/dpt/ - Daily Programming Thread

>>51526469
First for D

gotta GO fast

Image upload processing for a server I'm writing.

(defn handle-upload [{:keys [filename] :as file}]
  (let [filename-escaped (clojure.string/replace filename #" " "_")]
    (println file)
    (upload-page
      (if (empty? filename)
        {:errors "Please select a file to upload."}
        (if (.endsWith (clojure.string/lower-case filename) "gif")
          {:errors "GIF uploads are not supported."}
          (try
            (println (str (gallery-path) File/separator filename-escaped))
            (upload-file (gallery-path) (assoc file :filename filename-escaped))
            (save-thumbnail (assoc file :filename filename-escaped))
            (db/add-image ((session/get :user) :user_id) (gallery-path) filename-escaped)
            {:image (thumb-uri ((session/get :user) :uname) filename-escaped)}
            (catch Exception ex
              ;(error ex "an error has occured while uploading" name)
              {:errors (str "error uploading file " (.getMessage ex))})))))))

>>51526469
for a mudslime, both are pretty kawaii.

>>51526469
Trying to figure out if the latest change in my program caused a bug or just hit a case of bad luck.

File: 1406339541350.jpg (38KB, 300x232px) Image search: [Google]

38KB, 300x232px

>>51526469
My little nigger can't be this cute!

Procedural or OOP /g/?

>>51526650
A combination of both
I like to call it POOP

>>51526650
OOP

>>51526650

functional

you'll thank me in five years

>>51526650
Procedural

>>51526650
idiot

>>51526650
multi paradigm

>>51526650

OOP, obviously. You can avoid the OOP if you really want to.

so is /dpt/ making a game

>>51526650
OOP for the ultimate agile enterprise event-driven non-blocking I/O scalable pattern driven cattle methods.

File: 1405024184572.jpg (175KB, 1044x766px) Image search: [Google]

175KB, 1044x766px

>>51526469

yeah i know mohammed.

the cause of all world problems right now.

>>51526469
>>51526622
>>51526779
>>51526805
>>51526837
>programming

we /pol/ now

>>51526862
I like it

File: 1446828557138.png (157KB, 298x390px) Image search: [Google]

157KB, 298x390px

Want to talk about computers?

>What are you working on, /g/?

El Gamal Encryption using the discrete logarithm problem

>>51526868
kikes

>>51526862
but /pol/ hates muslims. right now, they hope russia invades turkey and removes kebab.

russia saviour of white people

File: 4560228202533a.jpg (35KB, 480x720px) Image search: [Google]

35KB, 480x720px

>>51526868
Yes =3
I am working on my Lisp interpreter in C, how about you?

>>51526887
And they are right.

>>51526905
>slavs
>white

putin save europe

>>51526931
he said Russia was the savior of white people, not that they were white. and hes right, Russia will invade turkey and save the white race.

>>51526887
/pol/ loves muslims because god is great

OP, delete this thread and start over.
Please.

>>51526930
Working on my optimizations skills, playing around with SIMD stuff (SSE,AVX etc) and checking out what works well on which processors. It's kinda fun but a hassle since you have to drag around hardware.

>>51526931
They may not be pure blood r1b bros but they are somewhat white.

More than white enough for me at least.

>>51526469
Started writing a Kernel running in 16 bit real mode for an old laptop I had lying around.

File: r1b.png (176KB, 2133x1245px) Image search: [Google]

176KB, 2133x1245px

>>51526991
>r1b
>white
Benin is white

File: islam peace 2.jpg (235KB, 635x875px) Image search: [Google]

235KB, 635x875px

>>51526469

/g/ should make their own game

>>>/pol/
Sage

>>51527103
About what?
Will everyone agree to use C?
2d? 3d?
Will it have jews in it?
What license?
lol good luck

How do you guys deal with knowing there is a more efficient way to do what you just wrote?
It really bugs me and I spend hours making it perfect.
What do you think, /g/?

>>51527143
dubs decide

>>51527103
great idea, i'll make the logo for it

>>51527143
>Will everyone agree to use C?

Clearly not.

>>51527170
I like 30min-hours on perfecting my code

>>51527103
>>51527143
what genre? top down shooter? platformer? RPG? JRPG like final fantasy? WRPG like wizardry?

>>51527189
post logo, inspire others
>one step at a time

>>51527192
Nobody cares about you, retard.

File: 1447355869328.png (847KB, 1200x978px) Image search: [Google]

847KB, 1200x978px

Lets make a game about Jews and write it in common lisp

>>51527214
what's the difference between wrpg and rpg where you can play as a wizard

>>51527227

In the same way that nobody cares about C, except for the contrarian mega-weenies on /dpt/.

>>51527230
okay
>dubs decide

>>51527243
No, really, not even one person, not even your mother cares about you. Leave retard.

File: 1439500439356.jpg (334KB, 800x800px) Image search: [Google]

334KB, 800x800px

>>51527243
>tripfag calls other people mega-weenies

>>51527103

We would never agree on the language and libraries to use, or even what the game should be called. Even if these factors were agreed upon, there still remains the fact that most of us aren't creative types. We might be able to write code really well, but a game requires more than that. It requires art, music, and in many cases, decent story writing.

In any case, if you want to do game development yourself, you would be better suited going to AGDG on /vg/. If you need help with some algorithmic problem, or fixing the bugs in your code, /dpt/ might be more applicable.

File: 1448347168344.png (418KB, 480x650px) Image search: [Google]

418KB, 480x650px

>>51527230
But Scheme is better.

>>51527278

It takes one to know one, and being a mega-weenie myself provides that insight.

>>51527265
>kill urself haha!

Gud one.

>>51527286
Who the fuck are you quoting retard?

>>51527243
>except for the contrarian mega-weenies on /dpt/.

so, everyone?

>>51526659
kekd

nah, some people don't push on just One language. Also, for the creative people, I've heard from more artists than coders

>>51527304

Who are you? Huh? Huh? Huh!?

File: gay_lambda.gif (8KB, 648x432px) Image search: [Google]

8KB, 648x432px

>>51526469
I'm building a Lisp, called Valutron. I am implementing its compiler-interpreter in Objective-C.

