/dpt/ - Daily Programming Thread

When I allocate a class in C++ using new, the objects in the class are on the heap as well right? Even if they're non-pointer members?

Repost:

I uploaded my language in its current state at: http://neetco.de/languagedev/language

1. There are design issues with the typenodes (MNode in pmf.h/cpp) which makes the code bloated, error-prone and leaky.
2. The primitives only work with int types. Beside, they're handled ad-hoc and bypass most of the code generation, which is also bug-prone and oogleh.
3. I haven't tested that it actually produces valid executables (but the llvm looks OK).
4. printf-and-continue-until-segfault is used instead of proper erroring,

build.sh provides the commands to run to build the entire thing. Only the last command is required if the .y (bison file) and .l (flex file) aren't touched.

The code is mostly written as almost-throwaway code so it's not pretty, but it should be understandable and workable.

The language's syntax is in the .y file. There is support for if-then, if-then-else, while, set!, let, lambda, type declaration (might not work, not sure) and function calls. Moreover, any symbol can be placed infix to represent an infix call without ambiguity under LALR(1):

let fn = lambda (x, y) x + y;
fn(3, 4) fn fn(8, 9);
\\should output the llvm IR for "7 + 17"

There are some simple to correct syntax flaws such as require ; after {} in if and while statements.

llvm_backend.h/cpp does code generation. No passmanager is used, so it outputs unoptimized code, which is especially useful for debugging.

I like the base syntax (beside some flaws that need to be addressed, as outlined above) so far, but the codebase is starting to be hard to work with. Next up is probably, implementing multiple dispatch and type translation (e.g. <number> = <u>(<long>, <double>) should unify the union type, not the <number> type, with whatever argument is used in a function call).

Any pull request, patches, or suggestions would be appreciated, especially in how to cleanup the code.

Pay attention to meeeee!-edition.

>>51577013
One of our customers is the largest television broadcasting compnany in our country. So we build companion apps around their shows and an VOD platform.
For some reason the clients tests on the day of the release and expect a new bugless build the same day.

>>51577095
Yes, where else would they be?

>>51577099
>Page Not Found
gr8 link m8

File: 1447643860612.jpg (94KB, 1274x1001px) Image search: [Google]

94KB, 1274x1001px

>>51577055
>What are you working on, /g/?
Trying to motivate myself to get off 4chan and do something productive

File: GBzPEMQ.jpg (105KB, 640x640px) Image search: [Google]

105KB, 640x640px

>>51577055
>trump

>>51577125
>and expect a new bugless build the same day

I know this feel, and the project managers think it's possible at my place. It's more annoying when we get an incomplete spec and told that a feature had different functionality then what we made, It's painful.

Salesguys sell stuff that doesn't exist, then we have to upgrade the system to include the added functionality, management thinks it's as simple as a "reskin" of another app. Apparently got in an argument with one of the developers because he tried to improve the system instead of spending time on other stuff.

>>51577174
Why are there so many people shilling against trump lately? even on pol

>>51577174
>live in sweden
>immigrants enabled by SJWs are literally destroying the country

>>51577128
Works just fine for me, maybe the site is blocked in your censorship-happy country or something.

File: CUcHOeSW4AA9KEN.jpg large.jpg (45KB, 630x630px) Image search: [Google]

45KB, 630x630px

>>51577201
BECAUSE THEY CAN'T STUMP THE TRUMP

anyway, I'm deciding if I'll use Common Lisp or Racket for my next project

>>51577229
No, you dipshit, the website works just fine, the particular repository 404s, did you even make it public?

>>51577233
Common lisp is a much better idea. It's more stable in both sense of the word and has a much more extensive library.

>>51577238
It's publish, works for me with or without an account logged in. Weird.

>>51577278
public*

>>51577174
top kek
reminder: sage goes in all fields

>>51577201
because trump is a retard. just as bad as hilary but in different ways.

>>51577278
It must be some sort of a bug with either the software or Firefox because I just went to your profile instead and clicked on the repo name and it loaded fine, and now I can access it from your link in the thread as well, before it'd just 404 one me whenever I clicked the link, weird shit.

>>51577225
Sweden migrants are from different culture/religion and are coming to breed your sweet daughters (+80% of them are young males). South America migrant have not much different culture, sex ratio is more equilibrated, and most of them marry intraracially

>>51577244
it seems so. I'd just love racket's ecosystem and documentation

>>51577304
BERNIE 2016

I'm putting together a function in python that's supposed to find the median of a list of numbers.

def median(n):
    num = sorted(n)
    length = len(num)
    x = len(num)/2
    index = length-1
    
    if length % 2 == 0:
        return (num[length/2] + num[index/2] / 2.0
    else:
        return num[x]

here's what i have so far. it keeps say there's an issue with else and I even found someone else's code for the exact same thing and everything was the same and it's still not working. can anyone tell me what's wrong with my 'else'?

>>51577188
We usually start off with a UI design. There's no technical spec. Gradually the client requests more and more features. The features are never documented so upon your next test build you usually get mails/tickets containing 'feature <worthless animation> missing when pressing <worthless button>'.
Anyway I like programming. It pays the bill. Most programmers are also cool, just steer clear from the kiss ass and the brogrammers.

File: 1448568845527.jpg (572KB, 891x600px) Image search: [Google]

572KB, 891x600px

>>51577391
>socialism
>voting for a SJW

File: 1410715789839.gif (3MB, 499x281px) Image search: [Google]

3MB, 499x281px

>>51577225
2018

>>51577401
You're missing a ).

Working on a search app for a simple test project.
Anyone know how do i get this to store the data, it should be working however when data connection is turned off. But the results meant to be stored dont come up.

http://pastebin.com/YKrcD36D

>>51577225
Kek. Sweden is condemned to suffer a similar fate as Sicily

see https://youtu.be/S3yon2GyoiM?t=6m15s

>>51577436
Okay thanks. That was stupid of me. It kept saying something was wrong with 'else' on line 9 that i didn't even notice that. thank you tho

>>51577391
Nope, I will just not be voting this year.

>>51577475
Don't worry about it. Typoes are the most common programming errors you can make and there's no real way of guarding against them.

>>51577443
>db.insert(...)
It returns a long. Log it. -1 indicates an error.
Also use the String.format method for queries.

String query = String.format(SELECT %s FROM %s WHERE %s=\"%s\"", COL_JSON, TABLE_NAME, COL_TITLE, keyword);

Is much more readable than.

String query = " SELECT "+ COL_JSON +" FROM " + TABLE_NAME + " WHERE "
                +COL_TITLE+ "='" + keyword + "'";

>>51577408
I like programming and web dev. I hate tight deadlines, all I want in this job is the time to make something I'm happy with. Not a rushed job that's complete shite. I've still learned lots of stuff, and yeah It pays the bills etc, it's also experience for your CV and not what to do and what to look out for in another job I guess.

Speaking of politics, Obama said that 'enough is enough' after the Planned Parenthood shooting.

Preparing for gun grab, guys. :(

>>51577572
>anarchist
How's that working out for ya?

>>51577508
Returns a long? Sorry I'm very inexperienced. What do you mean by this?

>>51577581

This is all Obama's fault.

>>51577594
The method db.insert returns a long. Log the return value. It most likely does a poo poo when you perform the insert.

>>51577601
Wrong thread fug

>>51577572
friendly reminder that the most dangerous city is Chicago where gun control is the most sever one in usa.

>>51577588
Resistance against immigration is increasing by the day to the point where governments can no longer ignore it. Even our EU-loving prime minister here has been talking about closing the borders and a "sober" immigration policy
The fire rises

>>51577569
I'm just waiting to become a medior developer before looking into other jobs. Hopefully by that time my hobby project is also in a decent state so I can add it to my CV. Maybe somewhere in Asia...

>>51577614
Thanks for the explanation but how would I go about logging it. More so the %s only returns me errors.

reposting from old thread. r8 my java swap function.

import java.lang.reflect.*;
public class T{
    public static void swap(int a, int b) throws Exception{
        Field f = Integer.class.getDeclaredFields()[3];
        f.setAccessible(true);
        char[] t=(char[])f.get(null);
        char temp = t[a];
        t[a] = t[b];
        t[b] = temp;
    }

    public static void main(String args[]) throws Exception{
        int a = 5;
        int b = 3;
        int c = 1;

        System.out.println("Values = a=" + a + " b=" + b + " c=" + c);
        System.out.println("swap(a, b);");
        swap(a, b);
        System.out.println("Values = a=" + a + " b=" + b + " c=" + c);
        System.out.println("swap(b, c);");
        swap(b, c);
        System.out.println("Values = a=" + a + " b=" + b + " c=" + c);
    }
}

output

Values = a=5 b=3 c=1
swap(a, b);
Values = a=3 b=5 c=1
swap(b, c);
Values = a=3 b=1 c=5

>>51577664

long rval = db.insert(TABLE_NAME, null, cv);
Log.d(getClass().getSimpleName(), "" + rval); // should NOT print -1

File: guns are the problem.png (100KB, 1533x1237px) Image search: [Google]

100KB, 1533x1237px

>>51577635

Yes, I already know. Being a certified gun nut means I already examined the statistical relationship between firearms and crime.

The truth is... less politically correct.

>>51577588
>>51577657
Also, I keked at quite some of the snippets in this video, seeing how they were from the USA, specifically footage of the knockout game

but back to technology

>>51577678
>needing reflection to swap numbers

>>51577657
Yeah, sure. You really do think that migrants will stop coming ?

>>51577588
>>51577225
Updated video:
http://www.liveleak.com/view?i=cb0_1447249820

>>51577713
>no r^2 for any of those lines
sure m8

>>51577741
>You really do think that migrants will stop coming ?
I don't, I am just saying that resistance is rising. Parties that are anti-immigrants are getting more votes than ever before

File: CPDBoj2XAAAxbPW.jpg (102KB, 1024x512px) Image search: [Google]

102KB, 1024x512px

>>51577423
AWW YEAHH

>>51577735
your favourite language does not have it? stay mad pleb

>>51577779
And what they will do ? Shoot migrants ? That not gonna happen because of ww2 holocaust.

>>51577765

I forgot to include them, made the graphs a while ago.

In any case, if you can't figure out the relation just by looking out the data points, then I have some bad news.

>>51577805
>having a favorite language
>not just programming with what suits your needs
>needing reflection
>for a swap function

>>51577055
I am not Muriccan so I don't know much about US politics

but from far apart, the only sensible Rep candidate is Jeb Bush, which is scary, because it's a brother of W that fucked up both the economy and Middle East in general

Also because it's the only one sensible without doing crazy Trump shit, nobody listens to him and he is like 6th or something

And democrats have Hillary, who is OK I guess, and Sanders, who is fucking crazy

So to me, the only two sensible candidates would be Bush and Clinton, and that is totally weird

Luckily, my country is totally insignificant and the politicians here are nutjobs too

File: 1hrOIbW.gif (3MB, 320x180px) Image search: [Google]

3MB, 320x180px

>2015
>Not using best language C++

Might as well go kill yourself.

>>51577839
>not just programming with what suits your needs
So you're shit in every language instead of not entirely terrible in a few?

>>51577678
You should create the class with a SwapperFactory.

>>51577839
I am too lazy to search for an image, so I will draw it here

------ joke ----->

        O  <- your head
        /|\
         /\

>>51577572
>enough is enough
>I've had it with these...

>>51577884
well, almost

>>51577741
>http://www.liveleak.com/view?i=cb0_1447249820
Kek'd when trump popped up (13:35)

>>51577689
My program doesnt load up for some reason.
What exactly is this code doing for me?
Would you mind showing me what it looks like in the whole code?

>>51577884
Wow, he doesn't look too healthy with his head off like that.

