Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

Control Flow

2 (2.2%)

Arrays, Strings, Pointers, and References

8 (9%)

Functions

4 (4.5%)

Basic object-oriented programming

30 (33.7%)

A bit more advanced OOP (Composition, Operator overloading, Inheritance, Virtual Functions)

18 (20.2%)

Templates

8 (9%)

Other (Explain)

4 (4.5%)

Working with files?  (Streams)

15 (16.9%)

Total Members Voted: 89

[olemars]

You generally don't need the extern keyword for anything in C++, just FYI.

[ILikePie]

a makefile usually (but doesn't have to) list each .o file as a target with its corresponding .c file as a requirement. if the .c has been modified since the .o was last built, it runs commands (usually just gcc, with a flag to stop before linking) to update it. the default target is generally the executable, which depends on all of the .o files

I think that mostly covers it... makefiles can be used for a number of other things too

A Makefile consists of "rules" which are a series of shell commands to build your application. Instead of manually running gcc a couple of times, you can use make to do it for you.

[Rose]

You generally don't need the extern keyword for anything in C++, just FYI.

you do for global variables, and that's about it.

[olemars]

And you shouldn't be using any global variables in C++.

[Rose]

And you shouldn't be using any global variables in C++.

I know, I'm sorry.

[Siquo]

And you shouldn't be using any global variables in C++.

I know, I'm sorry.

Yeah, you better be!

[Alexhans]

And you shouldn't be using any global variables in C++.

I know, I'm sorry.

Give me 10 hail Stroustrup's for penance.

[ILikePie]

More linked list fun. My linked list breaks when setting it to circulate (i.e. linking the first and last items together).

Code: [Select]

#include <stdio.h>
#define LIST  _Node

typedef struct __Node {
     int data;
     struct __Node *next;
} _Node;

LIST navigate (LIST *l, int i)
{
     LIST ret = *l;
     while(i-- > 0)
         if (ret.next != NULL) ret = *(ret.next);
     return ret;
}

void link (LIST *l, LIST *m) /* insert 'm' after 'l' */
{
     m->next = (l->next == NULL) ? NULL : l->next;
     l->next = m;
}


int main ()
{
     LIST a, b, c;
     a.data = (int)'a'; b.data = (int)'b'; c.data = (int)'c';
     link(&a, &b); link(&b, &c); link(&c, &a);

     printf("%d %d %d\n",
        a.data,
        navigate(&a, 1).data,
        navigate(&a, 2).data
      );
     return 0;
}
The correct output should be "97 98 99", but for some reason I get "97 1330556 1330556".

[TolyK]

And you shouldn't be using any global variables in C++.

I know, I'm sorry.

Give me 10 hail Stroustrup's for penance.

Have you read the (thick) book by him on C++?
...
I'm still reading 

[Shades]

Code: [Select]

void link (LIST *l, LIST *m) /* insert 'm' after 'l' */
{
     m->next = (l->next == NULL) ? NULL : l->next;
     l->next = m;
}


I haven't run this code so I might be missing something but it seems to me this will mean the third link statement will actually modified a's next as well to point at whatever c's next was set to.
If you running in debug c's next pointer will have been defaulted to NULL (depending on your compiler) but if your not it would have been random memory as you don't initialise any of the next values.

Also m->next = (l->next == NULL) ? NULL : l->next; is the same as typing m->next = l->next;
if (a = null) then b = null else b = a either way.

[Siquo]

What Shades said. *next is uninitialised, and therefore not-NULL. Make a constructor.

[qwertyuiopas]

The struct naming is a bit odd. Judging by the typedef, I would guess that that is C, and in C, "struct foo" and "foo" can be entirely different types (bad idea, though. It would be extremely confusing). typedef struct a {...} a; is valid, and would result in both a "struct a" and a "a" being types that mean the same thing.

Furthermore, link() inserts m between l and the node following l, so before you do any of the links, setting a.next to &a means that the list will automatically remain circular as nodes are inserted. Alternatively, directly setting the last node to have the first one as .next would also work. However, link() is not designed to re-insert a node, so it can't be used to link the last node to the first node properly, instead it would basically move the first node to the end of the list, and leave the list non-circular, or, if used on a node mid-list, you get something like a backwards tree.

[ILikePie]

The struct naming is a bit odd. Judging by the typedef, I would guess that that is C, and in C, "struct foo" and "foo" can be entirely different types (bad idea, though. It would be extremely confusing). typedef struct a {...} a; is valid, and would result in both a "struct a" and a "a" being types that mean the same thing.

I didn't know that, though, does it really matter? I mean, I wont be using struct __Node or struct _Node anywhere anyway.

e, Creating a constructor and modifying link did it.

Code: [Select]

...

LIST node () /* list node constructor */
{
     LIST l;
     l.next = NULL;
     return l;
}
...
void link (LIST *l, LIST *m) /* insert 'm' after 'l' */
{
     if (l->next != NULL) m->next = l->next;
     l->next = m;
}

int main ()
{
     LIST a = node(), b = node(), c = node();
     a.data = (int)'a'; b.data = (int)'b'; c.data = (int)'c';
     link(&a, &c);
     link(&a, &b);
     link(&c, &a);
     
     printf("%c->%c->%c->%c\n",
    a.data,
    navigate(&a, 1).data,
    navigate(&a, 2).data,
    navigate(&a, 3).data
  );
     return 0;
}

------

$ ./list
a->b->c->a


[bowdown2q]

Damn, shame I found this thread late at night >_o. I've been trying to code a game in VB and keep getting hung up. I'll read through this thread tomorrow, but for now here's a link to my current project http://www.bay12forums.com/smf/index.php?topic=81060.0 and I'll post to watch and remind myself this exists.

[lordnincompoop]

And you shouldn't be using any global variables in C++.

I know, I'm sorry.

Give me 10 hail Stroustrup's for penance.

Have you read the (thick) book by him on C++?
...
I'm still reading 

Mmm, same here.

I'm finding it really useful, though. Good purchase.