---------------

FIDO MESSAGE AREA==> TOPIC: 203 C++           Ref: F5G00113 Date: 04/27/98
From: ADAM MAJER                                            Time: 07:31pm
\/To: MATT SARNECKY                                       (Read 3 times)
Subj: Need borland help fast!

MS>I am using Borland Turbo C++ 4.5 for Windows. (Which is compatible with
MS>Borland C++ 4.02) and I created an application called lworks that uses 
WCC
MS>controls. I have a dialog box called IDCDLGCLIENT which is a resource for 
th
Post the function that is trubling you. I'll try to help.
 * SLMR 2.1a * I'm in shape ... round's a shape isn't it?
--- FMail 0.92
 * Origin: The Programmer's Oasis on FIDONET! (1:348/203)
---------------

FIDO MESSAGE AREA==> TOPIC: 203 C++           Ref: F5G00114 Date: 04/28/98
From: KURT KUZBA                                            Time: 01:13am
\/To: MARK HOOVER                                         (Read 3 times)
Subj: Interrupt Based Programmi

MH> KK>   ODOOR or RS232. Either is available as source for a fee.
MH>   Know where to find them?
   http://www.filepile.com
> ]<urt   (NFX)
> Yoda I am, of Borg. Assimilated you will be, hmmm???........
---
 * Origin: *YOPS ]I[* 8.4 GIG * RA/FD/FE * Milwaukee, WI (1:154/750)
---------------

FIDO MESSAGE AREA==> TOPIC: 203 C++           Ref: F5G00115 Date: 04/28/98
From: KURT KUZBA                                            Time: 01:13am
\/To: GLEN OLDENHUIS                                      (Read 3 times)
Subj: vesatest.cpp  1/3

GO>   activate 256 colors?
   C++ code for activating VESA modes 101h and 103h in Borland.
   This is not optimized for speed. It is set out as an example
   of the general method for using these modes.
   I don't know about 1024x1024x16M, having never attempted it,
   but if you have a list of VESA modes, you could add it.
   Handling the pixels would not be as simple as with 256 colors,
   of course, but a 24-bit color depth would be 3 bytes per pixel.
   Unless you want to use photographic quality graphics, 256
   colors should be all that you need, really. Try the PC Game
   Programmer's Encyclopedia. [ PCGPE ]
/*_|_|   VESATEST.CPP   PUBLIC DOMAIN   PART 1 OF 3
_|_|_|   Example of using 800x600 and 640x480 VESA modes.
_|_|_|   No warrantee or guarantee.  Kurt Kuzba.  (8/24/1997)*/
#include <stdlib.h>
#include <conio.h>
#include <time.h>
#include <iostream.h>
class Vesamodes   {
public:
   Vesamodes(int);
   ~Vesamodes(void);
   void Dot(int, int, int);
   void Line(int, int, int, int, int);
   void Circle(int, int, int, int, int, int, int);
private:
   void Disk(int, int, int r, int c, int hg, int wd);
   int High, Wide, Color;
};
Vesamodes::Vesamodes(int mode = 0)
{
   switch(mode)
   {
      case 800: High = 600; Wide = 800; mode = 0x103; break;
      default:  High = 480; Wide = 640; mode = 0x101; break;
   }
   Color = 7;
   _asm {
      mov ax, 4F02h;
      mov bx, mode;
      int 10h;
   }
}
void Vesamodes::Dot(int x, int y, int c)
{
   long po, pa, px = (long)x, py = (long)y;
   int o;
   static int a = 0;
   if(x < 0 || x >= Wide || y < 0 || y >= High) return;
   switch(Wide)
   {
      case 800: po = pa = (py << 9) + (py << 8) + (py << 5) + px;
         break;
      default: po = pa = (py << 9) + (py << 7) + px; break;
   }
   po &= 0xffffL;
   pa /= 0x10000L;
   o = (int)po;
   if(a != (int)pa)
   {
      a = (int)pa;
      _asm {
         mov ax, 4F05h;
         xor bx, bx;
         mov dx, a;
         int 10h;
      }
   }
   _asm{
      mov ax, 0A000h;
      mov di, o;
      mov es, ax;
      mov ax, c;
      stosb;
   }
}
/*_|_|   end   VESATEST.CPP   PUBLIC DOMAIN   PART 1 OF 3   */
> ]<urt   (NFX)
> God's side of the tire is Never Flat........................
---
 * Origin: *YOPS ]I[* 8.4 GIG * RA/FD/FE * Milwaukee, WI (1:154/750)
---------------

