/dpt/ - Daily Programming Thread

No.

>>59734442
Stop.

The real anime poster only makes another thread if the first one was posted before bump limit.

Yes.

nth for good thread

>>59734470
The other thread is dead, this is the official one.

Sage and report the other one.

Did you guys program in Prolog?

>>59734659
Both threads are dead.

[spoiler]But still, I agree.[/spoiler]

Can we just have a consistent /dpt/ instead of shitposting about which /dpt/ is right? It's worse than the constant "ryzen BTFO/intel is the botnet!" meme.

Unrelated: Is there a /dpt/ irc channel or discord yet?

>>59734711
Yes.

https://discord.gg/UEEKQ

>>59734711
Yes, it's called the Usenet.

>>59734442
why do you fucking faggots keep posting anime images for dpt? like wtf is your issue with another type of picture? shit like this and all the other anime bollocks that pervades this board is why the good posters moved to reddit

#5..6
>bollocks
oh look its another dumb yurucuck lost on his way from reddit

File: .jpg (36KB, 233x215px) Image search: [Google]

36KB, 233x215px

>>59734442
baka op

>>59735222
Baka janai yo!

>>59734670
Bumpung question

>>59736341
You may want to try the currently active thread.
>>59734157

>>59734796
>complaining about mongolian pictographs on Tibetan basketweaving imageboard

File: ! (2).gif (176KB, 500x711px) Image search: [Google]

176KB, 500x711px

First for ML master race

I will make robot take your job (ง ͠° ͟ل͜ ͡°)ง ≧◡≦

guide for ricing, pretty much done with bspwm, what other WM/DE should I do?

>>59736709
WindowMaker.

>>59736432
what?

>>59736692
quick have you implemented back-prop in js?

@59734796
>is why the good posters moved to reddit
Move you there as well.

what is the best way to structure your code when you are making a video game programming it all?

>>59736940
Put everything in one big file so it's easier to delete when you give up.

>>59736963
>meme answer

>>59736940
You're going to need to elaborate on what you're actually asking.

This highly depends on the language, how many people you're working with, and a myriad of other factors relating to the actual game.

>>59737055
C#, 1 man army, and a 3d puzzle horror type game like the old school Resident Evil games

File: image.png (54KB, 1265x861px) Image search: [Google]

54KB, 1265x861px

Need some help with Java OOP if anyone willing to help.

I have 3 Classes:
+ MainApp,
+ a Basic Class ( Primitive data types )
+ and a Container Class ( This class holds an ArrayList of Objects ( The objects being the instances of the Basic Class )

What I'm struggling with is how I get the Container Class to collect data from the Basic Class.
I can create instances of the Basic Class using variables e.g:

[script]
CarSpace a01 = new CarSpace(01,'a',"Standard",true,1.0,2.0,5.0);
[/script]

But I want this information added to the ArrayList in my Container Class.
So then my methods in the Container Class can use the ArrayList.
If I am being to ambiguous I'll add more information if need be.

Image:
Red : Container Class
Blue : Basic Class

Any help would be appreciated.

I really need some advice in creating a dynamic crypto key gen for selling my program using this (keys), would give bt on precious info or, just don't post deposit addresses. I'm a mediocre programmer who didn't go to any school, using c# .net

>>59737122
>rolling your own crypto

>>59737132
How?
I certainly can do it with just having minimal basic knowledge on it - which is what I need

>>59737178
don't

>>59737198
Too much money though

>>59737206
use a library and follow advice

>>59737206 (You)
Which I need.

>>59737117
you would use the get methods to get specific information of the objects

>>59737233
I'll try, but, is there any book I can gather knowledge from? White hatting here like an ethical dude

>>59737279
look up a page on the internet
don't do it yourself

>>59736744
cool what else. I got i3, herbs.., basic unity ricing, awesomewm, ncmpcpp

>>59737304
No security that way, anyway, thanks dude

OOP is the only way to program

>>59737335
Don't false flag.

>>59737323
look up what to do on a page on the internet
don't implement your own crypto shit

>>59737323
>>59737350
and even then that's telling you how to USE a library, definitely don't write the library yourself

>>59737350
What if my name is Ron Rivest?

File: 1486930683280.jpg (453KB, 1920x1080px) Image search: [Google]

453KB, 1920x1080px

>>59737249
I get that ( No pun intended ).

So I should call the getters from within the Container class? OR:
So everytime I create a new instance of 'CarSpace' I have the Container Class 'get' that info and add it to the 'ArrayList' by having a 'call' for the setter ( For the ArrayList ) in my Basic Class constructor? Because that's the only time I all the fields are together?

Sorry if I am missing something; I am new to this.

>>59737414
then you'll be someone i've never heard of with a ridiculous name

>>59737435
yes the second one everytime you create new instance of it have it retrieve the information and store it in the ArrayList or just store all the new instance objects in the ArrayList objects array whatever you are trying to do

>>59737466
https://en.wikipedia.org/wiki/RSA_%28cryptosystem%29

Are there any Java back end developers here? Do you like your job? I'm thinking about changing my job from Android developer to back end developer.

>>59737506
I have done some Java back end I would recommend you try it out see if you like it

>>59734721
>https://discord.gg/UEEKQ
the invite expired, can you post another

>>59737576
https://discord.gg/CqUF4

A Java Swing app for practise. Looking to get into android development.

Can someone link me the discord?

File: 1486930327993.jpg (385KB, 1920x1080px) Image search: [Google]

385KB, 1920x1080px

>>59737476
Thanks so much, Last question. Will I just use one variable to pass data from the MainApp ( Data Entered )
to the Basic Class. As a variable for evey instance seems unessasseary as I'm using the
ArrayList in the methods.

E.G:
[script]

CarSpace a01 = new CarSpace(01,'a',"Standard",true,1.0,2.0,5.0);

// Here the Basic Class is sending the data to the ArrayList

CarSpace a01 = CarSpace(CarSpace(26,'z',"Disabled",false,0.3,2.0,5.0);


[/script]

>>59737522
Do you have any idea how difficult Java back end is compared to front end like Android? I asked the same question to a C# coworker and we came to the conclusion that it's about the same in terms of difficulty.

>>59737957
Employed C# back-end dev here.

I'll take back-end dev 1000 times over before I ever touch UI again.

>>59737640
invites not working can you post a new one?

>>59737983
I've also looked into C# back end job offers, but I soon realized that I'd rather stick with the Java ecosystem. I don't really like the Microsoft world (not that Oracle is any better though). I would pick the C# job offer if no Java ones were available, but there are plenty to pick from. Java as a back end language seems to be especially popular in the banking industry.
>I'll take back-end dev 1000 times over before I ever touch UI again.
Why do you prefer back end over front end? Personally, I don't find UI all that interesting. I'd rather invest my time in writing proper business logic than in writing slick animations. I spend most of my time implementing video players, DRM included, because that's the most interesting task I have right now. Clicking a UI together isn't that interesting.

>>59736886
why the fuck would anyone implement back-propagation in javascript

do it in a real language like a normal person

>>59738715
javascript is a fine language to implement back-prop.

>>59736432
I think you meant "Swedish Antiques/Thrift IRC"

>>59737813
i think you meant

 [ "code" ] ["/ code "] 

but if this doesnt work im just making a fool of myself kek

>>59739703
Thanks I forgot what the correct tag was :^).

>>59738715
https://github.com/karpathy/convnetjs

Why do people love Java?

>>59741264
high level, statically typed, well tooled, state of the art gc, stable

>>59741264
In 1998 it was much easier to work with than C++. Once the hype was here, it never left.

How would i go about writting a function to freeing the memory used here?

struct node {
    struct node *so1,*so2,*so3;
    struct node *next;
    char *type;
    char *trm;
    int flag;
};

to allocate a new node i do

struct node *make_type_node(char *type)
{
    struct node *new_node = malloc(sizeof(struct node));
    new_node->so1 = new_node->so2 = new_node->so3 = NULL;
    new_node->next = new_node->trm = NULL;
    new_node->type = type;
    new_node->flag = 0;

    return new_node;
}

to print the tree (recursively) i do

void print_tree(struct node *tree)
{
    if (tree->trm && tree->flag == 1)
        printf("%s\n", tree->trm);
    else if (tree->trm && tree->flag != 1)
        printf("%s(%s)\n", tree->trm, tree->type);
    else
        printf("%s\n", tree->type);

    if (tree->so1) print_tree(tree->so1);
    if (tree->so2) print_tree(tree->so2);
    if (tree->so3) print_tree(tree->so3);
    if (tree->next) print_tree(tree->next);
}

>>59741602
What's the upper bound for your tree? Use a stack located storage pool and it'll go out of scope when your access type does.

>>59741870
>What's the upper bound for your tree?
my what?

i was just looking for a function that frees the tree
something like (obviously not this)

void free_tree(struct node *tree)
{
    free(tree);

    if (tree->so1) free_tree(tree->so1);
    if (tree->so2) free_tree(tree->so2);
    if (tree->so3) free_tree(tree->so3);
    if (tree->next) free_tree(tree->next);
}

>>59741902
Upper bound, meaning the max number of allowed or expected nodes. I told you how to do it the fast way. For plain C, move the free(tree) to the end of that function and you'll be fine.

>>59741968
>Upper bound, meaning the max number of allowed or expected nodes.
well i don't know.
i'm parsing real programs. depends on the size of the program

>>59734711
>It's worse than the constant "ryzen BTFO/intel is the botnet!"
No. the GPU/CPU Intel/AMD/Nvidia wars on /g/ are definitely the worst thing on the board. Worse than anything else on the board besides maybe that one SJW who I think is just a troll.

>>59741986
Your memory isn't infinite. There is always an upper limit. You just might have a high standard deviation.

>>59741968

==13387== 
==13387== HEAP SUMMARY:
==13387==     in use at exit: 17,140 bytes in 200 blocks
==13387==   total heap usage: 674 allocs, 474 frees, 43,684 bytes allocated
==13387== 
==13387== LEAK SUMMARY:
==13387==    definitely lost: 675 bytes in 196 blocks
==13387==    indirectly lost: 0 bytes in 0 blocks
==13387==      possibly lost: 0 bytes in 0 blocks
==13387==    still reachable: 16,465 bytes in 4 blocks
==13387==         suppressed: 0 bytes in 0 blocks
==13387== Rerun with --leak-check=full to see details of leaked memory
==13387== 
==13387== For counts of detected and suppressed errors, rerun with: -v
==13387== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 2 from 2)

its not freeing all memory i guess?

File: polbooks_fr.png (152KB, 780x382px) Image search: [Google]

152KB, 780x382px

hey guys, i need help.
im trying to make some kind of map(maybe like pic related? just a concept) which is navigable by the user through a 3D UI. The thing is that I want the program to be able to import and display images/text and play audio or video files. since i've never writen a program that involves 3D rendering of any kind I thought about using Unity3D. Is this a good idea? I've only used python, C++, C# and java. What would you guys recommend?

>>59742042
Is 'type' and 'trim' being deallocated anywhere?

>>59742071
no they are not.

i never malloc them.

i just pass them as parameters like

make_type_node("Function");

Debating becoming a dev vs going into finance

I like programming more than finance, but I hear dev work can end up as low-prestige codecuck with meh pay.

Finance is cuckery of a different kind but at least you get paid a ton and wear a cool suit.

Any advice?

>>59742104
My best guess is that you're replacing a node with a new node. Therefore losing memory.

>>59742169
you're going to figure it out.

don't worry

>>59742169
finance

How would you write a simple program in visual basic that calculates the sum of all odd numbers between 1 and another number? Not bait or anything, I'm just retarded.

>>59742169
both have excellent pipelines to alcoholism and suicide it's sort of an at your own pace kind of deal.

>>59742217
> visual basic
> I'm just retarded
checks out.

>>59742201
Why

>>59742235
more money

>>59742230
It's not up to me. I wouldn't use VB either if it was my choice.

>>59742217
You can do that in excel tbqhwu

File: 1490274520146.gif (1MB, 480x358px) Image search: [Google]

1MB, 480x358px

>>59742258
take control of your life man you're being made to write VB. At some point you have to take responsibility for how bad things have gotten.

>>59742217
Here's the steps to do it in any language
>Google hello world in X
>Google variable declarations in X
>Google arithmetic in X
>Google loop control in X
>Good mod operator in X

>>59742217
python

compute = lambda n: sum([i for i in range(n+1) if i&1])

print compute(5) # prints 9 because sum of 1,3 and 5

>>59742314
nice use of lambda.

Resposting in hope for answer
So i have this .dat file that has a dictionary in it (dumped with json)
the only thing that's written in the file is

{"Dan": [1500]}

but when i try to makes this file a dictionary in python this happens

, line 24, in load_reg
    register = dict(register_file)
ValueError: dictionary update sequence element #0 has length 15; 2 is required

full fuction

def load_reg():
    global register_file
    global register
    if os.path.exists("accreg.dat"):
        print("Debug[1] Reg File Found")
        register_file = open("accreg.dat","r+")
        print("Debug[2] File assigned to register_file")
        register = dict(register_file)
        print("Debug[3] Created register reference")
    else:
        print("Debug[1] Register File Not Found")
        register_file = open("accreg.dat","w")
        print("Debug[2] File created & assigned to register_file")
        load_reg()

>>59742334
thanks

>>59742219
>both have excellent pipelines to alcoholism and suicide
this

File: Screen Shot 2017-04-04 at 3.38.52 PM.png (37KB, 1232x164px) Image search: [Google]

37KB, 1232x164px

struggling with SICP
What am I doing wrong? I get the following error when I compile the code with clisp:

(defun cc (amount kinds-of-coins)
    (cond ( (= amount 0) 1) 
          ((or (< amount 0) (= kinds-of-coins 0)) 0)
          (else (+ cc amount (- kinds-of-coins 1)
                  (cc (- amount (first-denomination kinds-of-coins))
                kinds-of-coins)))))

(defun first-denomination (kinds-of-coins)
    (cond ((= kinds-of-coins 1) 1)
          ((= kinds-of-coins 2) 5)
          ((= kinds-of-coins 3) 10)
          ((= kinds-of-coins 4) 25)
          ((= kinds-of-coins 5) 50))))

(defun count-change (amount)
  (cc amount 5) )
(print (cc 100 5))

*** - READ from #<INPUT BUFFERED FILE-STREAM CHARACTER #P"coins.lisp" @13>: an object
cannot start with #\)

>>59742393
http://stackoverflow.com/questions/14302248/dictionary-update-sequence-element-0-has-length-3-2-is-required

>>59742408
>>59742408
I know why this error happens but I can't find a reason for it to happen in my code, i mean the dat file is a pure dictionary without any tipos

>>59742403
Nevermind... It was an extra closing parenthesis.

You know, I'm loving lisp... But my goodness... Is it hard to love

>>59742393
I'd recommend reading the python docs to see what is returned by open()

https://docs.python.org/2/tutorial/inputoutput.html#reading-and-writing-files

i'm adapting a data mining algo. to be used with spark + gpu acceleration. a FIM algo. to be precisely

>>59742403
Follow up: I appear to be getting a weird error:

(defun cc (amount kinds-of-coins)
    (cond ( (= amount 0) 1) 
          ((or (< amount 0) (= kinds-of-coins 0)) 0)
          (else (+ (cc amount (- kinds-of-coins 1))
               (cc (- amount (first-denomination kinds-of-coins))
                kinds-of-coins)))))

Despite my parentheses matching the sample code, I get the following error:

*** - COND: variable ELSE has no value

Removing the ELSE seems to remove any issues, but why would including something as simple as "else" make my code not work?

File: 04045558.png (95KB, 1069x1039px) Image search: [Google]

95KB, 1069x1039px

Started playing around learning more about this language. How am I doing?

>>59742616
>why would including something as simple as "else" make my code not work?
Because COND doesn't treat the ELSE specially; it's just an element that is evaluated to get a boolean value. Simple words are evaluated by reading the variable with that name. Try using T instead of ELSE.

>>59742680
nice code anon. is it x86? what does it do?

>>59742680
Here's the program's output btw.

21, 25, 37, 47, 85, 86, 88, 88, 92, 95, 98, 127, 129, 157, 169, 194, 203, 206, 217, 234,

>>59742765
Right at the top cpu x86 bits 64.

>>59741870
Why does Ada have such a hardon for the stack?

Why
Json is throwing some errors

def load_data():
    if os.path.exists("accounts.json"):
        with open("accounts.json", "r+") as database_file:
            database = json.load(database_file)
    else:
        with open("accounts.json","w") as database_file:
            load_data()

When executed this happens

AppData\Local\Programs\Python\Python36-32\lib\json\decoder.py", line 357, in raw_decode
    raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

>>59742951
how about you google your errors?

>>59742951
open("accounts.json", "r+")

should be +r no? at least with chmod it's like this

>>59742964
I always do, but I have ran into cases that have the same error but the solution was different or not fit for my script
>>59742970
Will try this, thanks

Just finished a programming project that I have been working on for a long time, feeling completely lost now..

>>59743008
bout what?

>>59743015
that i don't have a thing im working on

>>59742999
np bruh. i think it makes sense because if your permision argument was wrong it'd probabily throw some exception like this because the json could not be read.

>>59742885
It's a real embedded systems language

>>59743028
>Just finished a programming project
ABOUT WHATTTTT

>expecting to be a good programmer while only speaking one language

what's a good tool to make minesweeper clone?

>>59743433
emacs

>>59743444
language and library?

>>59734442
I'm working on a WPF app for my dad's business to help him organize jobs and such.
I just copy/pasted an entire RelayCommand class from some MVVM sample and started using it, and I feel pretty dirty because I have no idea how it works, only that it Just Werks.

>>59743488
And that's how the literal code monkey operates.

>>59743462
emacs-lisp

>>59743521
In my defense, this is literally my first non-fizzbuzz/project euler attempt at programming. I'm learning C#, WPF, and MVVM all in the same project. Just trying to get something working, going to worry about how RelayCommand works when I'm done.

>>59741602
>>59741902
You're still having trouble with this? You posted almost the exact same shit about a week ago. You also accidentally the original function names which showed this is an AST.

Maybe you should consider learning C before writing a parser in it.

File: 1473442521136.png (21KB, 900x900px) Image search: [Google]

21KB, 900x900px

>General Requirements: Divide your source code into the following six files

Why would anyone ever use C to implement a programming language? I can understanding writing a VM in C but everything else?

>>59743711
Because it works with memory directly.

>>59743711
yacc

>>59743711
C is portable. C is efficient.
That seems like a thing that basically every language wants for their interpreter or whatever.

File: 1485890385859.jpg (65KB, 700x700px) Image search: [Google]

65KB, 700x700px

>>59743711
asking the big questions.

>>59743740
>C is portable.
Wrong.
>C is efficient.
Wrong.

Is it easy to call Rust from Lisp?

>>59743750
You're going to need to provide an actual rebuttal.

>>59743761
Just go through the C ABI.
Every language that isn't complete garbage can use it.

>>59743750
lol how this is wrong?
C can run almost anywhere and is colse to the metal, thats why it is efficient

>>59743750
C is probably the most portable language in existence.

>>59743716
Lots of languages are capable of this, and it's not that useful for implementing a language outside of writing a VM, which I already mentioned.

>>59743726
Parser combinators >>> parser generators

>>59743740
Already said VM.

>>59743761
Does rust do C++ tier name mangling? If not, I'd assume it's like any C FFI.

>>59743777
Parser combinators 'are slower than' parser generators

>>59743740
Wrong, C CAN be portable and CAN be efficient, but rarely is because of the poo in the loos that use it.

>>59743726
Every worthwhile language has an equivalent.

>>59743798
poo in loos never use C lol

>>59743750
>>59743762
>>59743771
>>59743775
I like where this is going.

>>59743798
thats a cop out of a defense you dingus give the man an answer

>>59743777
>it's not that useful for implementing a language
Well, how do you think are you going to create a garbage collector if you can't work with memory?

>>59734442
Working on an imageboard: 4kev.org
Send banners xP

>>59743857
>outside of writing a VM
>making a language with a garbage collector

>>59743865
How will you allocate memory if you can't work with memory?

>>59743894
You don't have to write your compiler in C to make it output machine code.

>>59743902
Which compilers are you even talking about?
Most "popular" languages are interpreted, so having that in C makes sense.
Most of the compilers for "native" languages are self-hosted, and written in the language they compile.

>>59743921
How does one bootstrap a compiler?

>>59743921
>Most "popular" languages are interpreted, so having that in C makes sense.
I already said that C would be okay for the VM but nothing else.

>Most of the compilers for "native" languages are self-hosted, and written in the language they compile.
You still need to bootstrap it.

>>59743933
>>59743934
A lot of the time when I hear about bootstrapping compilers, it's usually OCaml or some shit.

>>59743933
You write in C first.

>>59743944
Exactly, not C.

>>59743944
First version of OCaml compiler was written in C, by the way.

Does anybody have a problem with json.load in python?
I have the simple statement of

with open('accounts.json', "r+") as databasefile:
                database = json.load(databasefile)

but when executed it gives me this error.

raw_decode
    raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

What to do?

How was the first C compiler bootstrapped?

>>59742044
Anon, these kinds of things aren't relegated to science fiction movies because they're impossible, but because they're a huge pain in the ass.

>>59743998
Seems like there's an error in the JSON file.

>>59744025
So what do i do? Change to pickle or csv?
Json came installed with python by the way

>>59744004
It was a B compiler written in B which was slowly tweaked into the C compiler.

>>59744033
Maybe you should fix your JSON file.

>>59744004
It was written in Lisp

>>59734442
Wirting a lisp2html converter in C
I don't know lisp
I don't know html

>>59744157
what're ya smokin bruh

>>59744157
Why not just write a HTML DSL in Lisp?

>>59744182
Not him, but multiple already exist

File: sicp.jpg (40KB, 912x561px) Image search: [Google]

40KB, 912x561px

Have you watched your SICP today?

>>59744130
No, it wasn't. I don't think anywhere along the bootstrap chain was lisp ever used. I think you would get at the bottom ASM -> ALGOL -> ...

>>59744181
Just lost in some parentheses man

>>59744182
I'm trying to regrasp parsing
I wrote a terminal emulator in VB .NET years ago and I want to rewrite it in C
One step at a time

>>59744231
And what was ASM first implemented in? Lisp.

It's funny that of all the languages that have been invented since Lisp, very few manage to equal Lisp, let alone surpass it.

>>59744240
In machine code actually.

>>59744240
Machine code.

File: language-comparison.png (8KB, 748x446px) Image search: [Google]

8KB, 748x446px

Reminder

>>59744365
A python job is almost 14 times higher paying than a C job?

>>59744365
A python programmer takes almost 14 times more cocks than a C programmer?

>>59744365
Reminder that this graph is trash?

>>59744365

There are 14 times more python jobs than C jobs?

>>59744365
python sucks 14x harder

>>59744399
>>59744397
>>59744386
Excuse me?

>>59744426
Excuse you?

>>59744426
You're not excused.

>>59744426
You're excused

>>59744426
You might be excused.

File: 1472256857087.gif (2MB, 320x240px) Image search: [Google]

2MB, 320x240px

>>59744222

>>59744426
Can you repeat the question?

File: bulletto.jpg (31KB, 600x450px) Image search: [Google]

31KB, 600x450px

>mfw I am reading "Introduction to algorithms"
>mfw I have to understand more complex math to understand every algorithm
>mfw it only gets harder from here

Hold me fellas, my sanity is waning.

