C/C++ Programming Questions Not Worth Their Own Thread

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Cavern of COBOL > C/C++ Programming Questions Not Worth Their Own Thread

«‹›639 »

Fecotourist: Nov 1, 2008

Use boost.format.

str(boost::format("%5.3f") % x)

prints the value of x formatted into a std::string as you would expect from experience with printf. It's type-safe, and there's no need to worry about overrunning a buffer.

# ? Apr 13, 2009 02:32

Adbot: ADBOT LOVES YOU

# ? Apr 29, 2024 18:27

Elswyyr: Mar 4, 2009

Is there any online guides to learn c++ that you can recommend? I apologize if this question already has been asked.

# ? Apr 13, 2009 18:17

Dijkstracula: Mar 18, 2003; You can't spell 'vector field' without me, Professor!

If you're serious about wanting to learn the language, you should buy a book instead of relying on lovely and outdated web tutorials.

See the first post for recommendations.

# ? Apr 13, 2009 19:09

narfanator: Dec 16, 2008

Is there any way to do have two functions identical, except for the order of default-valued parameters of different types?

code:

void foo(int x, double y = 1, float z = 2)
{
   return x*y*z;
}
void bar(int x, float z = 2, double y = 1);
{
   return x*y*z;
}

# ? Apr 13, 2009 22:10

Avenging Dentist: Oct 1, 2005; oh my god is that a circular saw that does not go in my mouth aaaaagh

I am assuming you mean for both of the functions to be named foo because there's nothing wrong with your code now (except that you can't return a value from a void function, and that your second definition has an extra semicolon).

In that case, foo(1) is ambiguous. There are lots of ways to resolve this, but the real answer is that you probably shouldn't be using default parameters like that. Any more than one default per function is just annoying to use.

Boost.Parameter probably resolves the core issue you have, at the expense of some hilarious templates.

# ? Apr 13, 2009 22:34

Contero: Mar 28, 2004

I'd like to have a vector class that has 3 floats. I'd like to be able to access them either as an array: myvec.val[0] or as individual members: myvec.x. Is there a way to do this with unions or something?

# ? Apr 13, 2009 23:50

chips: Dec 25, 2004; Mein Führer! I can walk!

You could always do:

code:

class Vec3
{
  double x, y, z;

  double operator[](int i)
  {
    return (&x)[i];
  }
};

Probably considered bad form.

# ? Apr 13, 2009 23:54

Zakalwe

May 12, 2002

Wanted For:

Terrorism
Kidnapping
Poor Taste
Unlawful Carnal Gopher Knowledge

code:


union
{
  struct
  {
    float x, y, z;
  };

  float val[3];
}

Anonymous union with anon struct. Not standard but supported by pretty much any compiler you'll use.

edit: VV Suck it AD :P

No. you just sucked off the trailing ; :P

Zakalwe fucked around with this message at 00:00 on Apr 14, 2009

# ? Apr 13, 2009 23:56

Avenging Dentist: Oct 1, 2005; oh my god is that a circular saw that does not go in my mouth aaaaagh

^^ At least mine compiles. :colbert:

Contero posted:

I'd like to have a vector class that has 3 floats. I'd like to be able to access them either as an array: myvec.val[0] or as individual members: myvec.x. Is there a way to do this with unions or something?

Yes, through non-standard extensions supported in gcc and MSVC (stolen verbatim from one of my projects):

code:

struct vec2d
{
	union
	{
		struct
		{
			float x;
			float y;
		};
		float coord[2];
	};
};

# ? Apr 13, 2009 23:57

Contero: Mar 28, 2004

Sweet. Thanks guys.

# ? Apr 14, 2009 00:09

narfanator: Dec 16, 2008

Avenging Dentist posted:

I am assuming you mean for both of the functions to be named foo because there's nothing wrong with your code now (except that you can't return a value from a void function, and that your second definition has an extra semicolon).

In that case, foo(1) is ambiguous. There are lots of ways to resolve this, but the real answer is that you probably shouldn't be using default parameters like that. Any more than one default per function is just annoying to use.

Boost.Parameter probably resolves the core issue you have, at the expense of some hilarious templates.

Yes, yes, and those aren't hilarious, they're kinda hideous. (Not as bad as the same for policy class templates!)

It technically resolves the core issue, but not nicely enough to warrant use. Thanks for the info, tho.

# ? Apr 14, 2009 01:37

Great Britain: Dec 17, 2005; & Northern Ireland

I have an SDL app with loads of bitmaps that I want to to include in the .exe and I'm using Dev-C++. Does anyone know where I can find a simple guide to writing resource files?