FIDO MESSAGE AREA==> TOPIC: 203 C++           Ref: F5G00116 Date: 04/28/98
From: KURT KUZBA                                            Time: 01:13am
\/To: GLEN OLDENHUIS                                      (Read 3 times)
Subj: vesatest.cpp  2/3

/*_|_|   VESATEST.CPP   PUBLIC DOMAIN   PART 2 OF 3
_|_|_|   Example of using 800x600 and 640x480 VESA modes.
_|_|_|   No warrantee or guarantee.  Kurt Kuzba.  (8/24/1997)*/
/*
   Circle is adapted from Bresnham.C in Bob Stout's SNIPPETS
*/
void Vesamodes::Circle(int xc, int yc, int r,
   int c, int WD = 0, int HG = 0, int dsk = 0)
{
   int x = 0, d = (1 - r) << 1, w;
   HG = HG ? HG : High;
   WD = WD ? WD : Wide;
   w = (WD / HG) << 1;
   if(dsk)
   {
      Disk(xc, yc, r, c, HG, WD);
   }
   else
   {
      while(r >= 0)
      {
         Dot(xc + x, yc + r, c);
         Dot(xc + x, yc - r, c);
         Dot(xc - x, yc + r, c);
         Dot(xc - x, yc - r, c);
         if(d + r > 0)
            d -= (w * --r) - 1;
         if(x > d)
            d += ((++x) << 1) + 1;
      }
   }
}
void Vesamodes::Disk(int xc, int yc, int r, int c, int hg, int wd)
{
   int l, x = 0, d = (1 - r) << 1, w = (wd / hg) << 1;
   while(r >= 0)
   {
      for(l = xc - x; l <= xc + x; l++)
      {
         Dot(l, yc + r, c);
         Dot(l, yc - r, c);
      }
      if(d + r > 0)
         d -= (w * --r) - 1;
      if(x > d)
         d += ((++x) << 1) + 1;
   }
}
void Vesamodes::~Vesamodes(void)
{
   _asm{
      mov ax, 0x03;
      int 10h;
   }
}
int rnd(int m){return (int)(((long)rand()*((long)m+1)-1)/RAND_MAX);}
/*_|_|   end   VESATEST.CPP   PUBLIC DOMAIN   PART 2 OF 3   */
> ]<urt   (NFX)
> Their numbers are 664 and 668, the neighbors of the beast...
---
 * Origin: *YOPS ]I[* 8.4 GIG * RA/FD/FE * Milwaukee, WI (1:154/750)
---------------

FIDO MESSAGE AREA==> TOPIC: 203 C++           Ref: F5G00117 Date: 04/28/98
From: KURT KUZBA                                            Time: 01:13am
\/To: GLEN OLDENHUIS                                      (Read 3 times)
Subj: vesatest.cpp  3/3

