New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Dump dump #15155
Comments
From @epaCreated by @epaThe dump() builtin has been deprecated and barely working for as long If that is too much, then at least the unqualified use of dump() (Personally I would like dump() to return the stringification of a Perl Info
|
From @AbigailOn Thu, Jan 28, 2016 at 04:27:51AM -0800, Ed Avis wrote:
I don't think it will be a great loss if dump() gets dumped. Has it ever worked in a useful way? I started using perl in 1995,
That I think should remain on CPAN. But removing dump() allows Abigail |
The RT System itself - Status changed from 'new' to 'open' |
From zefram@fysh.orgAbigail wrote:
Its usefulness always depended on the availability of the undump utility, The actual trick that dump+undump achieves has really gone out of You'd think there'd still be a niche for it around Perl, with some people -zefram |
From @epaEven for emacs, the undump / unexec approach might be going away: http://lwn.net/SubscriberLink/673724/a5165e8736a4fac2/ -- |
From @epaIn Perl 5.30, dump is no longer a reserved word. You have to say CORE::dump(). Out of interest what was the rationale for keeping the code around as CORE::dump() rather than removing it altogether? Does it have any users? |
From @demerphqI remember FC saying he used it for breakpointing when he is debugging. Yves On Mon, 23 Sep 2019, 11:48 Ed Avis via RT, <perlbug-followup@perl.org>
|
From @leonerdOn Mon, 23 Sep 2019 12:43:53 +0200
Isn't that what study() is used for too? -- leonerd@leonerd.org.uk | https://metacpan.org/author/PEVANS |
From @demerphqOh. Shoot. Maybe I'm confusing the two. Probably. Good catch. Yves On Mon, 23 Sep 2019, 12:47 Paul "LeoNerd" Evans, <leonerd@leonerd.org.uk>
|
From @AbigailOn Mon, Sep 23, 2019 at 02:48:39AM -0700, Ed Avis via RT wrote:
How would we know whether it has any users? Or rather, can we ever know Abigail |
From @LeontOn Mon, Sep 23, 2019 at 11:48 AM Ed Avis via RT
What would be the rationale of removing it? Is it standing in anyone's way? Leon |
From @dur-randirI occasionally use CODE::dump to actually generate a core file programmatically instead of sending SIGABRT to $$, as the latter is async and may arrive sometime later, not exactly where I need it. |
From @epaPerhaps the documentation should be changed? It currently says
It could instead say something like
|
From @eserteDana Tue, 24 Sep 2019 03:53:37 -0700, randir reče:
Would it be possible to use PERL_SIGNALS=unsafe to get the non-deferred behavior? |
From @dur-randirOn Tue, 24 Sep 2019 13:37:01 -0700, slaven@rezic.de wrote:
It's async kernel delivery, not async perl handling, what's the culprit here. |
From @LeontOn Wed, Sep 25, 2019 at 3:28 PM Sergey Aleynikov via RT
That doesn't sound right. POSIX requires that "If the value of pid Leon |
From @dur-randirOn Wed, 25 Sep 2019 10:23:46 -0700, LeonT wrote:
Never thought about this. Is it also true for windows builds (aka are they POSIX compliant)? But this still requires running under unsafe signals, which is not the default and not always possible to set. |
From @LeontOn Wed, Sep 25, 2019 at 7:34 PM Sergey Aleynikov via RT
No, Windows is very much not a POSIX, and in fact almost nothing about
No it doesn't. Deferred signaling only affects Perl's signal handlers, Leon |
Migrated from rt.perl.org#127405 (status was 'open')
Searchable as RT127405$
The text was updated successfully, but these errors were encountered: