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 »

Mr.Mojo: Sep 11, 2001; Nascita di Venere

I'm doing a problem that asks me to implement logical negation, !, without using !. My idea is to use "~x+1", but the tester says this is incorrect. What am I doing wrong?

# ? Sep 29, 2008 00:13

Adbot: ADBOT LOVES YOU

# ? May 7, 2024 03:42

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

Comedy x ? false:true option.

# ? Sep 29, 2008 00:22

rjmccall: Sep 7, 2007; no worries friend; Fun Shoe

Mr.Mojo posted:

I'm doing a problem that asks me to implement logical negation, !, without using !. My idea is to use "~x+1", but the tester says this is incorrect. What am I doing wrong?

Do you understand the semantics of logical negation? Your proposal doesn't satisfy them for... well, for anything.

# ? Sep 29, 2008 01:10

Mr.Mojo: Sep 11, 2001; Nascita di Venere

rjmccall posted:

Do you understand the semantics of logical negation? Your proposal doesn't satisfy them for... well, for anything.

Oh duh, I was doing a normal negate, not a logical negate. Back to thinking about it!

# ? Sep 29, 2008 01:45

Dire Penguin: Jun 17, 2006

I'm fairly new and starting to work a lot with derived classes. I've got a main class with a virtual function that should never be called, and then several derived classes each with their own different versions of said function that should always be called instead.

But, when I go through my queue calling my function on each class, it always calls the main class's function, not the derived class. Is there something simple that I'm missing?

# ? Sep 29, 2008 03:18

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

Post some code.

# ? Sep 29, 2008 03:20

Dire Penguin: Jun 17, 2006

Virtual function in the main class:

code:

virtual int hit(P,P,P&, P&) {printf("failure\n");};

Actual function in a derived class:

code:

int hit(P, P, P&, P&);

Declaring a new instance of the derived class & adding it to queue, where the templated type is the main class, Shape:

code:

template <class T>
void declare(Queue<T> *q){
Derivd newball(amb, mir, trans, azi, elev, roll, diff, centr, .25);
q->add(newball);
}

Calling the function from a linked list, which calls the virtual hit function instead of the Derivd:

code:

template <class T>
Shape* hitter(P sp, P dp, Queue<T> q){
P spot, tee;
Node<T>* walk = q->getfront(); //gets the front of the queue
walk->getdata()->hit(sp, dp, spot, t);

getdata() returns a pointer to whatever data is stored in the queue, which should be the Derivd newball I stuck in there earlier with q->add. Node contains a pointer to the next Node, and to data stored, getfront() would set this to be the very first Node in the queue.

# ? Sep 29, 2008 03:44

cronio: Feb 15, 2002; Drifter

Since you're storing T in your queue, and not T*, what's happening is that you're actually storing a *copy* of the object you're adding. And because the queue is actually of type Point, it's using the copy constructor to create a new Point object, not a derived one, and copy just the Point data from the derived object you're adding.

You need to store pointers inside your queue, and allocate the objects on the heap:

code:

template <class T> 
void declare(Queue<T> *q){ 
  Derivd* newball = new Derivd(amb, mir, trans, azi, elev, roll, diff, centr, .25); 
  q->add(newball); 
}

And of course don't forget to delete them when they are removed from the queue.

cronio fucked around with this message at 05:31 on Sep 29, 2008

# ? Sep 29, 2008 04:07

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.

Dire Penguin posted:

Queue<T>

Is your Queue holding Shapes instead of Shape*s? If so, you're not going to see polymorphic behavior because the queue isn't holding any Deriveds. Here's an ultraquick overview:

code:

Derived d(/* whatever */);
Shape s = d; // okay, copy the Shape members of d to s
s.hit(p,p,p,p); // uses the compile-time type of s, which is Shape
Shape *sp = &d;
sp->hit(p,p,p,p); // uses the run-time type of *sp, which is Derived
Shape& sr = d;
sr.hit(p,p,p,p); // uses the run-time type of the object reference by sr, which is Derived

Long story short, the member has to be called via either a pointer or a reference if you want polymorphic behavior, and assigning a derived object to a base object will* just copy the base class's members from the derived object to the base object. A book or the C++ FAQ can give you more detail.

* depending on how the copy constructor is defined, of course

# ? Sep 29, 2008 04:09

Dire Penguin: Jun 17, 2006

Awesome, that's exactly what I was doing. I went back through and changed the Queue to store T* and changed my declarations and the error goes away, as far as I can tell.

# ? Sep 29, 2008 04:36

cronio: Feb 15, 2002; Drifter

Whoops, noticed I had a slight typo... I didn't mean to have Queue<T*> in there.

What you really want to do is keep the queue holding T, but instantiate it with Shape* instead of Shape:

code:

Queue<Shape*> my_queue;

This lets it work for any kind of object instead of just pointers.

# ? Sep 29, 2008 05:34

JoeNotCharles: Mar 3, 2005; Yet beyond each tree there are only more trees.

Dire Penguin posted:

Virtual function in the main class:
code:
virtual int hit(P,P,P&, P&) {printf("failure\n");};

If you make this function a pure virtual instead, it will 100% guarantee that the base version will never be called, because it doesn't even exist:

code:

virtual int hit(P,P,P&, P&) = 0;

(You also won't even be able to create an instance of Shape, because the compiler will complain that it has pure virtual functions, which'll mean the problem with creating new Shape objects in Queue<Shape> instead of just copying the points would show up at compile time instead of at runtime. This may not be what you want if you actually do want to create a Shape object instead of a derived object at some point.)

# ? Sep 29, 2008 06:03

floWenoL: Oct 23, 2002

JoeNotCharles posted:

If you make this function a pure virtual instead, it will 100% guarantee that the base version will never be called, because it doesn't even exist:

That's not true.

# ? Sep 29, 2008 09:30

JoeNotCharles: Mar 3, 2005; Yet beyond each tree there are only more trees.

floWenoL posted:

That's not true.

It is if you consider crashing with an error to be "not calling" it.

# ? Sep 29, 2008 14:52

That Turkey Story: Mar 30, 2003

JoeNotCharles posted:

It is if you consider crashing with an error to be "not calling" it.

He means that pure virtual functions can be defined and called. All that pure virtual implies is that the child should specify an implementation and makes the base type abstract, not that you can't also define the function in the base and explicitly call that implementationt. Case in point, pure virtual destructors, which in that case must be defined and are always called during destruction.

That Turkey Story fucked around with this message at 20:56 on Sep 29, 2008

# ? Sep 29, 2008 20:53

JoeNotCharles: Mar 3, 2005; Yet beyond each tree there are only more trees.

Huh, I didn't know that. Yet another example of why C++ is so hard to master!

Anyway, it will 100% guarantee that the base version will never be called automatically - is that better?

# ? Sep 29, 2008 21:11

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.

JoeNotCharles posted:

Anyway, it will 100% guarantee that the base version will never be called automatically - is that better?

Like That said, not true in the case of destructors.

# ? Sep 29, 2008 22:40

JoeNotCharles: Mar 3, 2005; Yet beyond each tree there are only more trees.

Dire Penguin posted:

Virtual function in the main class:
code:
virtual int hit(P,P,P&, P&) {printf("failure\n");};

Yeah, C++ is supposed to have a feature that prevents this happening, but it doesn't loving work. Have fun!

# ? Sep 29, 2008 23:12

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.

It isn't supposed to have any such feature. If it's been advertised to you as such, you've been misled.

edit: I mean, I'm not blaming you for thinking that; that's just not what it was designed for.

Mustach fucked around with this message at 23:39 on Sep 29, 2008

# ? Sep 29, 2008 23:32

JoeNotCharles: Mar 3, 2005; Yet beyond each tree there are only more trees.

Uh, I'm pretty sure it's for exactly this situation - you have an abstract base class (Shape), and you want to always call the member from a subclass and not the abstract base class.