/*_|_|   VESATEST.CPP   PUBLIC DOMAIN   PART 3 OF 3
_|_|_|   Example of using 800x600 and 640x480 VESA modes.
_|_|_|   No warrantee or guarantee.  Kurt Kuzba.  (8/24/1997)*/
/*
   Line is adapted from Bresnham.C in Bob Stout's SNIPPETS
*/
void Vesamodes::Line(int x, int y, int x2, int y2, int c)
{
   int i, steep = 0, sx, sy, dx, dy, e, ex, ey;
   dx = abs(x2 - x);
   sx = ((x2 - x) > 0) ? 1 : -1;
   dy = abs(y2 - y);
   sy = ((y2 - y) > 0) ? 1 : -1;
   if(dy > dx)
   {
      steep =  x;
      x =  y;
      y = steep;  /* swap  x and  y */
      steep = dx;
      dx = dy;
      dy = steep;  /* swap dx and dy */
      steep = sx;
      sx = sy;
      sy = steep;  /* swap sx and sy */
      steep = 1;
   }
   e = 2 * dy - dx;
   ex = dx << 1;
   ey = dy << 1;
   for(i = 0; i < dx; i++)
   {
      if(steep)
         Dot(y, x, c);
      else
         Dot(x, y, c);
      while(e >= 0)
      {
         y += sy;
         e -= ex;
      }
      x += sx;
      e += ey;
   }
   Dot(x2, y2, c);
}
void test(Vesamodes *V, int W = 640, int H = 480)
{
   while(!kbhit())
   {
      V->Line(rnd(W), rnd(H), rnd(W), rnd(H), rnd(255));
      V->Circle(rnd(W), rnd(H), rnd(200), rnd(255));
      V->Circle(rnd(W), rnd(H), rnd(200), rnd(255),
         10 + rnd(100), 10 + rnd(100));
      V->Circle(rnd(W), rnd(H), rnd(20), rnd(255),
         10 + rnd(100), 10 + rnd(100), 1);
   }
}
int main(void)
{
   time_t TM = time(NULL);
   srand((unsigned)TM);
   cout << "Hit any key to test 800x600x256" << endl;
   getch();
   Vesamodes *VM = new Vesamodes (800);
   test(VM, 800, 600);
   delete VM;
   cout << "Hit any key to test 640x480x256" << endl;
   getch();
   getch();
   VM = new Vesamodes;
   test(VM);
   delete VM;
   cout << "Hit any key to exit Vesatest.cpp" << endl;
   getch();
   getch();
   return 0;
}
/*_|_|   end   VESATEST.CPP   PUBLIC DOMAIN   PART 3 OF 3   */
> ]<urt   (NFX)
> Opossums on the Information Rustic Road.....................
---
 * Origin: *YOPS ]I[* 8.4 GIG * RA/FD/FE * Milwaukee, WI (1:154/750)
---------------

FIDO MESSAGE AREA==> TOPIC: 203 C++           Ref: F5G00118 Date: 04/26/98
From: TIM HUTZLER                                           Time: 04:10pm
\/To: GLEN OLDENHUIS                                      (Read 3 times)
Subj: Re: 256 color

GO>What command do you have to enter to activate 256 colors?
TH>That's a kind of a wide open question, isn't it?
TH>Are you lookin' for a VESA controls, or what?
GO>The lot.
You mean the whole kit-and-kaboodle?
Technically, there's a difference. [grin]
___ Blue Wave/QWK v2.12
--- Maximus/2 3.01
 * Origin: Madman BBS * Chico, California * 530-893-8079 * (1:119/88)
---------------

FIDO MESSAGE AREA==> TOPIC: 203 C++           Ref: F5G00119 Date: 04/26/98
From: TIM HUTZLER                                           Time: 04:27pm
\/To: DARIN MCBRIDE                                       (Read 3 times)
Subj: Re: Default constructor