It has some unique features: first, it offers what are called V-expressions, an alternative notation for writing code. Second, it offers a comprehensive object system inspired mainly by that of SmallTalk-80, one of the earliest Object-Oriented languages, and by Common Lisp's CLOS, Common Lisp being the first object-oriented language to be standardised.

This object model is implemented atop the Objective-C runtime, which I have extended with support for Double Dispatch to enable the availability of CLOS' characteristics multimethods.

In the previous thread I received some criticism. Here are my responses:

>this is bullshit, Obj-C is worthless without Apple's foundation libraries, GNU-Step is a joke. the only reason that Apple got away with using Obj-C for so long is that Obj-C is a strict superset of C, so pretty much all the low level Foundation libraries are pure C. Obj-C is slow, unsafe, and unsuitable for any modern programming. It is not really even its own language, all Obj-C syntax gets translated into C and is compiled as C.

There are multiple class libraries for Objective-C. GNUstep is not a joke, it is actually very competent. Other foundation libraries include the Etoile GNUstep extensions, ObjFW, and my own in JXobjC (derived from the portable object compiler). And as I maintain an ObjC compiler, runtime, and class library, I'm well aware it's translated into C by my compiler. Not so for gcc and clang. Doesn't make it less of a language - you can compile Scheme or C++ to C as well. Or, conceivably, any language.

>>51526031
>Even Apple realised it's shit and dropped all that dynamic typing shit in Swift.
A stupid decision. Alan Kay was very clear: the keypoint of OOP for him was messaging, i.e. late binding, which is closely associated to dynamic typing.

>>51527344
Faggot

>>51527103
>>51527143
>making a game in C
LEL

>>51527344
dynamic typing = shit

even at apple, python, racket and clojure know it

>>51527344
You were enough of a faggot as is, but then
>objective-C
200% dropped!

>>51527430
this

>>51527419

what language should a game be made in

>>51527457
C++

>>51527457

Valutron.

>>51527457
D

>>51527457
C

>>51527459
k, I used C++, not C

How do you keep SJWs out of your open source projects /g/?

File: 1445427057249.png (87KB, 853x621px) Image search: [Google]

87KB, 853x621px

>>51527419
>tfw making a game in C

>>51527460
Nobody cares, faggot.

>>51527474
make them /pol/-tier

>>51527474
By calling them as "Free software" instead of "open source".
And by not caring if they are sjws or not, contributors are contributors, as long as they don't mix their shit into my projects.

>>51527474
By not using Github for repository hosting.

File: C++vsObjC.jpg (62KB, 728x546px) Image search: [Google]

62KB, 728x546px

>>51527430
I'd better go inform the scientific discipline of programming language theory that they can stop writing papers on typing, because this anon has with just two lines annihilated dynamic typing.

>>51527436
Such constructive contributions.

>>51527457
ML

>>51527484
I really like this image.

>>51527474
Anti semetic remarks

>>51527515
are we throwing most niche languages

>>51527507
BS image

>>51527457
C++

File: typing.png (46KB, 603x486px) Image search: [Google]

46KB, 603x486px

>>51527507

>>51527457
Jai

File: 1446032198208.gif (938KB, 500x281px) Image search: [Google]

938KB, 500x281px

What is the cutest programming language?

>>51527507
>because this anon has with just two lines annihilated dynamic typing.
That happened a long time ago. Fuck, even the vast majority of Obc-C code is statically typed (which is optional in obj-c), since everyone figured out pretty quick that code is far more bug prone when it's dynamically typed, and dynamic typing has more or less no advantage whatsoever. Again, this is why Apple dumped it completely with Swift, like everyone else did decades ago.

The only big advantage to dynamic typing is that it makes compilers and run times much less work to write. Thus JS and python being dynamically typed. That's pretty much the only reason.

that's three voices for C++

>>51527344
do you have any proof of what you're doing or are you just trolling?

>>51527564
C

Generating a random 1024 bit prime number using miller rabin to verify primaility seems to vary wildly in the amount of time it takes. It can be anywhere from 0.21 seconds to 10.46 seconds. Either ruby is just being slow as shit, or I need better algorithms for finding large primes.

>>51527457
All the major studios use C++, it's got the right balance of high level features and performance (due to it's C heritage).

>>51527586
Here's your reply.

>>51527586
Does Ruby have a good C FFI?
What I often do is write my higher order logic in Python and do maths in C, it's pretty nice.

looks like the language of /g/game should be C++

File: 1448393099788.jpg (98KB, 589x733px) Image search: [Google]

98KB, 589x733px

>>51527564
c: ofc!

>>51527564
BASIC

>>51527565
The main advantage is that you have full-spectrum generic programming by default. That is the true only advantage of dynamic typing and why automatically-inferred statically-typed languages are best when supported by an exclusive-union type.

>>51527586
>Not just torrenting your primes

>>51527607
Everyone uses C++ for vidya, it's already the de facto standard.

>>51527606

Ruby does have a good C FFI. But my assignment is due Sunday, and I can't be fucked to dip down into C if I need something done fast.

>>51527637
>generic programming by default
How is that an advantage?

>exclusive-union type.
What do you mean by union type in this case?

>>51527662
RIP.
If it makes you feel any better I'm in my computing lab at uni working on the assignment I'm due to present tomorrow.

>>51527490

Nice post. :)

>>51527643

I have a hard requirement to use (and implement myself) miller rabin for this assignment. I also have to implement modular exponentiation by squaring myself.

>>51527652
yh, that's my experience for sources of games I found too
>unless we make a game in Javascript/Html but that would be pretty simple

>>51526650
OOProcedural

>>51527669
>How is that an advantage?
He means you don't need templates since dynamic typing generally implies duck typing.

>What do you mean by union type in this case?
Discriminated union, algebraic data type, etc. that you pattern match over. Like "data" in Haskell or "enum" in Rust.

>>51527679

Well, in all fairness, I don't think my professor's going to give too much of a shit about the performance, so much as the correctness. His suggested language for this was Python (this was not a hard requirement, fortunately).

