever without clear guidelines to do things, which I have observed is actually a hindrance for programmers that are used to "stricter" languages. It's also easy to write inefficient lua programs if you are bad at programming. It also doesn't have the 39825823582502 libraries ready to import, without which java and python devs would often be helpless.

Contrary to popular opinion, it's also entirely possible to write big programs in lua, you just need good tools for linting and static analysis.

>>1069509

>Is Rust really the best alternative to C++?

Certainly not, Rust can't even compile itself on the "minimal hardware" you're likely thinking of (unless xu6_64 counts as minimal) without running out of registers.

>>1069482

It's useful if you don't have another language, but no one should be doing java ui's other than maybe android devs, and knowing how shit javafx and swing are I'd give a look at writing it in go first.

Java's great for plenty of things complex business systems, rest, soap message busses, any kind of data storage, many more things I don't feel like listing, but it's ui's are trash just slap up a web service and write a client for it in another language to do your ui with.

I'm in the midst of making that mistake with javafx now not a day goes by I don't wish I'd just said fuck it and made the thick client an electron app.

>>1069525

Hakase, is that you?

>>1069509

Golang really is a great language for a lot of use-cases, particularly when paired with Electron or a plain browser. I'm in the process of replacing a lot of Excel and Access crap with homemade Electron apps in my accounting work. Fuck VBA, life's too short for that shit.

The standard library is really solid, I love that I can basically jettison Apache and PHP entirely from my life. Good riddance to them. And I don't need to make sure some entire damn interpreter stack with hundred optional packages is installed on the machine on which my program runs, because it compiles to a single static binary. Drop the binary on the office server, restart the service, done. My life is simpler and calmer.

And I love love love how readable it all is. Golang has no style-guide debate bullshit, like certain other languages (coughCcough). I can't remember the last time I opened someone else's Golang source and wasn't able to instantly follow the logic. More languages could use something like gofmt.

>>1069525

that lua love2d engine is actually pretty OK on android

lua is almost a general purpose language

>>1069603

Android and iOS don't allow LuaJIT do they? Would be a pretty big penalty on performance to fall back on regular Lua.

>>1067819

Based Javascript tomboy.

>>1069509

Did Go ever get package management or is it still a joke?

>>1069626

https://blog.golang.org/using-go-modules

>>1069557

Glad to hear that it's going so well. I've found the template and net packages really useful for automating administrative tasks related to my job (civil engineering).

I'm not sure if I can write my own tools in Go to do some of the graphics and geometry processing I keep promising myself that I will do someday. I would love to never pay for an Autodesk license as long as I live.

>>1069626

What is good dependency management in your opinion? I don't know if NPM and its clone Cargo are good tools. They seem to lead to the ecosystem becoming a huge trash-heap of half-finished and poorly maintained packages. How secure is that third version of some random guy's code that's nested four levels deep in your dependency tree? That's way too much trust for me. I read a comment once that every programming language package manager is just a worse reimplementation of dpkg, and I tend to agree.

Go modules are available now, though, and will be standard in 1.13 (proposed release in August). So far it looks easier to manage, but it is a middle finger to the people who worked on godep. It's easy to forget that Go is a trademarked Google language and all decisions are underpinned by Google's needs and financial considerations.

>>1069636

>It's easy to forget that Go is a trademarked Google language and all decisions are underpinned by Google's needs and financial considerations.

Yeah, this is my one concern with the language. Granted, it's developed and shepherded by extremely smart and accomplished people. And Google's implementation is under a super-permissive license. And the reference implementation doesn't have any phone-home telemetry sketchiness. And it's such a perfect language for my beginner-skill-level and specific problem sets.

But yeah, it is very much a Google-owned project. Such is life...

>>1069486

It suggest whoever wrote the question didn't know what they were asking for. Which is super common for retarded coding questions.

Most of them are just guess what I'm thinking or come up with the same solution I did.

YOLOL https://www.youtube.com/watch?v=7lXK6_bktUk

>>1069482

>>1069508

>Behind the scenes, the Java compiler translates the operators into a StringBuilder with a bunch of appends

Only for Java 8 and older. Newer versions of Java generate bytecode which performance an invokedynamic on StringConcatFactory::makeConcatWithConstants(), and its quite a bit more performant than StringBuilder (avoids buffer reallocations) and even String::format (which does format parsing, possible boxing/unboxing, and array allocation for each call).

In summary, you should always use the + operator to concatenate strings in Java, and compile with JDK 9+.

>>1062447

altGr + right blue button

>>1069711

Do you believe that the current languages are inadequate for this purpose?

Best book to learn a useful assembly language? I have tried to pick up programming as a hobby a few times but I am more interested in the nitty gritty "bare metal" stuff than I am in making anything of real value yet.

>>1069711

>What the hell is multi-threading

>>1069731

>>1069730

I was hoping some company would try to replace C/C++ on the back of 'multi-core' programming with something memory safe. Go is my first web dev for Google and Rust is rusting away in its safe space.

>>1069733

Maybe GNAT is more to your taste? Swift is a thing if you'll marry yourself to iOS. It's linux support is . . . not good.

Golang is genuinely useful. You can also learn it in a weekend. Just don't expect to do a bunch of Clojuretricks. Rust is a good idea with a cautionary tale of bikeshedding to learn from it. I think in 2-5 years, we will see a lot of the cruft disappear as big tech pushes out the Steve Klabniks of the dev team.

That or somebody will fork it and drop all the bullshit and keep the good parts.

Go is 10 years old. Rust is 9 years old.

Language development died in the 2010s. So did software development. In fact a lot of shit just stalled totally with the great recession.

30 years of javascript and php. oh but don't listen to a fag like me who says we are in a dark age nah just around the corner in the next 10-20 years we will have a replacement for something that wasn't even well designed in the 70s. In the mean time why not revisit the JVM or write a fat webapp or write assembly for some chinko device, what a time to be alive.

>>1069721

I've not finished it yet, but I've found x86-64 Assembly Language Programming with Ubuntu by Ed Jorgensen to be helpful. You don't have to use Ubuntu, but you do need yasm.

http://www.egr.unlv.edu/%7Eed/x86.html

>>1069755

I hate to sound like a commie but I blame corporate America. They have systems that work so the best course of action is to just patch the old system with whatever is new. It explains why we still have C/C++ around.

>>1069758

Damn corporations and their working systems! Gnu Herd would be released and written in a truly proletariat language if it weren't for these greedy porkies.

>>1069758

nah the issue isn't that corps are slow and conservative to change.

I mean what is the business model of the modern tech corp? Exchange rep for money pretty much. Same as the video game industry, bankers etc.

Literally nothing to do with tech at all. It's no wonder most of their nigger cattle board are essentially political CFR faggots. What the fuck do you need to know about the internet to run jewgle? or faceberg? Is even 10% of their business technical.

>>1069762

literally this

>>1069762

But gamergate won.

I just want to throw my ten cents in. I've been in love with C since I was a kid. I still feel amateurish when using C and there are always more libraries constantly changing than one could ever know what to do with. The only reason I touch Python is for BeautifulSoup4 because it would be a nightmare to parse HTML in C unfortunately. It is only really a requirement for scraping websites anyway.

What makes C so great is that once you know what you're doing, most of everything in *IX operating systems is available for your own customizations and you can now contribute/submit PRs to actual prolific projects in the open source world. It's such a rush to have a grasp on something that matters and not some skiddie php copypasta. I used to spend months trying to figure out why bash scripts didn't behave the way I expected them to and now that I moderately understand C, I have no reason to play with scripts in the first place.

The most recent success I can claim is that I've delved into gdb (I used to use Visual Studio) and I'm now more confident in developing portable code as I debug existing projects listing code and probing variables to understand how my peers and predecessors do things.

>>1069811

>LARPing as a Cnile

pretty convincing desu

>>1069829

Rust

>>1069830

Yeah, no. I'd rather not go insane with Rust itself and emscripten and JNI and whatever iOS has to convolute the use of non officially sanctioned languages.

>>1069758

How much time have you spent porting bad legacy code to not-C for no pay at all?

