Re: [fpc-pascal] IntToStr implementation

2024-01-18 Thread Adriaan van Os via fpc-pascal 

Michael Van Canneyt via fpc-pascal wrote:



More accurately, I meant: you cannot write a function declaration to 
handle that syntax.


It is true that a regular (non-built-in) procedure/function with that syntax 
can not be declared.

Some more strange ducks

- Exit allowing Program as parameter (UCSD-Pasca)
- Scan allowing a partial-expression as second parameter (UCSD-Pascal) e.g. 
Scan( -26, =':', S)
- SizeOf allowing a type-identifier as parameter

etcetera.

Regards,

Adriaan van Os
___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
https://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] IntToStr implementation

2024-01-18 Thread Michael Van Canneyt via fpc-pascal 




On Thu, 18 Jan 2024, Adriaan van Os via fpc-pascal wrote:




Also something which cannot be handled by 'standard' syntax.
(unless you wish to count generics and implicit specialization as "standard 
syntax")


actual-parameter =  expression { ":" expression } .
actual-parameter-list = "(" actual-parameter { "," actual-parameter } ")" .


The compiler obviously can parse it.

More accurately, I meant: you cannot write a function declaration to handle 
that syntax.

procedure Str(O : Integer { what to write here for the :N:M ??? }; var S);


Michael.
___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
https://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] IntToStr implementation

2024-01-18 Thread Adriaan van Os via fpc-pascal 




Also something which cannot be handled by 'standard' syntax.
(unless you wish to count generics and implicit specialization as 
"standard syntax")


actual-parameter =  expression { ":" expression } .
actual-parameter-list = "(" actual-parameter { "," actual-parameter } ")" .

Regards,

Adriaan van Os

___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
https://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] IntToStr implementation

2024-01-18 Thread Michael Van Canneyt via fpc-pascal 



On Thu, 18 Jan 2024, Hairy Pixels via fpc-pascal wrote:





On Jan 18, 2024, at 1:51 PM, Sven Barth via fpc-pascal 
 wrote:

There is no source for Str() in that sense, because Str() is a compiler 
intrinsic implemented in the compiler's ninl.pas.



Well that explains everything now! What is this an intrinsic? I was curious how 
it was implemented and now I'm even more curious.


I asked myself the same question.

The reason is it needs to support the same syntax as Write(ln) for padding the 
output:

Str(I:N:M,S);

This is not a "standard" syntax and needs to be handled specially by the 
compiler.

Additionally, Str() can be used to convert an enum to a string value:

Type
  TMyEnum = (one,two,three);

var
  M : TMyEnum = one;
  S : String;

begin
  Str(M,S);
end.

Also something which cannot be handled by 'standard' syntax.
(unless you wish to count generics and implicit specialization as "standard 
syntax")

Michael.___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
https://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] IntToStr implementation

2024-01-18 Thread Hairy Pixels via fpc-pascal 


> On Jan 18, 2024, at 1:51 PM, Sven Barth via fpc-pascal 
>  wrote:
> 
> There is no source for Str() in that sense, because Str() is a compiler 
> intrinsic implemented in the compiler's ninl.pas.
> 

Well that explains everything now! What is this an intrinsic? I was curious how 
it was implemented and now I'm even more curious.

Regards,
Ryan Joseph

___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
https://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] Trash variables (-gt) - Translation

2024-01-18 Thread Pierre Muller via fpc-pascal 



Le 18/01/2024 à 08:57, Michael Van Canneyt via fpc-pascal a écrit :



On Thu, 18 Jan 2024, LacaK via fpc-pascal wrote:


Hi *,
I am translating "Trash variables" (from English to Slovak), but I can
not find appropriate words.
That's why I want to describe it in more words.
Is this correct: "Initialize local variables with random values" ? Or
better suggestion?


That seems most correct to me.


  I think it is more something like 'specific pattern values' which are 
different depending on the number of 't'
(extracted from current trunk):
./globals.pas:126:   trashintvalues: array[0..nroftrashvalues-1] of int64 = 
($,$,$EFEFEFEFEFEFEFEF,0);

Thus  -gt  will use $ pattern,
  -gtt will use $ pattern,
  -gtttwill use $EFEFEFEFEFEFEFEF pattern,
  -g   will use $ pattern
for 'trashing' of local variables.

Pierre
___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
https://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-pascal

Re: [fpc-pascal] Trash variables (-gt) - Translation

2024-01-18 Thread Adriaan van Os via fpc-pascal 

LacaK via fpc-pascal wrote:

Hi *,
I am translating "Trash variables" (from English to Slovak), but I can 
not find appropriate words.

That's why I want to describe it in more words.
Is this correct: "Initialize local variables with random values" ? Or 
better suggestion?


I once had a bug in a software package, written by me and running at a customer site, that turned 
out to be caused by an incorrect translation (into Dutch) of a book by Peter Norton. Since then, I 
refuse to read computer books in another language than the original one .


Regards,

Adriaan van Os

___
fpc-pascal maillist  -  fpc-pascal@lists.freepascal.org
https://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-pascal