headers
Father Chrysostomos via RT | 23 May 05:13
Picon
Favicon

[perl #111918] [PATCH] Fix thawing seen objects in STORABLE_attach hook.

On Wed Mar 21 10:21:43 2012, igor.zaytsev <at> gmail.com wrote:
> Before any thaw hook is called Storable creates a new blessed object
> that
> is stored in a seen cache and then is provided to the hook. That is
> fine
> for STORABLE_thaw which fills in this object and returns it.
> STORABLE_attach
> on the other hand can create entirely new object by itself, so one
> memoized before should be thrown out to be replaced by that new
> object.
> ---
>  dist/Storable/Storable.xs       |   18 +++++++++++++++++-
>  dist/Storable/t/attach_errors.t |   37
> ++++++++++++++++++++++++++++++++++++-
>  2 files changed, 53 insertions(+), 2 deletions(-)

Thank you.  Applied as ecc6a8caad.

--

-- 

Father Chrysostomos

---
via perlbug:  queue: perl5 status: new
https://rt.perl.org:443/rt3/Ticket/Display.html?id=111918
Permalink | Reply | 0 comments |
headers
Perl 5 commit summary | 23 May 03:14

Perl 5 Commit Summary

Perl 5 commit summary, activity since Saturday

Current branch blead
187 commits.  18 unique authors.  6 unique committers.
135 files changed, 6527 insertions(+), 5405 deletions(-)
Thanks, applied:  Father Chrysostomos (23) Ricardo Signes (1)
                  Karl Williamson (1)
Snapshot: http://perl5.git.perl.org/perl.git/snapshot/8465c88d32125678.tar.gz

  Make CV * typemap entry support overloading
  Steffen Mueller             1 file changed, 10 insertions(+), 10 deletions(-
  http://perl5.git.perl.org/perl.git/commit/8465c88d32125678

  Purported equivalency isn't.
  Eric Brine                  1 file changed, 3 insertions(+), 3 deletions(-)
  http://perl5.git.perl.org/perl.git/commit/f6c6dcb63681097e

  replace B::COP::stashflags by B::COP::stashlen
  Reini Urban                 2 files changed, 6 insertions(+), 6 deletions(-)
  http://perl5.git.perl.org/perl.git/commit/667d5932a778569c

  refactor macro to avoid compiler warning in regcomp.c
  Robin Barker (via RT)       1 file changed, 14 insertions(+), 9 deletions(-)
  http://perl5.git.perl.org/perl.git/commit/4f6289a38d7a3e24

  perlebcdic: make verbatim line fit in 79 cols
  Karl Williamson             2 files changed, 4 insertions(+), 4 deletions(-)
  http://perl5.git.perl.org/perl.git/commit/b693e169ef8b0319

  regen/mk_invlists.pl: Fail if inversion list not found
(Continue reading)

Permalink | Reply | 0 comments |
headers
Darin McBride | 23 May 00:08
Picon
Gravatar

Re: [perl #112990] Documentation for 'kill' should call more attention to using names instead of numbers

On Tuesday May 22 2012 1:43:01 PM you wrote:
> I haven‘t looked at the code, but I can tell you where it is, if that’s
> any help.  It’s the ‘case OP_KILL’ in doio.c:Perl_apply.

Yes, that helps.  I'm definitely no expert in perl guts (hey, that'd make a 
good name for a perldoc about this!  Oh hold on...), but it seems to me from a 
cursory read that:

> > # A
> > kill INT => 10, -11, 12;

This will send SIGINT to PIDs 10 and 12, and rely on kill(2) to kill the 
process group for 11.

> > # B
> > kill -INT => 10, -11, 12;

This will croak C<Unrecognised signal name "-INT">.

*EDIT*: TIAS says this is wrong.  I now suspect it's actually trying to do a 
C<kill 0 => 10, -11, 12>.  And warnings seem to indicate it:

$ ~/perl/blead/bin/perl -Mwarnings -lE 'kill -INT => $$'
Argument "-INT" isn't numeric in kill at -e line 1.

Note to self: "-" isn't ALPHA.  Duh.  :-)

> > # C
> > kill 2 => 10, -11, 12;
(Continue reading)

Permalink | Reply | 1 comment |
headers
demerphq | 22 May 23:16
Picon

Curious -i issue. Bug?

perl -i'foo e eval "warn q[bar]" '

http://perlmonks.org/?node_id=971866

Yves

--

-- 
perl -Mre=debug -e "/just|another|perl|hacker/"
Permalink | Reply | 5 comments |
headers
Steffen Mueller via RT | 22 May 23:06
Picon
Favicon

[perl #96872] CV * typemap entry should support overloading

Change applied as 8465c88d321256783d00b03482a840dab3ad16be. Sorry again 
for dropping the ball on this before and thanks for the reminder.

Best regards,
Steffen
Permalink | Reply | 0 comments |
headers
Father Chrysostomos via RT | 22 May 22:43
Picon
Favicon

[perl #112990] Documentation for 'kill' should call more attention to using names instead of numbers

On Tue May 22 12:21:04 2012, dmcbride <at> cpan.org wrote:
> On Friday May 18 2012 3:27:11 PM you wrote:
> > On Fri, May 18, 2012 at 6:19 PM, Darin McBride <perlbug-
> followup <at> perl.org>wrote:
> > > So, is it the signal being negative or the process ID or both?
> >
> > Both negative signals and negative pids send to the process group,
> IIRC.
> 
> Ok, so what happens with:
> 
> # A
> kill INT => 10, -11, 12;
> 
> # B
> kill -INT => 10, -11, 12;
> 
> # C
> kill 2 => 10, -11, 12;
> 
> # D
> kill -2 => 10, -11, 12;
> 
> Assuming, of course, proper permissions, these processes exist, they
> have
> process groups beneath them, etc.
> 
> I'm assuming A == C and B == D (assuming INT is signal "2" as it is on
> my
> linux box, though I'm unsure if that's guaranteed).  I'm not sure
(Continue reading)

Permalink | Reply | 0 comments |
headers
Father Chrysostomos via RT | 22 May 22:39
Picon
Favicon

[perl #113044] Return value of map shared when referencing a sub in CORE::

On Tue May 22 13:21:35 2012, gustavderdrache <at> gmail.com wrote:
> This is a bug report for perl from gustavderdrache <at> gmail.com,
> generated with the help of perlbug 1.39 running under perl 5.16.0.
> 
> 
> -----------------------------------------------------------------
> [Please describe your issue here]
> 
> When using a subref in CORE:: (e.g., \&CORE::oct), the return value is
> shared across all iterations of map.  This can be alleviated by using
> an expression such as 0+$foo.
> 
> The script below reproduces the issue:
> 
>     $oct = \&CORE::oct;
> 
>     @mapped = map &$oct($_), qw(0b10 010 0x10);
>     print "@mapped\n";
> 
>     @mapped = map 0+&$oct($_), qw(0b10 010 0x10);
>     print "@mapped\n";

Yes, I discovered this recently.  The same issue affects lvalue subs.  I
fixed it in commit 5811c07eab3.

--

-- 

Father Chrysostomos

---
(Continue reading)

Permalink | Reply | 0 comments |
headers
Father Chrysostomos via RT | 22 May 21:58
Picon
Favicon

[perl #113038] fc() doc claims equivalency where there isn't

On Tue May 22 12:38:09 2012, ikegami <at> adaelis.com wrote:
> This is a bug report for perl from ikegami <at> adaelis.com,
> generated with the help of perlbug 1.39 running under perl 5.16.0.
> 
> 
> -----------------------------------------------------------------
> [Please describe your issue here]
> 
> The docs for C<fc> say C<< $this =~ /\Q$that/i >> can be written
> as C<< fc($this) eq fc($that) >>.
> 
> Patch attached.

Thank you.  Applied as f6c6dcb.

--

-- 

Father Chrysostomos

---
via perlbug:  queue: perl5 status: new
https://rt.perl.org:443/rt3/Ticket/Display.html?id=113038
Permalink | Reply | 0 comments |
headers
Father Chrysostomos via RT | 22 May 21:56
Picon
Favicon

[perl #113034] [PATCH] 2d8d7b1 replace B::COP::stashflags by B::COP::stashlen

On Tue May 22 09:37:53 2012, rurban <at> cpanel.net wrote:
> 
> This is a bug report for perl from rurban <at> cpanel.net,
> generated with the help of perlbug 1.39 running under perl 5.15.7.
> 
> From 2d8d7b1c730932ee5a45c88f1ec23e5fbedd9b81 Mon Sep 17 00:00:00 2001
> From: Reini Urban <rurban <at> x-ray.at>
> Date: Tue, 22 May 2012 10:57:03 -0500
> Subject: [PATCH] replace B::COP::stashflags by B::COP::stashlen
> MIME-Version: 1.0
> Content-Type: multipart/mixed; boundary="------------1.7.10"

Thank you.  Applied as 667d5932a778.

--

-- 

Father Chrysostomos

---
via perlbug:  queue: perl5 status: new
https://rt.perl.org:443/rt3/Ticket/Display.html?id=113034
Permalink | Reply | 0 comments |
headers
Eric Brine | 22 May 21:38
Picon
Favicon

[perl #113038] fc() doc claims equivalency where there isn't

# New Ticket Created by  "Eric Brine" 
# Please include the string:  [perl #113038]
# in the subject line of all future correspondence about this issue. 
# <URL: https://rt.perl.org:443/rt3/Ticket/Display.html?id=113038 >

This is a bug report for perl from ikegami <at> adaelis.com,
generated with the help of perlbug 1.39 running under perl 5.16.0.

-----------------------------------------------------------------
[Please describe your issue here]

The docs for C<fc> say C<< $this =~ /\Q$that/i >> can be written
as C<< fc($this) eq fc($that) >>.

Patch attached.

[Please do not change anything below this line]
-----------------------------------------------------------------
---
Flags:
    category=docs
    severity=low
---
Site configuration information for perl 5.16.0:

Configured by eric at Tue May 22 01:57:47 EDT 2012.

Summary of my perl5 (revision 5 version 16 subversion 0) configuration:

  Platform:
(Continue reading)

Permalink | Reply | 0 comments |
headers
Darin McBride | 22 May 21:20
Picon
Gravatar

Re: [perl #112990] Documentation for 'kill' should call more attention to using names instead of numbers

On Friday May 18 2012 3:27:11 PM you wrote:
> On Fri, May 18, 2012 at 6:19 PM, Darin McBride <perlbug-
followup <at> perl.org>wrote:
> > So, is it the signal being negative or the process ID or both?
> 
> Both negative signals and negative pids send to the process group, IIRC.

Ok, so what happens with:

# A
kill INT => 10, -11, 12;

# B
kill -INT => 10, -11, 12;

# C
kill 2 => 10, -11, 12;

# D
kill -2 => 10, -11, 12;

Assuming, of course, proper permissions, these processes exist, they have 
process groups beneath them, etc.

I'm assuming A == C and B == D (assuming INT is signal "2" as it is on my 
linux box, though I'm unsure if that's guaranteed).  I'm not sure that's the 
case.

I don't mind attempting a doc patch, but I don't actually understand what it's 
doing, so maybe someone who does know can explain it roughly first?  I know
(Continue reading)

Permalink | Reply | 0 comments |

Gmane