>>51577862
but I am

>>51577861
>but from far apart, the only sensible Rep candidate is Jeb Bush

Do you even know who Rand Paul is?

>>51577678
>char
>for swapping ints
wot

>>51577876
>implying you need to know anything other than assembly

>>51577813
The best (or worst, depending on how you look at it) a right-wing government can do is reinstate border controls and deny entryto people without the proper papers while denying shit like welfare and social housing to the immigrants that are already here
they can just go to germany instead :^)

Any D programmers around? What's the performance of Vibe.d? And more importantly, can I use Vibe.d for a TCP server that's not web-related?

>>51577918
Yeah, his father is a libertarian, he takes the worse of libertarianism and adds in some general religious nutjobbery.

He has a nickname based on Ayn Rand. Instantly discarded.

File: 1437349834059.jpg (961KB, 1179x2854px) Image search: [Google]

961KB, 1179x2854px

>>51577965
>>51577918
>>51577861

>>51577965

He is, by far, the least shitty candidate. I already dislike government immensely, so anyone that would seek to reduce government is good by me.

>>51577508
>

String query = String.format(SELECT %s FROM %s WHERE %s=\"%s\"", COL_JSON, TABLE_NAME,

>Is much more readable than.
>>51577508
>

String query = " SELECT "+ COL_JSON +" FROM " + TABLE_NAME + " WHERE "
                +COL_TITLE+ "='" + keyword + "'";

Are you on drugs?
They are both bad because SQL injections, but former is WAY, WAY more readable.

>>51577678
optimized my code a little.

import java.lang.reflect.*;

public class T{
    static char t[];
    static {
        try{
            Field f = Integer.class.getDeclaredFields()[3];
            f.setAccessible(true);
            t=(char[])f.get(null);
        }catch(Exception e){}
    }
    public static void swap(int a, int b){
        char temp = t[a];
        t[a] = t[b];
        t[b] = temp;
    }

    public static void main(String args[]){
        int a = 5;
        int b = 3;
        int c = 1;

        System.out.println("Values: a=" + a + " b=" + b + " c=" + c);
        System.out.println("swap(a, b);");
        swap(a, b);
        System.out.println("Values: a=" + a + " b=" + b + " c=" + c);
        System.out.println("swap(b, c);");
        swap(b, c);
        System.out.println("Values: a=" + a + " b=" + b + " c=" + c);
    }
}

>>51577879
you are right. just rename class T to swapperFactory and you can do SwapperFactory.swap(int, int)

>>51578008
* I mean latter. Latter is way, way more readable.

>>51578008
> 2015
> people are still making queries by hand like this

please
why

even MySQL has prepared statements

>>51578029
Your inane.
But yeah, raw queries are bad.

>>51578056
Prepared statements are not in any way an improvement upon writing queries normally.

>>51578077
>in any way

>performance
>sql injection

>>51578077
They are. They protect you from SQL injections instantly

>>51577443
Bumping again. The emulator screen goes black and presumably crashes when the search is run with internet connection switched off. How will I fix this?

Is it possible to create a very basic website, that has an upload button and saves a file to its db without writing a single line of javascript?

>>51578092
Performance is literally the same, tested to hell and back. By me personally.
SQL injections are just as easy to avoid.
You can use database-provided quoting mechanism to quote your values that go into query just the same.
In fact, when you execute a prepared statement, it's a simple query, just as a select, only instead of "select <...> where a='test'" it's "execute prepared_statement_12345('test')"

>>51577735
with enough reflection, you can do anything my friend.

>>51578170
>Performance is literally the same
yeah no
if you send multiple prepared statements in one batch, smart sql servers will only analyse the command once
that cannot be done with simple formatted text

>>51578157
yes

prefer anime pictures over memepolitics

>>51578025
>yfw no one in /dpt/ is smart enough to understand your code.
I thought you guys were cool

>>51578353
>being this much of an attention whore
fuck off

>>51578353
to begin to understand things, I have to care
I saw the java syntax and skipped the entire post

>>51577713
>correlation is causation meme
Didn't austrailia ban guns, and it immediately ended random shooting massacres every few years?

>>51577713
How do you know those stats and yet still love nigresses? I don't know your race but I believe white people are best with white people, and that you can make exceptions for quality races like Asians or maybe Hispanics, but seriously, blacks are culturally, socially, and ethnically the bottom of the barrel. I am dating outside my race (a south american girl) but I would never date an actual black girl.

Performance vs Readability vs Portability

Order from most important to least important

>>51578353
This: >>51578398

>>51578416
>every few years
don't think there's been a 'massacre' in australia since the ban, though

weapons aren't 100% banned, you can still get them
people in australia are simply cool guys, so they don't go around shooting schools
apart from the obvious muslim terrorists, this place has been pretty chill

>>51578469
>heavily implying that you can't have all 3

>>51578469
Readability > performance > portability.

>>51578487
you can but it's not always feasible

>>51578416
>it immediately ended random shooting massacres every few years?

Nope. They've had a few after the 'ban', and the lion's share of shootings turned into arson events.

>>51578480
>don't think there's been a 'massacre' in australia since the ban, though

And you'd be wrong. I guess that's why the agitprop is so effective.

>>51578434

Because it's not black women going out and shooting people to death every day.

>>51578480
>don't think there's been a 'massacre' in australia since the ban, though
That's my point. I wasn't being sarcastic.

>people in australia are simply cool guys, so they don't go around shooting schools
They did before the gun controls. Hence the gun controls.

File: Screenshot from 2015-11-28 21-06-57.png (219KB, 1920x1080px) Image search: [Google]

219KB, 1920x1080px

how could I make this query faster?
I don't know much about MySQL

>>51578487
Absolute portability means you can't have absolute performance, and absolute readability means you can't have absolute performance

You might be able to have absolute readability and absolute portability though

>>51578532
>and the lion's share of shootings turned into arson events.
And you are confident the increase of the arson events was caused by the gun controls, why?