>>1069761

Spooky, but probably true.

>>1069833

Just compile to Wasm for Android/iOS/Web.

>>1069829

Ask around on /v/'s /agdg/ threads. One of the yesdevs there has shipped two games written in Haxe, one of them 3D.

>>1069829

>var foo:Bool

>foo is null on dynamic targets

>foo is false on static targets

If they can fuck up something this basic I've got zero faith in the rest of it.

https://www.jetbrains.com/lp/devecosystem-2019/

Parrot the popular responses here so you seem less like a nodev

>>1070033

>having a popular opinion

>/tech/

you came to the wrong board bucko

>>1070033

>doesn't realize this is polled from a majority of pajeets

>>1069829

SDL is the better meme, but HAXE werks.

How long did it take you guys to "git gud", I've been struggling to grasp the fundamental concepts of programming, like recursion and those sorts of things. I think I'm a brainlet. I've been studying SICP and learning about Data Structures and Algorithms but it just won't click for me. I can write some stuff in python and now I'm trying to learn C but I feel like I'm not getting anywhere. Any advice?

>>1070522

SICP is rather hard for a beginner.

>>1070522

it takes about 10 years.

This is a meme and also the truth.

>>1070522

practice. The first step is the art of solving problems with a poor (inefficient, naive, overcomplicated) solution. As you continue practising, you can revise your older solutions and consider ways to improve the logic in some way. It is only with practise and consistent critical evaluation that your skills will improve.

I recommend the Khan Academy programming course because it contains a significant number of programming challenges that are designed to build up your knowledge and experience piece by piece.

>>1070522

Stop reading memes and actually write code.

https://www.humblebundle.com/books/programming-packt-books

Are any of those worth jack shit?

>>1070669

wow, this is the worst programming bundle I've seen yet. Not sure how it could get worse and not actually be an "iOS dev bundle" or some other niche topic.

These are books designed, not to appeal on their own, but to be the only book on the shelf when you go to the only bookstore in the area to pick up something for your resume.

Should I learn Ada if I hate Jews?

>>1070689

No you should go to therapy.

>>1070701

>(((therapy)))

I guess I hit a nerve, huh? I will learn Ada now.

>>1070714

I've been mulling that idea. Some autistic chink has been spamming Ada praise on his resume-promotion blog and HN. I wasted 3 months on Rust just to learn that to do anything significant you end up using bindings or an FFI to C++ libraries anyway. So you avoid scary C++ by calling C++. The safety guarantees are ultimately "dude, trust me." The dependency growth is also Nodenigger tier. Especially because the standard lib doesn't support shit that is actually useful (like argparsing or unmarshaling data to a struct without a fucking opaque macro) and supports way too much shit you should not be using regularly (like async).

I do find a lot of appeal in a language with strong safety guarantees and performance, though. Not many options on the table other than Ada, it seems.

What's the point in learning other new languages in the context of actual applications when most modern OSs still use x86 based assembly for which C is the most optimized?

Yeah, for toying with interesting concepts other languages might be good, but unless there's some connection between a high-level language, the low-level language, and the hardware, it seems moot.

>>1070777

Developer productivity.

>>1070777

>modern OSes use assembly

>C is the most optimized for assembly

hell, C compilers are better at optimizing generated C code than they are hand-written C code. Even if you had some kind of phobia of abstraction, you could use D with -betterC

>>1070730

I figured that's basically what rust was.

I've took a stab at Ada but its syntax isn't clicking right now. Needed to mess with some tcl early this year though and fell in love and can't stop using it.

Still have Ada on the back burner, but I have too much stuff in C or C++ to work on.

>>1070792

generated C code is generated from something

C++ traditionally preprocessed into C as an intermediate step, LLVM may have an improved pathway for this but compiler optimizations tend to be language agnostic and C has a wide range of compilers. The difference between C and emerging languages is that decades of optimization work have been put in, you won't find that with newer languages so much, although if LLVM has front-ends for your preferred language you may be able to take advantage of compiler optimizations one usually only finds for C/C++

>>1070730

>to do anything significant you end up using bindings or an FFI to C++ libraries anyway.

wrong

>Especially because the standard lib doesn't support shit that is actually useful (like argparsing or unmarshaling data to a struct without a fucking opaque macro)

But that's a good thing. Just take a look at any language that has shit like that in its std.

