Re: On the case for exception-based error handling. by Bennett Todd- August 27, 2000 2000-08-22-16:40:13 Peter Scott:> >I&39;m not sure, but I think Chaim&39;s main point was just that, not> >that divide-by-zero should be ignored too.> > Well, it could be made user-selectable, right Bennett Do you envisage > being able to say> > use Fatal qw(Arithmetic IO etc)> > using classnames that fall out of RFC 80I hadn&39;t thought about the current cases where the core throwsexceptions. I agree that Fatal would be a fine linguistic hook...http://www.nntp.perl.org/group/perl.perl6.language.errors/2000/08/msg283.html RFC 88 version 3 draft 1 is available via http. by Tony Olekshy- August 27, 2000 I have added the remaining changes based on the traffic in -errorsand the batch of changes I received from Peter Scott. Productionformatting has been completed. RFC 88v3d1 is available at: http:www.avrasoft.comperl6rfc88.htmAn updated version of the Perl 5 reference implementation, in whichC<exception> does class hierarchies and $->show works, is at: http:www.avrasoft.comperl6try6-ref5.txtAssuming no major problems with this draft are reported in thenext couple days,...http://www.nntp.perl.org/group/perl.perl6.language.errors/2000/08/msg282.html Re: Structured exception handling should be a core module. by Tony Olekshy- August 26, 2000 Peter Scott wrote:> > Tony Olekshy wrote:> >> > In fact, not only would I be pleased and honoured to author the> > Perl 6 core Try.pm module, I&39;m already working on a Perl 5 standard> > reference implementation.> > > Peter, I think we should make this approach more clear in RFC 88.> > I&39;m not convinced that this can totally be implemented in a> module.- File Try.pm ----------------------------- Subject: Structured Exception Handling...http://www.nntp.perl.org/group/perl.perl6.language.errors/2000/08/msg281.html Re: $SIG__DIE__ should be localized and cleared at the start of an eval block by Tom Christiansen- August 25, 2000 >Why wouldn&39;t you want to just get rid of the whole construct Is there any >need for itIt seems safer to kill it.--tomhttp://www.nntp.perl.org/group/perl.perl6.language.errors/2000/08/msg280.html Re: $SIG__DIE__ should be localized and cleared at the startof an eval block by Peter Scott- August 25, 2000 >My statement: this shouldn&39;t be the programmer&39;s responsibility. If>you&39;re using other people&39;s modules that depend on eval, you&39;re in>trouble anyway. The "local $SIG__DIE__;" statement should be implicit>at the start of the eval block.Why wouldn&39;t you want to just get rid of the whole construct Is there any need for it--Peter ScottPacific Systems Design Technologieshttp://www.nntp.perl.org/group/perl.perl6.language.errors/2000/08/msg279.html Re: Structured exception handling should be a core module. by Bart Lateur- August 25, 2000 On Thu, 24 Aug 2000 17:57:55 -0700, Peter Scott wrote:>>I&39;ve read 151 a few times, and I don&39;t understand how it can impact>>the implementation of RFC 88 as a module. Please explain.>>If $ and $! are merged, then in code like>> try > system_call_that_fails();> more_stuff_that_succeeds();> > finally > >>does the finally block think there is a current exception or not $! was...http://www.nntp.perl.org/group/perl.perl6.language.errors/2000/08/msg278.html $SIG__DIE__ should be localized and cleared at the start of an eval block by Bart Lateur- August 25, 2000 Watch the behaviour of this code snippet in a current Perl (5.6): eval print STDERR "Testing...n"; warn "Oops!"; print STDERR "Still going...n"; die "Argh!!!"; print STDERR "I died, didn&39;t In"; ; print STDERR "Outcome: $n";--> Testing... Oops! at test.pl line 4. Still going... Outcome: Argh!!! at test.pl line 6.Now try again, but with this in front of the eval block: $SIG__DIE__ = sub print STDERR...http://www.nntp.perl.org/group/perl.perl6.language.errors/2000/08/msg277.html Re: Structured exception handling should be a core module. by Tony Olekshy- August 25, 2000 Peter Scott wrote:> > At 06:48 PM 82400 -0600, Tony Olekshy wrote:> >> >I&39;ve read 151 a few times, and I don&39;t understand how it can> >impact the implementation of RFC 88 as a module. Please explain.> > If $ and $! are merged, then in code like> > try > system_call_that_fails();> more_stuff_that_succeeds();> > finally > > > does the finally block think there is a.http://www.nntp.perl.org/group/perl.perl6.language.errors/2000/08/msg276.html Re: Structured exception handling should be a core module. by Peter Scott- August 24, 2000 At 06:48 PM 82400 -0600, Tony Olekshy wrote:>Peter Scott wrote:> >> > At 06:06 PM 82400 -0600, Tony Olekshy wrote:> > >> > >In fact, not only would I be pleased and honoured to author the> > >Perl 6 core Try.pm module, I&39;m already working on a Perl 5 standard> > >reference implementation.> >> > >Peter, I think we should make this approach more clear in RFC 88.> >> > I&39;m not convinced that this can totally be...http://www.nntp.perl.org/group/perl.perl6.language.errors/2000/08/msg275.html Re: Structured exception handling should be a core module. by Tony Olekshy- August 24, 2000 Peter Scott wrote:> > At 06:06 PM 82400 -0600, Tony Olekshy wrote:> >> >In fact, not only would I be pleased and honoured to author the> >Perl 6 core Try.pm module, I&39;m already working on a Perl 5 standard> >reference implementation.> > >Peter, I think we should make this approach more clear in RFC 88.> > I&39;m not convinced that this can totally be implemented in a> module. Particularly if RFC 151 passes :-)I&39;ve read 151 a few times, and..http://www.nntp.perl.org/group/perl.perl6.language.errors/2000/08/msg274.html |