>>51526873
>El Gamal
wasn't he muslim? why are you using a terrorist encryption method?

>>51527721
>dynamic typing generally implies duck typing.
you can do that with static typed languages if you like to. Just pass a string keyed dictionary of objects.

>Discriminated union, algebraic data type, etc. that you pattern match over
matching types is no easier in static typing than in dynamic, regardless of whether or not you have union types.

>>51527457
ruby

>>51527474
make an anti-CoC

File: 1407206320097.jpg (131KB, 824x1100px) Image search: [Google]

131KB, 824x1100px

>>51527771
>this entire post
I don't think you know what duck typing or pattern matching are.

>>51527746

https://en.wikipedia.org/wiki/Taher_Elgamal

>Elgamal is a recipient of the RSA Conference 2009 Lifetime Achievement Award,[2] and he is recognized as the "father of SSL,"[3] the Internet security standard Secure Sockets Layer.

Well it's not like he's anybody important or anything...

>>51527809
and I bet he joined ISIS so that he could help them atack france and other western countries
stop being a traitor to your white skin

Daily reminder that objects are a single data layout with multiple interface functions and closures are multiple data layouts with a single interface function.

If you want to fight over which is better you might as well fight over what color is 'better'

>>51527842
1st for blue

>>51527669
>How is that an advantage?
I won't lower myself to answering that.

>What do you mean by union type in this case?
I mean a type-operator which builds a type that unifies with its components. Exclusive means it's only one-way.

>>51527855
>not purple
P L E B
L
E
B

>>51527855
Red is lower frequency so it shows up better on film and it also generates more heat faggot.

>>51527931
>>51527937
blue blood = royal, noble blood

>>51527832

Why would the CTO of security of a company worth 50 billion dollars want anything to do with a group of nutheads like ISIS?

>>51527024
is not what it says.

Working on an HTML5 chatroom. Beta here:

http://raskie.com:85/

Been in here a few times looking for testers, but think I've cleared most of the bugs and vulnerabilities. Suggestions and bug reports are still welcome, nontheless.

>>51527957
>>51527937
>>51527931
>>51527855


There is a reason our eyes see green the best and everything natural is green.

Checkmate keks

>>51527804
I know exactly what they are.
Duck typing in dynamic languages basically means not needing to define a data structure type definition somewhere to use it in your code. This works because these languages just have object types, that are basically just string keyed dictionaries that contain arbitrary values. So in JS you can write:

var person = getPerson();
console.log(person.firstName);
console.log(person.lastName);

person is an object in this case, and person.firstName simply checks if the object contains a key "firstName", and then returns it's value. This works because it's just a string dictionary and nothing else. It's called duck typing because this code will always work okay so long as getPerson always returns a object with those two fields in it. It doesn't matter if it has other shit in it. So basically just has to "look like a duck, quack like a duck".

So yeah, if you're in a statically typed language and for some reason you really need to use a dynamic object like this (to add arbitrary fields at run time maybe), you can use a string keyed dictionary of objects that will behave the same. Like this C#:

var person = getPerson()
Console.writeLine(person["firstName"] as string);
Console.writeLine(person["lastName"] as string);

effectively the same thing. This is done pretty often when building dynamic data trees for example.

Union types are types that can be represent multiple other types. Pattern match works well with it because it's a very neat way of checking what type a given union type object is and saying what to do in the case of each possible sub type. Generally only functional languages support this. Imperative languages generally use

if (person instanceof Lady) {
    //blah
} else if (person instanceof Fella) {
    //jah
}

And you need to write the exact same sort of code in both static and dynamic languages.

>>51528006
it's nice. html5 nice

>>51527957
purple is the imperial colour
eg purpleborn

>>51528033
You have to be 18 or older to post on 4chan.org.

>>51528053
I'm 29

File: shapes.gif (8KB, 640x480px) Image search: [Google]

8KB, 640x480px

>>51527564
LOGO

>>51528010
Fuck off, primate. There's a reason why humans have evolved beyond living in mud huts in the forest. Green worst color!
Purple and green go well together because purple is the opposite of green: it's the color of PROGRESS!

>>51527975
because all muslism are terrorists. and by transitivity, you are supporting terrorists, too

>>51528033
In static languages you use things like OOP interfaces or FP type classes to do the polymorphism that duck typing does in dynamic languages. You don't hack it in using reflection or dictionaries.

>And you need to write the exact same sort of code in both static and dynamic languages.
In dynamic languages you have to be prepared that you got input that you weren't expecting, whereas with proper discriminated unions there is a finite set of possibilities. Also, pattern matching syntax is ubiquitously nicer than writing huge if-else chains.

>>51528074
Green and purple are not opposites.

>>51528124
Lrn2colorwheel

>>51528035

To be perfectly honest, most of it could be done without HTML 5. But who the fuck wants to hear that!!?

>>51528033
>Generally only functional languages support this
And multi-paradigm languages

enum Temperature {
    case Fahrenheit(Double)
    case Celsius(Double)
}

let hot = Temperature.Fahrenheit(212.0)

switch hot {
case .Fahrenheit(let temp):
    print("\(temp) degrees fahrenheit")
case .Celsius(let temp):
    print("\(temp) degrees Celsius")
}

File: 1439065064127.png (305KB, 640x974px) Image search: [Google]

305KB, 640x974px

>tfw thinking of switching to compsci from math because compsci is more fun

>>51528113
>In static languages you use things like OOP interfaces or FP type classes to do the polymorphism that duck typing does in dynamic languages.
I know. But you also get compile time errors if you misspell something, refer to a field that doesn't exist or treat something as if it's a type you weren't expecting. Pretty common issues in programming and can cause bugs that take a long time to figure out. It also enabled powerful IDE tools to be written like intellisense, infallible auto refactoring and built in documentation among other things.
>You don't hack it in using reflection or dictionaries.
reflection is 100% unneeded. Dictionaries are not a hack. That is literally what objects are in dynamic languages.

>In dynamic languages you have to be prepared that you got input that you weren't expecting
If you didn't expect that, then you didn;t write code to handle them and your code breaks. It doesn't matter if you're using static or dynamic typing. With dynamic it's likely to take longer to figure out what the problem was though.

>Also, pattern matching syntax is ubiquitously nicer than writing huge if-else chains.
Utterly agree. pattern matching is based. Nothing to do with static typing though. You could do it in dynamic languages, but I don't know of any that support it.

>>51528134
no u. Yellow is purple's complement, not green.

>>51528152
What the fuck is that language

>>51528198
Swift

>>51528152
well, I consider multi-paradigm languages that support functional paradigms to be functional languages. Just not pure ones.

>>51528152
>print("\(temp) degrees Celsius")
I wish you could just do

print(temp " degrees Celsius");

C# question.

I'm writing a small CLI application that reads a growing video file and writes the data to stdout where ffmpeg is used to transcode/playback the video.

The data is read and written every second using a Timers.Timer object.

The playback goes corrupt after a few seconds. If I don't use a timer the playback is ok.

I'm guessing this has something to do with threads,but I'm not sure.

Does anyone know what could be culprit?

File: Valutron.png (17KB, 750x216px) Image search: [Google]

17KB, 750x216px

"What is then type checker if not the Super-ego, placing impossible to meet demands on our code, and deriving pleasure from our failures?" - Slavoj Žižek

>>51527535
Factual image.

>>51527552
One image proves nothing; see http://c2.com/cgi/wiki?BenefitsOfDynamicTyping

>>51527565
Again, see http://c2.com/cgi/wiki?BenefitsOfDynamicTyping

>>51527565
Ruby, Python, C#, Scheme, Julia, Perl, Common Lisp, PHP, SmallTalk-80, PowerShell, Lua, Groovy - all these languages support dynamic typing. This is proof that dynamic typing has benefits. I could go on about them, but I'll just summarise what I like: language dynamism as a whole is made much easier, so you get things like reflection, transparent proxying, double dispatch, real late binding, ease of writing... I could go on.

Most importantly, it enables a language to implement OOP of the SmallTalk school, which is rather different to the Simula school that C++ epitomises.

>>51528225
We could all make wild guesses at the problem, or you could post the source.

>>51528237
>c2.com
>without irony
HAHAHAHAHAHAHAHAHAHAHAHAHAHAHA!

>>51528237
>Factual image.
Only if you are a retard.

>>51528237
Hey /g/, if an inbred like this guy can make a programming language, so can you! Don't be scared, just do it! Don't let your dream (language) be dream!

>>51528237
"Dynamic typing is better than static typing because you still need unit tests with static typing" is the biggest strawman argument ever.

With a powerful enough type system you can simply not let an incorrect program be compiled.

>>51528006
Did you kill the site?

>>51528245
k

class Program1
{
    static FileStream file;
    static byte[] buffer;
    static Stream consoleStream;
    static System.Timers.Timer timer;
    const Int32 BUFFER_SIZE = 64484;

    static Int64 lastFileLength;
    static Int32 newDataSize;
    static Int32 readLength;
    static Int32 writeLength;

    static Program1()
    {
        consoleStream = Console.OpenStandardOutput();
        buffer = new byte[BUFFER_SIZE];
        timer = new System.Timers.Timer(1000) { AutoReset = true };
        timer.Elapsed += new ElapsedEventHandler(OutputNewData);

        file = new FileStream(@"growing.ts",
            FileMode.Open, FileAccess.Read, FileShare.ReadWrite, BUFFER_SIZE);
        file.Position = file.Length;
    }

    static void OutputNewData(object sender, ElapsedEventArgs e)
    {
        lastFileLength = file.Length;
        newDataSize = (Int32)(lastFileLength - file.Position);

        while (file.Position < lastFileLength)
        {
            readLength = (newDataSize >= BUFFER_SIZE) ? BUFFER_SIZE : newDataSize;
            writeLength = file.Read(buffer, 0, readLength);
            consoleStream.Write(buffer, 0, writeLength);
            newDataSize -= writeLength;
        }
    }
    
    static void Main()
    {
        timer.Enabled = true;

        Console.ReadKey();
    }
}

As I said, if I use the reader/writer loop without eventing it works ok.

>>51526469
thank you for posting an anime ima.....

WHAT

>>51528237
>Again, see http://c2.com/cgi/wiki?BenefitsOfDynamicTyping
>Let me link you to this wall of text because I can't defend my shitty paradigm myself.
If you don't actually understand what you're defending you can step out the debate.
At the very least qoute what you think is the most relevant part of that page.

>This is proof that dynamic typing has benefits.
Yes. Like i said, it makes compilers and runtimes MUCH easier to write. Saves a lot of time and money for these projects that generally don't get much development funding. C# only supports dynamic typing for the benefit of interacting with dynamic languages easier. But generally speaking all static languages support dynamic typing, just cast everything to object or what have you. Now it's dynamic. You can always work around the type system by casting. Dynamic typing is simply the absence of static typing.

> so you get things like reflection, transparent proxying, double dispatch, real late binding, ease of writing... I could go on.
I don't think you understand what dynamic typing is, because you can get all these in static languages too. What makes you think you can't?

>>51527103
do you not remember grand theft /g/entoo?

>>51528237
>Ruby, Python, C#, Scheme, Julia, Perl, Common Lisp, PHP, SmallTalk-80, PowerShell, Lua, Groovy - all these languages support dynamic typing. This is proof that dynamic typing has benefits
all the languages youve listed are either Lisps or scripting languages, these languages have good reasons for being dynamically typed. Scripting languages end up having to use TDD to make up for the lack of compile time type checking.

Julia is an example of a mistake IMO, it tries to unify scripting languages and fast compiled languages like Fortran and C used in scientific computing by making a scripting language run almost as fast as compiled languages. But you lose a lot of power by just using a scripting language. There are compiled languages that can run interpreted mode like Swift, D, Ocaml, Haskell, etc. There should be more effort put into making compiled languages function as scripting languages while in interpreted mode.

>>51527457
stackless python

>>51528334
You can't even get double dispatch in dynamic languages by definition anyway.