# ? Apr 14, 2009 05:10

Avenging Dentist: Oct 1, 2005; oh my god is that a circular saw that does not go in my mouth aaaaagh

Please don't use Dev-C++, it is an absolutely terrible IDE. If you're hell-bent on not using Visual Studio, use windres to compile the .rc files.

# ? Apr 14, 2009 05:14

Madoushi: May 9, 2003; Some days, you just get up on the wrong side of the bed...

Can anyone tell me what the problem with this code is?

code:

#include <iostream>
#include <string>
using namespace std;

class myException{

	public:
	myException(string exType){cout << "ERROR: "<< exType;}
};

int main(){
throw myException("TEST");
return 0;
}

I can compile and run it, but after it prints the message, I get this error:

9 [sig] extest 236 _cygtls::handle_exceptions: Error while dumping state (probably corrupted stack)
Segmentation fault (core dumped)

# ? Apr 14, 2009 08:20

awesmoe: Nov 30, 2005; Pillbug

Madoushi posted:

Can anyone tell me what the problem with this code is?
code:
#include <iostream>
#include <string>
using namespace std;

class myException{

	public:
	myException(string exType){cout << "ERROR: "<< exType;}
};

int main(){
throw myException("TEST");
return 0;
}
I can compile and run it, but after it prints the message, I get this error:

9 [sig] extest 236 _cygtls::handle_exceptions: Error while dumping state (probably corrupted stack)
Segmentation fault (core dumped)

What do you expect it to do?

edit: what does the backtrace from the core file tell you?

# ? Apr 14, 2009 08:44

agscala: Jul 12, 2008

I'm trying to make a simple roguelike on windows using Visual Studio 2008. I want to use pdcurses for console manipulation. I have the .h files, the .lib, and the .dll. I added the .h files and the .lib to the project preferences, but how to I get it to use the .dll?

currently all I have is:

code:

#include "stdafx.h"
#include <curses.h>

int main(int argc, char *argv[])
{
	initscr();
	return 0;
}

and the error I get is:

code:

1>curses.obj : error LNK2019: unresolved external symbol _initscr referenced in function _main
1>C:\Users\Andrew Scala\Documents\Visual Studio 2008\Projects\curses\Debug\curses.exe : fatal error LNK1120: 1 unresolved externals

# ? Apr 14, 2009 19:02

Mustach: Mar 2, 2003; In this long line, there's been some real strange genes. You've got 'em all, with some extras thrown in.

If you're getting a link error at build time, it's not because the DLL can't be found, it's because you're not linking with the lib. Did you add the lib's path to the "additional places to look for libraries"* setting in Project Preferences, in addition to adding the name of the lib to "additional library dependencies"?

* Something like that, I can never remember

# ? Apr 14, 2009 19:21

agscala: Jul 12, 2008

I believe so:

Click here for the full 972x789 image.

# ? Apr 14, 2009 19:39

csammis: Aug 26, 2003; Mental Institution

Did you explicitly add the library to the Linker -> Input properties?

# ? Apr 14, 2009 20:35

Kamikaze!: Jul 10, 2001; On probation for postcount++

Go to linker->input and add the .lib to Additional Dependencies.

# ? Apr 14, 2009 22:10

agscala: Jul 12, 2008

I've been playing with this for a good hour and a half straight and I'm getting no luck. I've put all of the libraries straight into the visual studio 9 folder in the appropriate places even. Maybe I'll just abandon using this and find some other way to control the console.

# ? Apr 14, 2009 23:44

Mustach: Mar 2, 2003; In this long line, there's been some real strange genes. You've got 'em all, with some extras thrown in.

All you should need are curses.lib in Configuration Properties > C/C++ > Linker > Input > Additional Dependencies, and the path in Linker > General > Additional Library Directories. I can think of three things that are wrong, in order of rapidly decreasing likelihood: 1) The path needs quotes around it because it has spaces, 2) initscr() isn't a curses function and you didn't define it yourself, 3) curses.lib doesn't reference initscr().

# ? Apr 15, 2009 00:00

agscala: Jul 12, 2008

Bah, I'll just let it go. I recently found libtcod which does everything that I want and more, and I got it running with no problems. :shobon:

Thanks for the help though.

# ? Apr 15, 2009 01:13

huge sesh: Jun 9, 2008

I think I've been coding too long and I'm going crazy but for some reason this isn't working

code:

queue<int> q;
int x = 1;
q.push(x); //std::bad_alloc

Depending on chance I either get:

First-chance exception at 0x777f0ab0 in mapmaker.exe: 0xC0000005: Access violation reading location 0x00d43c42.
First-chance exception at 0x75b7f35f in mapmaker.exe: Microsoft C++ exception: std::bad_alloc at memory location 0x0017dfd0..
Unhandled exception at 0x75b7f35f in mapmaker.exe: Microsoft C++ exception: std::bad_alloc at memory location 0x0017dfd0..

or a message about heap corruption. It's almost exactly the same as the cplusplus.com example std::queue usage and it's making me feel like an idiot that I can't use a drat stl container right.

# ? Apr 15, 2009 05:54

POKEMAN SAM: Jul 8, 2004

huge sesh posted:

I think I've been coding too long and I'm going crazy but for some reason this isn't working
code:
queue<int> q;
int x = 1;
q.push(x); //std::bad_alloc
Depending on chance I either get:

First-chance exception at 0x777f0ab0 in mapmaker.exe: 0xC0000005: Access violation reading location 0x00d43c42.
First-chance exception at 0x75b7f35f in mapmaker.exe: Microsoft C++ exception: std::bad_alloc at memory location 0x0017dfd0..
Unhandled exception at 0x75b7f35f in mapmaker.exe: Microsoft C++ exception: std::bad_alloc at memory location 0x0017dfd0..

or a message about heap corruption. It's almost exactly the same as the cplusplus.com example std::queue usage and it's making me feel like an idiot that I can't use a drat stl container right.

Works fine for me:
http://codepad.org/9JxpwM65

# ? Apr 15, 2009 07:02

Mustach: Mar 2, 2003; In this long line, there's been some real strange genes. You've got 'em all, with some extras thrown in.

Is that all your program is doing? Or are you allocating lots of memory elsewhere? Did you disable your page file or something wacky like that? Do you get the same error when you try to use other containers, like vector?

# ? Apr 15, 2009 15:30

Mach Won: Jun 17, 2004; Is UNC playing? My post is either whining about (1) ref favoritism, (2) unfair recruiting/player caliber, (3) opposing team sucking or (4) the inevitable UNC win. The day you see me giving UNC credit for anything is the day someone hacked my account.

Roy era: 1-16 RIVALS!!!!!

I use Code::Blocks just because it's pretty straight forward. I found Visual Studio to be overly complicated.

# ? Apr 15, 2009 17:13

Vinterstum: Jul 30, 2003

Mach Won posted:

I use Code::Blocks just because it's pretty straight forward. I found Visual Studio to be overly complicated.

The code you posted is fine, there'll be something that's corrupting the heap preceding it.

# ? Apr 15, 2009 17:36

Plank Walker: Aug 11, 2005

I'm not all that familiar with C, but while working on a grad-level assignment came across something weird. The assignment was filling in a couple dummied out functions in a c program and we were given a bunch of pre-written functions to call. The source code was all written by the professor in a format I had never seen:

code:

void someFunction(arg1, arg2, arg3)
type1 arg1;
type2 arg2;
type3 arg3;
{
// Function here
}

Every function call worked fine, except for one:

code:

void somefunction(anInt, aFloat)
int anInt;
float aFloat;
{
    someGlobalFloat += aFloat;
}

No matter how I passed the float value to the function, aFloat never got set to a value. Debugging showed that immediately before my call to somefunction it was undefined, and immediately after it was a double with a garbage value like 3.90898089097e-293. I "fixed" the problem by rewriting the function definition as

code:

void somefunction (int anInt, float aFloat)

and it worked just fine, but I'd really like to know what "feature" of C made it behave the way it did when every other function written in the author's format worked normally.

# ? Apr 15, 2009 17:42

TOO SCSI FOR MY CAT: Oct 12, 2008; this is what happens when you take UI design away from engineers and give it to a bunch of hipster art student "designers"

Plank Walker posted:

I'm not all that familiar with C, but while working on a grad-level assignment came across something weird. The assignment was filling in a couple dummied out functions in a c program and we were given a bunch of pre-written functions to call. The source code was all written by the professor in a format I had never seen:

Your professor is using traditional C. If I remember correctly, float parameters are automatically converted to doubles in some cases. Is the procedure defined before any code that calls it?

# ? Apr 15, 2009 18:03

Plank Walker: Aug 11, 2005

Janin posted:

Your professor is using traditional C. If I remember correctly, float parameters are automatically converted to doubles in some cases. Is the procedure defined before any code that calls it?