The issue is that C++ is so fiddly that I can't give a one-line description of pure virtual methods because, surprise, there are confusing corner cases. A pure virtual destructor isn't at all related to this situation because I wasn't advising him to use a destructor, but it still makes what I said incomplete and in C++ that's dangerous because there are so many features that interact that you have to be a huge rules lawyer to understand how they work.

I mean, in any other language I'd say, "Pfft, you're being pedantic. Obviously I wasn't talking about destructors, and obviously if you actively try to make your pure virtual callable by implementing it after you declare it than you deserve what you get." But in C++ that kind of pedantry is actually necessary and admirable.

# ? Sep 30, 2008 00:16

more falafel please: Feb 26, 2005; forums poster

code:

#if DEBUG
#define PURE_VIRTUAL { assert(false, "Pure virtual called"); }
#else
#define PURE_VIRTUAL =0

class foo {
   void bar() PURE_VIRTUAL;
}

# ? Sep 30, 2008 02:35

JoeNotCharles: Mar 3, 2005; Yet beyond each tree there are only more trees.

So... you want it to be harder to find errors in debug mode?

# ? Sep 30, 2008 03:00

newsomnuke: Feb 25, 2007

That Turkey Story posted:

Case in point, pure virtual destructors, which in that case must be defined and are always called during destruction.

I've read that pure virtual destructors should always be empty, but can't think of a reason for this?

# ? Sep 30, 2008 14:54

Hubis: May 18, 2003; Boy, I wish we had one of those doomsday machines...

ultra-inquisitor posted:

I've read that pure virtual destructors should always be empty, but can't think of a reason for this?

There is no reason, they should do whatever you need them to do.

# ? Sep 30, 2008 20:35

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

I recall there being an idiom in C++ for "template polymorphism", but I'm probably calling it by the wrong name, since Google isn't much help. I think TTS posted it in #cobol, if that helps any.

Does anyone know what I mean?

# ? Sep 30, 2008 20:50

ehnus: Apr 16, 2003; Now you're thinking with portals!

Avenging Dentist posted:

I recall there being an idiom in C++ for "template polymorphism", but I'm probably calling it by the wrong name, since Google isn't much help. I think TTS posted it in #cobol, if that helps any.

Does anyone know what I mean?

The curiously recurring template pattern?

# ? Sep 30, 2008 22:48

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

ehnus posted:

The curiously recurring template pattern?

Nah, but that's a fun one.

The idea was that it worked somewhat like a virtual functor, but there was some template magic in place of the virtual function call. I don't even know if I'd have any use for it, but I was thinking about it the other day and couldn't find it again.

EDIT: it looked something like this

code:

class something_cool
{
public:
    template<typename T> something_cool(const T &whatever);
private:
    // Some weird template member stuff here
};

Avenging Dentist fucked around with this message at 23:10 on Sep 30, 2008

# ? Sep 30, 2008 23:05

That Turkey Story: Mar 30, 2003

I don't know if there is a special name for it, I think the common term is just vaguely "static polymorphism" which could apply to a number of different things. You do generally use the curiously recurring template pattern to do it, however.

code:

#include <cstddef>

template< typename ChildType >
struct container
{
  std::size_t size() const { return static_cast< ChildType const& >( *this ).size(); } 

  // etc.
};

struct vector
  : container< vector >
{
  std::size_t size() { return 1; }
};

template< typename ChildType >
void function_expecting_a_container( container< ChildType >& a )
{
  a.size(); // in practice you actually use it
}

You can also do some compile-time magic to make sure the child member function is overwritten.

# ? Oct 1, 2008 03:04

BigRedDot: Mar 6, 2008

Also seen it called "compile-time polymorphism".

# ? Oct 1, 2008 15:08

emf: Aug 1, 2002

I'm just getting started with gcc's built-in vector extensions, and I want to perform the following operation on a vector:

code:

c = 0.5*(a + b);

where a, b, and c are all vectors, but I get a compiler error saying

code:

error: can't convert value to a vector

The only way I can get it to work is by creating a vector "half" where each element is equal to 0.5, and writing the operation as:

code:

c = half*(a + b);

This is not ideal, but I'm having a lot of difficulty finding a good tutorial or documentation on how to do this correctly.

I'm kind of stuck with gcc 4.1.2 as I'm on Debian stable, but if this is "fixed" in later versions I might have reason enough to lobby for a version update.

# ? Oct 1, 2008 15:28

ehnus: Apr 16, 2003; Now you're thinking with portals!

That's correct because the concept of vector immediates generally doesn't exist. You'll need to splat that value across a vector variable before you can use it

# ? Oct 1, 2008 16:53

emf: Aug 1, 2002

That's poopy.

I wanted to use very large vectors for readability (and a few other minor reasons), and there are a minimum of 5 different constant values I need to use (depending on the vector). That's going to eat a ton of memory.

# ? Oct 1, 2008 17:11

Soldat: Jan 22, 2008; my name is soldat and I get respect, your cash and your jewelery is what I expect

Intro c++ here...

I'm trying to write a code that simply copies one word in a char array over to another, using pointers. I think I'm pretty close, but when I cout << str2 in my main function, all that comes out is garbage for some reason, even though the correct output comes when I cout << str2 in my copyString function. Any hints would be nice.

code:

#include<iostream>
using namespace std;

void copyString(char * str2, const char * str1){

 const char * origin = str1;

 int size = 0;
 while(*str1 != '\0')
   {size++;
     str1++;}

 str2 = new char [size+1];

 str1 = origin;

 cout << "size= " << size << endl;

 for (int i=0;i<size;i++)
   {
     str2[i]=str1[i];
   }

 str2[size]='\0';

 cout << str2 << endl;

}

int main()
{
 char * str1 = "Hello!";
 char * str2;
 copyString(str2, str1);
 cout << "str1= " << str1 << endl;
 cout << "str2= " << str2 << endl;
 // delete [] str2;
 return 0;
}

Edit: I just changed it, but for some reason I accidentally changed str2 to str3 when I copied it over here, sorry about that

Soldat fucked around with this message at 00:28 on Oct 2, 2008

# ? Oct 2, 2008 00:14

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

~~Ignoring the fact that your code doesn't even compile (str3 should be str2)~~, you aren't allocating space for the destination string. You're just changing what str2 points to in copyString. All of this is secondary to the more important questions though: why aren't you just using std::string, and if you feel you absolutely MUST use char*s, why aren't you just using the functions in the C standard library, like strdup?

Furthermore, in 99% of situations (especially in C++), it's bad practice to allocate memory in a function and then expect the calling function to free that memory.

# ? Oct 2, 2008 00:21

Adbot: ADBOT LOVES YOU

# ? May 7, 2024 03:42

Soldat: Jan 22, 2008; my name is soldat and I get respect, your cash and your jewelery is what I expect

Sorry I should have clarified, I think we're only supposed to use char arrays. I can post the spec if that helps:

code:

Name:  copyString
2 Arguments:  char * and const char *, respectively, representing the destination and source strings.
   These variables may or may not be references!
Return: none
State Change:  arg2 points to an array of char containing the same characters as the array of char
   pointed to by arg1.

Also, could you clarify what you mean by saying I'm not allocating space for the destination string? I thought I was dynamically allocating space for it with the line:

str2 = new char [size+1];

EDIT: nevermind, thanks for the advice dentist. I changed where I was allocating space and tweaked a couple of other things, it works now.

Soldat fucked around with this message at 00:39 on Oct 2, 2008

# ? Oct 2, 2008 00:27

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

«‹›639 »

Mr.Mojo posted:

rjmccall posted:

code:

code:

code:

code:

code:

Dire Penguin posted:

code:

code:

Dire Penguin posted:

code:

code:

JoeNotCharles posted:

floWenoL posted:

JoeNotCharles posted:

JoeNotCharles posted:

Dire Penguin posted:

code:

code:

That Turkey Story posted:

more falafel please posted:

code:

code:

code:

ultra-inquisitor posted:

Avenging Dentist posted:

ehnus posted:

code:

code:

code:

code:

code:

code:

code: