Hiya,
I was reading some perl5 source today and noticed that
Perl_sv_setsv which gets called on assignments, calls Perl_sv_upgrade,
now Perl_sv_upgrade from documentation:
Upgrade an SV to a more complex form. Generally adds a new body type
to the
SV, then copies across as much information as
On Wednesday, August 20, 2003, at 04:31 am, Kurt Starsinic wrote:
So it seems to me that, in Perl, a named subroutine is never a
closure, and an anonymous subroutine is always a closure. The fact
that perl optimizes the special case of an anonymous subroutine with 0
private bindings is an
Hi,
Herr Schwern pointed out to me that method calls are slow (big news!),
but that class method calls are nearly 2x as slow as object method
calls, aka
Test-foo(); taking 2x the time of
my $foo = bless {}, Test;
$foo-foo();
The reason for this is that for every single call we need to
a) See
Yes
Arthur
On måndag, mar 24, 2003, at 18:54 Europe/Stockholm, H.Merijn Brand
wrote:
Serious enough to follow up on?
Subject: Free to wrong pool
From: Suresh Rukmangathan [EMAIL PROTECTED]
Date: Fri, 21 Mar 2003 18:47:16 GMT
Newsgroups: comp.lang.perl.misc
X-Newsreader: Microsoft Outlook
On lördag, okt 5, 2002, at 11:00 Europe/Stockholm, Rafael Garcia-Suarez
wrote:
Arthur Bergman wrote:
Should this be released as a CPAN release of Safe.pm? If so I can
volunteer to do it.
This is a good idea. The new regression test is designed to work on
older Perls.
I don't think
The uploaded file
Attribute-Handlers-0.78.tar.gz
has entered CPAN as
file: $CPAN/authors/id/A/AB/ABERGMAN/Attribute-Handlers-0.78.tar.gz
size: 15392 bytes
md5: c935f240bee1baf4b46e7d69bf2f1636
Synced up with the latest repository change.
Arthur
On onsdag, augusti 14, 2002, at 01:11 , Mark-Jason Dominus wrote:
Perhaps this would be a good time for someone to do some real
research, rather than continuing to spread FUD. Does someone have the
Stevens book or the POSIX standard? If not I will try to obtain one
or the other.
I have
Hah, this sounds interesting,
I think you will find it reports the tid of the thread that does the
join :-), I don't know offhand how to fix it.
Arthur
On fredag, juli 12, 2002, at 05:31 , Jarkko Hietaniemi wrote:
*sigh* Why do people have to do this, find more and more bugs
just before a
On onsdag, juni 26, 2002, at 10:46 , Mark Mielke wrote:
Does this actually make the code execute faster?
mark
The code $i = $i + 1; executes around 10% faster on my machine.
Arthur
On torsdag, juni 13, 2002, at 04:55 , Nick Ing-Simmons wrote:
Arthur Bergman [EMAIL PROTECTED] writes:
Since Benjamin came up with objections on using FETCH when doing a lval
operation on a tied array I came up with this.
I don't agree with his position I think FETCH is just fine
On onsdag, juni 12, 2002, at 05:25 , Arthur Bergman wrote:
On måndag, juni 10, 2002, at 11:26 , Nick Ing-Simmons wrote:
From: Nick Ing-Simmons [EMAIL PROTECTED]
Date: mån jun 10, 2002 11:26:45 Europe/Stockholm
To: [EMAIL PROTECTED]
Cc: Nick Ing-Simmons [EMAIL PROTECTED], Jarkko
On söndag, juni 9, 2002, at 10:32 , Rafael Garcia-Suarez wrote:
On 2002.06.09 14:33 Arthur Bergman wrote:
For all of you that want to register actions to happen on scope leaves,
(Schwern!), here you go.
Cool! thanks.
PS. Isn't there a keyword in Perl 6 to do this ? I don't remember
On onsdag, juni 5, 2002, at 09:12 , Johan Vromans wrote:
From: [EMAIL PROTECTED] (Johan Vromans)
Date: ons jun 05, 2002 09:12:05 Europe/Stockholm
To: Arthur Bergman [EMAIL PROTECTED]
Cc: Johan Vromans [EMAIL PROTECTED], Gurusamy Sarathy
[EMAIL PROTECTED], [EMAIL PROTECTED]
Subject: Re
On måndag, juni 3, 2002, at 10:16 , Elizabeth Mattijsen wrote:
At 09:57 AM 6/3/02 +0200, Arthur Bergman wrote:
Well, the thread should have finished after the detach. I guess I'm
still thinking too linear in this respect and not parallel
enough... ;-)
No, actually the thread is running
On torsdag, maj 23, 2002, at 02:56 , Tim Bunce wrote:
From: Tim Bunce [EMAIL PROTECTED]
Date: tor maj 23, 2002 02:56:03 Europe/Stockholm
To: Arthur Bergman [EMAIL PROTECTED]
Cc: '[EMAIL PROTECTED]' [EMAIL PROTECTED], Simon Oliver
[EMAIL PROTECTED], Murali Kanaga [EMAIL PROTECTED
subzero:~/perl/build/t$ ./perl -le 'require Attribute::Handlers'
Too late to run CHECK block at ../lib/Attribute/Handlers.pm line 199.
Too late to run INIT block at ../lib/Attribute/Handlers.pm line 201.
Which makes threads::shared emit this warning when it is being required
in 0threads.t So
Here's a transcript of a recent debugging session, with some commentary.
Current directory is /home/mjd/src/perl/Tie-File2/t/
Loading DB routines from perl5db.pl version 1.15
Editor support enabled.
Enter h or `h h' for help, or `man perldebug' for
Okay. So some toplevel file it is, and I would prefer a new file. In
that case threadutil.c sounds okay-ish (which is a bit long... thread.c?)
I forgot to say I also prefer a new file.
But I think the Subject: line might be not too far off. There may be
non-thread-ish things that
Ok, trying to continue to find the cppflags issue.
Currently during the build on darwin there is nothing to uses the
cppflags during the make except makedepend.
The following patch fixes that and makes it possible to build without
all the warnings.
*NO APPLY*
--- ../perl-nothreads/cflags.SH
I just built perl on a 4way itanium machine and all tests passed.
Both 32 bit and -Duse64bitall
Arthur
ps. it seems to be a pretty speedy machine
Merijn tells me of the following breakage on AIX.
If any of the more XS knowledgable persons can figure out what might have gone wrong
or just chime in with a suggestion that would be great. We got all makedef issues
fixed.
Fatal:
Making B (dynamic)
make[1]: Entering directory
sharedsv_unlock_scope is not in the public API
fixed the function names in apidoc.
Arthur
--- embed.pl~ Mon Aug 13 14:48:55 2001
+++ embed.pl Mon Aug 13 14:57:49 2001
@@ -2475,7 +2475,7 @@
Adp|shared_sv*|sharedsv_find|SV* sv
Adp|void|sharedsv_lock
Ok, now we actually compile after I have run perl embed.pl, DOH.
Arthur
--- sharedsv.c.old Mon Aug 13 15:31:34 2001
+++ sharedsv.c Mon Aug 13 15:33:00 2001
@@ -42,7 +42,7 @@
*/
void
-Perl_sharedsv_init(pTHX)
+Perl_sharedsv_init(void)
{
sharedsv_space = PERL_GET_CONTEXT;
}
---
01-07-25 06.39, skrev Jeff 'japhy/Marillion' Pinyan på [EMAIL PROTECTED]
följande:
This dumps core (or gives a bus error, or some other nastiness). I
suspect the problem is based upon the need for ACTUAL EVALUATION of a
regex -- when we merely use /z/ or /za/, Perl can use an index()-like
I have two debian boxes, both run the same libc and gcc, only difference is
one is linux 2.2.17 and one is 2.4.4.
On the linux 2.4.4 I have a big file laying around.
Under both 5.6.1 and 5.7.2 make test fails because a file called t/big is
created, (created by ext/Fcntl/syslfs.t).
When I try
25 matches
Mail list logo
