Re: D7 Compiler Bug

From: Marco van de Voort (marcov_at_stack.nl)
Date: 10/06/04 

Date: Wed, 6 Oct 2004 07:32:11 +0000 (UTC)

On 2004-10-05, VBDis <vbdis@aol.com> wrote:
>>> to take it out.
>>
>>The compiler changed totally from D1 to D2. So they explicitely had to
>>code it into D2. So it is there on purpose.
>
> The documentation says that the Object type is supported for compatibility
> reasons. This might mean that dynamic types (AnsiString...) cannot occur in
> legacy (TP...) code and consequently these new types must not be supported in
> Object.

Yes, that is my opinion (and expectation) too. Even if it was supported, I
only expect it to work when constructed properly. Otherwise it would become
an automatic object.

> But it's a half-hearted policy to support Object even in D7 without
> mentioning or correcting hard limits in the implementation and use of that
> data type. The Delphi compiler IMO is very reliable, so it's very strange
> to learn that there exist situations where the compiler definitely
> produces broken code :-(

IMHO this is a documentation fault, and the compiler is fine. FPC
independantly has the same behaviour.

It's a pity that it hasn't been released to PD (or MPL) yet, but I still
want to try to get Turbo Vision compiling under Delphi one day. That would
really stress the object mechanism.

Relevant Pages

  • Re: Why forth is not popular
    ... CHForth and all other Forths ... distributed by the Dutch FIG have truly excellent documentation and tutorials. ... > error, you execute your marker, make changes to your definition and run ... but Perl doesn't claim to be a compiler. ...
    (comp.lang.forth)
  • Re: No Equals on interfaces
    ... > and the compiler is perfectly happy to let me do that. ... > Dim iref As I = Nothing ... the compiler applies implicit widening ... without taking a look at the documentation. ...
    (microsoft.public.dotnet.languages.vb)
  • Re: Weird string bug
    ... > the statement in Borland's printed documentation to the effect that ... are implicitly supplied by the compiler. ... to detect the need for the conversion. ... What may be obvious to an ex-C programmer ...
    (comp.lang.pascal.delphi.misc)
  • Re: What a translation unit is.
    ... >> programmers here, some of whom have used Microsoft compilers and ... >> Microsoft produced compilers or linkers, ... You have referred to Microsoft documentation ... Any compiler vendor can make mistakes in the documentation. ...
    (alt.comp.lang.learn.c-cpp)
  • Re: Why forth is not popular
    ... >documentation are costly to produce. ... with compiler writing, choose Forth, then proudly publish the result ... web site, and whip everybody whining about the lack of a truly ... There are also other new books, such as the two from FORTH, ...
    (comp.lang.forth)