>>51527586
>ruby
>slow as shit
shocking

The only reason that dynamic typing works in Smalltalk is that it is implied that Smalltalk will function in an enclosed ecosystem. All attempts at getting Smalltalk to be used in enterprise environments have failed horribly

>>51528237

People who use dynamic in C# deserve death.

>>51528356
rand Theft Gentoo
by carrot


"Stallman, wake up!"
>you wake up
You find yourself covered in stale cheetos from a long night of reading troll emails from /g/.
>you look at the clock, it's 2 in the afternoon
"What the hell! I'm not supposed to wake up for another 2 hours and twenty minutes, this better be fucking important!" You say.
>you shift over to look at shota-linus who woke you up.
"IT IS IMPORTANT! Ubuntu's gone proprietary." He says
....

>>51528402
People who use C# deserve death.

>>51528407
last updated 2013 March

>>51528402
I never used it much myself. I think i used it once for running JS code and getting the values back. It was kinda cool.. but i felt cleaner just to regular static typing to get the values back anyways.

Generally speaking I always notice that in languages that support both dynamic and static typing, people will generally always use the static typing.

>>51528418
C# is one of the best languages out there in popular use. Literally loved by all except those with hate boners for MS.

>>51528432

It's one of those things where C# provides the feature, but usage under any circumstance but extreme duress is thoroughly frowned upon.

For instance, you could litter your code with pointers and use allochglobal all the time.

That would make you a major cock, but you COULD do it.

>>51528418

It feels good.

>>51528334
>I don't think you understand what dynamic typing is, because you can get all these in static languages too. What makes you think you can't?
You're right. I maintain an Objective-C compiler. Dynamic & duck typing are implemented very simply, since every object pointer is opaque, exposing only a pointer to the class it's a member of. That is implemented in C - a statically typed language, albeit one with a very anhaemic type system. You can directly use the Objective-C runtime if you like; it will still have a dynamic object model.

The fundamental defence I am making of dynamic typing is that, when you defer type-related computation to runtime, it makes it possible to implement the features I mentioned in a 'native' manner, so to speak.

>>51528479
>For instance, you could litter your code with pointers and use allochglobal all the time.
That goes for many features. A bad programmer will write shit code in any language. A good programmer will know how to use each feature properly.

Though i sometimes feel dynamic in C# was a mistake. They made it out to be a HUGE feature when they added it, but it's almost never used. Not even when interacting when dynamic shit. I sort of feel like they added because JS was becoming a big deal at the time and they panicked and felt they had to make C# relevant somehow.

>>51528538

In any case, more features is better than less features. I know that sounds like I'm endorsing a C++-tier clusterfuck, but I don't think there's anything inherently wrong with adding more tools to the toolbox, even though you should use most of them sparingly.

>>51528371
Hahaha, what? Let me illustrate it to you in pseudocode:

Fun DispatchAMessage(Recipient, Message, Args)
MessageSignatures = LookupMessageSignature(Message)
Order Message Signatures by the Specificity of their Arguments
While a More Specific Method Signature Exists
MethodImplementation = the More Specific Method Implementation Asssociated With That Signature
End While

MethodImplementation(Recipient, Message, Args)
End Fun

>>51528495
>it makes it possible to implement the features I mentioned in a 'native' manner, so to speak.
That's very vague. What's the actual practical benefit of implementing those features more "natively"? SInce you yourself put "native" is qoutes, could you define exactly what you mean by that?

Anyone?

>>51528225
>>51528317

>>51528562
Not sure if clinically retarded or actually a bot.

Are trees just glorified linked lists?

>>51528562
Meanwhile, all it takes to do N-ary dispatch in languages with strong, static type systems is type classes, which in dependently-typed languages are just plain old values that can be passed around.

>>51528649
It's typically better to think that lists are just a special case of trees.

>>51528649
Are linked lists just glorified kademlia-based overlays?

>>51528561
>more features is better than less features
I disagree. There is a cost. I too love having an awesome toolbox filled with tools. But it becomes a burden when you're working with a team of people and some of them have struggled to learn how to use a lot of those tools. This can be a problem when a language has loads of tiny little features that each save a little time in certain situations, but you gotta learn all of them to be confident you can read any page of code in that language and know what it will do when you run it.

This is my biggest gripe with F#. Love it as I do, there's quite a bit in there. I think they picked features really well and kept it pretty slim. Every features adds tonnes of value. But at the same time you gotta know OOP, know everything in C# pretty well, know how everything in C# would be written in F#, then know all the new stuff in F#. And then you can start getting good at it. It's a lot when sitting beside a C# bro and showing him your sick F# code, but feeling like you need to explain what every second lines does tah used a feature they never saw. For that reason I try to limit the features I use in F#, so I can feel like I could explain it easily to a C# bro if the time came. Brevity is the soul of wit and all that.

The best thing about C# is that anders understand this very well. They are extremely selective of what features they add in each version, it's usually only one or two big things. Anders knows that all languages eventually become overcomplicated clusterfuck like C#, he he works hard to make sure C# stays fresh for as long as possible.
/blog

>>51528283

Yeah. The boot process for the service missed a step.

>>51528677
>lists are special cases of trees
How so? I mean the list structure is what trees are. I guess it could be said the other way too. Tree structures are linked lists but with special permissions set for child parent relationship

>>51528687

I can dig it. I guess I'm of that opinion because there haven't been any major missteps yet wrt to selecting features for new versions of C#.

Dynamic was seemingly bad, but nobody uses it anyway, so there's no risk involved.

Maybe Anders will drop a major stinker on it at some point. Only the future will tell.

>>51526429
Do u have bitbucket

>>51528713
Linked lists are trees where each node can only have one child.

Maybe it's not necessarily better to think of trees as fundamental, as opposed to lists as fundamental. Trees are closely linked to recursion, and lists are closely linked to iteration. So I guess it depends which you prefer in any given situation.

>>51528741
>>>/reddit/

File: KRC.jpg (125KB, 300x377px) Image search: [Google]

125KB, 300x377px

>>51527344

Fuck you and your fellow Functional faggots. I can see why you chose a rainbow flag for your logo, because functional programming is a gay trend.