>>51577911
Sorry mang. I only have Android Studio installed on my business laptop. Make sure the database is created, the table is created with the correct columns and the insertion is done with the correct column names.

>>51578564
>Absolute portability means you can't have absolute performance
Huh?

>and absolute readability means you can't have absolute performance
Huh?

>>51578544
>WHERE 1=1
why.3gp

>>51578532
>And you'd be wrong.
https://en.wikipedia.org/wiki/List_of_massacres_in_Australia

nothing comes close to Martin Bryant's death toll
all the others are mostly arson and family stuff (mostly aboriginals)

My assignment is to create this generic method class with an array but I haven't been able to get the push() and pop() methods to work. Any help?

public class GenericStack<E> {
    public static int size = 16;
    @SuppressWarnings("unchecked")
    private E[] list = (E[])new Object[size];
  
  public void add(int index, E e) {
      ensureCapacity();
      
      for (int i = size - 1; i >= index; i--) {
          list[i + 1] = list[i];
          
      list[index] = e;
      
      size++;    
    }
  }
  public int getLength() {
    return list.length;
  }

  public E peek() {
      E o = null;
      o = list[0];
      return o;
  }
  public E push(E o) {
      ensureCapacity();
      list.append(o);
        size++;
        return o;
  }
  public E pop() {
      E o = null;
      for (int i = 0; i > list.length; i++) {
          o = list[i - 1];
    }
          list[list.length - 1] = null;
        size--;
        return o;
      }
  private void ensureCapacity() {
      if (size >= list.length) {
        @SuppressWarnings("unchecked")
        E[] newlist = (E[])(new Object[size * 2 + 1]);
          System.arraycopy(list, 0, newlist, 0, size);
          list = newlist;
      }
  }
  public boolean isEmpty() {
      if (list.length > 0) {
        return false;
      }
      else {
          return true;
      }
   }
}

>>51578599
forgot to remove it. Was trying to put my fulltext into the where statement, but I'm still trying to figure out why (without having to execute it again)

>>51578487
this is what you end up with when you care about portability:
http://geoff.greer.fm/vim/

>>51578573
>Huh?
Different platforms have different conventions, different hardware, different optimal code paths.
>Huh?
Highly tuned and optimized code tend to be subtle and hard to understand.

>>51578566
>And you are confident the increase of the arson events was caused by the gun controls, why?

That's not what I said. I just said that shooting events were replaced by arson events as the leading mass murder tool. After the ban there was a crossover of shooting events and arson events. The latter went up and the former went down.

Not only that, but traditionally, arson is the most effective means of mass murder, even in the United States. For instance, the Happyland arson in 1990 killed 87 people.

>>51578572
Darn, I'm pretty sure it is. Thanks for the help so far though.

>>51578602
>nothing comes close to Martin Bryant's death toll

Which is irrelevant, because that's got nothing to do with whether or not they were massacres.

Just admit you were wrong and we can all move on.

>>51578532
>shooting people to death

only in OneSlow's hole

>>51578228
>smart sql servers will only analyse the command once
Can and is.

>>51578638
>The latter went up and the former went down.
Who cares so long as the overall murder rate is down?

>>51578652
>Martin Bryant
>irrelevant
obviously you don't even know what you're talking about, as that was the massacre which made Howard introduce gun controls
>Just admit you were wrong and we can all move on.
I'll do the latter of those.

>>51578674
Because you can't tell if it's due to the statistic being tracked or to an external factor without using tools like the R2 measure (or other).

File: fig012.png (9KB, 600x344px) Image search: [Google]

9KB, 600x344px

>>51578674
>Who cares so long as the overall murder rate is down?

That's what I'm saying, the 'buyback' did not affect the homicide rate meaningfully. There was no reason to take away Australians' guns.

The US homicide rate has gone down faster than that of Australia, but their's was already low to begin with.

It was just political posturing and that's all.

>>51578653

There hasn't been a killing in my town for the better part of a century. Shit, the last time it happened was over an early 1900s railroad dispute or something.

>>51578624
or this:
http://git.savannah.gnu.org/cgit/emacs.git/tree/src/keyboard.c

File: 25.jpg (331KB, 1247x814px) Image search: [Google]

331KB, 1247x814px

post comfy pics to lessen tension plox

>>51578734
>11785 lines
HOLY FUCK!

>>51578688

I didn't say Bryant was irrelevant (even though we all know he was a government plant such that the gov't could take away the guns), just that his kill count has no bearing on whether or not some other event qualifies as a massacre.

We don't base our definition of massacre on whether or not somebody beats Lanza's score.

Truth be told, it's a shame that Howard didn't make like Harold and fuck off for a really long swim.

File: 2.jpg (101KB, 1247x832px) Image search: [Google]

101KB, 1247x832px

>>51578748

File: 1447188509060.jpg (1MB, 3000x2250px) Image search: [Google]

1MB, 3000x2250px

>>51578756

>>51578748
This political bullshit usually starts when the burgers wake up. Before that it's nice and civilized.

File: 1448285507458.jpg (297KB, 1920x1200px) Image search: [Google]

297KB, 1920x1200px

>>51578769

>>51578782
Sockets

>>51578695
common sense would dictate if the main tool of mass murder is guns, and then you introduce gun controls, and then after that the overall murder rate is down, then the gun controls probably worked helped lower the murder rate.

Nothing like this can be proven to work or not work with an extremely high degree of dependability. In in the absence of evidence to suggest otherwise, it's best to assume the theory that makes the fewest assumptions.

>>51578726
That graph does look like it's on a downward trend though. I'd be curious to see it taking population growth into account too.

>>51578784

The problem only starts when Australians show up and shitpost like they're wont to do. They even come out with the old 'gun ban stopped massacres' chestnut!

Never gets old, mate.

>>51578782
>Good? y/n
n

>>51578782
yes fifo is fine

>>51578797
The theory that makes the fewest assumption is that there is no correlation between the reduced amount of crimes and gun controls, even if we were to use just the statistics posted here. Good job on skewering your own argument, though.

