[fpc-pascal] Turbo Pascal Mode Bug?

[Zaka E-Lab]

Hello,

I have a peace of code that demonstrates that in the TP mode of
freepascal, one can use a function as a procedure.

Is this a bug or a normal behavior?

My University have replaced ( other time ) freepascal.org, and we
must use the archaic Turbo Pascal 6.0, cause they say that freepascal
don't respect the ISO Pascal.

I let you the code here:

program func_as_proc(input,output);

function pru_func:boolean;
begin
   writeln ('If I am printed, then you can use a function as a
procedure.');

   pru_func := TRUE

end; { pru_func }

begin
   (*
*  Prove if a function can be used as a procedure.
*)
   pru_func;
end.

Thanks in advance.

___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] Turbo Pascal Mode Bug?

[Zaka E-Lab]

Daniël Mantione:
Funny. Free Pascal does not respect ISO Pascal, because it respects 
Turbo Pascal. :) Turbo Pascal in turn, does not respect ISO Pascal, 
because it respects UCSD Pascal.


Is very confusing when talking about Standard Pascal, cause it's not 
like C/C++ that have standard that is respected by most people ( at 
least, is what I feel ). My teachers talk about "Standard Pascal" ( no 
one knows what they are talking about, but you must accept their holy 
truth ), I knew that TP is not ISOP, but the pupil can't refuse the 
teacher unless he wants to be bitten.


Florian Klaempfl:

Normal because FPC emulates TP7, just add {$X+} in TP6 and you get the
same behaviour.


Is there a way to do the contrary? I mean, to avoid the use of a 
function as a procedure in freepascal.


Thanks.

Zaka.
___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

[fpc-pascal] Turbo Pascal mode!!

[Zaka E-Lab]

Some time ago, I posted about this.

On the TP mode FreePascal can use a function as a procedure. Florian
told me that that was because of the version of the TP emulation.
Well, after that, I was happy with that until now. I discovered that,
even adding {$x-} to the code, if you put a function inside a "for"
or an "if" you can use a function as a procedure:


program pru_funcion_como_procedimiento (input,output);

{$x-}

function pru_func:boolean;
begin
   writeln ('Hola programador.');
   pru_func := TRUE;
end; { pru_func }

procedure pru;
var
   i : integer;
begin
   for i:=1 to 10 do
  pru_func;
end; { pru }

begin
   pru;
end.
*

Is this still normal behaviour?

I hope this helps anybody.

Zaka.
___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

[fpc-pascal] Pointers in Pascal!!

[Zaka E-Lab]

I have something like this:

PStruct = ^Struct;
Struct = record
 num: integer;
 end;

I have a unit to handle that structure ( is a bit more complicated, but
is a good example), in this unit I have a procedure that deletes (
disposes ) the pointer. How can I know if a pointer are created or not?,
I mean, I don't wont to try to delete something that has not been
created yet, but I don't know how to do it, if that is possible.

Thanks a lot.

Zaka.
___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] Pointers in Pascal!!

[Zaka E-Lab]

Maybe I haven't explain it correctly. What I want is to construct a
Unit that contains some procedures, one of them is
create(var p : PStruct), and other is delete(var p : PStruct).

I don't create anything, the memory is allocated by the create
procedure. How can the procedure know if the argument is a pointer
already allocated or not?

I hope I've explain good enough.

Thanks.

Zaka.

Leonardo M. Ramé escribió:

You can assign "nil" to the pointer before using it, then check if PStruct = 
nil it wasn't
created.

Leonardo.



___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] Pointers in Pascal!!

[Zaka E-Lab]

It's possible to give a default initialization value to a defined type?
I suppose that it isn't, but it would be to easy to resolve that trouble
that way.

Anyway, I appreciate your help and support.

Thanks.

Zaka.

Jonas Maebe escribió:
It cannot know that without help. You have to do such bookkeeping 
yourself (e.g., as mentioned before, by assigning nil to all pointers 
when declaring them). That is one of the basic properties of a language 
with manual memory management.



Jonas



___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-pascal