Lisp, Haskell, Rust, Clozure, and all your other piece of shit crap languages will be dead soon. Deal with it.

Imperative programming is how computers work. Stop trying to be a stupid geek, you and all your freaky friends.

>>51528773
>The linear logic of J.-Y. Girard suggests a new type system for functional languages, one which supports operations that "change the world". Values belonging to a linear type must be used exactly once: like the world, they cannot be duplicated or destroyed. Such values require no reference counting or garbage collection, and safely admit destructive array update.
Philip Wadler, "Linear Types Can Change the World!"

>>51526484
Fucking faggot

>>51527474
By using bitbucket

>>51528741
Bitbucket's bad

>>51528589
ok implemented a mutex so it's working now somewhat

>>51526650
C

>>51528189
I know that feel all too well. Just switch, for god's sake just switch. I wish I had.

>2015
>not running your own buildserver with git
>ashiggydiggy

Anyone else here use D and GTKD?
First time, getting the error:

object.Error@(0): Access Violation
0x666F8AA8 in gtk_print_run_page_setup_dialog_async
0x0040216F in _Dmain at <<file>>(36)
(the line is "Main.run();")
(...)

Where should I be looking if I want to get into GUI (C++) ?

File: buildbot.png (25KB, 718x547px) Image search: [Google]

25KB, 718x547px

>>51528948
Word.

>>51528954
>using meme and memememe
You brought it upon yourself.

>>51528963
>slave lost
TRIGGERED

>>51528956
Qt?

>>51528956
lol just write your guy lib from scratch, its not hard

>>51528956

Qt jest works. The builder included with QtCreator is pretty nice.

>>51528998
For one second I thought you talked about Russians.

Is it okay if I make a linked list where the next and previous pointers are actually indexes into a fixed-size array of nodes?

>>51529042
Memory is just a fixed-size array of bytes, and pointers are just indices into that.

>>51529042
Why not? This is a way to allocate needed Memory somewhat deterministically.

>>51529042
yes
it is okay

>>51529042
https://en.wikipedia.org/wiki/Memory_pool

>>51529006
>>51529018
Yeah just what I was looking for, thx.

>>51528954
Anyone?

>>51529073
Cool. What if I remove the next and previous pointers from each node, and instead maintain the order of the nodes as a secondary array that's just an array of array indexes into the first array? Can I still call it a linked list?

>>51529150
If there's O(1) random insertion and O(1) random deletion, past location ( O(n) in the random case ), and O(1) access to the front, then it's a linked list.

>>51528947
I am just scared that comp sci is just a big fucking meme and that you should study software "engineering" or something else instead and that I havent programmed before taking the first compsci course (the first compsci course was the introductory chapters of SICP, bretty based imho)

>>51529150
>>51529168
a linked list needs to be linked

if your nodes are something like this
[content_index, node_nextindex]
that's a linked list

>>51529184
>I am just scared that comp sci is just a big fucking meme

It is.

>>51529150
No.

>>51529184
Compsci is a science. SoftE is being a codemonkey. IT is being a pooinloo. Choose your poison.

>>51529192
but if you just have an array of content, that's not a linked list

the key part is each element points to the next item

c# question. suppose i've got a master and slave class relationship like a form and a control. the form needs to pass mouse clicks to the control, so it needs to be able to call control.mousedown() and that method has to be marked internal. that has the unfortunate side affect of allowing any other class in the assembly to also call mousedown, which i don't like because i'm autistic. is there some other structure that makes sure only the controller class can call that method?

>>51529226
As for any data-structure, the key point is the access and modification properties. The implementation details don't matter.

>>51529168
All right, so it's not a linked list. Traversal is O(1) rather than O(n), insert and delete are O(n) (but with a smaller constant factor), but I can get O(1) front access by using a deque.

How about if I have the array of data nodes, plus more than one secondary array of indexes into the first array, each one sorted according to a different ordering function? Which data structure is that?

>>51529263
That doesn't mean you can call it a fucking linked list

>>51529280
It does.

>>51529263
If he keeps a linear order in a second Array, then it is not a linked List.

[Index of Elem 1][Index of Elem 2][Index of Elem 3]

>>51529294
I'm not >>51529192 but I suppose that specifically could be called a linked list

However, just because something might have similar performance characteristics to a linked list doesn't make it a fucking linked list

Define numbers as constants, not macros. C programmers have traditionally used #define to manage magic number values. The C preprocessor is a powerful but blunt tool, however, and macros are a dangerous way to program because they change the lexical structure of the program underfoot. Let the language proper do the work. In C and C++, integer constants can be defined with an enum statement.

Constants of any type can be declared with const in C++:

const int MAXROW = 24, MAXCOL = 80;

or enums in C:

enum {MAXROW = 24, MAXCOL = 80};

C also has const values but they cannot be used as array bounds, so the enum state-
ment remains the method of choice in C.

>>51529343
>C also has const values but they cannot be used as array bounds, so the enum state-
Yes, they can.

>macros are a dangerous way to program because they change the lexical structure of the program underfoot
And? This does not make them dangerous.

>>51529265
>All right, so it's not a linked list. Traversal is O(1) rather than O(n), insert and delete are O(n) (but with a smaller constant factor), but I can get O(1) front access by using a deque.

Your Big-O-Skills seem lacking.

>How about if I have the array of data nodes, plus more than one secondary array of indexes into the first array, each one sorted according to a different ordering function? Which data structure is that?

Permutations saved in Arrays just help displaying Stuff in a different Order.

You have not even talked about how you insert Stuff into the main Array.

>>51529312
By that logic, doubly-linkedlists are not linked lists, indexed linkedlists are not linked lists, and if the pointer to any element of a linked list is ever kept anywhere, it's no longer a linked list.

New to programming here...what IDEs do you guys use? I'm learning C++ and visual studio freezes on me whenever I try to run a program.

How do I permanently modify class variables in C++? For example,

class box
{
    public:
        volume = 10;
};

void function1 ()
{
    // Somehow permanently set value of box.volume to 20;
}

void function2 ()
{
    box thisIsABox;
    cout << thisIsABox.volume;
}

int main ()
{
    function1();
    function2();
}

Would output 20. As far as I know right now, anything I do in function1 is destroyed when function1 closes, and can only modify a specific instance of box, not the class itself.

>>51529332
It does. That's literally how you define a data structure.

Friendly reminder that the definition of a meme language is if none of the software you use is written in it.

>>51529458
Pass a pointer to the box to the function.

>>51529456
vim

>>51529459

class TotallyALinkedList<T> {
  T[] array;

  // other things omitted for brevity

  // O(n) access so it's a linked list
  T get(int i) {
    int j = 0;
    while(j < i) ++j;
    return array[j];
  }
}

>>51527170
Usually deal with it by rewriting it.

Sounds like lack of experience either with your language or the codebase you are working on in general. Eventually you should get to the point where more often than not you get it right the first time.

Spending the time to design / think about the solution up front before diving in helps too.

Work is a bit frustrating. We have no code formatting conventions. Everyone just does whatever. The big problem has been trying to get people to agree to a standard. We use Java, so I suggested following Sun's original Java conventions, but I was surprised at the number of people who disagreed.

>>51529562
>no set
You're a failure.

>>51529641
>// other things omitted for brevity
Brevity is the soul of wit, after all :^)

>>51527909
this is why I stay the fuck away from reddit

>>51529580
>We use Java
Found your problem. Java is truly bottom of the barrel for a rewarding work environment. No one used Java who truly cares about programming. It's all they know often. So challenging them to learn so become better coders will not go down so well.

> I suggested following Sun's original Java conventions
I understand they are incredibly verbose. Even for Java.

Famously at the time, when MS launched C# they rewrote some of the Sun's own JSP examples accessing database shit, in the equivalent for C#. The C# was 70% less code.

>>51529648
kek

>>51528649
binary search trees are degenerate

use AVL trees :)

let sum = 0;
let x = 0;
while (x < 10)
{
 set! sum = sum + x;
 set! x = x + 1;
};
sum;

->

define i8* @sequence() {
main:
  %0 = alloca i64, i64 0
  %1 = bitcast i64* %0 to i8*
  %sum = alloca i8*
  store i8* %1, i8** %sum
  %2 = alloca i64, i64 0
  %3 = bitcast i64* %2 to i8*
  %x = alloca i8*
  store i8* %3, i8** %x
  %x1 = load i8*, i8** %x
  %4 = alloca i64, i64 10
  %5 = bitcast i64* %4 to i8*
  %"<" = icmp ult i8* %x1, %5
  %6 = bitcast i1 %"<" to i1*
  br i1* %6, label %loop, label %loopend

loop:                                             ; preds = %loop, %main
  %seqeval = call i8* @sequence.1()
  %loopphi = phi i8* [ %seqeval, %loop ]
  br i1* %6, label %loop, label %loopend

loopend:                                          ; preds = %loop, %main
  %sum2 = load i8*, i8** %sum
  ret i8* %sum2
}

define i8* @sequence.1() {
main:
  %sum = load i8*, i8** %sum
  %x = load i8*, i8** %x
  %"+" = add i8* %sum, %x
  store i8* %"+", i8** %sum
  %x1 = load i8*, i8** %x
  %0 = alloca i64, i64 1
  %1 = bitcast i64* %0 to i8*
  %"+2" = add i8* %x1, %1
  store i8* %"+2", i8** %x
  ret i8** %x
}

W E W
E
W

I sure hope it's actually correct, though. Does anyone see anything wrong or just weird?

>>51527909
https://www.reddit.com/r/gtaonline/comments/3u5fqf/im_new_to_modding_check_out_my_mod/

>>51529731
What are AVL trees?

>>51529456
vim and the GNU compiler

>>51529736
>Sorry, I am French Canadian
Quebecshitter being cancer, as usual.

>>51529745
a tree where the depth of a sub tree is no deeper than 1 level than the other

solves the degeneration of BSTs

I got a Computer Science degree and I just realized I suck at programming and I still don't have a job ahahaha I'm so fucked

>>51529745
implement them yourself, find out
do it in asm for extra credit

>>51529826
Cryptography, quantum computing, machine learning, and optimization don't require much programming knowledge.

>>51529826
just flash ur degree and implement a few basic ass data structures and show them off to your interviewer

using python/flask

how would i get a server side validation with this form input?

<input type="text" required name="shipto" class="form-control"/>

if the user leaves it blank, i want to redirect to a 404 server error.

File: Screenshot from 2015-11-25 20:19:14.png (35KB, 1366x743px) Image search: [Google]

35KB, 1366x743px

>>51529456
Vim

>>51529458
Your functions are not methods here, so they do not have access to any object initialized outside of them. Morever, if you think that "volume" should be a property of the class itself, independently of the instances of the class, you should declare volume as "static".

>>51529938
good stuff, mike :)

>python
User input is allowed to make a file, what do I use to represent characters the file system can't use?
Do I have to re.sub into hexadecimal or something?
How would you guys fix this?

Pleb here, learning hash tables but not using a good book

I want to ask though, is the division key%CAPACITY the only way to generate a hash index ?

My book only mentions this one... I need better resources for this

>the only gtkd dlls i can get working are 64 bit
>the only d std lib dlls i can get working are 32 bit

>>51530064

 key % n 

is to keep the index in the bounds of the array

study up on probing (ex. quadratic probing)

>>51530101
>wants the D
>gets it
>complains

>>51530064
>Reads a book that's not focused on hash tables
>Expects wide, focused information on hash tables

>>51530101
Have you tried 48-bit as a compromise?

>>51530136
not regretting it though

>>51530200
no, but I might try 32 bit twice

>>51530168
Hey man I have this saying, if you're not gonna do a good job then don't do it at all

>>51530242
I for one am glad not every book is 100,000 pages in length and that if I want more information on something I can just refer to books or sites focused on said topic.

>>51530285
My friend, we have people who believe that concise explanations does not equal more pages.