File: 1448283726061.jpg (225KB, 1280x720px) Image search: [Google]

225KB, 1280x720px

>>51578789

>>51578782
>>51578791
?
What is the best way to use a dynamic variable between two (programs?) in python?
I'm about to use a hotkey to write it to a file and have the other one read it. Good? y/n

File: snake.jpg (648KB, 1299x1191px) Image search: [Google]

648KB, 1299x1191px

>>51578837

File: SDT-2013-05-gun-crime-1-2.png (14KB, 417x395px) Image search: [Google]

14KB, 417x395px

>>51578797
>That graph does look like it's on a downward trend though.

The same thing is happening in the United States though, and things have only become more free with respect to firearms, here.

The relation between homicide and firearms is pretty much fuck-all.

They neither assist in reducing it or assist in increasing it.

Whos programming their gun assembly line in lisp? cause I sure am.

>>51578870
Could you find a graph showing the relationship between murders and burgers? You'll be amazed. Protip: start with the Vietnam war.

File: kc.png (3MB, 3338x4294px) Image search: [Google]

3MB, 3338x4294px

>>51578868

File: 1447850078666.jpg (534KB, 1600x1028px) Image search: [Google]

534KB, 1600x1028px

>>51578934

File: hamburger.jpg (680KB, 772x622px) Image search: [Google]

680KB, 772x622px

>>51578917
>murders and burgers?

What have you got against hamburgers?

I want to re-learn calculus. What are some softwares that can aid me?

>>51578961
but did you create your image
i created mine

>>51578976
>using software

Have your mom drive you to the store and buy a book.

>>51578995
Right. Which book would you recommend?

>>51578976
khan academy

>>51579008
that's not really a /dpt/ question now, sorry m8.

>>51578976
youtube is software
https://www.youtube.com/watch?v=P9dpTTpjymE

File: 1447279581880.png (4MB, 1918x1084px) Image search: [Google]

4MB, 1918x1084px

>>51578992
I did create it

>>51579027
no, youtube is service.

>>51579031
then why would you create, upload it to 4chan, download and store that version, then upload it again

>>51579036
it's a Software Solution (TM)

File: 1447707565130.jpg (786KB, 1920x1080px) Image search: [Google]

786KB, 1920x1080px

>>51579050
I like 4chan so much I name all my creations by 4chan's convention.

>>51579078
fuck off, also
>url slugs
fucking disgusting

>>51579078
f http://u.no/rmi

>>51579078
desu makes no sense
boards should be around topics
anime content -> anime board
computing content -> computing board
anything else is dumb

>>51578544
How many rows are there in result? I see it's limited to 1000, but to retrieve those 1000, database may have to retrieve all rows for your request, then calculate vidmatch+tagmatch, then sort all rows by this value, and only after that throw away all but first 1000.

Try sorting by only one field, not sum of them, see if speed improves.

File: 6m4qq.png (10KB, 422x252px) Image search: [Google]

10KB, 422x252px

>>51577055
uhhh
just me, or did this thread just implode

>>51579313
It's just you.
>dat dynamic typing

>>51579313
hm, now it's back
sydney cloudflare had the APIs response as a near empty array
anyone else see it?

>>51579313
I had that too
just refresh

>>51579313
Nah, I had that happen to me too. Shit was strange.

>>51579347
>>51579344
coolbeans
mook better fix that

>>51578025
>reflection
>optimized

>>51578353
answer me faggot >>51577921

LINKED LISTS IN C

http://pastie.org/private/bvurtrw0uxidcdtxwxtww

/* Get a `slice' of a list.
 * A negative step reverses the list and then behaves 
 * the same as a positive step.
 */
struct node *
slicestep(const struct node *list, size_t start, size_t end, int step) {
    assert(list && start >= 0 && start <= end &&
            length(list) > mag(start, end)); 
    struct node *slice;
    if (step < 0) {
        struct node *reversed = reverse(list);
        slice = slicestep(reversed, start, end, -step);
        delete(reversed);
        return slice;
    }

    slice = newnode();
    const struct node *cur = seek(list, start);
    for (size_t i = start; i < end; i += step) {
        append(slice, dup(cur));
        for (int nsteps = 0; nsteps < step; ++nsteps)
            cur = cur->next;
    }
    return slice;
}

struct node *
drop(const struct node *list, size_t ndrop) {
    assert(list && ndrop <= length(list));
    return slice(list, ndrop, length(list));
}

...

int sum = 0;
for(unsigned int i = 10; i >= 0; --i)
    sum += i;
std::cout << sum;

>program takes ages to process
>kept it running for 3 minutes now
i mean i knew win10 was slow but damn

File: Screenshot from 2015-11-28 23-05-47.png (242KB, 1920x1080px) Image search: [Google]

242KB, 1920x1080px

>>51579227
you made me realize the main problem was fulltexting over the tags.
I'm getting the tag relevancy by a join now, it's much faster and the relevancy seems much more natural/expected.
thank you

>>51580390
lel

>>51580390
Epic

File: smug loli maid - why are you so smug, it appears you broke the glassware.jpg (3MB, 2560x2880px) Image search: [Google]

3MB, 2560x2880px

>>51580390

>>51577443
Bumping again. Really need a solution for this.

>>51578617
Can I get some help please?

>>51580417
c... can you dump that database?

Lose motivation to complete large projects what do

>>51580633
How many times are you going to post that?

>>51580704
Until someone gives me an answer.

>>51580390
I like it.

>>51580712
Keep posting. If you get dubs I'll help you.

>>51580712
/g/ isn't your personal stack overflow

>>51580759
Why can't you just help me

>>51580712
Either I'm a genious or you're completely retarded.
Pop

T item = list[list.length -1];
System.arraycopy(list, 0, list, 0, list.length -1);

>>51580771
Because it's obviously homework and you should be paying attention in class

>>51580777
If only I spelled genius correct...

>>51580818
It's an online class and professor gave us an assignment that wasn't covered in the chapter.

>>51580777
I'm retarded. What method is that supposed to be?

>>51577391
please don't ;(

>>51580838
What wasn't covered?

>>51580838
>wasn't covered in the chapter
That's what they all say. I'm pretty sure the chapter was about generics. Now it's up to you to actually put your knowledge to use.
Protip: drop your current code. It's shit. Start with writing push followed by pop.

>>51579499
As such a keen programmer can easily see, swap part does not use reflection. only the static initialization part does. after that, it is not any different than any other swap method.

gee, /dpt/. you is stupid

People hate on Go, but what should I use instead for a (performance-focused) webapp?

>Python is dogshit slow
>Ruby is even slower
>PHP is torturously inconsistent
>Javascript is Javascript
>Java is bulky and boilerplate-ridden

File: 1448664014675.jpg (401KB, 802x609px) Image search: [Google]

401KB, 802x609px

>>51581117
Go

>>51581117
c++, obviously

>>51581117
Elm

>>51581117
C

>>51581117
just use go
if you want a performant high level language, you want to build something concurrent and don't essentially want to use a functional language, then use go

File: wtfrudoin.jpg (213KB, 551x650px) Image search: [Google]

213KB, 551x650px

>>51581117
>performance-focused
>webapp

>>51581099
>implying i bothered to read your code
it's still shit

If I have python program that has output redirected to a file via pipe. How can I somehow print certain messages to terminal - logging info?

>>51581117
I don't hate Go, I have D&R on the way

>>51581117
ELIXIR or erlang.

>>51581238
Yeah, not everyone appreciates the beauty of reflections.

>>51581303
how is it beautiful? how is it useful?

>>51581325
It is a feature that lets you break everything Java stands for. It is beautiful

>>51581128
Yes, I know the GC is stop-the-world. I mean performance as in "handling many complex requests concurrently", not "realtime video game".

>>51581147
That's a kind of interesting design, actually. But it's too obscure. No support or libraries for it.

>>51581184
Basically what I'm thinking

>>51581224
Guess what, webapps are popular and make money. I prefer native code too but the market is not there right now.

>>51581255
>functional
errrr

>>51581353
Go routines are THE solution for your problem.

File: hmm, yes.jpg (34KB, 396x303px) Image search: [Google]

34KB, 396x303px

>>51581352

>>51581353
If you don't know shit about a language, at least take the time to look at it instead of
>hurrrrr

>>51581325
I sometimes use it for making private fields public in a 3rd party library.

>>51581358
No, elixir is. Elixir and erlang's whole game is handling many complex requests concurrently. There's a reason erlang is used everywhere in telephony.

>>51577678

swap(3, 5);
System.out.printf("5 == %d\n", 5);
System.out.printf("3 == %d\n", 3);

5 == 3
3 == 5

Good one.

>>51581449
if this actually happens in java
then oh man
java is worse than dumb

>>51581478
Reflection is magic. You can do anything if you believe in it.

>>51581478
http://www.tutorialspoint.com/compile_java_online.php?PID=0Bw_CjBb95KQMbk5Xc3J3cXdBbTg

>>51581478
NO U

stick to your python shit fucking tard

>>51577443
Is anyone able to help fix this? I just need that one genius anon to assist this problem.

>>51581478
I remember seeing a lovely writeup about how you can overload all of the basic operators with reflection and sun.misc.unsafe. It could be included as a dependency.

Like a language-level version of chaos monkey.

I want to write several 64b numbers to a file as bytes

Should I use little endian or big endian? It doesn't matter that much, the library I am using supports both, but what is more normal?

>>51581598
>sun.misc.unsafe

And to think, they want to get rid of this last bastion of freedom.

>>51581607
little endian is more common

I cannot decide if I should make vidya in C and take advantage of cache locality or C++ and take advantage of objects and std::vectors. I'm reasonably competent at both, which one should I use?

The entire game industry seems to have moved to c++.

>>51581621
didn't they consider making it a public API in like java 8 or 9 or something

>>51581646
you can have both. ignore std::vector, create an array implementation with templates. there is no reason to go C over C++

>>51581607
Little endian is more common (normal). Big endian is more natural to me though.

>>51581658

Not that I know of. Last I heard they were still trying to get people not to use it in anticipation of full removal.

>>51581728
>>51581658
They have been soliciting developers for feedback as to what parts they would like to see moved into other packages before full removing Unsafe.

>>51577055
I am a curry nigger but I want Trump to win in 2016 elections.

>>51581811
Are you Bobby Jindal?

>>51581607

If OpenSSL had to support big-endian x86, you should too.

created a new monad (at least i think its a monad), Voidable<T>, where T could be void
has a member of type T (or not if T is void)
Voidable<T>.bind!function
invokes function(T_member) (or function() if no member), wraps the return (or lack thereof) in another voidable

I want to write a program that takes one note files (.one) and converts them to xhtml so I can load them on an e-reader.

How would I go about getting the raw text from the .one files? They can only be edited from one note and google has yielded no results

>>51581880
useful for generics, don't need to specialise for void

>>51581895
>I want to convert proprietary format to something else
good luck

>>51581913
Exactly why I don't know where to start. Is it even possible?

>>51581895
>>51581913
>lol proprietary
[microsoft shilling]
https://msdn.microsoft.com/en-us/library/dd924743(v=office.12).aspx
[/microsoft shilling]

>>51581686
or std::array
Also, std::vector is exactly as cache-local as plain arrays.

>>51581966
The M$ programs don't adhere to the published standards so yes, lolproprietary.

>>51577055

getting my head wrapped around TCP/IP
coding sockets in C

>>51581945
std::vector can only allocate on heap. It will be local only in itself, not with other values on stack or other members of object for example

>>51581728
>>51581746
looks like they might keep it

http://openjdk.java.net/jeps/260

>>51581998
whoops, this one was for
>>51581971

>>51581998
>how do generics future proof
std vector takes an allocator type as a template

make a stack allocator

>>51582034
Write me an allocator within standards that allocates a dynamic-sized memory in stack.