>>59744475
YOU'RE NOT THE BOSS OF ME NOW

I really want to learn Rust, but I just can not get into it. Have you ever had that issue? I just keep thinking why am I not just doing it in C the entire time. There's just too much syntax for me.

>>59744548
>I really want to learn Rust
why?

>>59744548
Learn Lisp.

>>59744575
I already know CL.

>>59744587
Good. Your life is complete.

I'm creating a program that automatically downloads memes from 4chan and adds them to my collection making sure I never get duplicate memes.

I might make it categorise them eventually too by board and by meme type depending on how far I take this.

I'm going to have to figure out how to avoid downloading terabytes of black cock if it watches boards like pol but I should be able to do that by checking the image title and thread title.

Currently I have 3000 memes all harvested by hand but once the program is complete I want 10x that many at least.

>>59744548
Don't bother with Rust. You'll quickly realize that it's lacking in areas that are needed to make any complex system. The entire language as a whole is also poorly thought out. Just as an example they went and gave Traits (their interface system) inheritance, but completely left out any nice way to do inheritance on structs for data forcing you to either write unsafe code (defeating the entire purpose of Rust) or come up with some complex shit system.

>>59744701
>but completely left out any nice way to do inheritance on structs
That's good. Use composition.

>>59744718
Great idea in C, but too bad you can't write it the same way you can easily do in C since it results in unsafe code.

File: cncLarge.png (285KB, 517x538px) Image search: [Google]

285KB, 517x538px

>>59734442

Building a Arduino CNC; Starting with GRBL

Any advice?

>>59744745
Can you give an example of how composition over inheritance requires unsafe code?

>first week
>productive af
>second week
>can't write a single line of code

FUCK

>>59744814
hi me

>want to learn CL
>want to learn Scheme
>want to learn Clojure
>want to learn Rust
>want to learn Haskell
>want to learn Idris
>want to learn emacs
FUCK

>>59744749
is there anything in C that doesn't require unsafe code?

>>59744749
I can't even do struct composition like I can in C with Rust. It gives compile errors if I try to cast.

>>59744899
scratch all those memes and learn C++17.

>>59744963
I already know enough C++.

>>59744899
that is some serious case of ycombinator/dpt memedocrination

>>59744980
one can never know enough c++.

>oldfart Lee got himself a Turing award
>meanwhile Fields medalists are all ~40 or something
Is CS dead? Where are dem young NP=P scientists

>>59745142
Mathematics is dead.
CS is dead.

>>59734442
Deko's skin is so smooth and creamy!

File: lel.jpg (84KB, 829x675px) Image search: [Google]

84KB, 829x675px

>tfw making pass one of an assembler in java with regex
look at this abomination

File: 1469843979355.png (231KB, 304x366px) Image search: [Google]

231KB, 304x366px

>>59745275
Okay friend do yourself a favor and split the line up to start with rather than splitting it on every if case.


Secondly look into lex & yacc, Java has their own version of both out there on the web somewhere and it will make your assembler a cake walk.

File: 1439358089166.gif (606KB, 500x257px) Image search: [Google]

606KB, 500x257px

>>59745330
too late for that, assignment due in an hour
still have to write the docs

>>59744899
How about you learn something useful and learn about opencl/cuda instead?

>>59745351
Anon...

File: d0b.png (572KB, 600x681px) Image search: [Google]

572KB, 600x681px

>>59744478
If you dont make it you can always join poos writing java and node cruds at your local indian bowel movement

File: scren.png (32KB, 633x299px) Image search: [Google]

32KB, 633x299px

Why is pic related returning 1 more than it should? It is printing 232169 when the correct answer is 232168

>>59744905
So no?

>>59745401

if __name__ == '__main__':
    main()

What the fuck is this?

>>59745388
Never, everything but that. My sanity tis' just but a small price to pay, so that I don't join the poos in the loos.

>>59745275
bruh why 1 pass....
if it's mips you are gonna have to do 2 pass.

>>59745415
second pass is due next week

>>59745413
Python

File: 1469583171814.png (350KB, 621x393px) Image search: [Google]

350KB, 621x393px

>parsing assembly with regex in Java

>>59745403
Can you tell me how to do the following in Rust because everything I try it apparently not allowed.

struct Object {
    int type;
    int x, y;
};

struct Rect {
    struct Object obj;
    int w, h;
};


void f(struct Object *o)
{
    printf("Object of type %d @ %d, %d\n", o->type, o->x, o->y);
    if(o.type == 1)  
        g((struct Rect *)o);   //specifically this
}

void g(struct Rect *r)
{
    printf("Rect of width %d and height %d\n", r->w, r->h);
}

and likewise going in the opposite direction in C is as simple as casting to (struct Object *), but that's also not possible in Rust and instead you simply have to pass the r->obj which is fine I guess, but going up the chain is impossible it seems.

>>59744222
>>59744472
This video is too good to be something people glance over as a meme.

>>59745469
Why would you design your program like that, even in C?

>>59745469
Fucking retard. Kill yourself

>>59745413
Python programs entry point is the top of the file. This is a way of getting a entry point that's more familiar to programmers of other languages.
Containing the entry point in a function tends to be easier to follow because of how people generally structure programs.

>>59745556
It's just an example and valid C code. It may be a weird example and I'm sure there are systems that can show the usage far better but take that as you will.

>>59745557
Kill yourself edge lord. Now tell me how to do it in Rust.

>>59745559
And people consider Python a serious language?

>>59745469
Downcasting is bad. For this specific problem you should use a tagged union:

struct Object {
  x: i32,
  y: i32,
  body: Type
}

struct Rect {
  w: i32,
  h: i32
}

enum Type {
  Rect(Rect)
  // more variants go here
}

fn f(o: &Object) {
  println!("Object @ {}, {}", o.x, o.y);
  match o.body {
    &Type::Rect(ref r) => g(r)
  }
}

fn g(r: &Rect) {
  println!("Rect of width {} and height {}", r.w, r.h);
}

>>59745469
>>59745624
Note that this code can be compressed a lot because it is a 1:1 translation of your code and not idiomatic in the slightest.

>>59745401

Is the correct answer 232168 or is it 233168? Try

for i in range(3, ceiling):

instead.

>>59745596
Python is a scripting language. It isn't meant for more complex programs but when people insist on using it for more complex programs, retardation happens.

>>59745624
>>59745696
I suppose this could work but it just seems like a weird system, especially if it gets more complex with the compositions. And all the logic would need to be focused on going up rather than down or bi-directional (which doesn't seem possible anymore since once you're a Rect you have no idea of your Object data x,y)

>>59745590
What you've done is fucking retarded and unsafe as fuck. I would not even do that in C
Rust enums are just tagged unions, which should allow you to do what you want to do safely.

>>59745596
I do find it unnecessary. But for some reason people who do scripting find it onerous to write a main function for your program I suppose. I don't think anyone is realistically having that as an issue. But it's a common feature of scripting languages.
That you can get a more normal entry point in a way that's as easily recognizable as this is great though.

>>59745753
What I'm doing is basic concepts in OOP and safe in a lot of languages and C if you are careful and have a way to check for it to ensure you're not fucking up a cast to something that shouldn't exist.

>>59745753
>unsafe
I'm not a rust memer. What's the issue here really?

>>59745753
Consider the following

struct Widget {
    x: u32,
    y: u32,
    //various other shit that defines the basic widget
    body: WidgetType
}

struct Label {
    text: String
}

enum WidgetType {
    Label(Label)
    // many more advanced widget variants for a gui
}


fn draw(widget: &Widget) {
    math widget.body {
        &Type::Label(ref r) => draw_label(r);
    }
}

fn draw_label(lbl: &Label) {
    //Draw our text to screen, but wait
    //now we only have our 'text' data
    //and can't get our 'x', or 'y'
}

A very basic widget gui system following your composition method in Rust. Do you see my issue right away? Are you basically telling me I'll have to not only pass a reference to my Label type but also a reference to the Widget itself in order to make draw_label draw its text properly on screen where the Widget is located?

I feel like if we had more layers of a widget to build some complex Widget it's just going to be an annoying hassle.

>>59745875
in that case why don't you just compose the structs downwards?

>>59745733
>>59745783
That's just the tagged union method. If you really want an inheritance hierarchy, you use traits.

trait Objectish {
  fn x(&self) -> i32;
  fn y(&self) -> i32;

  fn g(&self);
}

struct Object {
  x: i32,
  y: i32
}

impl Object {
  fn f(&self) {
    println!("Object @ {}, {}", self.x, self.y);
    self.g();
  }
}

impl Objectish for Object {
  fn x(&self) -> i32 { self.x }
  fn y(&self) -> i32 { self.y }

  fn g(&self) {}
}


struct Rect {
  obj: Obj,
  w: i32,
  h: i32
}

impl Objectish for Object {
  fn x(&self) -> i32 { self.obj.x }
  fn y(&self) -> i32 { self.obj.y }

  fn g(&self) {
    println!("Rect of width {} and height {}", self.w, self.h");
  }
}

There's boilerplate involved, same as there would be were you trying to do it without inheritance in any language. You also only have single inheritance unless you do some generics tricks.

What you should ask yourself is "do I really need this?" because inheritance is well known to be bad for other reasons. I've never not found a way to do something without it, and I'm always happier without it.

>>59745875
Yeah, in practice you would compose the other way, which better matches how inheritance works in other languages. The tagged union is a possible approach even in an object-oriented language, if you want to trade an open system for a closed one.

How do I declare a variable in lisp?
I want to do the following: Given values for a1, a2, and a3,

 temp = (a1 + 2*a2 + 3*a3)

I read that I should either do a "(define)" statement or a "(let)" statement, but when I do:

    (let (temp (+ a1 (* 2 a2) (* 3 a3) ))) 

Clisp yells at me saying "variable LET has no value", even though I think LET is pretty valuable... What am I doing wrong?

>>59745946
Here's another way to do it:

struct Widget {
  x: u32,
  y: u32,
  body: Box<WidgetBody>
}

trait WidgetBody {
  fn foo(&self);
}


struct Label(String);

impl WidgetBody for Label {
  fn foo(&self) {
    println!("Label: {}", self.0);
  }
}

>>59745946
>>59746021
Forgot a bit:

impl Widget {
  fn bar(&self) {
    println!("Widget @ {}, {}", self.x, self.y);
    self.body.foo();
  }
}

>>59745946
>inheritance is well know to be bad
How so?

>>59745910
>>59745946
>>59746045>>59746045


I think I get it now. Whether I compose my struct bottom up or top down I can always just use a tagged enum as an argument to a function if there's a possibly that I need to go upward anyway like:

struct Object {
    x: i32,
    y: i32,
}

struct Rect {
    obj: Object,
    w: i32,
    h: i32
}

enum Type {
    Object(Object),
    Rect(Rect)
    // more variants go here
}

//going back to the Object/Rect example but top-down
fn print(obj: &Type) {
    match obj {
        &Type::Object(ref r) => println!("Object @ {}, {}", r.x, r.y),
        &Type::Rect(ref r) => {
            println!("Object @ {}, {}", r.obj.x, r.obj.y);
            println!("Rect of width {} and height {}", r.w, r.h);
        }
    }
}

would work right?

>>59746078
You could also do it that way, I suppose. I'd never thought of it before but you could define the entire inheritance hierarchy as a closed system with a recursive algebraic data type, exactly as you would define any tree type. That's pretty clever actually.

You'd tweak the function a bit to use recursion to call up the tree:

fn print(obj: &Type) {
    match obj {
        &Type::Object(ref r) => println!("Object @ {}, {}", r.x, r.y),
        &Type::Rect(ref r) => {
            r.obj.print(); // changed from a duplicate of the first plane
            println!("Rect of width {} and height {}", r.w, r.h);
        }
    }
}

>>59746049
https://www.quora.com/Is-inheritance-bad-practice-in-OOP

>>59746104
>>59746078
Wait, never mind this. I missed something, and you can have this or the other thing and not both.

Learning lisp, might make a website with it to host some APIs, also maybe a game.

>>59745469
Rust is teaching you good practices actually.

>>59746104
I really didn't even think of the whole Enum thing as unions at first when reading their documentation and examples. Thanks a lot anon.

>>59746128
Yeah it would only be usable like that in certain conditions since you can't go back up I guess. Either way I think here's lots of ways to tackle it and it just comes down to what's needed during the project and coming up for a better idea to begin with or designing with that in mind.

Also what is all the fuss about Borrowing being hard and a huge roadblock for rust? It's just a matter of knowing when to pass something to a func/var as a reference rather than non-& in order to not give ownership away prematurely, right?

Thinking about array primitives for my dependently typed system language. Arrays are dependent, i.e. the type of each element can depend on the index, but the types have to be able to be compiled as if they're in a union (to keep the array homogeneous). Because of linearity in the type system, arrays can be mutated and also change their type (strong update).

Pseudocode:

Array : (Nat -> Type) -> Nat -> Type

// U is the equivalent of size_t
write : Array F n -> (i : U) -> 

>>59746222
>Also what is all the fuss about Borrowing being hard and a huge roadblock for rust? It's just a matter of knowing when to pass something to a func/var as a reference rather than non-& in order to not give ownership away prematurely, right?
The difficulty arises when you have moving, unique borrowing (&mut), and shared borrowing (&) happening together with non-linear control flow, since they're all mutually exclusive. You need to prove to the compiler that it all works using let statements, scope brackets, etc.

>>59746114
I've found the quality of answers on qoura to be incredibly poor most of the time.
I don't find this as an exception.
It's basically
>person who coined OO doesn't agree its important
>>Childnaturally relies on some behavior fromGrandParent, but in reality, it should only beawareof theParentclass.
But the Child only relies on the parent. It just happens that Parent is defined in terms of grandparent. What's the problem there?
Apparently nothing as the tangent is not even addressed. It's deemed self evident that inheritance in multiple steps is bad.
>In practice, the more you inherit, the more behavior you’re pulling in which you may not need and it’s hard to be sure exactly where that behavior is coming from.
Sure. You have to have tools for that realistically IDEs often do this. At worst this is a hassle for editing reasons.

A method foo() is established. First parent wants it. Later child. And grandparent made one too. Odd order of change here but sure. It's not an issue for inheritance.
>>The Child class, meanwhile, realizes they needGrandParent.foo()and notParent.foo(), but whether or not they’re able to call that method directly depends on the language.
OK so we have languages that are dealing with inheritance poorly. Where's the inheritance complaint? It's an incredibly basic problem you run into which has simple solutions.
>>if you call grandparent.foo() that violates encapsulation
Yes. But you've explicitly structured the code in a way to violate encapsulation. You've required we call the grandparent then claim its breaking encapsulation because of the requirement you put up.
To any reasonable person this would be viewed as a code duplication problem. Not an encapsulation problem. Because no other system can achieve the same goals.

It's really not worth it to keep reading this.
I don't even like OOP but this is not its issues. And it's not the issues I find with inheritance.

File: lel.jpg (105KB, 815x817px) Image search: [Google]

105KB, 815x817px

finished
professor better give me full points for this shit

>>59746380
I don't even know how you would rate a task like this. Unless he has given some metrics for you to judge (performance, code malleability etc) it seems fairly arbitrary how you judge it. If it works.

>>59746380
Why not a switch instead of a line of if elses?

>>59746410
real programmers don't use switches

File: $.gif (1002KB, 250x251px) Image search: [Google]

1002KB, 250x251px

>>59746421

Anyone have that program roulette pic? I need a list of programs to go through to help me learn this language I'm using.

File: bst.png (36KB, 867x406px) Image search: [Google]

36KB, 867x406px

I feel full on fucking retarded because I can't figure my assignment out for a binary search tree.

https://pastebin.com/CAC7Pdxm

is the given BST.

i've done a dozen other assignments over 3 semesters and for some reason my brain does not get this.

any steps in the right direction would be lovely, even though the instructions say what to recycle..

>>59746490
nigga this is like retard shit, how did you do the insert assignment and not know how to do find?

find is literally insert except not inserting anything and just checking the node until it's determined no such value can exist after expanding the tree branch.

>>59746490
use recursion

>>59746490
1. check if tree is empty, print tree is empty if it is, return
2. visit all the nodes, check to see if key matches
3. print result if found, print not found if not found, return
there are youtube videos on this shit
perhaps your problem is understanding pointers

>>59746574
>2. visit all the nodes, check to see if key matches
Pretty sure you get 0 points for visiting all the nodes of a search tree, unless it has one node.

>>59746490
Do you understand how a binary tree is laid out? If you do this should be extremely simple.

>>59746490
>pastebin has 61 views
what's with all these lurking fags

>>59746709
every pastebin has >60 views, those are bots scanning the new paste list.

File: idontlikethis.png (99KB, 1082x1079px) Image search: [Google]

99KB, 1082x1079px

holy shit graphics programming sucks

I feel retarded working on this one problem and I'm hating myself for it.

Is prime searching memory bound?

I can write a sieve with a bunch of optimizations that can go up to 40 billion but it eats up 1.5GB memory, and it only grows from there.

If I started writing the lookup table to disk, it would kill performance, but I could probably go much further.
Is this how large primes are found?

File: gaisdmcdlas.png (29KB, 635x521px) Image search: [Google]

29KB, 635x521px

Please help me become less retarded. How do I fix "cannot find symbol" error?

>>59746846
Is Store defined?

File: store.png (5KB, 597x233px) Image search: [Google]

5KB, 597x233px

>>59746853
I have this within the Assignment class.

>>59746761
why in the world are you doing graphics via asm? Is this for a special embedded software?

>>59746846
compile from the command line instead of using an ide

>>59746255
Fuck, I always get distracted with another reply and forget to finish my first thought.