For everyone else, we have >>>/lit/

>>51529963
thanks man :^)

>>51530057
If the user input is live, you should probably reject the filename containing invalid characters and ask them to input another one:

def isvalid(filename):
"""Rejects invalid filenames. This function is for POSIX. """
    invalid = set('\0/')
    for c in filename:
        if c in invalid:
            return False
    return True

Otherwise you can just fail noisily or do whatever you want. The user really shouldn't be inputting bad filenames.

Reverse engineering is fun, too bad IDA is garbage though (unions will literally delete themselves when modified, if the program has been running long enough.)

Any usable alternatives?

>>51530242
>search engines shouldn't exist because they don't provide enough of the listed website's content

>>51530359
nobody likes botnets, bro

>>51530315
What if it is to log something like "/g/" and that is actually important?

>>51530381
what is the file being used for

>>51530313
So you want to know all about something, but want it concise, but without more pages.

>>51530394
to log interests
>Wed.25.Nov.15.06PM_N_jewel lady_I_ r9k, r9komegle and omegler9k.PNG

>>51530381
If it's really important that the file is named "/g/", and your file system does not support naming a file "/g/", then it is clear that you should either get a new file system or store that important information elsewhere.

The problem is that if I, the user, pick an invalid name and the program accepts it, I expect to be able to go into the directory and find my file with the exact name I chose. Illustratively, say I automate the use of your program with a script: my script doesn't know that you're going to be mangling the filename into "\x44g\x44" (or whatever) and will assume your program succeeded in creating the exact filename "/g/". It will look for "/g/" and fuck up.

>>51530416
map the characters into escape sequences
optionally add the intended name inside the file

>>51530416
Consider appending that information to the end of the file.

metadata='2015-11-25T15:06:N_jewel lady_I_ r9k:r9komegle and omegler9k'
printf "\n\n$metadata\n" >>some_filename.png

>>51530453
I don't know, I already log some of the information into another file. Maybe this.
>>51530455
...
>>51530541
Might as well put it in the other text file I have at this point, I like that though.

ImageGrab.grab().crop(box).save(cd+time.strftime("%a.%d.%b.%y.%I%p")+"_N_"+input("Name: ")+"_I_"+input("Interests: ")+".PNG", "PNG")

this is the actual code by the way
cd is the current directory

Know that i know the basics of Python what should i jump into? Interfaces development(tkinter)? Web(Django)?
I'm pretty tired of scripts that runs on terminal.

>>51529456
I generally use Jetbrains IDEs for everything, they're perfect for newbies and very comfy to write in.
Cool people use Vim and it's not a bad program but it's just easier and faster to use something bigger that does half of the work for you.

>>51530592
pls use

strftime("%FT%H:%M:%S")

>>51530610
what do you want to do?

>>51530610
Your question implies some kind of misunderstanding. You should do what you want to do. Ultimately, why do you program? Do you want to make interfaces? Do you want to make web applications?

any libgdx user here?

how do I get the number of frames of an animation?

>>51528225
Sounds like the read isn't keeping up with the playback. Maybe you could have the video info be read from a dedicated buffer instead of directly from stdout?

As one anon said, we can't be sure unless we have a look at the code.

>>51530325
there are a shitload of alternatives. from google:
https://reverseengineering.stackexchange.com/questions/1817/is-there-any-disassembler-to-rival-ida-pro

don't ask me for recommendations, though

>>51529826
Should have got gud Iad.

Best linux distro for serious devs?

I'm fucking bored m90s.

>>51531046
Linux From Scratch

>>51531054
learn Go, it's fun
tour.golang.org

File: game 2015-11-25 21-40-51-96.png (1MB, 1280x720px) Image search: [Google]

1MB, 1280x720px

Working on fixing the networking in my MMO

I had a hacked together networking that wasn't proper client/server, I tore it down a few months ago and started rebuilding, this is probably my third commit over the past 4 months.

The map is procedurally generated and needs some work. Probably next on the list after I get networking sorted out.

The robots are currently made from pre-modelled parts, but procedural is how I want to do them for a borderlands style bazillions of parts. That's after I get the map a little better.

If he's still here, I'd like to thank the anon who helped explain pushdown automata to me like I was an idiot. It was very helpful.

Does knowing Erlang make it easier to get into Elixir?

>>51531095
Can you link his explanation so I can learn

I've started a project an CLI online chess with C++ and boost.asio, but kind of hard and didn't have the time to finish. Any advice guys? Some nice reference manual would be good. Never started to hate a library in less then 1 hour

>>51531113
Yes, elixir is essentially erlang with a fuckload of QoL improvements.

>>51530325
radare2 if you're not retarded

>>51531046
what kind of dev?
anyway, I'd say... any distro. just learn how to use the package manager(s), how to configure/fix things by yourself, etc. get used to these things and you'll be OK. also, if you know some C, linux makes a lot of sense IMO

>>51531123
The only advice you need is: stop being retarded.

>>51531117
It was about a week ago in another DPT

>>51531123
kek u only lasted an hour before you gave up?

consider a different hobby

>>51531046
NixOS is amazing

https://wiki.libsdl.org/SDL_RenderDrawRect
https://wiki.libsdl.org/SDL_Rect

int SDL_RenderDrawRect(SDL_Renderer*   renderer,
                       const SDL_Rect* rect)

is it possible for me to pass in a struct anonymously for the coords?

SDL_RenderDrawRect(renderer,
{ 0, 00, 50, 50 });

>>51531143
oh, and if you want to make packages for linux... be prepared to suffer ;)
although if you made a good app, people will probably package it by themselves and/or help you distribute it. also, most programs are usually packaged for ubuntu.

anyway, go ask >>>/g/flt , I'm far from knowledgeable

It's time for a new thread. Please make it a thread celebrating women in programming.

>>51531243
>celebrating women in programming

You mean like rape simulators?

>>51531255
Dull that blade shortstuff

>>51531294
>>51531294
>>51531294

>>51530708
That isn't a valid file name, Dick.

>>51526805
>Boko Haram
More like Boku no Harem.

>>51527986
no shit

>>51531672
It is on UNIX which you should be using