>>51581607
Most formal standards use big-endian (e.g. all of the low-level internet protocols are big-endian: IP, TCP, UDP, DNS, ...).

Little-endian is probably more common due to the sheer number of formats that arise from Windows/x86 software just dumping the in-memory representation to a file.

If you're trying to create a cross-platform format and/or portable code, use big-endian for the simple reason that if you screw up, you'll find out sooner rather than later. Because:

1. Little-endian architectures are more popular at the moment (x86 and ARM are both little-endian, and that's 99-point-something percent of the market already), so code which fails to deal with endianness is more likely to get overlooked if the format is little-endian.

2. If you make invalid assumptions about the sizes of integral types, you'll often get away with it with little-endian architectures. E.g. if you read a 32-bit integer value into a 64-bit "long", with a little-endian you'll get away with it if the extra bytes are zeroes. With big-endian, "extra" bytes go at the start, which changes the position of the low bytes.

>>51577987
> Fell for the reduce government meme

>>51582122
can someone else
http://stackoverflow.com/questions/8049657/stack-buffer-based-stl-allocator/28574062#28574062

>>51582175
>fags should take most of your money and spend most of it on themselves and queer shit

>python, please kill me
>threading

   thread2 = threading.Thread(loop())
   thread2.start()
   thread1 = threading.Thread(pidz())
   thread1.start()

The1st thread has a loop in it so the 2nd thread never starts, how do I run them both at the same time?
The loop also takes user input from time to time, Will that mess the 2nd process up?

>>51582213
You're not supposed to RUN THE LOOP OUT OF THE THREAD!

>>51582190
>This allocator uses a user-provided fixed-size buffer as an initial source of memory
genius, first allocate your whole stack using char buff[1000000] then use it in your allocator. you can call it 'dynamic' allocation I guess.

>>51582245
Teach me, senpai.

>>51582213
Why would thread1 never start? Thread2 is a difference thread than the thread where you call start on both threads.

>>51582213
this is your average python programmer /dpt/

>>51582213
You're calling loop, not passing it in. You can thank Python's lack of a type system for not giving you any indication :^)

why is this allowed

myDict = {}
myString = "asdf"

def changeDict(d):
    d["key"] = 100

def changeString(s):
   s = "ayy lmao"

changeDict(myDict)
changeString(myString)

print myDict    # {'key': 100}
print myString    # asdf

>people will defend this
>MUH MUTABILITY

File: Untitled3.png (21KB, 685x514px) Image search: [Google]

21KB, 685x514px

>>51582213

Actually, both threads have loops in them.
Doesn't matter witch I start with.
If I keyboard-interrupt the loop, The next thread doesn't even start.
Confused.

>>51582175
>reduce government meme

reducing government isn't a meme. It's the ultimate goal.

>>51582315
this has nothing to do with mutability

>>51582333
>>51582290

>>51582213
This:

       thread2 = threading.Thread(loop())

executes loop() and passes its return value to threading.Thread(). You want

       thread2 = threading.Thread(loop)

which passes the "loop" function itself to threading.Thread().

>>51582253
ok how about taking the address of recursive function

File: Untitled3.png (7KB, 383x185px) Image search: [Google]

7KB, 383x185px

>>51582345
>>51582290
>>51582358
?...

>>51582375
any memory you get would be invalid when that function returns

>>51582381
Well what does the exception say?

>>51582205
> Still thinks anyone is going to actually reduce government except for the companies spending money on lobbying

>>51582392
Interestingly I have made 'stack allocated' data structure that inlined the constructor and used alloca.

Of course... this is not exactly reliable.

File: Untitled3.png (9KB, 477x181px) Image search: [Google]

9KB, 477x181px

>>51582393
>Exception! <class 'AssertionError'>!

>>51582315
Assignment is not mutation.

The former modifies the passed-in dictionary object. The latter assigns to the variable which initially held a reference to the passed-in string object.

A clearer example:

list1 = [1,2,3,4]
list2 = [2,4,6,8]

def changeList(l):
    l[:] = [99]

def dontChangeList(l):
    l = [99]

changeList(list1)

dontChangeList(list2)

print list1 # [99]
print list2 # [2, 4, 6, 8]

The first modifies the list (via either the __setitem__ or __setslice__ methods). The second assigns to the variable, leaving the list untouched.

>>51582444
>Python
Well, I'm done here.

If you find programming to be frustrating as hell does that mean you're just not cut out for it?

>>51582451
python is looking more and more like a meme

a = [1, 2, 3]
b = a
b.remove(3)
print(a) # [1, 2]

Why the fuck does removing 3 from b also remove it from a here?

>>51582484
you're aware that every imperative language that abstracts pointers is like this, right?

>>51582484
any language that passes by value does the same

>>51582521
Reference types.

>>51582392
you do everything within the function and do it very carefully

>>51582546
How do I not do that?

>>51582607
language?

>>51582607
You have to copy it.

File: Untitled3.png (31KB, 685x503px) Image search: [Google]

31KB, 685x503px

>>51582444
working with
thread1 = threading.Thread(target=pidz)
Now for thread locking print.
Fuck.

>>51582632
>unknown variable errors are runtime exceptions in Python
JUST

Reminder that this compiles and runs under gcc:

#define ayy(x) (x)(lmao(x))
#define lmao(x) (x)(ayy(x))

int main()
{
    ayy(lmao);
}

File: tripfag enters thread.gif (3MB, 370x208px) Image search: [Google]

3MB, 370x208px

>>51578806
The problem only starts when tripfags enter the thread.

>>51582676
There's literally nothing wrong with this.

How do I do arithmetic in Hexadecimal?
I'm taking in 1 input at a time for a 4 digit number (so inputs are stored in a3a2a1a0) and I'm trying to figure out how to convert from ASCII Hexadecimal inputs to a 4-digit hexadecimal number, then output it back as a 4 ASCII Hexidecimal characters.

Alright so I'm trying to do this ants stimulation in python (matplotlib,numpy etc) where basically I decide to move an ant into an adjacent cell in an array based on a probability distribution of the adjacent cells' pheromone number. The problem is that I want to try to evade out-of-bound errors while keeping the code simple (I don't want to write clauses for every boundary condition).

Can I sort of tell the arrays to wrap around so that a[n+1] = a[0] ?

are these literally equivalent:
C

struct MyStruct
{
    int someNumber;

    void (*init)(struct MyStruct *mystruct);
    void (*destroy)(struct MyStruct *mystruct);
    void (*setNumber)(struct MyStruct *mystruct, int number);
    int (*getNumber)(struct MyStruct *mystruct);
};

C++

class MyClass
{
    private:
    int someNumber;
    
    public:
    void init();
    void destroy();
    void setNumber(int number);
    int getNumber();
};

or does c++ do it differently behind the scenes? If they are then every call to a method in c++ has a jump to a function which is slower and not great for locality.

>>51582839
They're not equivalent at all. What would require function pointers like that in C are virtual methods. If you just have regular methods, it's the same cost as free functions in C.

>>51582839
>every call to a function has a jump if it's not inlined

>>51582839
language designers are a little bit smarter than that

>>51582806
Use modulo.

File: 1395141058647.jpg (25KB, 782x127px) Image search: [Google]

25KB, 782x127px

>>51582761

No, it's definitely Australians. This isn't some conspiracy theory I've come up with, it's a well-known phenomenon.

It got so bad that kike-moot was forced to sell this garbage dump to Hirohito.

>>51582616
python

>>51582806
Write a get(x,y) function

File: Tripfag.jpg (52KB, 692x623px) Image search: [Google]

52KB, 692x623px

>>51582888
Nah, it's tripfags.

>>51582521
so many pythonfags confused

yet they still believe it's a good language

>>51582839

class MyClass
{
    private:
    int someNumber;
    
    public:
    void init();
    void destroy();
    void setNumber(int number);
    int getNumber();
};

is

struct MyStruct
{
    int someNumber;    
};

void MyStruct__init(struct MyStruct *mystruct);
void MyStruct__destroy(struct MyStruct *mystruct);
void MyStruct__setNumber(struct MyStruct *mystruct, int number);
int MyStruct__getNumber(struct MyStruct *mystruct);

with different name mangling

when does the new thread get created?

>>51582945
bump limit is 310

>>51582945
It doesn't, this is the last one.

Network stack in Python

>>51582958
kek'd

>>51582942
ok good

>>51582981
Virtual inheritance uses function pointers though (which is why nobody likes it, but it's an occasional necessity)

new DPT gogogo

>>>/trash/

>>51582521
Because a and b *are the same list*. That's what "b = a" does.

>>51582932
Practically every language which isn't C or C++ behaves in the same way as Python. Java, C#, JavaScript, ....

>>51583074
when did we get a /trash/ board?

>>51583103
>not using a god-tier pure language where it functionally doesn't matter

>>51579008
Egghead's Guide to Calculus is pretty good if you just need a refresher.

>>51583109
just for you

>>51583109
is this literally your first day here?

>>51582839
> are these literally equivalent:
Nope.

1. Only virtual methods use function pointers. Non-virtual methods are called directly. So obj->init is basically syntactic sugar for MyClass::init(obj);

2. If a class has virtual methods, they're the same for all instances of the class. So it doesn't store multiple function pointers in each object; it stores the function pointers in a virtual method table (vtable) and stores a pointer to the virtual method table in the object (only for classes which have at least one virtual method). If init() was virtual, obj->init would be equivalent to obj->vptr->init(obj).

>>51583103
python hides everything to do with types as much as possible. confusion about things like that don't even occur in statically typed languages.

>>51582632
How do I thread lock input?

>>51583109
>11/09/15 New trial board added: /trash/ - Off-Topic

>>51583181
>ask about a new thing
>lol newfag

on a semi-related note... people are actually replying to my garbage thread at >>>/trash/331114

>>51583339
japmoot seems to be trying new things, which is kinda cool.

>be hiroshima nagasaki
>add /trash/ board
>becomes populated by /mlp/
>remove /mlp/
>/mlp/ is now officially trash

>>51577099
How does the infix function syntax work with higher order functions? For example,

map fn xs

would be

fn(map, xs)

wouldn't it?

>>51583422
Won't it not be a problem since you'd call it like map(fn, xs)? fn map xs for infix?

File: Cirno - Answer Is Cat.png (87KB, 500x500px) Image search: [Google]

87KB, 500x500px

I'm procrastinating working on my large projects (which are still just for fun) by making a dumb script that translates an integer into a Lojban string.

#lang racket/base
; num2jbo.rkt
(require racket/contract
         racket/string)

(define nums 1437)
(define jbo (list "no" "pa" "re" "ci" "vo" "mu" "xa" "ze" "bi" "so"))

(define (helper s n)
  (if (> n 9)
      s
      (helper (string-replace s (number->string n) (list-ref jbo n)) (+ n 1))))

(define/contract (number->jbo n)
  (exact-nonnegative-integer? . -> . string?)
  (define str (number->string n))
  (helper str 0))

(number->jbo nums)
; => "pavocize"

>>51582839
No.

NEW THREAD


>>51583493


NEW THREAD


>>51583493

>>51583422
Correct. The language doesn't use space delimitation (it's not an ML descendant), so "map fn xs" would be expressed as map(fn, xs); instead.

>>51578992
>>51578934
If you're still here, may I know the concept/math behind your picture? Is it like a lorenz equation or attractor or something?

>2015
>not using telegaertner

http://www.telegaertner.com/en/info/highlights/mfp8/

>>51583667
what's telegaertner

>>51584245
A shilled site trying to infect you with a drive-by downloaded keylogger.

>>51584317
really? good thing i didn't even visit the link this time around