DM>Now, in practical terms, this means that there are two types of
DM>default constructors: - those with no actual arguments, and -
DM>those with all default arguments.
TH>Incorrect.
TH>There is only one default constructor - no arguments. It doesnt even
DM>You seem to be misinterpreting what I said. We're talking about
DM>how it looks to the new C++ programmer
Hi Darin;
Maybe the discussion *should* be about how it looks to the developers
of C++, namely Benjemin what's his name. [grin]
DM>- in particular, how it looks(looked) to Rene. There are a lot of
DM>concepts in following the two "types" that need to be ironed out.
Yes, and it is important to get those concepts in correct
*perspective* and accurately defined. There are certain rules no
matter what level of programming one is at. And in C++ one of the
rules is that there is only *one* _default_ constructor, no
arguments, no returns. All other constructors are there to provide
the desired functionality.
DM>Since most C++ programmers have come from other languages, they
DM>may not be used to the two concepts required here: overloading and
DM>default arguments.
True, overloading is an uncommon but nifty feature. But not default
arguments - which, BTW, are not the same thing as a default
constructor. Default arguments (or parameters, more correctly) can
exist in *any* function whether it's a method to a class or not. A
default constructor *must* exist in every class. If it is not
specified (ie. one that takes no arguments), then the compiler will
create one - behind the programmers back.
DM>The easy default constructor to understand, the one that is coded
DM>such that it cannot take arguments, is best seperated conceptually
DM>from the more difficult one until the student can understand the
DM>concepts enough to unite them again.
Default constructors don't take arguments - never.
Try it. Create a class with *one* constructor having one or more
parameters. Specify an instance with no arguments. You will find that
it both compiles *and* runs. Reason, the default constructor that was
ommitted from the code was created by the compiler.
DM>JMNSHO.
???
I'm not up on this acronym. [grin]
cya around, friend.
___ Blue Wave/QWK v2.12
--- Maximus/2 3.01
 * Origin: Madman BBS * Chico, California * 530-893-8079 * (1:119/88)
---------------

FIDO MESSAGE AREA==> TOPIC: 203 C++           Ref: F5G00120 Date: 04/28/98
From: RENE HERMAN                                           Time: 02:59am
\/To: TIM HUTZLER                                         (Read 3 times)
Subj: Default constructor

Hello Tim ...
Tuesday April 21 1998, you wrote Rene Herman
 RH>> According to ANSI C++, will a constructor with nothing but default
 RH>> arguments do as a default constructor?
 TH> It *is* the default constructor.
Thanks, that was indeed what I was asking, sorry if I formulated badly.
 TH> Every class MUST have a default constructor. But, if a default
 TH> constructor (no args) is not specified in your code, the compiler
 TH> will create one for you. That's why there was no error.
But now you're confusing me again. Not only was there no error, but declaring 
a TTime object as "TTime Time;" really resulted in the compiler calling my 
constructor with all arguments set to -1, not in calling an automatically 
generated no argument constructor.
Furthermore, why must every class have a default constructor? When I never 
declare an object of the class without explicitly stating which constructor 
to use (ie only declare a TTime Time(1, 1, 1980, 0, 0);) there is no use for 
a default constructor, is there?
Also, with simple classes that don't declare any virtual functions, there's 
really nothing to construct. When I have a
  class TClass {
  private:
    int i;
  public:
    void Set(int);
  };
declaring a TClass object doesn't generate any code. For the following main 
function
  int main()
  {
    TClass Class;
    Class.Set(0);
    return 0;
  }
the compiler generated (part of -S output)
  _main proc near
     ;
     ; int main()
     ;
   enter 2,0
     ;
     ; {
     ;   TClass Class;
     ;
     ;   Class.Set(0);
     ;
   push 0
   lea ax,word ptr [bp-2]
   push ax
   call near ptr @TClass@Set$qi
   add sp,4
     ;
     ;   return 0;
     ;
   xor ax,ax
     ;
     ; }
     ;
   leave
   ret
  _main endp
[ sidenote: making member function Set virtual, the compiler *does* indeed 
insert an (inline) constructor, initializing the objects VMT pointer:
     ;
     ;    int main()
     ;
   enter   4,0
     ;
     ;    {
     ;      TClass Class;
     ;
   mov word ptr [bp-4],offset @@TClass@
     ;
     ;   Class.Set(0);
     ;
]
Or did I get you wrong?
Rene
---
 * Origin: postmaster@rene.demon.nl (2:282/1.11)
---------------

FIDO MESSAGE AREA==> TOPIC: 203 C++           Ref: F5G00121 Date: 04/28/98
From: RENE HERMAN                                           Time: 03:06am
\/To: DARIN MCBRIDE                                       (Read 3 times)
Subj: Default constructor