>>1070777

>being this cnile

Have fun with your UB.

>>1070840

>rust has defined behavior

lol

>>1070860

Who are you quoting?

But yes, Rust currently has UB because it uses LLVM. LLVM is full of braindamage.

>>1070865

Rewrite LLVM in Rust

>>1070865

as opposed to what? Visual Studio? GCC?

LLVM is the only option afaik

>>1070868

https://github.com/CraneStation/cranelift

>>1070840

GUI is still basically all GTK or QT bindings. Conrod really isn't there. Math boils down to BLAS or LAPACK bindings because nalgebra's API is a bad joke. And languages that can unmarshal structured data like JSON are actually useful. Shit, dude, that's the basis of every fucking REST API! But, hey, I can overload operators and define macros until my code is an unintelligible DSL! Can I get a fucking PCRE, please?! Can I get a god damn reference implementation of HTTP routing and abstraction for headers?!

>Rust is great for CLI!

<How do I handle flags/switches?

>uhh, check Cargo

>Rust is great for web!

<How do I consume data from a REST API?

>uhh, check Cargo

>Rust is great for embedded!

<I have microcontroller that isn't a raspberry pi

>uhh...

>Rust is great for web assembly!

<But web assembly is actually slower than regular JS and webGL

>uhh...

It's not that Rust is awful, but it grossly under-delivers on its promises and has questionable priorities.

>>1070938

>GUI is still basically all GTK or QT bindings.

That is true for every language though.

>Math boils down to BLAS or LAPACK bindings because nalgebra's API is a bad joke.

True, nalgebra is pretty horrific. Why would you use Rust for math though? Use Julia.

>JSON

https://github.com/serde-rs/json

https://github.com/maciejhirsz/json-rust

>Can I get a fucking PCRE, please?!

https://github.com/rust-lang/regex

>Can I get a god damn reference implementation of HTTP routing and abstraction for headers?!

I don't really know what HTTP routing is.

https://github.com/hyperium/http

>muh CLI

Just take a fucking look at Python. 3 fucking different approaches to argument parsing in its std alone.

Fuck off. Argument parsing has NO place in Rust's std.

>muh REST APIs

Wtf does that even mean? Stop repeating webdev buzzwords. I already told you how to handle JSON. If you just want a simple HTTP client API use reqwest.

>muh embedded

https://forge.rust-lang.org/platform-support.html

>muh web assembly

Fuck off webdev nigger.

>Rust grossly under-delivers on its promises

Rust never promised to have webdev tier shit in its std though.

>>1069853

>rust can be used for the web just use webassembly lmao rust 4 lyfe

<but webassembly is shit

>>1070941

>lol why do you even want to do webdev lmao rust 4 lyfe

Rust shills, everyone.

>>1070943

>>lol why do you even want to do webdev lmao rust 4 lyfe

Who are you quoting?

https://github.com/actix/actix-web

https://github.com/hyperium/hyper

sage negated btw

>>1070946

cuckchan begone

How do I make sure I get a high paying job in computer science? I just started college.

The goal is to get a job over 100,000 asap and then a job 200,000+ eventually

>>1070960

Take a graphic design course (make sure the rims in your teacher's glasses are at least 7mm thick), then learn all (see: ALL) the latest javascript frameworks and how to use Electron.

It helps to know some basic computer science so you can avoid things that result in fast software.

>>1070957

>projecting

sage negated

>>1070714

Ada is great, but your hate is irrational and being irrational is not useful in a programming context.

>>1070963

>be super concerned about safety

>cut off own genitals

what did the rustranny mean by this?

>>1070970

>hating kikes is irrational

get out

>>1070973

nice meme

btw GCC is allowed to cut off your genitals if you don't terminate your source files with a line feed.

>>1070941

Nodenigger, the whole argument that you need to lrn2rust is because C++ is inherently unsafe and your programs will be full of RCE bugs . . . but C++ is OK for my GUI. Retarded.