// Nat is the type-level unbounded unsigned integer
// value-level unsigned integers coerce to it when depended on
// not sure how to specify that the (Nat => Type) family must be a union yet
Array : (Nat => Type) => Nat => Type

// U is the equivalent of size_t
// i must be less than n, obviously this is good to statically verify
write : Array F n => (i : U | i < n) => A => Array (\j -> if (j == i) then A else (F j)) n


// would be used when implementing an array list
VecBuf : Type => (cap : Nat) => (len : Nat | len <= cap) => Type
VecBuf A cap len = Array (\i -> if (i < len) then A else Unit) cap

// relies on the fact that (if (i < len + 1) then A else Unit) is equivalent to (if (i == len) then A else (if (i < len) then A else Unit)))
// this can be automatically proven, more advanced formulae could be asserted at runtime or proven using tactics or something
push : (VecBuf A cap len | len < cap) => A => VecBuf A cap (len + 1)
push buf a = write buf len a


// struct
// (len <= cap) is automatically enforced from "VecBuf"
Vec : Type => Type
Vec = {
  cap : U,
  len : U,
  p-buf : Box<VecBuf A cap len>
}

// assume this increases the capacity of the vector when it's full like usual
resize : Vec A => Vec A

// these are both true, again automatically proven using the definition of "resize"
theorem-1 : (vec : Vec A) => (vec.len = (resize vec.len))
theorem-2 : (vec : Vec A) => (vec.len < (resize vec).cap) // crucial in order for "push" to work

push : Vec A => A => Vec A
push vec a = do {
  vec' <- let (resize vec); // let might look a bit funky, but it's for a good reason
  vec' { p-buf = map vec'.p-buf (\buf -> write buf vec.len a ) } // nice mutation
}

>>59746889
>>59746846
Since you defined them within another class your correct symbol is actually:

// package     class         inner-class
Assignment.Assignment.Officer
Assignment.Assignment.Store

Also you made them non-static inner classes so you need a reference to a object that's of type Assignment in order to make Assignment.Officer/Store.

hey thinking of switching to EE for that semi-comfy lifestyle

I know jack shit about coding, will the school teach me what I'll need to know to get an EE job? If no, what extra?

>>59746966
>EE
>Semi comfy lifestyle
You must be fucking retarded. What are you switching from? Please don't say CompSci

File: 1469656490556.png (14KB, 166x166px) Image search: [Google]

14KB, 166x166px

https://lists.torproject.org/pipermail/tor-dev/2017-March/012088.html

They are actually doing it. They are rewriting Tor in Rust.

>>59746966
Are you even good with electricity?

>>59746993
They should wait for my language and reimplement with it.

>>59746993
That would make a good advertisment for Rust.

>>59746993
Reminder that using Tor puts you on a FBI list and makes you a suspicious person

>>59746989
biochemistry

>>59747001
meh, I'm good with numbers and drawing but I don't have any passion beside drawing porn, so I wanna choose something semi-comfy that pays a little better than the biochem bsc 30k/yr while allowing me to draw porn on the side

>>59746789
I think the bounding factor is that you have to store somewhere a list of primes you've found so far. You could use a fixed size sieve like a moving window, but the list of primes is still going to grow.

Large primes are found with some sort of abracadabra mathemagic fuckery.

>>59747029
>implying using 4chan doesn't put you on FBI list

>>59746989
w-why, anon?

>>59747032
EE isn't easy. I am not sure about biochemistry but I guarantee you that EE as a major can be pretty fucking bullshit as well. Post grad you have a lot of work options but none of them will be comfy. You can't name me one junior engineer that has a comfy life.

>>59747032
Well, if you're good with physics and calculus, go for it. Even then, you'll put a bit of effort into it. So I wouldn't really consider it "comfy."

>>59747032
math/bio teacher

>>59747048
Engineer courses are consistently ranked as one of the hardest majors year after year and Electrical Engineer is probably one of the hardest among all the engineering majors. Why do people still ask this question?

>>59747056
>You can't name me one junior engineer that has a comfy life.
My friend pulled it off in civil engineering. Most of them work +60h/week though.

>>59746993
So Rust is finally getting it's flagship software like Go and C/++ have.

>>59747097
>+60 h/week
>Comfy
What? Yeah civil engineering is one of the "easier" engineering discipline. Not that any of them are actually easy but I would consider EE, ME and CE the "harder" ones. Just don't touch ChemE if you want a job.

>>59746993
All this means that FBI is going to find exploits in Rust and Rust is not safe anymore.

>>59747119
I wonder how much the US gov is secretly paying them for this.

I'm taking input from a user that will either be their age or date of birth.

How would I best check for which one they entered through the program? I'm using c#

>>59747128
Probably nothing, since this is bad for the US Gov, less exploits for them

>>59747130
Don't leave that choice to the user.

>>59747032
It's not really something you just do on the side.
HOWEVER, I think engineering students are up their own ass about how hard it is. I'm a comp e major and my ee classes are easier than my cs classes. I think this is because I'm used to doing lots of math, but not programming. The biggest time sink is labs. My school is ABET certified so they're religious about labs.
Also ee jobs are super competitve, more and more majors, less and less jobs.

>>59746935
it's for msdos, where graphics drivers don't exist. so I need to write my own.

New thread:

>>59747138
>>59747138
>>59747138

>>59747130
When programming for user input always assume that the user is a fucking moron

>>59747133
>>59747127
Rust is still very immature, it probably has bugs hidden away.

>>59747120
I said most of them have to do this amount of work, but if you want to just do your 9 to 5, then it's possible in CE. No crazy salaries though.

>>59747141
>>59747149
the dpt roll said us either 1.
Do I just go with the one that's easier to work with?
Or do I make it harder on myself because I'm learning?

>>59747141
>mfw I'm 40499 years old

>>59747159
CE has on site work tho. I won't consider that comfy at all.

>>59747166
If the input is has 3 or more digit, assume it is date.
If the input is lesser than 2 digits, assume it is age.

I am sure you can come up with actual ways to check the validity of both

>>59747172
yes, but if you go through it to get the papers, you can get a gov job stamping papers and be forgotten in some office for the rest of your life

>>59747184
What if I'm 100 years old?

>>59734442
is that an actual photo