Hello Darin ...
Friday April 24 1998, you wrote Rene Herman
 DM> Your TTime objects encapsulates a real object (in this case, an
 DM> abstract object that humans believe we understand): time.
I see what you mean, and regret having to disagree with you. My TTime objects 
really didn't do anything remotely resembling encapsulating the real world 
concept of time, or even my notion of such a concept. What's the substance of 
time (its data members) and what influences it, that is, which methods should 
I provide? If I knew, I would have won a Nobel prize. :-)
I hope I'm not over-interpreting your comments, but in my humble opinion 
object orientation shouldn't be made out to be more than a way of organizing 
computer code with advantages such as clarity and modularity, and 
disadvantages such as size and speed of the resulting code. Although it at 
times might provide for a useful abstraction, I feel that most real world 
objects, let alone concepts such as time, are far too complex to be 
encapsulated by, or even described by, a set amount of data and functions to 
operate on that data.
 DM> Abstract objects or abstract classes?  I've never heard the term
 DM> 'abstract object' before.
You were the one introducing the phrase, I just copied it. :-) But I must 
admit I thought you were referring to abstract classes.
 DM> A class is merely a description of what an object is and what
 DM> messages can be sent to it.
I just browsed the chapter on Windows programming using OWL, but I guess here 
you mean "which methods it has for operating on an object of the class" when 
you say "what messages can be send to it"? :-)
 DM> I always put main last with everything implemented either in other
 DM> modules (where I have a header that is #include'd) or above it.  I use
 DM> prototypes only in header files.
Very glad to hear it, I really prefer it that way. Of course, style is a to a 
some degree a matter of personal taste, but I wouldn't want to start of by 
getting into habits that are frowned upon by most other programmers.
 [ void function() / void function(void) ]
 DM> Both work, but the empty parenthesis is the "right" way for
 DM> everything in C++.
Thanks, I switched.
 [ NULL / 0 ]
 DM> I vastly prefer NULL.
Hate to admit it, since I don't, but you do of course have a valid point. 
Will see if I'm able to switch. :-)
 DM> Use the one in comments for both - but, yeah, spreading it out will
 DM> make it more readable.
Glad to hear it's "allowed", really disliked Tom Swan's way one-line way of 
doing things. I'm typing in most listings from the book as I go along, and 
everytime had to struggle with myself to *not* spread them out. :-)
 DM> One exception may be functions that merely redirect calls...
 DM>   int baz(int i)   { return b->baz(i); }
Agreed.
 DM> Any time I see "m_" in front of a variable, I know I can find its
 DM> definition in the class - it is local to the class, but still global
 DM> in that it can be modified by multiple functions.
Will have to consider this. I don't really like the "this->" method myself, 
but naming a constructors arguments the same as the class's data members does 
lend some clarity to the code, as far as I'm concerned. By the way, what does 
the "m" signify? I'm not sure I understand your comment about the "multiple" 
functions. Is there any other type? All class data members can be modified by 
all (non const) member functions, can they not?
 DM> Another possibility in this case is:
 DM> TExample(char* s) : s(strdup(s))
 DM> {}
I just might like this approach... :-)
 DM> I also have a "cchar" class that handles the strdup/delete paradigm
 DM> for you - it puts an array on the heap but it looks like any other
 DM> array on the stack - you don't have to delete it.  :-)
I am somewhat of a minimalist, I really prefer doing things such as that 
explicitly. Keeps me in touch with what's going on. :-)
 [ (draft) ANSI standard obtainable? ]
 DM> Someone else may help you there - I just go by what my compiler
 DM> allows me to do ... if it doesn't allow it (yet), there's no point in
 DM> doing it even if the standard allows it.  :-)
I agree, but what if you compiler *does* allow it, yet you aren't sure if 
others will? Or worse still, what if you aren't even sure yours will? For 
instance, I was wondering if it were safe to pass a NULL pointer to strdup, 
which would allow me to write something like
  TExample::TExample(const char *s)
  {
    this->s = strdup(s);
  }
without having to check for s being NULL first. Neither Borland's online help 
nor Tom Swan's function reference said one way or the other, so I had to dig 
into the RTL source to find out it *wasn't* safe, at least not on my 
compiler. I'm assuming the standard would have explicitly told me so.
 RH>> Hope I didn't ask too many questions at once. Please be
 RH>> aware that if you do reply I will very likely have some
 RH>> more lined up for you by that time... :-)
 DM> And I probably won't be the only one answering... good luck!
Thank you! I'll be addresing the next bunch to all again, hardly seems fair 
to make you into my personal teacher for having been the first one to answer 
that first question. Don't let it keep you from answering though, you're 
helping me lots! :-)
Rene
---
 * Origin: postmaster@rene.demon.nl (2:282/1.11)
---------------

FIDO MESSAGE AREA==> TOPIC: 203 C++           Ref: F5G00122 Date: 04/28/98
From: RENE HERMAN                                           Time: 03:19am
\/To: ALL                                                 (Read 3 times)
Subj: Questions

Hello All ...
I'm a newcomer to the C++ language, and have some questions. Thanks in 
advance to anyone answering, please feel free to pick only question(s) that 
you feel like answering.
First off, would anyone happen to know if/where one might obtain a copy of 
the (draft) ANSI C++ standard? If not that, I would also very much welcome an 
URL for a copy of the draft ANSI C standard, assuming it also includes the 
standard libary functions. I also read the C_ECHO, and remember a new draft 
being released? I thought I saved a message announcing it, but now I seem to 
have misplaced it. :-)
Should I allow the compiler to allocate enums as single bytes? Or to use 
unsigned characters by default? It allows me to set these as options, but I 
believe both could get me into portability trouble? What's standard?
When overloading operators for a class, when should I implement the functions 
as member functions, and when as friend functions? My textbook demonstrates 
both methods, but doesn't say when nor why to use either.
When overloading the "=" operator for class X, should I use
  void X::operator=([const] X&);
as my textbook does, or
  X& X::operator=([const] X&);
returning "*this", as Borland's online help says I should do? Is there a 
difference?
In an exercise which asked to to overload the ++ and -- prefix and postfix 
operators for a class, I wrote the following test code:
  cout << "a == " << a.GetValue() << "; ++a == " << ++a << endl
       << "a == " << a.GetValue() << "; a++ == " << a++ << endl
       << "a == " << a.GetValue() << "; --a == " << --a << endl
       << "a == " << a.GetValue() << "; a-- == " << a-- << endl
       << "a == " << a.GetValue();
First I was quite surprised by the really wrong output this produced but 
after some minutes of hard thought (and cursing buggy compilers), I concluded 
this really was to be expected as a result of the cdecl (by the way, is that 
an acronymn?) right-to-left style of argument passing.
Indeed, splitting the single cout statement into 9 seperate cout statements 
solved the problem, and quite content that I had managed to solve the mystery 
I went back through the chapter since I remembered it did something similar, 
only to be rewarded with an example where the autor *did* string multiple 
statements onto a single cout (not everything as I did, but one line at a 
time, one GetValue call and one ++/-- call, which still produced wrong 
output, as I now expected) providing the following note:
"Due apparently to a bug in Borland C++ 4.5, using that compiler did not 
produce the preceding results. If you experience this problem, repaired in 
Borland C++ 5, use the following main function, which divides the multipart 
output statements and ensures that the expressions are evaluated in strict 
left-to-right fashion: [ 9 couts]"
Pardon? Bug? Annoying as it is, this behaviour is to be expected of cdecl 
isn't it? Furthermore, if it is a bug, it doesn't seem fixed...
I guess this is a Borland specific question, but is there any way to stop 
name mangling for class member functions? I know that normally, when 
implementing functions in an external assembly language module I can declare 
them as extern "C" to stop name mangling, but the compiler doesn't seem to 
let me do that with member functions.
Rene
---
 * Origin: postmaster@rene.demon.nl (2:282/1.11)