On 3/1/07 16:44, "Paul Miller" <[EMAIL PROTECTED]> wrote:
>>> Oh, and I suppose I failed to mention that part. I'm running perl
>>> v5.8.8 and POE 0.9917.
>> ** spam (200/200): 35..-"30(:*/>6+.61:[EMAIL PROTECTED](,0#/-1' =! "),>0$+
>> %0";/"#1-8.! *.3:=' '6/;=,!$A
>> test>
>>
>> This is perl, v
The docs were broken. I've fixed them.
-- Rocco
On 3/6/07 05:27, "Jmax" <[EMAIL PROTECTED]> wrote:
> According to the pod, ->put should take an ARRAYREF_OF_RECORDS, however,
> it expects an array of records. Shall I post a bug report on RT?
On 3/12/07 04:21, "Sun Yi-Ming" <[EMAIL PROTECTED]> wrote:
> Hi POEr,
As far as I see, the Component::Client::DNS dosn't have a DNS
> cache
for the remote host address, and the Component::Client::HTTP use
> this
component, so how can I give the Component::Client::HTTP DNS
> caching
function with
nect.
Sorry for so many questions, I can use POE just fine but I want to
understand what's going on under the hood.
I'm all for that. :)
--
Rocco Caputo - [EMAIL PROTECTED]
am (2/200): ->2'@>>2=4%>?'C2'59"A>[EMAIL PROTECTED]&:' C.B-%$/<7#(2/ *23!.4
+2?8=@<2,A/C*6 +;?> 68%1C>#?
...
** spam (199/200): ?3& 2-A)B8(&28)9"3%?711)%B>2?9*A3$*46"!21
551,:8'( 6-0#="5:&?$%&+<+)"@67?78-474B
** spam (200/200): 35..-"30(:*/>6+.61:[EMAIL PROTECTED](,0#/-1' =! "),>0$+
%0";/"#1-8.! *.3:=' '6/;=,!$A
test>
This is perl, v5.8.8 built for darwin-thread-multi-2level
POE is from the svn repository. Maybe something's been fixed since
0.9917?
--
Rocco Caputo - [EMAIL PROTECTED]
POE will not know you've handled the signal. It'll
continue on to reap your sessions prematurely.
--
Rocco Caputo - [EMAIL PROTECTED]
d dig into the problem myself.
--
Rocco Caputo - [EMAIL PROTECTED]
On Feb 22, 2007, at 10:32, Michael Hare wrote:
I've done a little more digging and now I'm really confused.
Consider the following '_stop' handler for a Client::Ping session.
sub handler_stop {
pri
I think I mentioned this in IRC. You can do this from
ClientDisconnected callbacks if you maintain a semaphore and a
shutdown flag. Catching a callback from the server's _stop is
probably a lot easier, though.
--
Rocco Caputo - [EMAIL PROTECTED]
Please send a small test case, and cc bug-poe-component-server-
[EMAIL PROTECTED] The author/maintainer for this component may not
be watching the list.
--
Rocco Caputo - [EMAIL PROTECTED]
On Feb 19, 2007, at 15:12, Nicest of the Damned wrote:
Allo,
I think I'm seeing an edge
file, which effectively disables .inputrc redefines.
--
Rocco Caputo - [EMAIL PROTECTED]
On Feb 20, 2007, at 18:33, David Davis wrote:
I don't know much about readline, but perhaps someone else could
chime in.
That test runs fine here. For now you can force install POE on the
grained intervals, at the expense of FIFO event throughput.
POE's I/O dispatch granularity is defined by POE::Resource::Events'
_data_ev_dispatch_due() method. It may be possible to redefine the
dispatch in a subclass, but POE::Kernel would need a way to let
people load their own replacements.
--
Rocco Caputo - [EMAIL PROTECTED]
anged.
Generally stuff like this isn't implemented because $you are the
first to need it. If you use some of the existing cursor movement or
command line painting functions instead, and you document this, I'll
commit it for the next release.
--
Rocco Caputo - [EMAIL PROTECTED]
You might try submitting this as a bug to bug-curses-ui-
[EMAIL PROTECTED] I'm not sure whether the author is reading this list.
--
Rocco Caputo - [EMAIL PROTECTED]
On Feb 17, 2007, at 18:27, lanas wrote:
Folks,
I'm making a simple UI for a networked utility and have found the
Thanks for the notice. I've committed a fix.
--
Rocco Caputo - [EMAIL PROTECTED]
On Feb 11, 2007, at 16:40, Ton Hospel wrote:
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Ton Hospel) writes:
A few correction to the previous mail:
using POE loses $^W
De
($session,$heap) = @_[SESSION, HEAP];
$_[KERNEL]->delay('Timer_10', 10);
print POETLOG "10 Second Iteration: $iter_10 Count: $cnt_10\n";
$cnt_10 = 0;
$iter_10++;
}
... etc.
--
Rocco Caputo - [EMAIL PROTECTED]
ot;got $signal signal\n");
$kernel->post( server => "shutdown" );
$kernel->sig_handled;
},
},
);
--
Rocco Caputo - [EMAIL PROTECTED]
On Jan 3, 2007, at 15:27, Kevin Scaldeferri wrote:
Okay, I'm glad to see I was making this more complicated than I had
graveyard. You should track its progress at http://thirdlobe.com/svn/
poe-stage/trunk/ ... just in case it's behind you.
Thanks for reading, and enjoy!
--
Rocco Caputo - [EMAIL PROTECTED]
nnected callback.
e. When the active session count reaches 0, and the shutdown flag is
set, begin complex shutdown.
--
Rocco Caputo - [EMAIL PROTECTED]
You'll need to patch it. POE::Wheel::ReadWrite wasn't originally
designed for unbuffered writes. There's a patch to add it, but I
haven't included it because it fails one of POE's tests. You can
find the patch in POE's queue at rt.cpan.org.
--
Rocco Cap
We discussed this briefly in IRC. If I understand correctly, the
consensus was to use the POE::Component namespace, or your own
namespace (YOURCPANID::Module) and release the code with an in-
development version number.
--
Rocco Caputo - [EMAIL PROTECTED]
On Dec 14, 2006, at 11:54
ostback
("aio_open_event");
Whether you "should" use IO::AIO is a matter of your application's
requirements. Quite a lot of systems don't require AIO, so they
shouldn't use IO::AIO.
I hope this helps.
--
Rocco Caputo - [EMAIL PROTECTED]
On Dec 11, 200
on, so you can
only use perl's select on sockets. Even if you do properly configure
the port, you can't multiplex it...
--
Rocco Caputo - [EMAIL PROTECTED]
http://poe.perl.org/?POE_Cookbook/Serial_Ports
--
Rocco Caputo - [EMAIL PROTECTED]
or set_filter(). Let me know if it
works, and I'll commit it.
--
Rocco Caputo - [EMAIL PROTECTED]
set-filter.patch
Description: Binary data
- Doc patches by and for: Joel Bernstein, Frolian Zumbiehl, and
Apocalypse.
- Fixed POE::Wheel::ReadLine's terminal corruption on exit.
Enjoy!
--
Rocco Caputo - [EMAIL PROTECTED]
thing
with streams.
--
Rocco Caputo - [EMAIL PROTECTED]
On Nov 3, 2006, at 12:29, Tinia wrote:
"Rocco Caputo" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
On Oct 29, 2006, at 06:11, Tinia wrote:
I'd kindly ask if anyone more experienced can give some advice
where to
start to get as clean as possible skeleton
response to parent.
2. Parent receives response, and buffers it.
3. 100ms later, waitpid() triggers the parent to process the response.
--
Rocco Caputo - [EMAIL PROTECTED]
RS232 examples.
They won't tell you how to organize your program as a whole, but
they'll probably get you up to speed on the basics.
--
Rocco Caputo - [EMAIL PROTECTED]
ting
without something being delivered where it's supposed to go. I can't
really tell if this is what you mean, or what the details are,
without more information.
--
Rocco Caputo - [EMAIL PROTECTED]
able to
do something similar from POE::Component::Server::TCP's
ClientConnected callback if POE gave you the filehandle of the
connecting socket.
--
Rocco Caputo - [EMAIL PROTECTED]
ned up when
they're not needed. The new ReadLine test should stop blowing up
when it's fixed.
--
Rocco Caputo - [EMAIL PROTECTED]
On Sep 19, 2006, at 07:48, Nick Williams wrote:
I now have a new problem (don't I always ;-).
ReadLine has a modification that I didn'
;s okay, though. There's a CHANGES file inside the
tarball, and it'll be more than happy to explain everything in much
greater detail.
--
Rocco Caputo - [EMAIL PROTECTED]
els itself once the interesting PID has been delivered.
Maybe some other stuff, but those are the big ones. No other changes
are planned before 0.38, but some minor patches might sneak in.
Expect it on CPAN early this week.
--
Rocco Caputo - [EMAIL PROTECTED]
don't see why an event posted after detaching the child
should be lost. The session still exists. It still has its alias.
It should be receiving its events.
If you can whip up a quick test case, attach it to a message to bug-
[EMAIL PROTECTED] Thanks.
--
Rocco Caputo - [EMAIL PROTECTED]
se both as 0.38. That'll be one less CPAN release to test or
ignore before everything's fixed. I hope.
Are there any other signals issues/quirks/requirements I should
address while I'm in here?
--
Rocco Caputo - [EMAIL PROTECTED]
le in POE's distro lists the gory, gory
details for the changes since version 0.3601 was released on 2006-08-13.
I expect the next release to be in about a week and include the
rollback for the sig() reference counts. It may be longer if the
rollback exposes new issues.
--
Rocco Caputo - [EMAIL PROTECTED]
Committed as revision 2110. Thanks.
--
Rocco Caputo - [EMAIL PROTECTED]
On Sep 4, 2006, at 10:06, Pedro Melo wrote:
Hi,
around line 273 of PoCo::Client::TCP, there is a call to
alias_remove().
273 $kernel->alias_remove($alias) if defined $alias;
Inside an aplicatio
unclear or flat out wrong---and, frankly, who hasn't---post a note
(and maybe a patch) to [EMAIL PROTECTED] I'll squeeze the fixes
in among the planned releases.
Thanks.
--
Rocco Caputo - [EMAIL PROTECTED]
#x27;t want sig()'s semantics
flapping back and forth, so a reversion will most likely be permanent.
--
Rocco Caputo - [EMAIL PROTECTED]
don't think he proposed
it on the list, though.
--
Rocco Caputo - [EMAIL PROTECTED]
sig-refcount.text.gz
Description: GNU Zip compressed data
On Aug 22, 2006, at 14:06, Nick Williams wrote:
Rocco Caputo wrote:
Do you have a use case where it's impossible to do something
under the new behavior? I'm working under the assumption that a
session can always find a way to call sig(YOUR_SIGNAL_HERE =>
undef) when it&
tied to _signal, which delivered signals
without sessions explicitly asking for them. _signal is gone now,
so we can tie the explicit interest of sig() into a reference count
to keep the session alive.
Nobody else responded. 17 days later I replied with a public go-
ahead to make the c
y Adam Kennedy.
Applied Peter Farmer's fix for TRACE_FILENAME files gaining unneeded
quotes.
=== And all the rest:
What changed besides the highlights? A lot! Check the CHANGES for
details.
--
Rocco Caputo - [EMAIL PROTECTED]
Hi, Ed.
First, upgrade to POE 0.3502. Maybe the problem is already fixed.
If the problem persists, change the "die" to "confess" and let us
know the failure's full stack trace.
Thanks.
--
Rocco Caputo - [EMAIL PROTECTED]
On Jul 5, 2006, at 13:52, Ed Heil wrote
environment
run in two different processes (threads?) and communicate
between each others
It may be possible to write a POE::Loop::Camelbones. See the
POE::Loop documentation for the API you'll need to implement.
--
Rocco Caputo - [EMAIL PROTECTED]
Can you be more specific about the location and type of overhead
you're seeing?
--
Rocco Caputo - [EMAIL PROTECTED]
On Jul 6, 2006, at 02:22, Michael Giambalvo wrote:
I'm working on a large scale TCP server application.
Based on profiling, I see that more than half the CPU
I uploaded a new version of the HTTP client component. This one lets
you cancel streaming responses without shutting down the entire
component.
--
Rocco Caputo - [EMAIL PROTECTED]
erously close to becoming
empty (low watermark). Or patch the components to expose the
appropriate callbacks.
--
Rocco Caputo - [EMAIL PROTECTED]
POE 0.34 and 0.35 include a number of POE::Filter fixes. When you
have a chance, can you check the leak still exists? A test case
would rock, too.
--
Rocco Caputo - [EMAIL PROTECTED]
On Apr 30, 2006, at 03:22, Donovan Allen wrote:
Oops, didn't reply to group as well:
Glancing at
And just to note: POE 0.3501 was released early this week with
David's fix.
--
Rocco Caputo - [EMAIL PROTECTED]
On May 23, 2006, at 16:00, David Davis wrote:
Gil,
I've fixed that test in svn. You can either check out the lastest
version
or force install POE.
David
On 5/
tch to fix an edge case in
POE::Wheel::ReadLine's tab completion.
- Prefer POE::XS::Queue::Array if it's installed.
--
Rocco Caputo - [EMAIL PROTECTED]
ld not be tied. Resolves rt.cpan.org ticket 1648.
if (tied *STDOUT) {
carp "Cannot redirect into tied STDOUT. Untying it";
untie *STDOUT;
}
if (tied *STDERR) {
carp "Cannot redirect into tied STDERR. Untying it";
untie *STDERR;
}
--
Rocco Caputo -
suggest it here or post it to
POE's wiki (<http://poe.perl.org/?gsoc/2006>). I'm totally drawing a
blank today, and the enrollment clock is running.
Thanks!
--
Rocco Caputo - [EMAIL PROTECTED]
On Apr 18, 2006, at 07:18, lanas wrote:
Was anything ever done using CORBA with POE ?
Hi. Not that I'm aware of. I know a person who's writing an RMI-
like library, but it's nothing like CORBA.
--
Rocco Caputo - [EMAIL PROTECTED]
first. I suspect that recent patches to the module, or to
POE::Filter modules, have introduced a memory leak.
--
Rocco Caputo - [EMAIL PROTECTED]
$heap->{ECHOVAR}{$_[ARG0]}++;
$kernel->yield('shutdown');
$kernel->alias_remove($heap->{my_alias});
},
#InlineStates => { } # end inline states
); # end TCP Server
POE::Kernel->run();
--
Rocco Caputo - [EMAIL PROTECTED]
to already allocated memory in user
process ?
Maybe. Implementation details are undefined and subject to change.
If you post a reference, however, it will still be a reference when
it's delivered.
--
Rocco Caputo - [EMAIL PROTECTED]
res that
work now. I'm only supporting the latest version (currently build
817), so backward compatibility's not an issue.
While you're bugging ActiveState, ask them why POE's tests fail in
the build environment but not under the CPAN shell. That seems
arbitrary to me.
--
Rocco Caputo - [EMAIL PROTECTED]
m should
gleefully pass through whatever you care to send. If you can re-
create the problem in a small, stand-alone test case, I would love to
try it out here and possibly add your code to the regression test suite.
--
Rocco Caputo - [EMAIL PROTECTED]
else.
My guess is that another module is calling alarm(), but you aren't
stepping through the code fast enough to clear it before the signal
arrives. Perl's built-in alarm() is highly dependent on real time.
The operating system doesn't care if your code is paused on a
breakpoint.
--
Rocco Caputo - [EMAIL PROTECTED]
e_kernel->run();".
Fixed and committed. Also released, but sh! That hasn't been
announced yet.
I'm waiting for http://ppm.activestate.com/BuildStatus/5.8-windows/
windows-5.8/POE-0.34.txt to appear so I can check out any issues with
the new installer.
--
Rocco Caputo - [EMAIL PROTECTED]
eleased. If not,
there may be some point releases to resolve minor installer issues.
--
Rocco Caputo - [EMAIL PROTECTED]
ons, including its use of ExtUtil::MakeMaker. Version 0.34 won't
be using that module anymore, so we'll see what other problems crop up.
--
Rocco Caputo - [EMAIL PROTECTED]
On Mar 22, 2006, at 16:26, Steven Mackenzie wrote:
Rocco Caputo wrote:
I'd like a console watcher session with a "check_console" event
to fire
every second or so while there are still active worker sessions,
but I
don't know how to code that ...
This is untested c
On Mar 22, 2006, at 09:26, Steven Mackenzie wrote:
Rocco Caputo wrote:
ActivePerl's select() can't watch STDIN because it's based on
Winsock's select(), which only works with sockets. So passively
watching the console for regular input is out.
How insightfuly w
d console.
Someone with more Windows experience should chime in.
--
Rocco Caputo - [EMAIL PROTECTED]
POE would do, or hated that it does.
Everything else.
--
Rocco Caputo - [EMAIL PROTECTED]
ntation. Thanks for
the forward.
--
Rocco Caputo - [EMAIL PROTECTED]
connections you can have.
My guess is that either ActivePerl or Winsock has a descriptor limit
in its select() implementation. POE doesn't care, and it'll happily
try to use more descriptors than your OS supports, if your
application lets it.
--
Rocco Caputo - [EMAIL PROTECTED]
x27;t know what that would be.
--
Rocco Caputo - [EMAIL PROTECTED]
On Feb 26, 2006, at 13:56, sungo wrote:
On (02/26 03:12), Rocco Caputo wrote:
Does anyone have suggestions for naming its stand-alone distribution?
Filter::PreProcessor
Ah, Filter! I had forgotten about that. I settled on
Filter::Template after the usual round of overthinking
n.
sub input_handler {
goto &moo;
}
sub input_handler {
&moo;
}
sub input_handler {
moo(@_);
}
Yeah, it's an extra level of indirection. You're almost better off
with POE::Wheel::SocketFactory and POE::Wheel::ReadWrite here.
--
Rocco Caputo - [EMAIL PROTECTED]
limitations are your system's resources.
--
Rocco Caputo - [EMAIL PROTECTED]
(COMMAND) will wait until COMMAND completes. So if system() is
returning too soon, it means COMMAND has completed early.
Your question seems odd. Most people asking about system() and POE
are looking for ways to run commands without waiting for them to
complete.
--
Rocco Caputo - [EMAIL
uggestions for naming its stand-alone distribution?
--
Rocco Caputo - [EMAIL PROTECTED]
On Feb 25, 2006, at 10:43, lanas wrote:
Rocco Caputo <[EMAIL PROTECTED]> wrote:
{Inline,Object,Package}States apply to each connection session, not
the main server session for the component. You can certainly send
arbitrary messages to the server session, but it will not handle
them
modify the other wheels so that put() is immediate.
Inconsistent flush timing will bite someone down the road.
--
Rocco Caputo - [EMAIL PROTECTED]
ckage}States apply to each connection session, not
the main server session for the component. You can certainly send
arbitrary messages to the server session, but it will not handle them.
--
Rocco Caputo - [EMAIL PROTECTED]
On Feb 2, 2006, at 18:26, Dan McCormick wrote:
Any objections to adding a check for bad params in Client::TCP? It
would've
saved me some recent typo-related head-scratching. I just stole
the code
from Server::TCP.
[patch]
Patch applied. Thanks!
--
Rocco Caputo - [EMAIL PROTECTED]
a CR/LF terminated line. You may want to use POE::Filter::Block
rather than POE::Filter::Line (the default) for your client input.
--
Rocco Caputo - [EMAIL PROTECTED]
aliases by
default, but you can register them on connection and remove them upon
disconnect, similar to the way http://poe.perl.org/?POE_Cookbook/
Chat_Server does it.
There are different ways to address connections individually. Which
you use depends on what you're doing.
--
Rocco Caputo - [EMAIL PROTECTED]
safe thanks to a pointer from Nick
Williams.
--
Rocco Caputo - [EMAIL PROTECTED]
length or with
a header that indicates the length, you can use POE::Filter::Block.
The LengthCodec parameter defines subroutines that are used to parse
(and build) the length header.
--
Rocco Caputo - [EMAIL PROTECTED]
On Jan 27, 2006, at 08:10, lanas wrote:
Hi !
Using the TCP Components (
emantics may need to change as more features are added.
Thanks.
--
Rocco Caputo - [EMAIL PROTECTED]
pment comes from
people on this list, whether through comments, bug reports or
patches. POE wouldn't be nearly as cool without you.
--
Rocco Caputo - [EMAIL PROTECTED]
--
Rocco Caputo - [EMAIL PROTECTED]
--
Rocco Caputo - [EMAIL PROTECTED]
a DNS resolver, but it's not as
featureful as POE::Component::Client::DNS.
By the way, your cleaner example contains an error. :)
--
Rocco Caputo - http://poe.perl.org/
this because I tried using subroutine attributes
once. It ended badly, but that was mainly because I had no idea
where I wanted the feature to go. *Splurge!*
--
Rocco Caputo - [EMAIL PROTECTED]
On Oct 19, 2005, at 14:08, Jonathan Steinert wrote:
Quickly, my ideas on how things should be are as follows:
Looks like there are no objections. I say go for it.
--
Rocco Caputo - [EMAIL PROTECTED]
27;t comment on that. :)
--
Rocco Caputo - [EMAIL PROTECTED]
is listening
for a
few versions to help migrate people, or perhaps do this instead during
kernel shutdown? I personally will opt for the former.
I prefer the latter. While it means that a process will leak zombies
at runtime, the reason for the leakage should become apparent when
the program's stopped. Screwing up like this should be at least a
little painful.
--
Rocco Caputo - [EMAIL PROTECTED]
se an end-of-run
zombie reaper to detect zombies and warn the user that they need a sig
("CHLD") handler.
Similar to how the "you forgot to call run()" warning works. Let
people screw up, and then let them know the results aren't POE's fault.
--
Rocco Caputo - [EMAIL PROTECTED]
On Oct 10, 2005, at 08:15, Mario Minati wrote:
So my problem is how to just read the pipe when some data is in and
just loop and wait if no data is present?
Is there a way to reactivate that handle?
It sounds like POE::Wheel::FollowTail will do what you want.
--
Rocco Caputo - [EMAIL
throw another
signal at everything, that's the way to do it. If, however, it must
signal certain sessions specifically, then yes. It will need to know
something about those sessions to address them.
--
Rocco Caputo - http://poe.perl.org/
got_signal (from /opt/perl-5.8.5/lib/site_perl/5.8.5/POE/Loop/
PerlSignals.pm at 31)
got sigint at lib/POE/Component/Client/Artajasa.pm line 110.
4 -> got_signal (from /opt/perl-5.8.5/lib/site_perl/5.8.5/POE/Loop/
PerlSignals.pm at 31)
4 -> _stop (from /opt/perl-5.8.5/lib/site_perl
On Sep 22, 2005, at 06:31, Rachwal Waldemar-AWR001 wrote:
[...]
call() works from _stop handlers.
--
Rocco Caputo - http://poe.perl.org/
number of jobs. This cuts back the number of processes
overall, and each process is not itself overloaded with jobs. If you
use something like POE::Component::IKC or POE::Component::Lightspeed,
you can even throw more machines on the rack as you hit hard limits.
Best of luck.
--
Rocco Caputo - http://poe.perl.org/
ls. For example, this kills the wheel associated with an
error:
sub handle_error {
my ($heap, $wheel_id) = @_[HEAP, ARG3];
delete $heap->{connection}{$wheel_id};
}
I hope this helps.
--
Rocco Caputo - [EMAIL PROTECTED]
subsumed under it, and developers can focus on more interesting
issues than how they'll be passing around messages.
--
Rocco Caputo - [EMAIL PROTECTED]
On Sep 8, 2005, at 15:32, David Davis wrote:
Whatever dude... If you wanted to know what Rocco would do, then
just email him, don't ask the list.
That's a recipe for disaster. Rocco says the craziest things.
--
Rocco Caputo - http://poe.perl.org/
201 - 300 of 1064 matches
Mail list logo