Regex crate doesn't support look around, which is a big part of PCRE.

Python arg parsing exists and that alone makes it better than Rust. Go's flag package is even better.

And how do you not know what a REST API is? Holy shit, dude, you sound like a highschool nodev.

Each of those points are from Rust-lang.org's front fucking page. I didnt pick them. And none of them are even practical without using random people's v0.x shit off Cargo. Nobody else seems to have a problem shipping standard solutions to these extremely common problems in their nulangs. Fucking nim has JSON parsing. How much budget does that dev team have? How is one guy with a couple sponsors more productive than Mozilla who has half a billion in annual revenue? I know that defending Rust is a troll, but have some self-respect.

>>1070988

>Nodenigger

I'm the Rustnigger though

>but C++ is OK for my GUI

I did not say that.

>Regex crate doesn't support look around

Good. Rust's regex crate guarantees linear time matching on all inputs.

>Python arg parsing exists

Yes. 3 times in the std alone.

>Go's flag package

Is stripped down to the point of unusability, just like the language itself.

>And how do you not know what a REST API is?

Are you retarded? Read my post again. Then read https://en.wikipedia.org/wiki/Representational_state_transfer. REST is a meaningless buzzwords.

>And none of them are even practical without using random people's v0.x shit off Cargo.

CLI: Use clap. Currently on major version 2.

JSON: Use serde-sjson. Currently on major version 1.

HTTP: Use actix-web. Currently on majoir version 1.

Kill yourself you lying faggot.

JSON has no place in Rust's std. Rust is neither a scripting language nor a webdev language. It is a system programming language. Guess what other system programming languages also do not have JSON/HTTP/argument parsing in their std? C and C++.

>but muh Nim

some random person's v0.x language. epic

>I know that defending Rust is a troll

<everyone who isn't Cnile is a troll

Your cnility is showing.

>>1071001

>REST is meaningless

Your Google education is impressive. It's almost as if you've written a real program once.

>CLAP, SerDe, Actix

Check the dependency tree, Mr. Gender Dysphoria. And name one project actually using Actix. M$'s mystery project? It's all hype. There's already a well established actor model for networking, and it's Erlang. Everything old will be new again, it seems. Just like how hipster devs are discovering standard MLs 36 year old type system and getting woke on functional programming. Runtime errors moved to compiler errors (so I can be lazy and write no tests)? Far out maaaaaaaan.

C++ was standardized in 1985. JSON was standardized in 2013. No shit STD doesn't support it, but ISO doesn't have some gay website purporting that C++ is the one true lang for web development and servers. Rust does.

>>1071007

>Your Google education is impressive

I have a master's degree in computer science. You're just a Cnile LARPer.

>Check the dependency tree

Check the source of anything written in C/C++. UBs everywhere.

>Mr. Gender Dysphoria

Now that's just sad. What a weak insult.

>name one project actually using Actix

https://bible.rs/

Also Cloudflare uses it for the image resizing feature.

>There's already a well established actor model for networking, and it's Erlang.

actix-web v1.0 isn't based on the Actor model anymore btw. It is now based on something similar to this: https://monkey.org/~marius/funsrv.pdf

You have yet to make a convincing case for why Rust should have JSON support in its std btw.

>>1071013

>i have a master's degree in computer science

lmao the day has finally come a pajeet tranny on /tech/.

>>1071013

>Masters from university of Mumbai.

>No mention of production code

Sounds about right.

>bible.rs

>JSON unnecessary

Check out all the xhr's spawned by the search bar in that example you just looked up on github because you couldn't think of a serious example. Huh. A REST API? I wonder if that's something web servers need to deal with on a regular basis...

And your shining example would have been much more simply implemented as a static site with elastic search.

>>1071030

>incoherent babbling sprinkled with misquotes

Just admit that I destroyed you with facts and logic. Stop behaving like a child (you probably are one though).

>>1071044

The fact that you have demonstrably no knowledge of what you're talking about. I've been arguing with the top 3 Google results this entire time. Post your git repo or GTFO nodev.