The function prototype is written as somefunction(int, float);. I've sent it the actual parameter as

code:

20
20.0
20.0f
(float) 20.0

tried declaring a const float and sending that, and basically every format I could think of. The value that the function actually used as its parameter differed every time, making me think it allocated room for a double and half of it was overwritten with the float I was passing. Casting float on the value I was given always yielded 0.0.

# ? Apr 15, 2009 18:36

Mustach: Mar 2, 2003; In this long line, there's been some real strange genes. You've got 'em all, with some extras thrown in.

Janin posted:

Your professor is using traditional C.

"K&R" C is not a tradition. I would be suspicious of the quality of any compiler that quietly accepts it, and extremely suspicious of the quality of any man that willingly writes in it.

# ? Apr 15, 2009 18:43

TOO SCSI FOR MY CAT: Oct 12, 2008; this is what happens when you take UI design away from engineers and give it to a bunch of hipster art student "designers"

Plank Walker posted:

The function prototype is written as somefunction(int, float);. I've sent it the actual parameter as
[...]
tried declaring a const float and sending that, and basically every format I could think of. The value that the function actually used as its parameter differed every time, making me think it allocated room for a double and half of it was overwritten with the float I was passing. Casting float on the value I was given always yielded 0.0.

You could always just ask your professor -- it works fine on my system, so it might be a compiler bug.

Mustach posted:

"K&R" C is not a tradition. I would be suspicious of the quality of any compiler that quietly accepts it, and extremely suspicious of the quality of any man that willingly writes in it.

waaaah compilers aren't modified to gratuitously reject perfectly valid code waaah

# ? Apr 15, 2009 19:11

Mustach: Mar 2, 2003; In this long line, there's been some real strange genes. You've got 'em all, with some extras thrown in.

Yep I'm cryin' my eyes out here.

# ? Apr 15, 2009 19:20

Glasgerion: Jul 25, 2007; Strike on strike on

Does anyone know of any function or other way to have a program resize the terminal window it's running in?

# ? Apr 15, 2009 19:23

POKEMAN SAM: Jul 8, 2004

Glasgerion posted:

Does anyone know of any function or other way to have a program resize the terminal window it's running in?

That doesn't sound very portable to you, now does it?

# ? Apr 15, 2009 19:34

Glasgerion: Jul 25, 2007; Strike on strike on

Ugg boots posted:

That doesn't sound very portable to you, now does it?

Are you saying there's no good way to do it?

# ? Apr 15, 2009 20:12

huge sesh: Jun 9, 2008

Glasgerion posted:

Does anyone know of any function or other way to have a program resize the terminal window it's running in?

libncurses?

# ? Apr 15, 2009 20:31

huge sesh: Jun 9, 2008

Mustach posted:

Is that all your program is doing? Or are you allocating lots of memory elsewhere? Did you disable your page file or something wacky like that? Do you get the same error when you try to use other containers, like vector?

It's certainly not all the program's doing but it's really not that complicated. I don't allocate that much memory, page file is normal. std::vector works fine elsewhere in the program. I guess I'll look around for possible heap corruption. Just seems weird that it would happen every time.

# ? Apr 15, 2009 20:34

Adbot: ADBOT LOVES YOU

# ? Apr 29, 2024 18:27

Lexical Unit: Sep 16, 2003

Glasgerion posted:

Does anyone know of any function or other way to have a program resize the terminal window it's running in?

How you'd resize the window depends on a million factors. What OS are you using? Assuming *NIX, what Window Manager are you using? It may also depend on what Terminal application you want to resize. Will you interact with other applications the user may have running? It also doesn't have anything to do with C/C++.

Just for grins, let's assume you're using Linux with the Metacity WM. You might be able to pull it off like this:

code:

// Somehow find the X Window ID of your terminal window, this is left as an exercise to the reader.
Window window = magic (); // the X Window ID of your terminal window
int new_widget, new_height;
XWindowChanges changes;
changes.width = new_width;
changes.height = new_height;
XConfigueWindow (display, window, CWidth | CHeight, &changes);
XFlush (display);

Of course all that code presupposes that you've opened a X Display correctly and have code in place to handle spurious BadMatch, BadValue, or BadWindow errors.

Like huge sesh said, you might also be able to leverage ncurses's resizeterm() function. It all depends on what you mean by "resize the terminal window" and what environment you're working with.

# ? Apr 15, 2009 20:46

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Cavern of COBOL > C/C++ Programming Questions Not Worth Their Own Thread

«‹›639 »