public inbox for gcc@gcc.gnu.org
 help / color / mirror / Atom feed
* I want try...finally
@ 2003-05-15 16:02 Andrew V. Fionik
  2003-05-15 16:41 ` Matt Austern
  2003-05-15 19:06 ` Branko Čibej
  0 siblings, 2 replies; 3+ messages in thread
From: Andrew V. Fionik @ 2003-05-15 16:02 UTC (permalink / raw)
  To: gcc

Hi, guys.

I think that you doing great job by making GCC and G++. However i feel that
C++
compiler lacks support for one great language construct "try...finally". I'm
sure you know about it. It widely used in prooperitary Microsoft and Borland
C++/Delphi compilers. This construct allows to easy write safe, elegant, and
well readably code.

I hope you will add this language feature.

P.S. Of course i know that it not exist in standard, but i think in that
case standard deffinitely outdated.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: I want try...finally
  2003-05-15 16:02 I want try...finally Andrew V. Fionik
@ 2003-05-15 16:41 ` Matt Austern
  2003-05-15 19:06 ` Branko Čibej
  1 sibling, 0 replies; 3+ messages in thread
From: Matt Austern @ 2003-05-15 16:41 UTC (permalink / raw)
  To: Andrew V. Fionik; +Cc: gcc

On Thursday, May 15, 2003, at 08:06  AM, Andrew V. Fionik wrote:

> Hi, guys.
>
> I think that you doing great job by making GCC and G++. However i feel 
> that
> C++
> compiler lacks support for one great language construct 
> "try...finally". I'm
> sure you know about it. It widely used in prooperitary Microsoft and 
> Borland
> C++/Delphi compilers. This construct allows to easy write safe, 
> elegant, and
> well readably code.
>
> I hope you will add this language feature.
>
> P.S. Of course i know that it not exist in standard, but i think in 
> that
> case standard deffinitely outdated.

I would oppose adding the "finally" construct to our C++ compiler 
unless the
standard is updated to include it.

I don't have strong feelings about whether the C++ Standard should be 
updated
to include "finally", but I do feel strongly that this sort of language
extension should be done by the standards committee, not on a patchwork 
basis
by gcc maintainers.

I suggest that you write up a proposal for "finally" (what are the exact
semantics of the feature, what coding styles does it support that are
impossible with the existing language, how it wouild interact with 
existing
language features, how it would affect user code, what would need to 
change in
the standard).  Then you should get that proposal to the standards 
committee.
The easiest way to get a proposal to the standards committee is to post 
it to
the Usenet newsgroup comp.std.c++.  A number of committee members also 
read
this mailing list, so, if you write up a detailed proposal, we can make 
sure
it gets seen by the appropriate people.

			--Matt


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: I want try...finally
  2003-05-15 16:02 I want try...finally Andrew V. Fionik
  2003-05-15 16:41 ` Matt Austern
@ 2003-05-15 19:06 ` Branko Čibej
  1 sibling, 0 replies; 3+ messages in thread
From: Branko Čibej @ 2003-05-15 19:06 UTC (permalink / raw)
  To: Andrew V. Fionik; +Cc: gcc

Andrew V. Fionik wrote:

>Hi, guys.
>
>I think that you doing great job by making GCC and G++. However i feel that
>C++
>compiler lacks support for one great language construct "try...finally". I'm
>sure you know about it. It widely used in prooperitary Microsoft and Borland
>C++/Delphi compilers. This construct allows to easy write safe, elegant, and
>well readably code.
>
Anything that can be done with try...finally can also be done with
destructors without extending the language.


-- 
Brane ÄŒibej   <brane@xbc.nu>   http://www.xbc.nu/brane/

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2003-05-15 19:06 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-05-15 16:02 I want try...finally Andrew V. Fionik
2003-05-15 16:41 ` Matt Austern
2003-05-15 19:06 ` Branko Čibej

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).